Abstract
The purification of membrane proteins can be challenging due to their low solubility in conventional detergents and/or chaotropic solutions. The introduction of fusion systems that promote the formation of inclusion bodies has facilitated the overexpression of membrane proteins. In this protocol, we describe the use of perfluorooctanoic acid (PFOA) as an aid in the purification of highly hydrophobic membrane proteins expressed as inclusion bodies. The advantage of utilizing PFOA is threefold: first, PFOA is able to reliably solubilize inclusion bodies, second, PFOA is compatible with nickel affinity chromatography, and third, PFOA can be efficiently dialyzed away to produce a detergent free sample. To demonstrate the utility of employing PFOA, we expressed and purified a segment of the extremely hydrophobic membrane protein caveolin-1.
Introduction
Membrane proteins are major players in cellular biology. They are responsible for a plethora of cellular functions such as signal transduction and transport [1]. Additionally, a large number of drug targets have been identified as membrane proteins, indicating that these proteins are heavily involved in normal cell function [2]. However, the purification and analysis of these membrane proteins can be challenging because of their highly hydrophobic characteristics and strong propensity to aggregated [3]. In addition, recombinant expression of membrane proteins usually results in low yields due to stresses put on the host membrane which results in toxicity. However, the overexpression of membrane proteins into inclusion bodies has emerged as a powerful tool to achieve high levels of protein in E. coli cells by eliminating the toxicity issues mentioned above which limit protein production [4]. Inclusion bodies can be isolated through a series of wash treatments that separate them from the soluble and membrane components of the host cell. While this insolubility is an attractive feature in aiding the isolation of the protein, it can be an obstacle when it comes to solubilizing the protein. Typically strong chaotropic solutions such as 8 M urea or 6 M guanadinium hydrochloride are utilized. However, for highly hydrophobic membrane proteins, these solutions are often not powerful enough to completely dissolve the inclusion bodies. Alternatively, strong detergents such as sodium dodecyl sulfate are attractive, but many of these detergents are not compatible with widely used purification techniques such as nickel affinity chromatography. Furthermore, it is often difficult, if not impossible, to remove these harsh detergents from the sample, which can be problematic, as it is often desirable to acquire experimental results in the presence of a native-like detergent and lipid systems.
Perfluorooctanoic acid (PFOA) is a powerful detergent that has been shown to have the ability to solubilize membrane proteins (Figure 1) [5, 6]. In this report, we extend the utility of PFOA by showing that it can dissolve highly hydrophobic membrane proteins expressed as inclusion bodies. Furthermore, it is compatible with nickel affinity chromatography, and can be easily removed by dialysis, providing a detergent-free precipitate that can then be solubilized in a detergent or lipid system of choice. To demonstrate the usefulness of PFOA, we detail the purification of the integral membrane protein caveolin-1 from inclusion bodies. Caveolin-1 is the preeminent protein in membrane invaginations called caveolae, which have been shown to crucial for caveolae formation, signal transduction, mechano-protection, and endocytosis; however studies of this protein have been hindered by its extremely hydrophobic character [7–10]. Because of this, caveolin-1 is an ideal candidate to demonstrate the utility of PFOA.
Figure 1.
Structure of perfluorooctanoic acid (PFOA)
Materials and methods
Perfluorooctanoic acid was purchased from Synquest laboratories (Alachua, FL). PFOA has been shown to exhibit relatively low toxicity effects; therefore, standard chemical hygiene protocols are sufficient [11]. Dialysis tubing was purchased from Spectrum laboratories (Rancho Dominguez, CA). Ni Sepharose 6 Fast Flow resin was purchased from GE Health Sciences (Piscataway, NJ). All other reagents were of standard ACS grade.
Protein expression
H9_TrpLE_caveolin-1_62-178 was cloned into the pET-24a vector, and transformed into BL21(DE3) cells. 1 mL of an overnight culture (20 hours) in MDG media was used to inoculate 1 L of ZYM-5052 media [12]. The culture was shaken at 250 rpm on an orbital shaker at 37°C for 12–14 hours. Cells were harvested at 8200 × g for 15 min at 4°C, resuspended in 0.9% (w/v) NaCl, and re-centrifuged at 5000 × g for 30 min at 4°C. Pellets were stored at −80°C until needed.
Protein purification
1 L cell pellets were resuspended in 200 mL of a buffer containing 20% (w/v) sucrose, 10 mM Tris pH 8.0, 1 mM EDTA and 50 mM BME. Cells were lysed by sonication in a Branson Sonifier 450 for 15 minutes (power level 40 and duty cycle 5) with stirring at 4°C. Next, the lysis was centrifuged for 2 hours at 27,500 × g at 4°C. The supernatant was removed, and the pellet was resuspended in 200 mL of a buffer containing 1% (v/v) Triton X-100, 10 mM Tris pH 8.0 followed by sonication for 15 minutes with stirring at 4°C. The mixture was centrifuged at 27,500 × g for 1 hour at 4°C. The supernatant was discarded, and the remaining pellet contained the isolated inclusion bodies.
Solubilization of inclusion bodies
Isolated inclusion bodies were dissolved in 40 mL of 8% (w/v) PFOA, 25 mM phosphate pH 8.0 and homogenized using a dounce homogenizer. The solution was then centrifuged at 50,000 × g for 30 min at 22°C. The supernatant contained the solubilized inclusion bodies.
Ni-NTA purification
After solubilization into 8% (w/v) PFOA, the supernatant was filtered through a 0.2 μm filter and loaded onto a column containing 20 mL of Ni sepharose 6 resin. The column was washed with approximately 5 column volumes of 1% (w/v) PFOA, 25 mM phosphate pH 8.0, or until the absorbance at 280 nm was steady, to remove any unbound protein. Samples were eluted in the presence of 1% (w/v) PFOA, 25 mM phosphate pH 8.0, 250 mM imidazole.
Dialysis
The most concentrated column fractions were pooled, and placed in 10,000 MWCO dialysis tubing. Samples were dialyzed against 20 L of 50 mM ammonium sulfate for 24 hours at room temperature with stirring. Precipitated protein was isolated by centrifugation at 4000 × g for 30 min at 22°C. The pellet contained the purified precipitated protein.
Results and Discussion
Inclusion body production and solubilization
Inclusion bodies have emerged as a powerful tool to obtain high levels of membrane proteins expressed in E. coli cells. One of the challenges of membrane protein expression, especially of a non-native membrane proteins, is that the overexpressed protein can crowd the membrane and become toxic to the bacterial cell [13]. This leads to low protein expression that can make protein isolation very challenging due to the high background of endogenous host proteins. However, the fusion of a membrane protein to particular proteins will cause the protein of interest to be rapidly expressed in an unfolded state, and incorporated into insoluble cytoplasmic aggregates (i.e. inclusion bodies). One of the common proteins utilized to promote inclusion body formation is trp leader (trpLE) which has been shown to result in significantly enhanced membrane protein expression [14]. Advantageously, the properties of these aggregates can then be exploited to extract the protein of interest. First, the cells are lysed in a buffer containing sucrose which removes soluble cellular components. After centrifugation, this leaves a pellet that contains only the inclusion bodies and other hydrophobic membrane components. These hydrophobic membrane components can be removed via a second lysis step that utilizes a buffer containing a mild detergent (in this case Triton X-100). Since Triton X-100 will not solubilize the inclusion bodies, after centrifugation, the majority of the pellet contains inclusion bodies.
Normally, once the inclusion bodies have been isolated from the whole cell milieu, they are solubilized in either 8 M urea or 6 M guanidinium hydrochloride. However, in the case of highly insoluble transmembrane domains, even solutions of these strong chaotropic agents cannot effectively solubilize the inclusion bodies. For example, when the membrane interacting domain of caveolin-1 is expressed with the fusion protein trp leader, the inclusion body pellet is not soluble in either 8 M urea or 6 M guanidinium hydrochloride (data not shown). The membrane interacting domain of caveolin-1 has a GRAVY score of 0.659 which is an index of the overall hydrophobicity of a protein [15]. Higher GRAVY scores indicate a more hydrophobic protein while lower scores (negative) indicate a more hydrophilic protein. Therefore, the GRAVY score can be used as a benchmark, and based on our data, membrane proteins with GRAVY scores greater than 0.5 are ideal candidates for PFOA solubilization. For this reason, the conventional methodology for processing inclusion bodies was not applicable. It was determined that a solution of 8% (w/v) PFOA was optimal for efficiently and rapidly solubilizing inclusion bodies of all types and hydrophobicities (Figure 2). However, caution should be taken when employing solutions with lower concentrations of PFOA as they may not completely solubilize the inclusion body pellet. Additionally, we find that in some cases, the addition of 1% (w/v) PFOA to a solution containing 8 M urea can significantly enhance the ability of 8 M urea to solubilize hydrophobic inclusion bodies, thereby decreasing the need for a very high detergent concentration.
Figure 2.
SDS-PAGE analysis of caveolin-1 (62-178) inclusion bodies dissolved in 8% (w/v) PFOA. Lane 1, Molecular weight ladder; lane 2, caveolin-1 in PFOA buffer pre-centrifugation; lane 3, caveolin-1 in PFOA buffer post-centrifugation (50,000 × g).
The pKa of PFOA has been reported to be between 2.8 and 3. However, there is a study that has determined the pKa to be substantially lower approximately −0.5 [16, 17]. Whichever value is most representative, the low pKa of PFOA is an advantage because it will not interfere with typical buffering agents in the 4 to 12 range. It is also important to note that PFOA can form precipitates with potassium counterions, therefore, buffer conditions are limited to the use of sodium or ammonium counter ions. Additionally, it should be noted that due to the exclusionary fluorous nature of PFOA, it is incompatible with certain detergents, such as Empigen BB, so one should test miscibility before working with other detergents. In particular, detergents with positively charged groups are likely to form insoluble precipitates. However, non-ionic detergents such as Triton X-100 are compatible. Similarly, guanadinium hydrochloride is not compatible with PFOA as it forms a precipitate.
Purification
Nickel purification has arisen as a key tool for the purification of proteins. The poly-histidine tag imparts specificity for binding to the nickel-bound nitrilotriacetic acid resin, but due to its small size it generally will not perturb the protein structure or its function [18]. Most of the time, this eliminates the need to do additional cleavage steps to remove the affinity tag. The purification of membrane proteins using nickel affinity can be challenging. Membrane proteins need detergents for solubility, and not all detergents are compatible with the resin chemistry. Many milder detergents that are compatible with nickel affinity chromatography are not sufficient to keep the highly insoluble membrane proteins in solution much less solubilize them from an inclusion body state. In contrast, several stronger detergents (e.g. SDS) are able to solubilize inclusion bodies and keep the protein in solution, but interfere with the binding affinity of the column. This makes PFOA an attractive choice as it is able to both solubilize membrane proteins, and is compatible with the nickel affinity column chemistry. In this case, we show the nickel chromatography trace of a nona-histidine tagged Trp leader fusion of caveolin-1 residues 62–178 as an example of an insoluble membrane protein that can be purified using nickel affinity chromatography with PFOA buffers (Figure 3a). Although 8% (w/v) is needed to initially solubilize the inclusion bodies, the column can be run in 1% (w/v) PFOA which is sufficient to keep the protein in solution. In addition, the high ionic strength of the PFOA detergent mediates any ion exchange effects with the resin so the addition of NaCl to the buffer (typically 300 – 500 mM) is not needed. Imidazole can be added to the wash to enhance the purity of the finally product (0 mM – 40 mM), but the maximum tolerated level before there is significant protein loss must be determined for each protein individually. However, for caveolin-1 we have been able to obtain very high purities without using imidazole in the wash step (Figure 3b). Elution is accomplished using a 1% (w/v) PFOA solution containing 250 mM imidazole. The elution can also be done with a 1% (w/v) PFOA solution at a pH of 4.5. However it is important to keep in mind that the solubility of PFOA decreases dramatically below a pH of 4.0.
Figure 3.
A) Representative chromatogram of the nickel affinity column purification of caveolin-1 (62-178). B) SDS-PAGE analysis of nickel purified caveolin-1 in PFOA. Lane 1, molecular weight ladder; lane 2, caveolin-1 (62-178) after elution.
Dialysis
One of the major challenges of membrane protein purification is that often times, the detergents that are necessary for purification are not desirable for the downstream characterization of the protein. For example, PFOA may be used for purification, but the final experiments are desired to be done in phospholipid vesicles. Also, there are cases where even the residual presence of even a few detergent molecules can cause erroneous results [19]. One common method for the removal of detergents is dialysis. However, many detergents that are commonly used in membrane protein expression have very low CMC values making them difficult if not impossible to remove using this method. This again highlights the need for a detergent that meets three fundamental requirements: protein solubilization, compatibility with purification techniques and the ability to be readily removed. PFOA has been shown to be removed by slow dialysis over time [5]. This is due to the relatively high cmc of PFOA (reported as 13–30 mM, depending on the buffer system [5]). When the protein is loaded onto the nickel column, it can be washed into 1% (w/v) PFOA. This is advantageous as the 1% (w/v) PFOA can be more readily dialyzed than an 8% (w/v) solution. We have found that the dialysis of a 50 mL solution against 20 L of water at room temperature is sufficient to cause 100% precipitation of the protein (Figure 4). In addition to caveolin-1 a number of other membrane proteins (e.g. OmpX and OmpF) have been successfully precipitated from dialysis using PFOA. Therefore, the precipitation methodology is expected to be general. After precipitation, the protein is recovered by gentle centrifugation (4000 × g). Next, the precipitated protein can be washed several times to remove all traces of PFOA. Importantly the addition of 50 mM ammonium sulfate to the dialysis buffer significantly enhances the rate of protein precipitation. We have found that this precipitate is extremely easy to redissolve in whatever downstream detergents and/or lipids that are desired. The ease at which the precipitated protein re-dissolves, suggests that the protein is in a non-aggregated state. In addition, the use of ammonium sulfate, which is protein structure stabilizing, during the dialysis should favor the native fold of the protein. However, further studies will be needed to determine the exact state of the precipitated protein. Unfortunately, experiments aimed at reconstituting PFOA solubilized protein directly into liposomes by the addition of lipids during the dialysis step were not successful; only very low levels of protein was reconstituted (data not shown).
Figure 4.
SDS-PAGE analysis of caveolin-1 after precipitation. Lane 1, molecular weight ladder; lane 2, caveolin-1 (62-178) after precipitation before centrifugation, lane 3, supernatant after centrifugation of precipitated caveolin-1.
Obtaining sufficient amounts of insoluble membrane proteins at high levels is crucial to the understanding of protein structure and function through biophysical characterization. However, there are several challenges to obtaining high levels of purified membrane proteins, most notably the need for a detergent system that is compatible with both purification techniques and downstream characterization. In the procedure described herein, we demonstrate that PFOA has a threefold advantage when applied to the purification of membrane proteins. First, it is able to efficiently dissolve inclusion bodies. Second, it is compatible with Ni-NTA purification. Finally, it can be easily removed through dialysis, which produces a detergent free sample that can be carried through for additional purification such as cleavage of the fusion protein or incorporation into other membrane mimics.
Highlights.
PFOA can be used to dissolve highly hydrophobic inclusion bodies.
PFOA can be easily removed by dialysis.
PFOA is compatible with nickel affinity chromatography.
Acknowledgments
This work was supported by NIH RO1 GM093258-01A1 awarded to K.J.G.
Abbreviations
- PFOA
perfluorooctanoic acid
- CMC
critical micelle concentration
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Author Contributions
SMP performed the experiments. SMP and KJG analyzed the data. SMP, KTR and KJG wrote the manuscript. KJG designed the experiments.
References
- 1.Smart EJ, Graf GA, McNiven MA, Sessa WC, Engelman JA, Scherer PE, Okamoto T, Lisanti MP. Caveolins, liquid-ordered domains, and signal transduction. Mol. Cell. Biol. 1999;19:7289–7304. doi: 10.1128/mcb.19.11.7289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Tang X, Wang Y, Li D, Luo J, Liu M. Orphan G protein-coupled receptors (GPCRs): biological functions and potential drug targets. Acta Pharmacol. Sin. 2011;33:363–371. doi: 10.1038/aps.2011.210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Molina DM, Cornvik T, Eshaghi S, Haeggstrom JZ, Norland P, Sabet MI. Engineering membrane protein overproduction in Escherichia coli. Protein Science. 2008;17:673–680. doi: 10.1110/ps.073242508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Diefenderfer C, Lee J, Mlyanarski S, Guo Y, Glover KJ. Reliable expression and purification of highly insoluble transmembrane domains. Anal Biochem. 2009;384:274–8. doi: 10.1016/j.ab.2008.09.038. [DOI] [PubMed] [Google Scholar]
- 5.Shepherd FH, Holzenburg A. The Potential of Fluorinated Surfactants in Membrane Biochemistry. Anal. Biochem. 1995;224:21–27. doi: 10.1006/abio.1995.1003. [DOI] [PubMed] [Google Scholar]
- 6.Kadiyala CSR, Tomechko SE, Miyagi M. Perfluorooctanoic Acid for Shotgun Proteomics. Plos One. 2010;5:e15332. doi: 10.1371/journal.pone.0015332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Root KT, Glover KJ. Reconstitution and spectroscopic analysis of caveolin-1 residues 62–178 reveals that proline 110 governs its structure and solvent exposure. Biochim. Biophys. Acta. 2016;1858:682–688. doi: 10.1016/j.bbamem.2016.01.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Rieth MD, Lee J, Glover KJ. Probing the caveolin-1 P132L mutant: critical insights into its oligomeric behavior and structure. Biochemistry. 2012;51:3911–3918. doi: 10.1021/bi3001853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Plucinsky SM, Glover KJ. Secondary Structure Analysis of a Functional Construct of Caveolin-1 Reveals a Long C-Terminal Helix. Biophys. J. 2015;109:1686–1688. doi: 10.1016/j.bpj.2015.08.030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Root KT, Plucinsky SM, Glover KJ. Chapter Nine - Recent Progress in the Topology, Structure, and Oligomerization of Caveolin: A Building Block of Caveolae. Curr. Top. Membr. 2015;75:305–336. doi: 10.1016/bs.ctm.2015.03.007. [DOI] [PubMed] [Google Scholar]
- 11.Domingo JL, Nadal M. Per- and Polyfluoroalkyl Substances (PFASs) in Food and Human Dietary Intake: A Review of the Recent Scientific Literature. J. Agric. Food Chem. 2017;65:533–543. doi: 10.1021/acs.jafc.6b04683. [DOI] [PubMed] [Google Scholar]
- 12.Studier FW. Protein production by auto-induction in high density shaking cultures. Prot. Expr. Purif. 2005;41:207–34. doi: 10.1016/j.pep.2005.01.016. [DOI] [PubMed] [Google Scholar]
- 13.Rosano GL, Ceccarelli EA. Recombinant protein expression in Escherichia coli: advances and challenges. Front. Microbiol. 2014;5:172. doi: 10.3389/fmicb.2014.00172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Hwang PM, Pan JS, Sykes BD. Targeted expression, purification, and cleavage of fusion proteins from inclusion bodies in Escherichia coli. FEBS Lett. 2014;588:247–252. doi: 10.1016/j.febslet.2013.09.028. [DOI] [PubMed] [Google Scholar]
- 15.Kyte J, Doolittle RF. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 1982;157:105–132. doi: 10.1016/0022-2836(82)90515-0. [DOI] [PubMed] [Google Scholar]
- 16.Burns DC, Ellis DA, Li H, McMurdo CJ, Webster E. Experimental pKa Determination for Perfluorooctanoic Acid (PFOA) and the Potential Impact of pKa Concentration Dependence on Laboratory-Measured Partitioning Phenomena and Environmental Modeling. Environ. Sci. Technol. 2008;42:9283–9288. doi: 10.1021/es802047v. [DOI] [PubMed] [Google Scholar]
- 17.Goss K. The pKa Values of PFOA and Other Highly Fluorinated Carboxylic Acids. Environ. Sci. Technol. 2008;42:456–458. doi: 10.1021/es702192c. [DOI] [PubMed] [Google Scholar]
- 18.Hoffmann A, Roeder RG. Purification of his-tagged proteins in non-denaturing conditions suggests a convenient method for protein interaction studies. Nucleic Acids Res. 1991;19:6337–6338. doi: 10.1093/nar/19.22.6337. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Allen TM, Romans AY, Kercret H, Segrest JP. Detergent removal during membrane reconstitution. Biochim. Biophys. Acta. 1980;601:328–342. doi: 10.1016/0005-2736(80)90537-4. [DOI] [PubMed] [Google Scholar]




