Strategic optimization of conditions for the solubilization of GST-tagged amphipathic helix-containing ciliary proteins overexpressed as inclusion bodies in E. coli

Shendge, Amruta A.; D’Souza, Jacinta S.

doi:10.1186/s12934-022-01979-y

Research
Open access
Published: 12 December 2022

Strategic optimization of conditions for the solubilization of GST-tagged amphipathic helix-containing ciliary proteins overexpressed as inclusion bodies in E. coli

Amruta A. Shendge¹ &
Jacinta S. D’Souza¹

Microbial Cell Factories volume 21, Article number: 258 (2022) Cite this article

3986 Accesses
4 Citations
Metrics details

Abstract

Expression of affinity-tagged recombinant proteins for crystallography, protein–protein interaction, antibody generation, therapeutic applications, etc. mandates the generation of high-yield soluble proteins. Although recent developments suggest the use of yeast, insect, and mammalian cell lines as protein expression platforms, Escherichia coli is still the most popular, due mainly to its ease of growth, feasibility in genetic manipulation and economy. However, some proteins have a spontaneous tendency to form inclusion bodies (IBs) when over-expressed in bacterial expression systems such as E. coli, thus posing a challenge in purification and yield. At times, small peptides undergo degradation during protein production and hence using suitable tags could circumvent the problem. Although several independent techniques have been used to solubilize IBs, these cannot always be applied in a generic sense. Although tagging a GST moiety is known to enhance the solubility of fusion proteins in E. coli, resulting in yields of 10–50 mg/L of the culture, the inherent nature of the protein sequence at times could lead to the formation of IBs. We have been working on a Myc Binding Protein-1 orthologue, viz. Flagellar Associated Protein 174 (FAP174) from the axoneme of Chlamydomonas reinhardtii that binds to an A-Kinase Anchoring Protein 240 (AKAP240) which has been annotated as Flagellar Associated Protein 65 (FAP65). Using an in-silico approach, we have identified two amphipathic helices on FAP65 (CrFAP65AH1 and CrFAP65AH2) that are predicted to bind to FAP174. To test this prediction, we have cloned the GST-tagged peptides, and overexpressed them in E. coli that have resulted in insoluble IBs. The yields of these over-expressed recombinant proteins dropped considerably due to IB formation, indicating aggregation. An integrated approach has been used to solubilize four highly hydrophobic polypeptides, viz. two amphipathic helices and the respective proline variants of FAP65. For solubilizing these polypeptides, variables such as non-denaturing detergents (IGEPAL CA-630), changing the ionic strength of the cell lysis and solubilization buffer, addition of BugBuster^®, diluting the cell lysate and sonication were introduced. Our statistically viable results yielded highly soluble and functional polypeptides, indiscreet secondary structures, and a yield of ~ 20 mg/L of the E. coli culture. Our combinatorial strategy using chemical and physical methods to solubilize IBs could prove useful for hydrophobic peptides and proteins with amphipathic helices.

Introduction

Proteins are thought of as the ‘workhorses’ of the cell, playing important roles in biological activities like metabolism, DNA replication, transcription, translation, DNA repair, gene regulation, and signalling. They also make up the structure of the cell. Genetic modification methods that produce physiologically active recombinant proteins are typically used to fully comprehend each of their distinct activities. An ever-expanding field of recombinant DNA technology has developed to achieve this goal, allowing for the design or modification and production of high-yield proteins. With the aid of these techniques, scientists have created valuable proteins that can be used to study the physicochemical, molecular, structure–function, and biological functions of many proteins [1]. Steps for obtaining a recombinant protein include selecting the gene of interest and cloning it in a suitable vector, transforming it into an expression host, and inducing and purifying it to homogeneity. A recent development suggests the use of yeast, mammalian, and insect cell lines over traditional expression systems such as bacteria and fungi for acquiring eukaryotic proteins that require post-translational modifications [2]. In contrast to E. coli, mammalian cell lines provide a further advantage by their ability to secrete the protein in the medium making the down-stream process simpler [2]. Although being the most popular over-expression platform, E. coli poses several drawbacks, it lacks a post-translational modification system, has a distinct codon bias, exhibits RNA-instability, forms IBs, may contain lipopolysaccharides that are associated with the purified protein and may interfere with some of its uses, etc. [2]. Earlier, IB formation was a valuable tool for obtaining active recombinant proteins, however, its extraction and solubilization depend on numerous factors such as the presence of osmolytes, molecular chaperones, stabilizing enzymes and effective post-translational modification machinery leading to biologically active proteins. Despite the development of several tools to predict protein solubility, the tendency of a protein to form IB cannot be accurately determined. These IBs can cause intrinsic toxicity and conformational stress to the cells producing them. Therefore, the formation of IBs can pose an important obstacle during the production of recombinant proteins. Since IBs tend to aggregate, they can eventually reduce the solubility of the target protein [3]. In spite of these disadvantages, E. coli remains the most preferred prokaryotic expression system because it has the advantages of genetic manipulation, well-optimized expression, rapid growth, and low cost, thereby making it the most sophisticated prokaryotic expression system [4]. To achieve efficient protein expression, the promoter should allow the recombinant protein to accumulate up to 10–30% of the total cellular protein. As described earlier, the formation of IBs during the induction of the recombinant protein is one of the major disadvantages of using E. coli as an over-expression platform. These IB proteins lack their inherent biological activity because they form aggregates and are difficult to purify and refold into their native conformation(s). To improve the expression, solubilization, and purification of recombinant proteins, it is therefore advised to alter the culture conditions by switching the medium or the temperature, co-expressing chaperones, adding large hydrophilic fusion tags as well as adding sucrose, raffinose, glycine, betaine and sorbitol during growth for enhancing the recombinant protein expression, solubilization and purification [5, 6].

Recombinant proteins have a variety of fusion tags attached to their N- or C-termini. These fusion proteins or tags serve as chaperones, assisting their protein folding and solubilization of the target proteins. Another fusion tag with the albumin binding domain, streptococcal protein G (SPG), stabilizes short-lived proteins as a result of the binding of serum albumin, which has a longer half-life [7]. By translocating the fusion protein to various cellular regions with fewer proteases, other tags, including maltose-binding proteins and tiny ubiquitin-related modifier fusion partners can prevent it from being degraded [8, 9]. To stop self-aggregation, these chaperones tend to bind to the hydrophobic portion(s) of the partially folded proteins. An 8 kDa calcium-binding protein extracted from the parasite Fascicola hepatica is employed as a tag for the synthesis of soluble proteins in one such fusion system (Fh8). The stability and ability to purify proteins are attributes of the first 11 amino acids at the N-terminus region of the Fh8 tag. As a result, a smaller (1 kDa) H tag was created to match these corresponding to these 11 amino acids. This tag is also well known for its ability to increase solubility [10]. E. coli over-expression vectors have fusion tags, the translational initiation site, the 5′ untranslated region (5′ UTR), the antibiotic selection marker, the transcription promoter site, and the origin of replication, just like any other expression vector. For successful protein expression, the promoter must be strong and enable accumulation of the recombinant protein up to 10–30% of the total cellular protein. In the past, tags were big proteins that were resistant to proteolytic degradation, which improved the production and solubility of heterologous proteins. For example, a fusion protein with a 1024 amino acid tag, such as LacZ, might be affinity purified on p-amino-phenyl-β-d-thio-galactosidase (APTG) column and eluted with a high pH borate buffer. Most proteins with this tag, though, are insoluble [11]. Poly histidine tag is the most prevalent and cost-effective affinity tag for generating significant quantities of recombinant proteins from E. coli, as compared to certain widely used tags like FLAG tag and Strep-II tag [12]. The majority of commercially available polyhistidine tags can span from 2 to 10 histidines, which can form coordination bonds with metal ions such as Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Ca²⁺, etc. More than 80% of pure protein is obtained by overexpressing proteins with this affinity tag in E. coli than compared to mammalian and insect cell lines wherein there are more chances of proteins having stretches of histidine residues thereby purifying nonspecific proteins in the affinity chromatography step [12]. Glutathione S-transferase (GST), a 26 kDa protein, isolated primarily from the parasitic helminth Schistosoma japonicum, is one of the oldest tags and has a high affinity for glutathione [13]. The GST fusion tag has been widely utilized to enhance protein solubility and protect the protein from proteolytic degradation while achieving native protein folding. It is typically positioned at either the C- or N-terminus of the protein of interest [13]. Efficient initiation of translation is demonstrated in a wide range of overexpression platforms such as E. coli [14, 15], yeast [16, 17], plants [18, 19] and mammalian cells when full-length or truncated proteins are expressed as fusion tags [20, 21]. The simplicity with which GST-fusion proteins can be purified to homogeneity in a one-step affinity chromatography using glutathione immobilized to a matrix (or support) and glutathione itself serving as the eluant further lends support to the choice of GST. Denaturing electrophoresis, followed by western blotting and immunoprobing with commercially available anti-GST antibodies, can also be used to quickly evaluate purified GST-tagged proteins [13]. Another disadvantage of the GST tag is its large size (218 a.a.) and the propensity to dimerize in a solution that may affect the properties of the fusion protein [13]. Small and large-sized tags, including poly-Histidine (19 a.a.), Myc epitope (11 a.a.), maltose binding protein (396 a.a.), Glutathione S-transferase (GST) (211 a.a.), small ubiquitin-like modifier (SUMO) (100 a.a.), galactose binding protein (GBP) (509 a.a.), etc., have all been reported to increase protein expression and solubility [11]. Calculating the pI of the recombinant protein is also necessary because it can be used to tune the buffer pH for ion exchange chromatography [22].

As was previously described, one of the biggest downsides of using E. coli as an over-expression platform is the formation of IBs during the induction of recombinant proteins. The formation of IB protein is not an abnormal phenomenon; rather, it is one of the manifestations of protein misfolding which may result from partially folded polypeptides that fail to achieve their native stable conformation or aggregation of native protein which has less solubility [23]. It is clear that a number of circumstances, such as a high concentration of overexpressed protein, a lack of post-translational mechanisms, reducing conditions in the cytoplasm, a lack of interaction with enzymes or chaperones necessary for protein folding, etc. [23] contribute in the formation of IB. Hofmeister discovered more than ten decades ago that differences in the solubility of proteins occur with a variety of different salts. Lindwall et al. optimized the buffer composition in an effort to extract and solubilize non-aggregated proteins. Arranging the ions from the least to the most chaotropic ones, he observed that ammonium sulfate can both stabilize the proteins in the folded state as well as extract them in the solution [24]. The high salt concentration of sodium sulfate, sodium acetate, and magnesium sulfate can stabilize protein or reduce its solubility or salts such as potassium thiocyanate, magnesium chloride and calcium chloride can denature the protein or increase its solubility whereas salts such as sodium chloride and potassium chloride may or may not act as a stabilizer [25, 26]. Any given salt’s ability to stabilize or destabilize a protein depends on the ratio of the exposed polar or nonpolar groups on its surface [26]. On the other hand, the propensity of divalent salts as agents of salting-in or salting-out is produced by a delicate balance between the preferential hydration or exclusion exhibited by the surface free energy of water and the binding of the cation to the protein [27]. During protein purification, the use of high concentrations of chaotropic agents such as urea (> 4 M) and guanidine hydrochloride (> 1.5 M) results in protein denaturation and leads to aggregation, thus creating misfolded protein during the refolding process [22]. Proteins with more hydrophobic amino acids are more likely to form aggregates. Several chromatographic techniques are typically used to separate soluble proteins with > 50% yield. However, utilizing solubilization buffers with a range of pH and ionic strength (0.01–1 M NaCl concentration) as well as detergents such as 1–5% of Triton X-100 and 10 mM CHAPS, IB proteins are recovered up to low to a decent amount of yield (5–20%) [22]. These IBs form electron refractile particles that take cylindrical to ovoid shapes to fit in the E. coli cells and are dispersed throughout the cytoplasm and periplasm [28]. However, the addition of glycerol, sucrose and other polyhydric alcohols increases the stability of the protein by helping in protein folding [29]. Although IB protein purification is considered undesirable, they have some major advantages too. They are resistant to proteolytic degradation by cellular proteases and can be easily separated based on their density with a high expression level as compared to other cellular proteins.

A Myc Binding Protein (FAP174) which is an RII-like protein and an A-Kinase Anchoring Protein (FAP65) from the axoneme of the green chlorophyte, Chlamydomonas reinhardtii are the two ciliary proteins that are the subject of the current study. We are trying to map the domains of interaction between these two ciliary proteins as a part of an ongoing research program. A-kinase anchoring proteins use their amphipathic helices to bind with high affinity to the regulatory subunit of Protein kinase A. The Dimerization and Docking (D/D) domain which is the characteristic feature of these regulatory subunits (RII) has been classically used to detect cellular AKAPs [30]. The 93 a.a. long stretch of FAP174 has been shown to interact directly with an A-Kinase Anchoring Protein, viz. AKAP240 [31]. The protein has now been annotated as FAP65 [32]. This protein (FAP65) from C. reinhardtii has also been predicted to have seven abnormal spindle-like microcephaly-associated or ASPM-SPD-2-Hydin (ASH) domains (Fig. 1a). These domains are a part of the Pap-D-like superfamily [33]. As an AKAP interactor, it is hypothesized that FAP174 harbours the D/D domain that binds to the amphipathic helices predicted to be present on FAP65. Hence, using in silico strategies and the Heliquest tool, two amphipathic helices were identified [33], cloned, and over-expressed as GST fusion polypeptides in the hope that these highly hydrophobic sequences would not form IBs. However, CrFAP65AH1, CrFAP65AH2 and their proline variants (CrFAP65AH1V12P and CrFAP65AH2V12P) all produced IBs when overexpressed in E. coli. As a result, a systematic strategy was adopted to solubilize the GST-tagged proteins and bypass the refolding step, thereby shortening the time between the growth of cells until dialysis of the purified protein. The purified proteins would serve as a source in the domain mapping of the amphipathic helices of FAP65 with FAP174.

Materials and methods

All chemicals and media components were procured from Sigma or Millipore-Merck or SRL, India. The bacterial strains E. coli DH5α and E. coli BL21 DE3 used for this study were procured from Genei (Bangalore, India) and Stratagene (C607003, Thermofisher Scientific, USA).

Bacterial strains, plasmids, and transformation of E. coli cells

The target genes CrFAP65AH1, CrFAP65AH2 and their variants were gene synthesized by GenScript (https://www.genscript.com/) in the pGEX-4T-1 vector. E. coli DH5α strain was transformed with the expression vector pGEX 4T-1 with individual commercially synthesized (Genscript USA Inc.) genes of CrFAP65AH1, CrFAP65AH1V12P, CrFAP65AH2 and CrFAP65AH2V12P giving rise to Glutathione S-transferase (GST) tagged fusion polypeptides (for details see plasmid construct in Additional file 1: Fig. S1). Plasmids were individually extracted from the transformed colonies and were subsequently used to transform competent E. coli BL21 cells. E coli transformation was carried out using the chemical transformation method. Briefly, 100 μL of overnight E. coli broth culture was inoculated in 10 ml fresh LB broth. Aliquots of culture were collected at an interval of 1 h and the optical densities were determined at 600 nm. When the optical densities reached 0.4, the culture was chilled for 10–15 min following which they were transferred to pre-chilled microfuge tubes and centrifuged at 4950g/10 min at 4 °C. The supernatant was discarded carefully and the cells were resuspended in 3 ml of chilled (80 mM CaCl₂ and 50 mM MnCl₂) buffer and kept on ice for 30 min. The cells were again centrifuged at 2200g/10 min at 4 °C. The supernatant was discarded carefully and the cells were resuspended in 200 µL of chilled CaCl2 (80 mM) and kept on ice for 2 h. These cells were transformed with 1.0 μL of the DNA (100 ng/μL) by the heat shock method (42 °C for 90 s). The colonies obtained after selection on LB medium (5 g/L yeast extract, 5 g/L NaCl, 10 g/L tryptone, 20 g/L agar) containing ampicillin were patched onto agarified LB medium containing 100 mg/L ampicillin.

Protein sequence analysis

Analysis of protein sequences such as CrFAP65AH1, CrFAP65AH2 and their respective variants CrFAP65AH1V12P and CrFAP65AH2V12P was carried out using an in-silico approach. To determine the hydrophobicity of the sequence, a PROT Scale tool based on Kyte and Doolittle hydrophobic scale was used (https://web.expasy.org/protscale/). Hydrophobicity indices were determined as the Grand average of Hydropathicity (GRAVY) values of hydrophobic regions using the ProtParam tool (https://web.expasy.org/protparam/) [34].

Cell growth and induction of protein expression

One single colony of each of the four transformed clones was transferred to a 10 ml LB liquid medium containing 100 mg/L of ampicillin and grown overnight at 37 ℃. The next day, 2 ml of each of the cultures was added to 100 ml LB liquid medium containing 100 mg/L of ampicillin and allowed to grow at 37 ℃ until the optical density at 600 nm (OD₆₀₀) reaches between 0.4 and 0.6. Following this, 1 mM iso-propyl β-D-1-thiogalactopyranoside (IPTG) was added. Cells were harvested at 1, 3, and 6 h after induction, and an uninduced sample was also harvested before the addition of IPTG.

Observation of inclusion bodies in E. coli using transmission electron microscope (TEM)

Escherichia coli cells harvested after 6 h of IPTG were centrifuged at 2000g for 5 min. and resuspended in 1X PBS (0.137 M Sodium chloride, 0.0027 M Potassium chloride, 0.01 M Sodium phosphate dibasic, 0.0018 M Potassium phosphate monobasic) just before the samples were analyzed. ~ 2 µL of the sample, corresponding to ~ 200 cells were placed on the TEM copper grid and analyzed using the Tecnai G² instrument.

Buffers used for the various steps in the solubility of IBs

LSB-1: 50 mM Tris–Cl pH 7.4, 1 mM EDTA, 10 mM β-mercaptoethanol, 1% Triton-X 100, 300 mM NaCl, 1 mM Phenylmethanesulfonyl fluoride (PMSF) (Sigma, Catalogue-P7626) and 500 μL BugBuster^® Protein Extraction Reagent (70,584 Merck-Millipore).

LSB-2: 10 mM Tris–Cl, pH 7.4, 1 mM EDTA, 10 mM β-mercaptoethanol, and 1 mM PMSF.

LSB-3: 10 mM Tris pH 7.4, 1 mM EDTA, 10 mM β-mercaptoethanol, 150 mM NaCl, and 1 mM PMSF.

Formation of inclusion bodies and solubility testing

Following induction with IPTG, E. coli cells were centrifuged at 2000g/10 min., supernatant discarded and cells washed 1X with fresh Luria Bertani liquid medium. For the solubility test, cell pellet from 250 ml (⁓2 × 10¹² cells) culture was harvested and 1250 μL of a lysis and solubilization buffer containing 50 mM Tris–Cl pH 7.4, 1 mM EDTA, 10 mM β-mercaptoethanol, 1% Triton-X 100, 300 mM NaCl, 1 mM PMSF and 500 μL BugBuster^® Protein Extraction Reagent was added, maintaining the ratio of cells to buffer as ~1.6 × 10¹² cells/ml of buffer. This lysis and solubilization buffer is called LSB-1. The homogenate was incubated for 1 h/RT and sonicated for 10 cycles of 30 s each with 60 s interval at 4 ℃ followed by centrifugation at 20,000g/30 min. The pellet and supernatant were re-suspended in an SDS-PAGE sample buffer and all samples were electrophoresed on 12% denaturing gel at a constant voltage.

Solubilization strategy

For solubilization, changes were made to the LSB-1 mentioned earlier. The components that were retained were 10 mM Tris–Cl, pH 7.4, 1 mM EDTA, 10 mM β-mercaptoethanol, and 1 mM PMSF. This lysis and solubilization buffer was called LSB-2. To this LSB-2, different NaCl concentrations (75, 150 and 300 mM), IGEPAL CA-630 (1%) and increased buffer volume (1250, 1875 and 3750 μL, respectively giving rise to ratios of ~ 1.6 × 10¹², ~ 1.06 × 10¹², ~ 0.53 × 10¹² cells/ml of buffer) for a 250 ml culture pellet was used along with BugBuster^® Protein Extraction Reagent and Sonication conditions were tested.

Purification of the GST-fusion polypeptides

In a 2 ml sterile Eppendorf tube, 1000 μL (bed volume 800 μL) of Glutathione S-sepharose beads were centrifuged at 375g at 4 °C/1 min. Glutathione Sepharose™ 4B affinity chromatography resin was procured from Cytiva (Product code 17-075-605). The supernatant was discarded and the beads were washed thrice with LSB-2 containing 150 mM NaCl which is termed LSB-3 (10 mM Tris pH 7.4, 1 mM EDTA, 10 mM β-mercaptoethanol, 150 mM NaCl, and 1 mM PMSF) with or without IGEPAL CA-630 followed by incubation at 4 °C/1 h. After 1 h of equilibration, the beads were added to the supernatant and were incubated on a cell mixer at 4 °C/1 h for binding. Meanwhile, the column was rinsed with MilliQ and LSB-3. Following 1 h of binding, the beads along with the supernatant were added to the rinsed column (Econo-Pac^® Disposable Chromatography Columns, Bio-Rad Catalogue-732-1010) and allowed to settle down for 2 min. The flow-through containing the unbound proteins was collected. Lysis and solubilization buffer-3 was used as the wash buffer so that the non-specifically bound contaminants are washed out. The GST-fusion protein was then eluted with an elution buffer [10 mM reduced Glutathione (Sigma Catalogue no. G4251) in 150 mM NaCl, 10 mM Tris pH 7.4, 1 mM EDTA, 10 mM β-mercaptoethanol and 1 mM PMSF]. The eluted protein was then dialyzed against 10 mM Tris and 150 mM NaCl (pH 7.4). For dialysis, sacks (Sigma Catalogue no. D6191) of 12 kDa cut-off were used. After dialysis, the protein concentration was estimated using Bradford’s reagent with BSA as a standard followed by aliquoting and storing at − 80 °C until further use.

Circular dichroism of proteins

All the purified proteins were individually dialyzed at 4 °C, checked for the GST tag using an anti-GST antibody, and CD spectra were measured (JASCO-CD Spectropolarimeter J-815 Serial no. A029961168) in a cuvette with 1 mm path length, 195–260 nm with a bandwidth of 1 nm, and scanning speed of 100 nm/min. The protein samples (0.1 mg/ml) were dialyzed with a buffer containing 10 mM Tris and 50 mM NaCl, pH 7.4. The chamber was continuously flushed with N₂ gas. The resultant spectral values are expressed as Molar ellipticity. The BeStSel tool was used to understand the detailed secondary structure of the protein by analyzing the CD spectra (http://bestsel.elte.hu/index.php). This tool uses CD spectra of protein which is nothing but differential absorption between the right and left circularly polarized light. It also provides an estimate of the secondary structure of elements of the protein such as helix, β sheet, turn and disordered.

Pull-down assay

In this assay, a GST fused polypeptide (bait) immobilized on a glutathione-conjugated resin is used to determine its interacting partners (prey) from unpurified samples or unknown protein samples. This rapid, in vitro method, helps not only to purify the recombinant protein but also to determine its binding partners [35]. For this purpose, 2.5 mg each of purified CrFAP65AH1, CrFAP65AH2 and their respective variants were incubated for 2 h/4 °C with Glutathione-Sepharose beads (~ 50 µl bed volume) and the flow-through allowed to drain. This was followed by washes with LSB-3 until no protein was present in the washes. The beads were then incubated overnight at 4 ℃ with purified FAP174 followed by collecting the flow-through and washes. GST was used as a positive control and FAP174 was used as a negative control, to eliminate any non-specifically bound proteins to the beads during pull-down. Finally, an aliquot of the beads was then mixed with 2X SDS-PAGE sample buffer, heated, and electrophoresed using SDS-PAGE.

Dot blot overlay

To demonstrate the biological activity of the purified recombinant proteins, an overlay assay was performed wherein increasing concentrations of purified 6XHis-tagged CrFAP174 protein was spotted directly onto a nitrocellulose membrane. The dot blot was then allowed to air dry followed by 1 h of blocking with 3% Bovine Serum Albumin (BSA, SRL Catalogue no. 9048-46-8) in TBST [10 mM Tris, 150 mM NaCl and 0.05% Tween20® (Sigma Catalogue no. P9416)]. Proteins such as CrFAP65AH1, CrFAP65AH2, CrFAP65AH1V12P, CrFAP65AH2V12P and GST were affinity purified and individually overlaid on the blots at a concentration of 5 μg/ml in TBST containing 1% BSA), overnight (16 h at 4 °C). Following which the blots were washed thrice with TBST. They were then incubated with shaking conditions for 1 h at RT with primary antibody (Monoclonal Anti-GST tag antibody produced in rabbit Merck, Catalogue no. A7340). After three 10 min. washes in TBST, a secondary antibody (Goat anti-Rabbit IgG Antibody-HRP conjugate, Sigma Catalogue no. 12-348) was added and incubated at RT for 1 h. The bands or spots were detected using Clarity™ Western ECL Substrate (Biorad Catalogue no. 1705060). Controls included blots that were not overlaid with the bait but incubated with primary and secondary antibodies or secondary antibodies alone.

MALDI-TOF mass spectrometry

One μL of 10 mg/ml α-Cyano-4-hydroxycinnamic acid (HCCA, Sigma Catalogue no. C8982) made in TA30 solvent (30:70 [v/v] acetonitrile: 0.1% TFA in water) was mixed with 1 μL of 1 μM each of CrFAP65AH1, CrFAP65AH2, CrFAP65AH1V12P, CrFAP65AH2V12P. 0.5 μL of this mixture was deposited on the MALDI-TOF target plate and allowed to dry. Once dried, the MALDI-TOF plate was inserted into the mass spectrometer followed by laser ablation of the samples under vacuum. Spectra generated for the samples allowed the identification of the molecular sizes of the respective proteins.

ImageJ analysis

The gels that are marked for ImageJ analysis were converted into a high-quality jpg image and dragged to open onto the ImageJ window. To adjust the histogram of the image, the subtract background is used from the ‘Process’ menu thereby reducing the background for further processing. Although quantification is not considered very accurate, a relative percentage is always reliable. For quantifying the bands of interest, the rectangular selection tool is used to select the band of interest usually in the first lane (UI). After this selection, the corresponding induced protein (band of interest) in the second lane (I) is further selected followed by the third and subsequent lanes. When all the lanes are selected, the Analyze/Gels/Plot lanes menu is used to obtain semi-quantitative values. The wand tool is then used to select the area under the curve, this leads to an absolute value for each band. For calculating the relative values, the induced band is considered as 100 and the other values are plotted as a percentage relative to the ‘induced' value.

Statistical analysis (ANOVA and Tuckey)

The mean and variance for each of the total soluble protein concentrations in the supernatants and yields/L of E. coli culture were calculated. To investigate whether the differences in the concentration of total soluble protein and the yields were statistically significant for the different combinatorial strategies of purification, One-way ANOVA was performed with Tuckey’s method that uses studentized range distribution [36]. The significance level was set at α = 0.05. The results so obtained for each parameter are shown in the Tables for each graph. It may be noted that each condition has been repeated three times (3 biological replicates) and each time three technical replicates have been performed.

Results

In silico insight into the amphipathic helices of CrFAP65

The FAP65 sequence was identified from Phytozome (https://phytozome-next.jgi.doe.gov/) with a gene ID of Cre07.g354551 and was shown to harbour two such helices henceforth termed CrFAP65AH1 (25 a.a., 2.74 kDa) and CrFAP65AH2 (38 a.a., 4.22 kDa) (Fig. 1a, d). Two proline variants at the 12th position for each of these sequences were also synthesized. Such variants were designed since proline ‘kinks’ in the peptide thereby inhibiting the interaction. These helices and their variants when purified as recombinant proteins would serve as a useful resource in protein interaction studies. Knowing that these are amphipathic in nature and that overexpression in E. coli might pose a problem, we set out to investigate in silico parameters to gain further insights. Various physicochemical parameters such as the total charge (− 6 for CrFAP65AH1 and − 7 for CrFAP65AH2), the pI (3.66 for CrFAP65AH1 and 3.89 for CrFAP65AH2), the number of polar (40% for CrFAP65AH1 and 44.74% for CrFAP65AH2) and non-polar residues (60% for CrFAP65AH1 and 55.26% for CrFAP65AH2) were estimated by HeliQuest (https://heliquest.ipmc.cnrs.fr/). The Grand average of hydropathicity (GRAVY) index score was also determined. It may be noted that the GRAVY index score is a measure of the average hydrophobicity and hydrophilicity of proteins measured by the Kyte-Doolittle Formula [37]. The GRAVY index and other in silico parameters for each protein sequence were also identified using the ProtParam tool. The GRAVY index measures the ratio of the sum of hydropathy values of all a.a. to the length of the protein. A hydrophobicity score is an arbitrary unit in which, below zero indicates the likelihood of the protein of interest to be globular (i.e. more hydrophilic), while scores above zero indicate the proteins be membranous (i.e. more hydrophobic). When applied to CrFAP65 amphipathic helices, the GRAVY index for CrFAP65AH1 and CrFAP65AH2 were 0.412 and − 0.089, respectively indicating that the former was more hydrophobic than the latter. To further ascertain the hydropathicity for these sequences, hydropathy values of each a.a. residues were plotted for CrFAP65AH1, CrFAP65AH2 and their variants (CrFAP65AH1V12P and CrFAP65AH2V12P; Fig. 1b, c). These plots indicate that the CrFAP65AH1 and its variant are largely hydrophobic in nature as compared to CrFAP65AH2 and its variant. The number of hydrophobic amino acids in CrFAP65AH1 are 11 in number, while those in CrFAP65AH2 are 23, thus making both the peptide sequences highly hydrophobic [ProtParam (https://web.expasy.org/protparam/)]. Our analysis showed two prominent hydrophobic peaks in all four proteins. Since we are dealing with amphipathic helices, we set about using Heliquest to make the helical wheel and determine the hydrophobic moment (μH, Fig. 1d). The latter is the mean vector sum of the side-chain hydrophobicities of a given helix with N number of a.a. residues. It is 0.303 and 0.307 for CrFAP65AH1 and its variant, respectively. The hydrophobic moment drops to 0.103 and 0.107 for CrFAP65AH2 and its variant, respectively (Fig. 1d). This indicated that CrFAP65AH1 and its variant were more amphipathic in nature as compared to CrFAP65AH2 and its variant. Given all these in silico inputs, it was decided to choose a tag such as GST (pGEX-4T-1 vector) which could in principle solubilize the polypeptides and refrain it m forming IBs when overexpressed in a host such as E. coli.

Induction and inclusion body formation of the CrFAP65 clones

Once the sequences of the transformed clones were verified for their authenticity, cells were grown and induced with IPTG (1 mM) and the cell pellet was checked for induction on a denaturing gel (Fig. 2a). All four independent clones of CrFAP65AH1, CrFAP65AH2, CrFAP65AH1V12P and CrFAP65AH2V12P showed abundant overexpression at 1, 3 and 6 h of induction. The molecular weights with the GST tag (26 kDa) on the denaturing gel for CrFAP65AH1 and CrFAP65AH1V12P (both, 29.4 kDa), and CrFAP65AH2 and CrFAP65AH2V12P (both, 30.64 kDa) were as expected. The next obvious step was to check for the solubility of the overexpressed recombinant fusion proteins (Fig. 2b). For the amphipathic helices of FAP65, the conventional laboratory procedure of lysis using sonication in LSB-1 (containing 300 mM NaCl and BugBuster^®) was used. The latter treatment did solubilize the recombinant proteins to some extent (Fig. 2b). However, when the individual supernatants for the fusion proteins were used for purification, the yields obtained were ~ 4–6 mg/L of E. coli culture (data not shown). This yield was found to be substantially lower than those reported for most GST-tagged fusion proteins [21]. It was observed that most of the protein remained in the pellet which means most of the protein did not solubilize due to IB formation. Therefore, sub-cloning these genes in a vector (pET28a) having a smaller tag (6X His) was attempted in parallel. However, no visible induction was obtained (data not shown). Hence, it was decided to pursue using pGEX-4T-1 for expressing these genes in E. coli. To confirm that the protein forms IBs in E. coli cells, TEM analysis was performed of uninduced E. coli BL21-DE3 cells, E. coli cells inducing GST, CrFAP65AH1, CrFAP65AH1V12P, CrFAP65AH2 and CrFAP65AH2V12P gene. The electron-dense particles (IBs) were observed at the extreme corners in the induced E. coli BL21-DE3 cells expressing CrFAP65AH1, CrFAP65AH1V12P, CrFAP65AH2 and CrFAP65AH2V12P protein. In the E. coli cells inducing GST, these electron-dense particles were seen throughout the cytoplasm whereas no such electron-dense particles were seen in any uninduced E. coli cells (Fig. 2c). Induction was also performed at a lower temperature i.e. at 20 °C using 0.1 mM IPTG. However, as observed under a transmission electron microscope, IBs were still evident (Additional file 2: Fig. S2). To improve their respective yields, a systematic strategy involving the ‘design of experiments approach' for solubilization was developed by first changing the ionic strength of the LSB-1 that contains 300 mM NaCl. Decreasing the NaCl concentration will give rise to a salting-in effect thereby solubilizing the protein further. These improvisations were first tested on CrFAP65AH1. Using different reagents for each solubilization experiment, the target outcomes for each were monitored after incubating the induced cell pellets with the solubilization buffer containing the reagent followed by centrifugation and electrophoresis on a denaturing SDS-PAGE gel. An outcome was monitored by semi-quantitative ImageJ analysis of the band corresponding to the GST-fusion polypeptide (mentioned as the band of interest) from the treatments of the various supernatants.

Effect of increasing the ionic strength and diluting the cell lysate in the presence and absence of BugBuster^® without sonication

Induced cells were next used in triplicates and re-suspended independently in LSB-2 containing 75, 150 or 300 mM NaCl keeping all the other components (10 mM Tris, 1 mM EDTA, 10 mM β-mercaptoethanol and 1 mM PMSF) intact, followed by incubation with BugBuster^® and processed without any sonication. The pellets after solubilization and supernatants for each treatment were electrophoresed on a denaturing gel (Fig. 3a, b). Although the salting-in effect was observed in samples treated with LSB-2 containing 75 and 150 mM NaCl, the total soluble protein content of the supernatants revealed that samples treated with LSB-2 containing 150 mM NaCl extracted maximum total soluble protein content in the supernatants (Fig. 3c), and those treated with 300 mM showed an inhibiting and therefore a salting-out effect. The overall solubilization for all proteins significantly increased (~ eightfold) in the presence of BugBuster^® (Fig. 3c) indicating that the proprietary formulation provided by the manufacturer worked to disrupt E. coli cells and released several proteins as it contains non-ionic detergents. At this stage, it was unclear if all the E. coli cells were disrupted and whether the retained CrFAP65AH1 in the pellet was because of compromised disruption of cells or because the protein was not fully solubilized. Semi-quantitative ImageJ analysis of the fusion polypeptide band indicated that the LSB-2 containing 150 mM NaCl with BugBuster^® increased the solubilization albeit not very significantly (Fig. 3d). Since it was decided to avoid most of the harsh chaotropic reagents, the next step was to increase the volume of the solubilization buffer. We observed that there was a statistically significant difference between the group means F(5,12) = 2140.42, p = 2.89E⁻¹⁷, F_{critical value} = 3.10 (Fig. 3c; Total soluble protein values), wherein F statistic (between,df_within) = F ratio, p-value, F-critical value.

In the presence of 150 mM NaCl in LSB-2, the volume of the buffer was increased from 1250 to 1875 and 3750 μL for a cell density of ~ 2 × 10¹² cells (250 ml culture pellet), (Fig. 4a) avoiding sonication at this point. Close to twofold solubilization (increase) was observed with an increasing volume of the buffer, maximum solubilization was seen when the buffer volume was raised to 3750 μL (Fig. 4b). At this stage, the effect of NaCl and increased buffer volume was monitored by electrophoresis of the pellet and supernatant post the respective treatments. The total protein in the supernatants of respective treatments and the intensity of the solubilized/insolubilized bands across gels was calculated. Nevertheless, to ascertain the success of this experimental design, the diluted cell lysates were also monitored for the yield of pure protein post-affinity purification (Fig. 4b, grey histogram). We observed that there was a statistically significant difference between the group means F(2,6) = 380.59, p = 4.78E⁻⁰⁷, F_{critical value} = 5.14 (Fig. 4b; Yields, grey bars; Fig. 4c), wherein F statistic (df_between,df_within) = F ratio, p-value, F-critical value. The effect of sonication was further determined in the presence and absence of a mild detergent, IGEPAL CA-630. In addition, sonication would also be carried out in the presence and absence (see below) of BugBuster^® to determine if the latter could be used as a substitute for sonication.

Effect of sonication in the presence and absence of BugBuster^® and another variable, IGEPAL CA-630

Since increasing the volume to 3750 μL solubilized > 50% of the protein from the IBs, it was decided to combine BugBuster^® with NaCl, a mild detergent (IGEPAL CA-630), and sonication either singly or in combination. The solubilization procedure was followed by centrifugation and affinity purification of the individual recombinant proteins. The yield of the pure protein was calculated after purification for each treatment. Hence, keeping constant the volume of 3750 μL LSB-3, the various parameters that were tested were: BugBuster^® (BB), BugBuster^® followed by Sonication (BB + SN), LSB-2 containing 150 mM NaCl, henceforth called, Lysis and sonication buffer-3 [LSB-3]; LSB-3 with BugBuster^® [LSB-3 + BB]; LSB-3 followed by Sonication [LSB-3 + SN]; LSB-3 with BugBuster^® followed by Sonication [LSB-3 + BB + SN]; LSB-3 with BugBuster^® and IGEPAL CA-630 [LSB-3 + BB + IG]; LSB-3 with Bugbuster^® and IGEPAL followed by Sonication [LSB-3 + BB + IG + SN] (Fig. 5a). When the eluates were pooled and dialyzed, the protein concentration was measured, and yield (per litre of E. coli culture) was calculated and plotted. The yield for all these combinations showed a broad range, indicative of the solubilization abilities of BugBuster^® alone to sonication in the presence and absence of IGEPAL CA-630 and NaCl. The treatment of LSB-3 increased 20-fold when it contained BugBuster^® and was sonicated. The presence of BugBuster^® alone with LSB-3 did not make a significant difference. On the other hand, substituting sonication for BugBuster^® did make a difference, and a ~ 3.5-fold increase in the yield with sonication was observed (see LSB-3 + BB versus LSB-3 + SN in Fig. 5a). Therefore, BugBuster^® could not be used as a substitute for sonication. Nevertheless, BugBuster^® alone with and without sonication did not make a significant difference in the yields (see BB versus BB + SN in Fig. 5a). Hence, retaining BugBuster^® along with sonication proved useful in solubilizing most of the fusion polypeptide resulting in the highest yield (~ 20 mg/L of E. coli culture) among all the conditions tested in this study (Fig. 5a). It may be emphasized that the presence of IGEPAL CA-630 reduced the yield of CrFAP65AH1 as compared to the treatment without IGEPAL CA-630 (see LSB-3 + BB + SN versus LSB-3 + BB + IG + SN in Fig. 5a). To test if this condition was inhibitory for other proteins as well, individual purifications were carried out for other protein pellets in the presence and absence of IGEPAL CA-630 (CrFAP65AH1V12P, CrFAP65AH2 and CrFAP65AH2V12P). The purified protein yields were calculated and a comparison indicated that the best yield was seen for CrFAP65AH2 and its variant which were less hydrophobic than CrFAP65AH1 and its variant (Fig. 6a). To further demonstrate the inhibitory effect of IGEPAL CA-630, all the fusion proteins were solubilized using LSB-3 with BugBuster^® and sonicated in the presence of IGEPAL CA-630. When compared with the treatment that did not contain IGEPAL CA-630, inhibition to the tune of 20–30% was observed in all the cases tested (Fig. 6a, dark grey histograms). We observed that there was a statistically significant difference between the group means F(7,16) = 90.83, p = 1.11E-11, F_{critical value} = 2.65 (Fig. 5b), F(1,4) = 8.56, p = 0.0429, F_{critical value} = 7.70 (Fig. 6b) (CrFAP65AH1 ± IGEPAL CA-630), F(1,4) = 302.17, p = 0.000064, F_{critical value} = 7.70 (Fig. 6b) (CrFAP65AH1V12P ± IGEPAL CA-630), F(1,4) = 28.53, p = 0.0059, F_{critical value} = 7.70 (Fig. 6b) (CrFAP65AH2 ± IGEPAL CA-630) and F(1,4) = 9.60, p = 0.036, F_{critical value} = 7.70 (Fig. 6b) (CrFAP65AH2V12P ± IGEPAL CA-630) wherein F statistic (df_between,df_within) = F ratio, p-value, F-critical value.

Quality checks, secondary structures of the purified proteins and bioactivity

The proteins thus purified were checked for their molecular weights by electrophoresis on a denaturing gel followed by the sensitive silver staining method (Fig. 6a, inset). Additionally, they were subjected to purity checks and verified for their precise molecular weight (Mr) using MALDI-TOF (Additional file 3: Fig. S3). It was observed that the molecular weights for all the fusion proteins were as expected. Biophysical characterization using a far UV CD spectrum of the purified fusion proteins post-solubilization was carried out (Fig. 7a, b). It indicated that the alpha-helicity increased twofold (from 29.2 to 54.1%) with the substitution of the valine at the 12^th position in CrFAP65AH1; while there is only a ~ 15% increase (from 66.4 to 76.8%) in the alpha-helicity when the same is done with CrFAP65AH2. The values for the alpha helicity were determined after processing the CD spectra with the BeStSel tool (Fig. 7c). It is known that proline residue causes a disruption in the protein's secondary structure and conforms to an alpha-helix or beta-sheet structure.

Taken together, our solubilization conditions with the use of mild denaturing reagents resulted in high-yield pure protein with the expected molecular weight and alpha-helical content. Further, the functionality of the purified recombinant proteins was ascertained using an interaction assay. Using highly purified FAP174 full-length protein as bait, a pull-down assay was performed individually with CrFAP65AH1, CrFAP65AH2 and their respective variants. These, which were bound to the Glutathione Sepharose beads worked as the prey. When the bands of these individual pull-downs were analysed on a denaturing gel, it was observed that CrFAP65AH1 bound to both the monomer and dimer of FAP174 (Fig. 8d); the variant on the other hand did not show any interaction (Fig. 8e). When CrFAP65AH2 was used, binding was evident but was weaker than that seen with CrFAP65AH1 (Fig. 8f). Similarly, the variant of CrFAP65AH2 did not show any interaction (Fig. 8g). The appropriate controls (only GST and GST with FAP174) showed no interaction with the respective recombinant proteins of CrFAP65 (Fig. 8a–c). Dot blot was further performed to ascertain the binding between FAP174 and CrFAP65AH1, CrFAP65AH2 and their respective variants (Fig. 8h). It was observed that CrFAP65AH1 and CrFAP65AH2 bound to FAP174 whereas no binding was observed in the respective variants, GST, Primary and Secondary controls. We, therefore, surmise that the procedure used for purification yields bioactive recombinant proteins.

Statistical analysis

We applied one-way ANOVA to determine whether there is a statistically significant difference between the protein yields within the eight purification techniques (independent variables). As the F statistic value for all was greater than the F critical value, we concluded that the test is significant. As the p-value was less than 0.05 we rejected the null hypothesis which was there is no significant difference between the protein yields within the respective purification techniques and accepted the alternative hypothesis which concludes that the difference between the protein yields within the purification techniques was statistically significant. Further, Tuckey’s post-hoc test confirmed the level of significance (Figs. 3e, 4c, 5b, 6b).

Discussion

Formation of inclusion bodies of GST-tagged amphipathic helices and their proline variants and initial steps of solubilization

The current study is an offshoot of an ongoing research work wherein FAP174 was used as bait to identify a protein complex from the flagella of C. reinhardtii. One of the direct interactors of FAP174 is an AKAP that was later annotated as FAP65 [31, 32]. It was shown to harbour two amphipathic helices, a signature sequence present in all AKAPs. Present in several stably folded proteins, amphipathic helices are found across kingdoms (viruses, bacteria, and eukaryotes) with a unique role in membrane targeting via protein-lipid and protein–protein interaction [38]. The important cellular functions they participate in include sensing membrane curvature, formation of tubular or spherical membrane intermediates, protecting membranes and lipid droplets, promoting membrane anchorage, and mediating membrane fission or scaffolding signalling components [38,39,40,41,42]. The list of proteins (membrane-bound, ion channels, apolipoproteins, AKAPs and lung surfactant) that are known to harbour amphipathic helices is long, and based on a detailed analysis of their physicochemical properties are categorized as 7 distinct classes (A, H, L, G, K, C, and M) [43, 44]. CrFAP65 has been identified as a ciliary protein with a likely role in motility. CrFAP65 is an AKAP, and keeping in mind the hydrophobic nature of the sequences, (Fig. 1) we used in silico tools to study the GRAVY index, a parameter that projects the hydrophobicity of the amphipathic helix. It was seen that CrFAP65AH1 and its variant showed 0.412 and 0.18 GRAVY index, respectively indicating the hydrophobic nature of the protein fragments; whereas CrFAP65AH2 and its variant showed − 0.089 and − 0.242 with not-so-high hydrophobic nature (Fig. 1d). However, as these values of CrFAP65AH2 and CrFAP65AH2V12P are closer towards zero, they may tend to display hydrophobic characteristics. Also, the amphipathic content was evident with the helical wheel projected using Heliquest and the hydrophobic moment indicated high hydrophobicity for CrFAP65AH1 and its variant as compared to CrFAP65AH2 and its variant (Fig. 1d). This feature has been observed for all amphipathic helices so far reported. Based on all these features, we have placed CrFAP65AH1 and CrFAP65AH2 into the globular (G) class [44]. There are different reports of amphipathic helices or amphipathic helix-containing proteins that form IBs upon over-expression in E. coli. These are also known to drive IB formation or simply form IBs in cells leading to diseases. For example, amphipathic polymers such as amphipols (Apols) help in stabilizing membrane proteins and are generally known to form IBs [45]. α-Synuclein, a 140 amino acid (a.a.) α-helix-rich protein contains an amphipathic helix and is known to form IBs in the cytoplasm. Such an aggregation of the protein is the main cause of Parkinson’s disease (PD) and Lewy body dementia (LBD) [46]. On the other hand, a self-assembling hydrophobic peptide GFIL8 (GFILGFIL) has been used to induce IB formation in E. coli [47]. However, the recombinant protein containing the coiled-coil domain does not need to maintain its helical structure in the IBs. Hence, we decided to fuse these amphipathic helices of FAP65 protein to a solubility-enhancer tag such as GST whose GRAVY index is -0.446, i.e. hydrophilic in nature [13]. Several studies have attempted the use of GST as a tag to produce recombinant amphipathic helices. Dengue virus non-structural protein 4A (NSP4A) whose 1–48 a.a. and its variant were cloned and overexpressed as GST-tagged fusion protein with yields of 4–5 mg/L of E. coli culture [48]. Yeast Bud3p involved in bud formation harbours an amphipathic helix and was GST-tagged for pull-down assays. However, no purification has been done with the Bud3p-GST protein [49]. Amphipathic helix-containing proteins belonging to the membrane binding BAR (Bin/Amphiphysin/Rvs) categories have also been produced with the GST tag, these being PICK1 (Protein Interacting with C Kinase), and ICA69 (Islet Cell Autoantigen 69 kDa). However, the buffers used were very different and the mention of IB is not evident [50]. Therefore, the amphipathic helices of CrFAP65 (WT and the variants) were individually gene synthesized in pGEX-4T-1 and the choice of the host was E. coli, as it is with many investigators too. While GST-tagged fusion proteins rarely pose challenges in E. coli, the absence of a post-translational modification system and the formation of IBs are the two most unpopular aspects of this overexpression platform. To confirm whether these recombinant proteins form IBs we carried out a TEM analysis of E. coli cells overexpressing these amphipathic helices. It is known that the presence of hydrophobic patches in a recombinant protein causes it to aggregate due mainly to misfolding, thus leading to IB formation. These IBs generally appear as an electron-dense structure under the transmission electron microscope [28]. While in most cases, the ease of isolating the IB itself serves as one step of purification, the isolation of such aggregates might lead to bio-inactive recombinant protein. This might prove to be an undesirable bet. Therefore, we avoided the isolation of these IB aggregates that are already in their unfolded state. As expected, the overexpression in E. coli did not pose any challenges (Fig. 2a) with very high overexpression seen in samples induced for 6 h/37 °C. The induction was confirmed by determining the molecular weight of the protein and comparing it with the induced samples after electrophoresing them on an SDS-PAGE denaturing gel. While the over-expression for all four fusion proteins was abundant, the molecular weights were also as expected (Fig. 2a). We made use of a solubility enhancer tag (i.e. GST), low concentrations of the inducer (viz. IPTG), and a lysis and solubilization buffer that is regularly used for the solubilization of other recombinant flagellar proteins from C. reinhardtii. Despite these conditions, the fusion proteins were not completely soluble (Fig. 2b). By visual examination, it was quite evident that they all formed IBs (Fig. 2c), the least IB formation was seen with CrFAP65AH2. We attribute the IB formation to several factors, such as the strong promoter (Tac) on the pGEX vector, probably the high copy number of the target genes and the hydrophobicity of the translated proteins [51]. Since these fusion proteins would eventually serve as baits in protein interaction assays, recovery of bioactive purified products by using milder treatments was sought. Hence, avoiding the use of harsh chaotropic or utilizing mild denaturing conditions for solubilization was the primary goal with the hope to obtain bioactive fusion protein. Studies on the use of non-denaturing agents such as N-lauryl sarcosine, dimethyl sulfoxide (DMSO), 5% n-propanol, mild non-ionic detergents, high pH buffers, and low denaturant concentration that preserve the native-like state of the fusion proteins have been reported [52,53,54,55,56,57]. To solubilize IBs a combination of denaturants such as Sodium do-decyl sulphate (SDS), urea and organic solvents such as 40% (v/v) 1-Propanol and 20% (v/v) 2-Butanol in an acidic pH (range of 2–3) are used [58]. Chaperones have the ability to transiently bind to the hydrophobic region of a protein and this binding avoids IB formation. Recently, the use of nanobodies as what is referred to as solubilization chaperones are preferred. These nanobodies can detect the discontinuous amino acids of a native protein structure thereby stabilizing it. Together, over-expressing an epitope EPEA tag (Glutamic acid-Proline-Glutamic acid-Alanine) bound to the recombinant protein and an anti-EPEA conjugated nanobody which is supposed to recognize each other thereby aiding in soluble protein production [59]. Our aim involved the use of a systematic design of experiments that tested the effect of salt, increasing buffer volume, adding a commercial solubilizing concoction (BugBuster^®), using a mild detergent such as IGEPAL CA-630 and mechanical lysis using sonication. This design, singly or in combination, was aimed at gently disrupting the interactions in the IB aggregates that are ionic, hydrophobic, and have disulphide bridges or van der Waals forces.

When the induced cell pellets for each fusion protein were individually tested for solubilization in the LSB-1 containing 300 mM NaCl, solubilization was not complete and > 50–70% of the fusion proteins were still found to be associated with the pellets (Fig. 2b), indicating that the conditions in this buffer were unable to unfold the fusion proteins from the aggregates of IBs. Hence, the systematic strategy for solubilization that we adopted was to first target the salt present in the lysis and solubilization buffer and replace it with three different concentrations, viz. 75, 150 and 300 mM. Several inorganic and organic salts have in the past been used to denature or solubilize proteins from IBs. Inorganic salts are known to denature proteins when used at concentrations > 1 M. Of these, NaCl and KCl are the most popular as these are not only easily dialyzable but are successful in selectively extracting membrane proteins, as well [26]. NaCl at the concentration we used (300 mM) may contribute to a salting-out effect. It is known that lower concentrations of NaCl (< 200 mM) create a salting-in, an effect that is useful for solubilization. Hence, we used three concentrations of NaCl (75, 150 and 300 mM) and compared the solubilization in the presence and absence of a cell-lysing commercial reagent, viz. BugBuster^®. It may be noted that the lysis of cells is one of the major contributory factors to the yield of purified proteins. The more the lysis, the more accessibility would be for the components in the buffer towards accessing the IBs and therefore solubilization. For effective solubilization, we used BugBuster^® which contains non-ionic and zwitterionic detergents (see user protocol TB245 Rev. F 1108, pages 1–7 of BugBuster^® Protein Extraction Reagent, Novagen). Next, we decided on the means for gauging or monitoring solubilization. Besides using denaturing gel electrophoresis, we also determined the total soluble protein content in the supernatants using Bradford’s reagent. The gels were used to semi-quantitate the band intensity using ImageJ. When such an experiment (with/without BugBuster^® and increasing NaCl) was performed, (Fig. 3a, b), the induced bands were quantitated for their respective intensities (Fig. 3d). The induced band intensity and the total soluble protein in the supernatants (Fig. 3c) were highest in the treatment that received BugBuster^® and 150 mM NaCl. Although there was no apparent difference in the three treatments, a one-way ANOVA with Tuckey analysis revealed that the total soluble protein in the supernatants obtained using 150 mM NaCl in LSB-2 was most significant over the other treatments (Fig. 3e). It was therefore decided to use this condition (lysis buffer containing 150 mM NaCl, i.e. LSB-2) for further solubilization. At this stage, two possibilities exist, either all the cells might not have been lysed with BugBuster^®, or given the ionic strength of the buffer with 150 mM NaCl and the buffer volume (1250 μL), there already exists an equilibrium between the aggregated protein molecules in the IBs versus the folded protein molecules in the soluble fraction (supernatant). The latter is true when non-denaturing buffers are used for solubilizing IBs, especially without the use of any drastic or strong solubilization agent(s). This has been reported in a few cases, such as N-acetyl-d-glucosamine 2-epimerase IBs have been solubilized using Tris–HCl buffer, pH 7.0 [60], Granulocyte Colony-Stimulating Factor (G-CSF), His7dN6TNF-α (His-tagged, N-terminally truncated form of tumour necrosis factor) and GFP (green fluorescent protein) wherein a lower temperature (25^oC) of induction was used [61]. It might also be possible that as in non-classical IBs, the unfolded aggregates might exist along with native-like structures, the latter being easily soluble. At this stage, we suspected that the amphipathic helix fusion proteins were partitioned between the insoluble aggregates and soluble fractions. Earlier studies have shown that protein solubility is also affected by other co-solvents that can bind to the protein or change the structure of water [62]. The best starting point was therefore to increase the lysis and solubilization buffer volume in the presence of BugBuster^®. We, therefore, used 1875 and 3750 μL in the hope that detergents in the BugBuster^® would solubilize the fusion proteins from the IBs and the increased volume would in turn shift the equilibrium towards solubilization. Having tested this design, we found that the band intensity increased ~ 10–12% with every rise in the volume from 1250 through 1875 to 3750 μL (Fig. 4b). The total soluble protein content in the supernatants was also calculated (Fig. 4b), it was found that the highest protein content was seen when the LSB-2 volume was 1875 μL. On the other hand, the total soluble protein content in the supernatant dropped by ~ 50%, partly because of the dilution. However, the band intensity indicated a selective increase in the fusion protein. This incongruency in the band intensity and total soluble protein content in the supernatant prompted us to purify the protein to near homogeneity using affinity chromatography. The yield of the finally dialyzed pure protein was calculated as mg protein/L of E. coli culture. The yield for 3750 μL showed > two-fold increase over that of 1875 μL and > four-fold increase over that obtained from 1250 μL (Fig. 4b). Further, these values were found to be statistically significant when one-way ANOVA with Tuckey analysis was performed (Fig. 4c). However, this increase in the pure protein yield was not as high as one would anticipate for a GST-tagged fusion protein. This also tempted us to believe that the IBs are probably non-classical as they were solubilized in the presence of mild detergents and low concentrations of NaCl with a shift in the equilibrium of folded protein from the IBs upon dilution of the cell lysate. However, the question that remained unaddressed was that of the lysis of the E. coli cells. Hence, the next step was to try these designs either singly or in combination. Since purification of the protein from the supernatant was the confirmatory test of selective solubility of the fusion protein, subsequent tests used the supernatants for purification and thus calculation of the purified protein yields.

Comparison of the final yields of the purified fusion proteins from all treatments for an accurate understanding of the optimum solubilization condition

Once assured that the fusion protein was soluble in the conditions mentioned earlier (3750 μL of LSB-3 with BugBuster^®) albeit, with not very high yield, we wanted to ensure complete cell lysis using sonication. Knowing that these IBs behave like the non-classical ones, we also tested the effect of a mild non-ionic detergent, IGEPAL CA-630. Hence, in the subsequent exhaustive design, we used the following conditions for solubilization:

1.
BugBuster^® (with and without sonication)
2.
LSB-3 (with and without sonication)
3.
LSB-3 incubated with BugBuster^® (with and without sonication)
4.
LSB-3 incubated with BugBuster^® treated with IGEPAL (with and without sonication).

The supernatants from these treatments were individually subjected to fusion protein purification, and the purified dialyzed protein was used to estimate the total yield per litre of E. coli culture. The graph so obtained indicated that while solubilization with BugBuster^® increased the yield to ~ 6.5 mg/L of E. coli culture, the conditions wherein sonication was used increased the protein yield to > 7.5 mg/L of E. coli culture. In contrast to the lowest yield (~ 2 mg/L of E. coli culture), a tenfold yield (~ 20 mg/L of E. coli culture) was obtained with LSB-3 incubated with BugBuster^® (with sonication; Fig. 5a). We also observed that IGEPAL CA-630 decreased the yield which meant that solubilization or cell lysis was inhibited (Fig. 5a). IGEPAL CA-630, like all detergents, is being used as a surfactant which by virtue of its amphiphilic property aids the process of cell lysis by disrupting the cell membrane thereby releasing intracellular material. IGEPAL CA-630 (IUPAC name octylphenoxypolyethoxyethanol) (Sigma-Aldrich Catalogue no. 56741) is a non-ionic, non-denaturing mild detergent and is completely miscible in water. The hydrophobic-hydrophilic balance for the detergent is 13.1 with a crucial role for both octylphenol and ethylene oxide. This allows it to break protein-lipid and lipid-lipid interactions, but not protein–protein interactions. IGEPAL CA-630 has bulky non-polar heads that generally do not exhibit cooperative binding, much as seen with ionic detergents. Due to this property, it will efficiently disrupt membranes and not penetrate native structures. Our results indicate that IGEPAL CA-630 is certainly not a substitute for sonication; in fact, it acts as an inhibitor of solubilization. Due to its interference with protein estimation assays, we ensured that dialysis completely removed the detergent before we embarked on the protein estimation assays. Hence, the values calculated for the pure protein yields are authentic and significant as analyzed using one-way ANOVA with Tuckey (Fig. 5b).

When we compared the yields of GST-tagged fusion proteins as reported in the literature with the current study, we found few reports which suggested that GST is a poor solubility tag as compared to commonly used fusion tags as their protein yields were very low (Efna1 − 0.06 mg/L, CDK2 − 2.17 mg/L) [63]. Another GST-tagged amphipathic helix from the N-terminal of the dengue virus non-structural protein 4A (NSP-4A) produced low yields (4–5 mg/L of E. coli culture [48]. In yet another report, > 27 genes (without any amphipathic helices) were cloned with the GST tag. Of these, only seven produced 50% solubility (SMPX_HUMAN, HBP1_HUMAN, IPKA_HUMAN), the others were either present in very low amounts (MAR1_HUMAN, APR_HUMAN) or were not solubilized at all (STP1_HUMAN, BTG1_HUMAN, MGN_HUMAN) [64, 65]. An immunomodulatory protein from Ganoderma tsugae containing an N-terminal amphipathic helix when over-expressed in E. coli as a GST-tagged fusion protein gives rise to 20 mg/L of E. coli culture [66].

Quality checks and structural analysis of fusion proteins by Circular Dichroism

Since the pure protein yield was maximum with LSB-3 incubated with BugBuster^® followed by sonication, this treatment was applied with CrFAP65AH2 and the variants (CrFAP65AH1V12P and CrFAP65AH2V12P). The yield for CrFAP65AH1 was lower than that obtained for CrFAP65AH1V12P, and the reason for this is not known (Fig. 6a). It may be that the substitution of proline introduces more helicity in the sequence. On the other hand, CrFAP65AH2 and its variant, CrFAP65AH2V12P exhibited a very high yield (30–45 mg/L of E. coli culture; Fig. 6a). However, upon SDS-PAGE analysis and silver staining, it was observed that the CrFAP65AH2 (WT and variant) was susceptible to degradation, and we attribute this high yield to the intact and degraded protein. It may be added that this degradation was seen even in the induced pellets and has no correlation to the solubilization process (data not shown), and the degraded 26 kDa protein is GST (Fig. 6A, inset). Taken together, the average yield for all the fusion proteins could be estimated as ~ 20 mg/L of E. coli culture. Such degradation has been observed in GST-tagged fusion proteins generated for truncated versions of a small G-protein (ArfGAP1) that harbours amphipathic helices. These seemed quite soluble in the extracts but showed degradation of the GST [67]. The authors attributed this to the fusion protein being soluble in the E. coli cells and being simultaneously vulnerable to proteases. It is also reported that the GST tag co-purifies itself along with the recombinant fusion protein as translation stops prematurely [68]. Given that CrFAP65AH2 and its variant are less hydrophobic, we too extend this reasoning. Additionally, the use of IGEPAL-CA630 in the buffer inhibited the protein yield. This was ascertained statistically using one-way ANOVA with Tuckey analysis (Fig. 6b).

Since we expect the native-like structures of the fusion proteins to be preserved in the solubilization experiments that we tested, which also bypassed the refolding step (and, saved time), we wanted to ascertain the intactness of the secondary structure. Amphipathic helices have both hydrophobic and hydrophilic a.a. residues arranged in such a manner that the helix so formed creates two faces, one being hydrophobic and the other hydrophilic facing the opposite side. Such sequences have an inherent property of folding into helical structures upon contact with polar/non-polar interfaces. Hence, we estimated the alpha-helical content of amphipathic helices of CrFAP65 using circular dichroism and the BestSel tool (Fig. 7a–c). The alpha-helical content was 29% and 66% respectively for CrFAP65AH1 and CrFAP65AH2 which increased with the substitution of the valine at 2020 and 2231 positions, respectively with proline from 29% for CrFAP65AH1 to 54% for the variant (CrFAP65AH1V12P). Again, the same was observed for the alpha-helical content for CrFAP65AH2 that increased from 66 to 77% for the variant (CrFAP65AH2V12P) (Fig. 7c). This was the first time that we estimated the alpha-helical content of the amphipathic helices. We further compared these values with those reported in the literature. For example, when a model amphipathic peptide (Ac-Gly-Ala-Glu-Lys-Ala-Ala-Lys-Glu-Ala-Glu-Lys-Ala-Ala-Lys-Glu-Ala- Glu-Lys-amide) was designed, and its alpha-helical content was measured using CD and 2D-NMR spectroscopies, it was shown to contain 65% alpha-helical structure [69]. The N-terminus Myristoylated-ADP Ribosylation Factor 1 (Myr-ADPR1) peptide has been shown to adopt a nearly 100% α-helical structure as determined by CD [70]. The human apolipoprotein C-1 has an amphipathic helix with alpha-helical content that increases to 65–75% when bound to phospholipids [71]. Certain GST-tagged amphipathic helices of ArfGAP1 (a small G-protein) are quite unstructured in solution, however, upon binding to liposomes, they do exhibit 25–48% alpha-helical content. ArfGAP1 responds to membrane curvature through the folding of a lipid packing sensor motif [70]. As for the increase in the alpha-helical content with substitution of value with proline, we note that despite proline being accepted as the helical breaker in an aqueous medium, reports of its function otherwise have been observed when they are in membrane environments than in water. The presence of proline in long alpha helices also helps in the proper folding of the proteins [72]. Proline has also been observed to increase the thermal stability of the protein as well as the alpha-helical conformation of the protein at high temperatures in presence of 2-propanol [73].

Since the secondary structure of the recombinant proteins was found intact, the functionality was further verified using a pull-down assay. FAP174 is an established interactor of AKAP240 i.e. CrFAP65 [31, 32]. Hence, the full-length purified 6XHis-tagged recombinant protein purified using Ni–NTA affinity chromatography was used as prey for the interaction with CrFAP65AH1, CrFAP65AH2 and their respective variants (Fig. 8a–g). Using appropriate controls (Fig. 8a–c), it was observed that CrFAP65AH1 bound strongly with FAP174, while CrFAP65AH2 bound weakly (Fig. 8d, f). As expected, the proline variants did not exhibit any binding with FAP174 (Fig. 8e, g). In order to further confirm the bioactivity of these recombinant protein, dot blot was also carried out wherein increasing concentration of FAP174 on the blot was overlayed with CrFAP65AH1, CrFAP65AH2 and their respective variants. The results obtained were like the pull-down assay wherein the CrFAP65AH1 strongly bound to FAP174 as compared to CrFAP65-AH2. However, no binding was seen in the variants as expected. These results indicate that the procedure used to purify CrFAP65 amphipathic helices and their variants yields functional recombinant proteins.

Conclusion

The current study enumerates a step-by-step design of experiments for the solubilization of IB proteins (probably of the non-classical type) for two amphipathic helices harboured on an AKAP (FAP65) that were surprisingly insoluble even after fusion with GST, a commonly used solubility-enhancer tag while producing recombinant proteins. We have successfully expressed the GST-tagged fusion proteins and using mild denaturing conditions have solubilized and purified the amphipathic helices in CrFAP65. Although the tag is supposed to aid in solubilization, the proteins form IBs and it is the optimized lysis and solubilization buffer and mechanical breakage of cells that assist in the solubilization and further purification of the proteins. Amphipathic helices are a very important class of alpha helices present in membrane proteins, are useful antimicrobial and anticancer peptides, have an anti-inflammatory effect, are inhibitors of DNA and RNA viruses, are useful lipid droplet coaters, etc. Hence, its medical and pharmaceutical use relates to producing highly pure and bioactive molecules in this category. Given the amphipathic nature, its production can become very challenging. We, therefore, offer a mild denaturing procedure (Fig. 9) that not only helps bypass the re-folding step but also produces high-yield (~ 20 mg/L of E. coli culture) functional fusion proteins. If the GST tag is to be removed, a protease cleavage site may also be introduced if not present in the vector.

Availability of data and materials

All the data generated or analyzed during this study are currently not available because a relevant unsubmitted manuscript is in progress.

References

Hutchison CA, Phillips S, Edgell MH, Gillam S, Jahnke P, Smith M. Mutagenesis at a specific position in a DNA sequence. J Biol Chem. 1978;253(18):6551–60.
Article CAS Google Scholar
Demain AL, Vaishnav P. Production of recombinant proteins by microbes and higher organisms. Biotechnol Adv. 2009;27(3):297–306.
Article CAS Google Scholar
De Marco A, Ferrer-Miralles N, Garcia-Fruitós E, Mitraki A, Peternel S, Rinas U, Trujillo-Roldán MA, Valdez-Cruz NA, Vázquez E, Villaverde A. Bacterial inclusion bodies are industrially exploitable amyloids. FEMS Microbiol Rev. 2019;43(1):53–72.
Article Google Scholar
Ferrer-Miralles N, Saccardo P, Corchero JL, Xu Z, García-Fruitós E. General introduction: recombinant protein production and purification of insoluble proteins. Methods Mol Biol. 2015;1258:1–24.
Article CAS Google Scholar
Terpe K. Overview of bacterial expression systems for heterologous protein production: from molecular and biochemical fundamentals to commercial systems. Appl Microbiol Biotechnol. 2006;72(2):211–22.
Article CAS Google Scholar
Makrides SC. Strategies for achieving high-level expression of genes in Escherichia coli. Microbiol Rev. 1996;60(3):512–38.
Article CAS Google Scholar
Nilvebrant J, Hober S. The albumin-binding domain as a scaffold for protein engineering. Comput Struct Biotechnol J. 2013;6(7): e201303009.
Article Google Scholar
Nikaido H. Maltose transport system of Escherichia coli: an ABC-type transporter. FEBS Lett. 1994;346(1):55–8.
Article CAS Google Scholar
Kishi A, Nakamura T, Nishio Y, Maegawa H, Kashiwagi A. Sumoylation of Pdx1 is associated with its nuclear localization and insulin gene activation. Am J Physiol Endocrinol Metab. 2003;284:E830–40.
Article CAS Google Scholar
Costa SJ, Coelho E, Franco L, Almeida A, Castro A, Domingues L. The Fh8 tag: a fusion partner for simple and cost-effective protein purification in Escherichia coli. Protein Expr Purif. 2013;92(2):163–70.
Article CAS Google Scholar
Kimple ME, Brill AL, Pasker RL. Overview of affinity tags for protein purification. Curr Protoc Protein Sci. 2013;73(1):9–9.
Article Google Scholar
Lilius G, Persson M, Bulow L, Mosbach K. Metal affinity precipitation of proteins carrying genetically attached polyhistidine affinity tails. Eur J Biochem. 1991;198(2):499–504.
Article CAS Google Scholar
Harper S, Speicher DW. Purification of proteins fused to glutathione S-transferase. In Protein chromatography. Humana Press; 2011, 259–280.
Schrödel A, de Marco A. Characterization of the aggregates formed during recombinant protein expression in bacteria. BMC Biochem. 2005;6(1):1–11.
Article Google Scholar
Fang M, Wang W, Wang Y, Ru B. Bacterial expression and purification of biologically active human TFF3. Peptides. 2004;25(5):785–92.
Article CAS Google Scholar
Mitchell DA, Marshall TK, Deschenes RJ. Vectors for the inducible overexpression of glutathione S-transferase fusion proteins in yeast. Yeast. 1993;9(7):715–22.
Article CAS Google Scholar
Rodal AA, Duncan M, Drubin D. Purification of glutathione S-transferase fusion proteins from yeast. Methods Enzymol. 2002;351:168–72.
Article CAS Google Scholar
Zhang N, Qiao Z, Liang Z, Mei B, Xu Z, Song R. Zea mays Taxilin protein negatively regulates opaque-2 transcriptional activity by causing a change in its sub-cellular distribution. PLoS ONE. 2012;7(8): e43822.
Article CAS Google Scholar
Song CP, Galbraith DW. AtSAP18, an orthologue of human SAP18, is involved in the regulation of salt stress and mediates transcriptional repression in Arabidopsis. Plant Mol Biol. 2006;60(2):241–57.
Article CAS Google Scholar
Thomson RB, Wang T, Thomson BR, Tarrats L, Girardi A, Mentone S, Soleimani M, Kocher O, Aronson PS. Role of PDZK1 in membrane expression of renal brush border ion exchangers. Proc Natl Acad Sci. 2005;102(37):13331–6.
Article CAS Google Scholar
Rudert F, Visser E, Gradl G, Grandison P, Shemshedini L, Wang Y, Grierson A, Watson J. pLEF, a novel vector for expression of glutathione S-transferase fusion proteins in mammalian cells. Gene. 1996;169(2):281–2.
Article CAS Google Scholar
Wingfield PT. Overview of the purification of recombinant proteins. Curr Protoc Protein Sci. 2015;80(1):6–1.
Article Google Scholar
Baneyx F, Mujacic M. Recombinant protein folding and misfolding in Escherichia coli. Nat Biotechnol. 2004;22(11):1399–408.
Article CAS Google Scholar
Lindwall G, Chau MF, Gardner SR, Kohlstaedt LA. A sparse matrix approach to the solubilization of overexpressed proteins. Protein Eng. 2000;13(1):67–71.
Article CAS Google Scholar
Von Hippel PH, Schleich T. Ion effects on the solution structure of biological macromolecules. Acc Chem Res. 1969;2(9):257–65.
Article Google Scholar
Von Hippel PH, Wong KY. On the conformational stability of globular proteins: the effects of various electrolytes and nonelectrolytes on the thermal ribonuclease transition. J Biol Chem. 1965;240(10):3909–23.
Article Google Scholar
Arakawa T, Timasheff SN. Mechanism of protein salting in and salting out by divalent cation salts: balance between hydration and salt binding. Biochemistry. 1984;23(25):5912–23.
Article CAS Google Scholar
Bowden GA, Paredes AM, Georgiou G. Structure and morphology of protein inclusion bodies in Escherichia coli. Biotechnology. 1991;9(8):725–30.
CAS Google Scholar
Asakura T, Adachi K, Schwartz E. Stabilizing effect of various organic solvents on protein. J Biol Chem. 1978;253(18):6423–5.
Article CAS Google Scholar
Herberg FW, Maleszka A, Eide T, Vossebein L, Tasken K. Analysis of A-kinase anchoring protein (AKAP) interaction with protein kinase A (PKA) regulatory subunits: PKA isoform specificity in AKAP binding. J Mol Biol. 2000;298(2):329–39.
Article CAS Google Scholar
Rao VG, Sarafdar RB, Chowdhury TS, Sivadas P, Yang P, Dongre PM, D’Souza JS. Myc-binding protein orthologue interacts with AKAP240 in the central pair apparatus of the Chlamydomonas flagella. BMC Cell Biol. 2016;17(1):1–10.
Article Google Scholar
Zhao L, Hou Y, McNeill NA, Witman GB. The unity and diversity of the ciliary central apparatus. Philos Trans R Soc B. 2020;375(1792):20190164.
Article CAS Google Scholar
Gautier R, Douguet D, Antonny B, Drin G. HELIQUEST: a web server to screen sequences with specific α-helical properties. Bioinformatics. 2008;24(18):2101–2.
Article CAS Google Scholar
Gasteiger E, Hoogland C, Gattiker A, Wilkins MR, Appel RD, Bairoch A. Protein identification and analysis tools on the ExPASy server. The proteomics protocols handbook; 2005, 571–607.
Kim SY, Hakoshima T. GST pull-down assay to measure complex formations. Methods Mol Biol. 2019;1893:273–80.
Article CAS Google Scholar
Sobhnamayan F, Adl A, Shojaee NS, Zarei Z, Emkani A. Physical and chemical properties of CEM cement mixed with propylene glycol. Iran Endodontic J. 2017;12(4):474–80.
CAS Google Scholar
Kyte J, Doolittle RF. A simple method for displaying the hydropathic character of a protein. J Mol Biol. 1982;157(1):105–32.
Article CAS Google Scholar
Drin G, Antonny B. Amphipathic helices and membrane curvature. FEBS Lett. 2010;584(9):1840–7.
Article CAS Google Scholar
Čopič A, Antoine-Bally S, Giménez-Andrés M, Garay CLT, Antonny B, Manni MM, Pagnotta S, Guihot J, Jackson CL. A giant amphipathic helix from a perilipin that is adapted for coating lipid droplets. Nat Commun. 2018;9(1):1–16.
Article Google Scholar
Shih YL, Huang KF, Lai HM, Liao JH, Lee CS, Chang CM, Mak HM, Hsieh CW, Lin CC. The N-terminal amphipathic helix of the topological specificity factor MinE is associated with shaping membrane curvature. PLoS ONE. 2011;6(6): e21425.
Article CAS Google Scholar
Zhukovsky MA, Filograna A, Luini A, Corda D, Valente C. Protein amphipathic helix insertion: a mechanism to induce membrane fission. Front Cell Dev Biol. 2019;7:291.
Article Google Scholar
Giménez-Andrés M, Čopič A, Antonny B. The many faces of amphipathic helices. Biomolecules. 2018;8(3):45.
Article Google Scholar
Jones MK, Anantharamaiah GM, Segrest JP. Computer programs to identify and classify amphipathic alpha helical domains. J Lipid Res. 1992;33(2):287–96.
Article CAS Google Scholar
Segrest JP, De Loof H, Dohlman JG, Brouillette CG, Anantharamaiah GM. Amphipathic helix motif: classes and properties. Proteins. 1990;8(2):103–17.
Article CAS Google Scholar
Le Bon C, Marconnet A, Masscheleyn S, Popot JL, Zoonens M. Folding and stabilizing membrane proteins in amphipol A8–35. Methods. 2018;147:95–105.
Article Google Scholar
Dettmer U, Ramalingam N, von Saucken VE, Kim TE, Newman AJ, Terry-Kantor E, Nuber S, Ericsson M, Fanning S, Bartels T, Lindquist S, Levy OA, Selkoe D. Loss of native α-synuclein multimerization by strategically mutating its amphipathic helix causes abnormal vesicle interactions in neuronal cells. Hum Mol Genet. 2017;26(18):3466–81.
Article CAS Google Scholar
Wang X, Zhou B, Hu W, Zhao Q, Lin Z. Formation of active inclusion bodies induced by hydrophobic self-assembling peptide GFIL8. Microb Cell Fact. 2015;14(1):1–8.
Article Google Scholar
Hung YF, Valdau O, Schünke S, Stern O, Koenig BW, Willbold D, Hoffmann S. Recombinant production of the amino terminal cytoplasmic region of dengue virus non-structural protein 4A for structural studies. PLoS ONE. 2014;9(1): e86482.
Article Google Scholar
Guo J, Gong T, Gao XD. Identification of an amphipathic helix important for the formation of ectopic septin spirals and axial budding in yeast axial landmark protein Bud3p. PLoS ONE. 2011;6(3): e16744.
Article CAS Google Scholar
Herlo R, Lund VK, Lycas MD, Jansen AM, Khelashvili G, Andersen RC, Bhatia V, Pedersen TS, Albornoz PBC, Johner N, Ammendrup-Johnsen I, Christensen NR, Erlendsson S, Stoklund M, Larsen JB, Weinstein H, Kjærulff O, Stamou D, Gether U, Madsen KL. An amphipathic helix directs cellular membrane curvature sensing and function of the BAR domain protein PICK1. Cell Rep. 2018;23(7):2056–69.
Article CAS Google Scholar
Singh A, Upadhyay V, Upadhyay AK, Singh SM, Panda AK. Protein recovery from inclusion bodies of Escherichia coli using mild solubilization process. Microb Cell Fact. 2015;14(1):1–10.
Article CAS Google Scholar
Kudou M, Ejima D, Sato H, Yumioka R, Arakawa T, Tsumoto K. Refolding single-chain antibody (scFv) using lauroyl-l-glutamate as a solubilization detergent and arginine as a refolding additive. Protein Expr Purif. 2011;77(1):68–74.
Article CAS Google Scholar
Singh SM, Sharma A, Upadhyay AK, Singh A, Garg LC, Panda AK. Solubilization of inclusion body proteins using n-propanol and its refolding into bioactive form. Protein Expr Purif. 2012;81(1):75–82.
Article CAS Google Scholar
Villar-Piqué A, Ventura S. Protein aggregation propensity is a crucial determinant of intracellular inclusion formation and quality control degradation. Biochem Biophys Acta. 2013;1833(12):2714–24.
Article Google Scholar
Antil M, Gouin SG, Gupta V. Truncation of C-terminal intrinsically disordered region of mycobacterial Rv1915 facilitates production of “difficult-to-purify” recombinant drug target. Front Bioeng Biotechnol. 2020;8:522.
Article Google Scholar
Ban B, Sharma M, Shetty J. Optimization of methods for the production and refolding of biologically active disulfide bond-rich antibody fragments in microbial hosts. Antibodies. 2020;9(3):39.
Article CAS Google Scholar
Greenfield EA, DeCaprio J, Brahmandam M. Sarkosyl preparation of antigens from bacterial inclusion bodies. Cold Spring Harb Protoc. 2020;2020(9): 100032.
Article Google Scholar
Nekoufar S, Fazeli A, Fazeli MR. Solubilization of human interferon β-1b inclusion body proteins by organic solvents. Adv Pharm Bull. 2020;10(2):233–8.
Article CAS Google Scholar
Yao G, Huang C, Ji F, Ren J, Luo X, Zang B, Jia L. Nanobodies as solubilization chaperones for the expression and purification of inclusion-body prone proteins. Chem Commun. 2022;58(17):2898–901.
Article CAS Google Scholar
Lu SC, Lin SC. Recovery of active N-acetyl-d-glucosamine 2-epimerase from inclusion bodies by solubilization with non-denaturing buffers. Enzyme Microb Technol. 2012;50(1):65–70.
Article CAS Google Scholar
Peternel Š, Bele M, Gaberc-Porekar V, Menart V. Nonclassical inclusion bodies in Escherichia coli. Microb Cell Fact. 2006;5(1):1–2.
Google Scholar
Arakawa T, Bhat R, Timasheff SN. Preferential interactions determine protein solubility in three-component solutions: the magnesium chloride system. Biochemistry. 1990;29(7):1914–23.
Article CAS Google Scholar
Dyson MR, Shadbolt SP, Vincent KJ, Perera RL, McCafferty J. Production of soluble mammalian proteins in Escherichia coli: identification of protein features that correlate with successful expression. BMC Biotechnol. 2004;4(1):1–18.
Article Google Scholar
Hammarström M, Hellgren N, van Den Berg S, Berglund H, Härd T. Rapid screening for improved solubility of small human proteins produced as fusion proteins in Escherichia coli. Protein Sci. 2002;11(2):313–21.
Article Google Scholar
Hammarström M, Woestenenk EA, Hellgren N, Härd T, Berglund H. Effect of N-terminal solubility enhancing fusion proteins on yield of purified target protein. J Struct Funct Genomics. 2006;7(1):1–14.
Article Google Scholar
Lin WH, Hung CH, Hsu CI, Lin JY. Dimerization of the N-terminal amphipathic α-helix domain of the fungal immunomodulatory protein from Ganoderma tsugae (Fip-gts) defined by a yeast two-hybrid system and site-directed mutagenesis. J Biol Chem. 1997;272(32):20044–8.
Article CAS Google Scholar
Bigay J, Casella JF, Drin G, Mesmin B, Antonny B. ArfGAP1 responds to membrane curvature through the folding of a lipid packing sensor motif. EMBO J. 2005;24(13):2244–53.
Article CAS Google Scholar
Kohl T, Schmidt C, Wiemann S, Poustka A, Korf U. Automated production of recombinant human proteins as resource for proteome research. Proteome Sci. 2008;6(1):1–10.
Article Google Scholar
Zhou NE, Kay CM, Sykes BD, Hodges RS. A single-stranded amphipathic alpha-helix in aqueous solution: design, structural characterization, and its application for determining alpha-helical propensities of amino acids. Biochemistry. 1993;32(24):6190–7.
Article CAS Google Scholar
Harroun TA, Bradshaw JP, Balali-Mood K, Katsaras J. A structural study of the myristoylated N-terminus of ARF1. Biochem Biophys Acta. 2005;1668(1):138–44.
Article CAS Google Scholar
Meyers NL, Wang L, Gursky O, Small DM. Changes in helical content or net charge of apolipoprotein CI alter its affinity for lipid/water interfaces [S]. J Lipid Res. 2013;54(7):1927–38.
Article CAS Google Scholar
Woolfson DN, Williams DH. The influence of proline residues on α-helical structure. FEBS Lett. 1990;277(1–2):185–8.
Article CAS Google Scholar
Li SC, Goto NK, Williams KA, Deber CM. Alpha-helical, but not beta-sheet, propensity of proline is determined by peptide environment. Proc Natl Acad Sci. 1996;93(13):6676–81.
Article CAS Google Scholar

Download references

Acknowledgements

The authors express their gratitude for the funding provided by the Department of Atomic Energy. The authors would like to acknowledge Tata Memorial Centre Advanced Centre for Treatment, Research and Education in Cancer (ACTREC) for providing the Circular dichroism instrument facility, Tata Institute of fundamental Research (TIFR) for providing Mass Spectrometry service and Icon Labs Pvt. Ltd. for providing use of Transmission Electron Microscope.

Funding

AAS and JSD thank the Department of Atomic Energy, Govt. of India for the funding of this work and the stipend provided to AAS.

Author information

Authors and Affiliations

School of Biological Sciences, UM-DAE Centre for Excellence in Basic Sciences, Kalina campus, Santacruz (E), Mumbai, 400098, India
Amruta A. Shendge & Jacinta S. D’Souza

Authors

Amruta A. Shendge
View author publications
You can also search for this author in PubMed Google Scholar
Jacinta S. D’Souza
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

AAS carried out all the experiments and helped in compiling the manuscript and figures. JSD conceived the idea, attracted funds, wrote the manuscript, and finalized the figures. All authors agree to be accountable for the content of the work. Both authors read and approved the final manuscript.

Corresponding author

Correspondence to Jacinta S. D’Souza.

Ethics declarations

Ethics approval and consent to participate

No animal or tissue culture work was carried out in the study. However, local Institutional Biosafety Committee (IBSC-CEBS) approval has been sought for the genetic engineering work.

Consent for publication

Not applicable.

Competing interests

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Figure S1.

Plasmid vector map of pGEX-4T-1 and the position where the inserts were cloned.

Additional file 2: Figure S2.

Transmission electron microscopy images of E. coli BL21 DE3 (a) Induced CrFAP65AH1, (b) Induced CrFAP65AH2, (c) Induced CrFAP65AH1V12P and (d) Induced CrFAP65AH2V12P.

Additional file 3: Figure S3.

MALDI-TOF spectra for the affinity-purified and dialyzed recombinant proteins (a) CrFAP65AH1, (b) CrFAP65AH1V12P, (c) CrFAP65AH2 and (d) CrFAP65AH2V12P.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article

Shendge, A.A., D’Souza, J.S. Strategic optimization of conditions for the solubilization of GST-tagged amphipathic helix-containing ciliary proteins overexpressed as inclusion bodies in E. coli. Microb Cell Fact 21, 258 (2022). https://doi.org/10.1186/s12934-022-01979-y

Download citation

Received: 16 September 2022
Accepted: 24 November 2022
Published: 12 December 2022
DOI: https://doi.org/10.1186/s12934-022-01979-y

Strategic optimization of conditions for the solubilization of GST-tagged amphipathic helix-containing ciliary proteins overexpressed as inclusion bodies in E. coli

Abstract

Introduction

Materials and methods

Bacterial strains, plasmids, and transformation of E. coli cells

Protein sequence analysis

Cell growth and induction of protein expression

Observation of inclusion bodies in E. coli using transmission electron microscope (TEM)

Buffers used for the various steps in the solubility of IBs

Formation of inclusion bodies and solubility testing

Solubilization strategy

Purification of the GST-fusion polypeptides

Circular dichroism of proteins

Pull-down assay

Dot blot overlay

MALDI-TOF mass spectrometry

ImageJ analysis

Statistical analysis (ANOVA and Tuckey)

Results

In silico insight into the amphipathic helices of CrFAP65

Induction and inclusion body formation of the CrFAP65 clones

Effect of increasing the ionic strength and diluting the cell lysate in the presence and absence of BugBuster® without sonication

Effect of sonication in the presence and absence of BugBuster® and another variable, IGEPAL CA-630

Quality checks, secondary structures of the purified proteins and bioactivity

Statistical analysis

Discussion

Formation of inclusion bodies of GST-tagged amphipathic helices and their proline variants and initial steps of solubilization

Comparison of the final yields of the purified fusion proteins from all treatments for an accurate understanding of the optimum solubilization condition

Quality checks and structural analysis of fusion proteins by Circular Dichroism

Conclusion

Availability of data and materials

References

Acknowledgements

Funding

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Ethics approval and consent to participate

Consent for publication

Competing interests

Additional information

Publisher's Note

Supplementary Information

Additional file 1: Figure S1.

Additional file 2: Figure S2.

Additional file 3: Figure S3.

Rights and permissions

About this article

Cite this article

Share this article

Keywords

Microbial Cell Factories

Contact us

Effect of increasing the ionic strength and diluting the cell lysate in the presence and absence of BugBuster^® without sonication

Effect of sonication in the presence and absence of BugBuster^® and another variable, IGEPAL CA-630