Accepted Manuscript High-yield soluble expression, purification and characterization of human steroidogenic acute regulatory protein (StAR) fused to a cleavable Maltose-Binding Protein (MBP) Dr. Nikolai N. Sluchanko, Kristina V. Tugaeva, Yaroslav V. Faletrov, Dmitrii I. Levitsky PII:
S1046-5928(15)30094-2
DOI:
10.1016/j.pep.2015.11.002
Reference:
YPREP 4819
To appear in:
Protein Expression and Purification
Received Date: 10 September 2015 Revised Date:
21 October 2015
Accepted Date: 4 November 2015
Please cite this article as: N.N. Sluchanko, K.V. Tugaeva, Y.V. Faletrov, D.I. Levitsky, High-yield soluble expression, purification and characterization of human steroidogenic acute regulatory protein (StAR) fused to a cleavable Maltose-Binding Protein (MBP), Protein Expression and Purification (2015), doi: 10.1016/j.pep.2015.11.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
3C
20
25 18 StARD1
15 10
StARD1
+ NBD -cholesterol
5
0
AC C
EP
TE D
C
Graphical abstract
SC
MBP
25
kDa 68 45 35
M AN U
His6
Fluorescence
N
RI PT
ACCEPTED MANUSCRIPT
0.0
0.2
0.4
0.6
0.8
Concentration of ligand
1.0
1.2
ACCEPTED MANUSCRIPT
High-yield soluble expression, purification and characterization of human steroidogenic acute regulatory protein (StAR) fused to a cleavable Maltose-Binding Protein (MBP)
Nikolai N. Sluchanko1,*, Kristina V. Tugaeva1, Yaroslav V. Faletrov2, Dmitrii I. Levitsky1,3
1
SC
RI PT
A.N. Bach Institute of Biochemistry, Research Center of Biotechnology of the Russian Academy of Sciences, Moscow, Russia 2 Research Institute for Physical Chemical Problems, Belarusian State University, Minsk, Belarus 3 A.N. Belozersky Institute of Physico-Chemical Biology, Moscow State University, Moscow, Russia
Dr. Nikolai N. Sluchanko A.N.Bach Institute of Biochemistry, Research Center of Biotechnology of the Russian Academy of Sciences Moscow 119071 Russian Federation Tel/Fax 7-495-9521384 E-mail:
[email protected]
M AN U
*Corresponding author:
AC C
EP
TE D
Abbreviations used: 20NP, 20-((7-nitrobenzoxadiazole-4-yl-)amino)-pregn-5-en-3β-ol; CD, circular dichroism; EDTA, ethylenediaminetetraacetic acid; FRET, fluorescence resonance energy transfer; HPLC, high performance liquid chromatography; IMAC, immobilized metal affinity chromatography; IPTG, isopropyl-β-thiogalactoside; MBP, Maltose-Binding Protein; ME, β-mercaptoethanol; PMSF, phenylmethanesulfonyl fluoride; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel-electrophoresis; StAR, Steroidogenic acute regulatory protein (=StARD1).
1
ACCEPTED MANUSCRIPT Highlights StAR protein has limited solubility and is mostly obtained through denaturation/refolding This is time-consuming, costly and may affect protein characteristics Fusing cleavable MBP to StAR confers its full solubility and facilitates purification The developed protocol results in native and aggregates-free StAR Capable of binding cholesterol, it is suitable for future investigations and biotechnology
AC C
EP
TE D
M AN U
SC
RI PT
• • • • •
2
ACCEPTED MANUSCRIPT
Abstract
TE D
M AN U
SC
RI PT
Steroidogenic acute regulatory protein (StAR) is responsible for the rapid delivery of cholesterol to mitochondria where the lipid serves as a source for steroid hormones biosynthesis in adrenals and gonads. Despite many successful investigations, current understanding of the mechanism of StAR action is far from being completely clear. StAR was mostly obtained using denaturation/renaturation or in minor quantities in a soluble form at decreased temperatures that, presumably, limited the possibilities for its consequent detailed exploration. In our hands, existing StAR expression constructs could be bacterially expressed almost exclusively as insoluble forms, even upon decreased expression temperatures and in specific strains of Escherichia coli, and isolated protein tended to aggregate and was difficult to handle. To maximize the yield of soluble protein, optimized StAR sequence encompassing functional domain STARD1 (residues 66-285) was fused to the C-terminus of His-tagged Maltose-Binding Protein (MBP) with the possibility to cleave off the whole tag by 3C protease. The developed protocol of expression and purification comprising of a combination of subtractive immobilized metal affinity chromatography (IMAC) and size-exclusion chromatography allowed us to obtain up to 25 mg/1 L culture of completely soluble StAR protein, which was (i) homogenous according to SDS-PAGE, (ii) gave a single symmetrical peak on a gel-filtration, (iii) showed the characteristic CD spectrum and (iv) pH-dependent ability to bind a fluorescently-labeled cholesterol analogue. We conclude that our strategy provides fully soluble and native StAR protein which in future could be efficiently used for biotechnology and drug discovery aimed at modulation of steroids production.
AC C
EP
Keywords: steroidogenic acute regulatory protein; soluble expression; maltose-binding protein; fluorescence; steroid hormones; cholesterol
3
ACCEPTED MANUSCRIPT
Introduction
AC C
EP
TE D
M AN U
SC
RI PT
Steroidogenic acute regulatory protein (StAR) [1] is responsible for the rapid delivery of cholesterol from cellular pools to mitochondria where the lipid serves as a source for steroid hormones biosynthesis in adrenals and gonads [1-4]. Its action is manifested upon activation of cAMP signaling, and StAR is believed to be responsible for ~6-fold increase in steroidogenesis during the acute phase [5]. The importance of StAR is underlined by the fact that mutations in Star gene result in potentially lethal disorder, lipoid congenital adrenal hyperplasia (LCAH), associated with almost complete inability to synthesize steroid hormones and accumulation of high levels of cholesterol in steroidogenic cells [2]. StAR is synthesized as a 37 kDa precursor protein of 285 residues long having typical Nterminal leader sequence which determines its targeting to mitochondria while being nonessential for the activity of the protein [6]. StAR has a remarkably short half-life [7, 8] and is regulated by changing its expression and degradation levels and by phosphorylation at specific sites that was reported to increase protein activity [9]. Human StAR1 belongs to the so-called StAR-related lipid transfer (START) domain protein family having 15 different members characterized by the presence of a relatively conservative α/β structured lipid binding domain (START domain), ~210 residues long [10, 11], forming the hydrophobic binding pocket for steroids [10]. The detailed mechanism of StAR's action during the acute phase of steroidogenesis is not completely understood, however some interesting results have been obtained. It was argued that StAR acts at the outer mitochondrial membrane [12, 13], is capable of binding and releasing cholesterol in a pH-dependent manner [5, 14] and that at reduced pH values it possesses characteristics of a molten globule [5, 14]. This feature is likely to dictate the direction of cholesterol transfer from cytosol (neutral-to slightly alkaline pH) to the outer mitochondrial membrane and intermembrane space of mitochondria (acidic pH) [5]. The results of molecular dynamics simulations and studies on StAR mutants indicated that the releasing of cholesterol occurs upon protonation and opening of the C-terminal α-helix of StAR serving as a special gatekeeper at the steroid binding pocket [14-16]. Nevertheless, existing data of literature are not sufficient to provide detailed understanding of the mechanism of StAR’s action, which clearly requires further investigations. In fact, there is only a limited number of studies devoted to physical characterization of StAR and its properties, and we believe that this is at least partially because until recently StAR protein was mainly obtained using insoluble expression and the following denaturation/renaturation scheme with prolonged gradual dialysis to remove urea [17] – the procedure which is time-consuming, costly and rather difficult to standardize. Therefore, the primary goal of the present study was to develop a protocol for obtaining large quantities of soluble, native and functional StAR protein for further studies without the necessity of denaturation and consequent refolding. We show here that fusing the StAR sequence encompassing functional domain STARD1 (residues 66-285) to the C-terminus of MaltoseBinding Protein (MBP) allows to obtain large amounts of completely soluble protein in different strains of Escherichia coli. The treatment of the fusion with highly specific 3C protease and the following immobilized metal affinity chromatography (IMAC) allows to obtain homogenous StARD1 eluted predominantly as a symmetrical peak on a size-exclusion chromatography. The isolated protein is stable for at least several weeks at +4 °C, shows characteristic circular dichroism (CD) and intrinsic fluorescence spectra and pH-dependent ability to bind fluorescently-labeled cholesterol analogue and therefore is functional and suitable for future biotechnology studies as well as drug discovery aimed at modulation of steroidogenesis. Materials and methods Materials 4
ACCEPTED MANUSCRIPT
TE D
M AN U
SC
RI PT
Steroid binding activity of StAR (=StARD1) was assessed using an 7-nitrobenzoxadiazole-4-yl- (NBD)-labeled cholesterol analogue, 20-((7-nitrobenzoxadiazole-4-yl)amino)-pregn-5-en-3β-ol (20NP), synthesized according to ([18] and Faletrov et al. (2015), submitted) via steps of reductive amination of pregnenolone and further reaction with 7nitrobenzoxadiazole-4-yl –chloride [19-21]. Two dominant fractions with absorbance maxima at 470 nm (retention time of 30.0 min and 33.5 min, respectively) were collected separately after performing the additional HPLC procedure as follows: HPLC was conducted using an LC-10AT (Shimadzu) system equipped with the SPD-M10A photodiode array detector and the LiChroCART C18 (250x4 mm, 5 µm) column (Merck). In all cases the eluent flow rate was 1 ml/min. For separation of 20NP isomers, anacetonitrile (A) and water (B) gradient was used according to the following protocol: 0-5 min, 20 % A; 5-25 min, 20-100 % A; 30-40 min, 100% A; and 40-45 min, 100-20 % A. According to the data of [21], the substances were designated to be 20S- and 20R- (cholesterol-like, 20α-H) isomers of 20NP, respectively. Only 20R-isomer was used for further studies. Luria Bertani (LB) media and isopropyl-β-thiogalactoside (IPTG) were from Amresco. All regular chemicals were of the highest purity and quality available. All water solutions in the study were prepared on the milliQ-quality water (18.3 MΩ/cm) and were filtered through the 0.22 µm Millipore filter system before use. H-MBP-3C vector [22] containing a multiple cloning site after Maltose-Binding Protein (MBP) and linker sequences and the same plasmid encoding uncleavable MBP-tagged rhinoviral 3C protease were kindly provided by York Structural Biology Laboratory (York, UK). pQE30N62 StAR plasmid was a generous gift by Professor Himangshu S. Bose (Mercer Health Sciences Center, School of Medicine, Memorial University Medical Center, Savannah, GA, USA). pNIC-Bsa4-StAR.66-274 was kindly provided by Dr. Herwig Schuler (Structural Biology, Dept. of Medical Biochemistry and Biophysics (MBB), Karolinska Institutet, Stockholm, Sweden). ArcticExpress(DE3) cells were from Agilent Technologies. Human skeletal α-tropomyosin (Tm) was kindly provided by Alexander M. Matyushenko (A.N. Bach Institute of Biochemistry of the Russian Academy of Sciences).
AC C
EP
Cloning of MBP fusion of human StAR protein To maximize solubility in bacterial system we chose widely known expression and solubility enhancer, Maltose-Binding Protein (MBP) [23], and cloned the StAR sequence (residues 66-285) into the H-MBP-3C plasmid [22] using BamHI and HindIII restriction sites to get the insert in which the C-terminus of MBP was connected to StAR through an Asn-rich linker followed by a human rhinoviral 3C protease cleavage site (i.e., LEVLFQ|GP) (see also Fig. 3A). StAR sequence (66-285) was amplified from the pQE30-N62 plasmid using a genespecific StAR-BamHI forward 5’-ATATAGGATCCACTCTCTACAGTGACCAG-3’ (corresponding restriction site is underlined) and a vector-specific pQE30 reverse 5’CATTACTGGATCTATCAACAGGAG-3’ primers by Pfu DNA-polymerase and then in-frame cloned into H-MBP-3C vector by conventional restriction/ligation protocols. The integrity and correctness of the obtained insertion were verified by DNA sequencing. 3C protease (residues 1182) was also cloned into H-MBP vector [22] but in this case there was no 3C cleavage site between MBP and 3C sequences to allow separation of the His.MBP-tagged protease from cleaved target protein by the second IMAC step. Expression of StARD1 or His.MBP-StARD1 proteins Initially StAR was expressed as simple, unfused His-tagged forms from known molecular biological constructions in either BL21(DE3) (no resistance), C41(DE3)-RIL (chloramphenicol resistance) or ArcticExpress(DE3) (gentamicin resistance) strains of E. coli. In this case, pNICBsa4-STAR.66-274 (ampicillin resistance) plasmids were used to transform corresponding cells made competent with a help of calcium treatment and plated on LB-agar with the corresponding antibiotics. In the case of pQE30-StAR-N62 (ampicillin resistance), M15(pREP4) cells 5
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
(kanamycin resistance) were used. Single colonies were picked up to grow overnight cultures at 37 °C which were then used for large-scale protein expression in 2 L flasks using 500 mL of LB media per flask with the corresponding antibiotic at 12-37 °C (in case of BL21(DE3), C41(DE3)RIL, M15(pREP4) cells) or 6–10 °C (in case of ArcticExpress(DE3) cells). IPTG was used in a 0.1–1 mM final concentration to induce protein expression at the optical density equal to 0.6–0.8 at 600 nm for 4–20 h (BL21(DE3), C41(DE3)-RIL cells) or 24–72 h (ArcticExpress(DE3)). The His.MBP-StARD1 fusion was expressed in a similar way. Either C41(DE3) (no resistance) or ArcticExpress(DE3) (gentamicin resistance) strains of E. coli were transformed by H-MBP-3C-StAR.66-285 plasmid (ampicillin resistance) obtained in the present work and then single transformant colonies were taken to grow overnight cultures. Large-scale expression was performed in 2 L flasks using 500 mL of LB media per flask with the corresponding antibiotic and was induced at the optical density equal to 0.6–0.8 at 600 nm by addition of IPTG up to a final concentration of 0.1–1 mM for periods of time and temperature indicated above. To harvest the cells with overexpressed proteins, bacterial culture was centrifuged (1h, 3,000g), the pellet was suspended in 30 mL of the lysis buffer (30 mM Tris-HCl (pH 8.0) containing 300 mM NaCl, 10 mM imidazole, 15 mM ME, 0.01 mM PMSF) and subjected to sonication and purification immediately afterwards to prevent possible problems after freezing/thawing. Protein composition was analyzed by SDS-gel electrophoresis [24].The identity of the fusion protein was additionally confirmed by excising the 69 kDa band, its in-gel trypsinolysis and further analysis by tandem mass spectrometry on a MALDI TOF/TOF ultrafleXtreme mass-spectrometer (Bruker, Germany).
AC C
EP
TE D
Purification of proteins In the beginning, StARD1 expressed as a simple His-tagged protein was purified as follows. Cell suspension in lysis buffer was sonicated on ice and then centrifuged for 1 h at 12000g and at 4 °C to separate soluble and insoluble fractions. The former, filtered through the 0.22 µm membrane, was directly loaded on a HisTrap 5 mL column (GE Healthcare) equilibrated with 30 mM Tris-HCl buffer (pH 8.0) containing 300 mM NaCl and 10 mM imidazole and after thorough washing with 10 mM and then 45 mM imidazole (optional) an eluate was collected using a shallow linear 10–510 mM gradient of imidazole on the same buffer. Insoluble fraction was resuspended in lysis buffer and sonicated again. After centrifugation for 30 min at 12000g, the pellet containing StARD1 protein was resuspended in 5 mL of lysis buffer additionally containing 8 M urea and 0.05% Triton X-100. After 6 h of intensive shaking at room temperature, the suspension was centrifuged for 1 h at 12000g to remove mechanical impurities and then yellowish but transparent supernatant (4.5 mL) was added at a rate of ~1 drop/min to 250 ml of the ice-cold 5 mM Tris buffer pH 7.5 at 4 °C at a constant stirring. After 2 h the solution was centrifuged for 20 min at 12000g and then supplemented with 1 mM ME and 0.1 mM PMSF and left for 2 days at 4 °C for refolding. The refolding solution was clarified by centrifugation and then slowly loaded on a HisTrap 5 ml column, equilibrated with 30 mM Tris-HCl buffer (pH 8.0) containing 300 mM NaCl and 10 mM imidazole, overnight. After washing with 20 mM imidazole, bound protein was eluted with a linear 20–510 mM gradient of imidazole and fractions containing StARD1 protein were concentrated on Amicon concentrators (Millipore) with a 10 kDa cut-off. The flowthrough containing major part of StARD1 protein could not bind to the resin and most probably corresponded to aggregated StAR. His.MBP-StARD1 fusion was expressed in a completely soluble form. After sonication and 1 h centrifugation at 12000g to remove cell debris, the supernatant was filtered and directly loaded on a HisTrap 5 ml column. The column was washed with 10, 20 and then 50 mM imidazole and bound protein was eluted using a linear 50–510 mM gradient of imidazole. HRV14 3C protease was obtained also as His.MBP-fusion [22] according to the identical scheme. Fractions containing a 69 kDa His.MBP-StARD1 fusion protein were combined and then StAR was released by mixing the fusion protein preparation with 3C protease (substrate/protease ratio 6
ACCEPTED MANUSCRIPT
SC
RI PT
of 500-1000/1 estimated using optical density at 280 nm) directly in a dialysis tube and dialyzed at 4 °C against 2 L of 20 mM Tris-HCl buffer (pH 7.5) containing 300 mM NaCl, 0.1 mM EDTA, 0.1 mM PMSF, 2 mM ME. Dialysate was clarified by centrifugation for 20 min at 12000g and then subjected to the second IMAC to remove His-tagged MBP and 3C protease from StARD1 preparation. The protein sample was finally polished by loading on a HiLoad 26/60 Superdex 200 column (GE Healthcare) equilibrated with a 20 mM Tris-HCl buffer (pH 7.6) containing 300 mM NaCl, 0.1 mM EDTA, 0.1 mM PMSF and 2 mM ME. Size-exclusion chromatography was performed at a flow rate of 2.6 ml min-1 and one fraction per minute was collected. Protein composition was analyzed by SDS-gel electrophoresis [24]. Fractions from the last symmetric peak with the highest StARD1 content were combined and protein sample obtained was aliquoted and stored (1) at 4 °C in the presence of 0.01% sodium azide or (2) flash frozen in liquid nitrogen and stored at -80 °C. Protein concentration was determined spectrophotometrically on a Nanophotometer P330 (Implen) using calculated molar extinction coefficients at 280 nm equal to 27080 M-1 cm-1 for StAR.66-285 and 91800 M-1 cm-1 for the His.MBP-StARD1 fusion.
TE D
M AN U
CD spectroscopy Protein sample (0.3–0.6 mg mL-1) was dialyzed overnight against 15 mM phosphate buffer (pH 7.2) containing 1.5 mM ME and then centrifuged for 10 min at 4 oC and 14200g. Protein concentration in supernatants was determined spectrophotometrically at 280 nm and then far-UV CD spectra of the samples were recorded at 20 oC. Prior to registration, samples were allowed to equilibrate at this temperature for 5–7 min. Five spectra were recorded in the range of 180–260 nm at a rate of 1 nm min-1 with 0.5 nm steps in 0.02 cm quartz cuvette on a Chirascan circular dichroism spectrometer (Applied Photophysics) equipped with temperature controller, and then the signal from buffer, filtered through 0.22 µm membrane, was subtracted. Secondary structure elements were estimated using DichroWeb server [25] by CONTIN and CDSSTR algorithms with a set 3 of reference proteins optimized for the range 185–240 nm and a mean residue weight for StARD1 equal to 110.96 Da.
AC C
EP
Intrinsic fluorescence of StARD1 Fluorescent measurements were performed on a Cary Eclipse fluorescence spectrophotometer (Varian Inc.) in either buffer F1 (20 mM Hepes-NaOH (pH 7.1), 300 mM NaCl, 0.1 mM EDTA, 2 mM ME) or buffer F2 (15 mM citrate buffer (pH 3.7), 300 mM NaCl, 0.1 mM EDTA, 2 mM ME). Intrinsic fluorescence spectra of purified StARD1 (0.5 µM) in the absence or in the presence of 1.2 µM NBD-labeled cholesterol analogue (20NP) in a total volume of 600 µl were recorded in the range of 305–400 nm upon excitation at 295 nm. The slits width was 5 nm and typically absorbance at the excitation wavelength was less than 0.1 to exclude the effects of inner filter. The spectra were corrected by subtracting a spectrum of a buffer. Cholesterol-binding activity Functionality of StARD1 obtained in our work was checked by its ability to bind the fluorescent cholesterol analogue 20-((NBD)amino)-pregn-5-en-3β-ol (or 20NP). StARD1 (1.0 µM) in either buffer F1 (pH 7.1) or in buffer F2 (pH 3.7) in a total volume of 600 µl was preincubated for 10 min at 37 °C and then mixed with an aliquot of the 20NP ethanol stock solution (final concentration – 1.2 µM) for 20 min at 37 °C. As a negative control, instead of StARD1 protein we used human skeletal α-tropomyosin which is coiled-coil hydrophilic protein with neither reported surface hydrophobicity nor lipid-binding activity. After pre-incubation, fluorescence spectra excited at 460 nm were recorded from 475 to 600 nm (slits width 5 nm). To assess apparent binding parameters, StARD1 protein (0.09–1 µM) was pre-incubated for 10 min at 37°C in a buffer F1 (pH 7.1) in a total volume of 600 µl and then titrated with increasing quantities of stock solution of 20NP in ethanol. Throughout the titration experiment, 7
ACCEPTED MANUSCRIPT
intensity of fluorescence excited at 460 nm was registered at 475 nm (slits 5 nm) after each addition of the 20NP aliquot (0.5–1 µl). After additions, the samples were mixed by microsyringe without disturbing the cuvettes and allowed to equilibrate for 10 min at 37 °C prior to the measurements. Titration curves against total ligand concentration were fitted using a quadratic equation as described in [26]. Calculations and non-linear curve fitting were performed using Origin 9.0 Pro software.
RI PT
Results and discussion
AC C
EP
TE D
M AN U
SC
Expression and purification of a His-tagged unfused StARD1 from available constructs At the beginning we tried to obtain StARD1 protein from genetic constructs available in other laboratories. Initially, StARD1 was expressed from pNIC-Bsa4 plasmid in BL21(DE3) or C41(DE3)-RIL cells or from pQE30 plasmid in M15(pREP4) cells of E. coli by traditional approach using induction with IPTG at the log phase of bacterial growth. Upon induction huge levels of expression of a ~25–27 kDa band corresponding to StARD1 could be reached from any of the pNIC-Bsa4-STAR.66-274 and pQE30-StAR-N62 plasmids, however, in our hands, the protein was observed almost exclusively in the insoluble fraction (Fig. 1A). In order to improve the solubility, we tried different IPTG concentrations (down to 0.1 mM), lowered the growth temperature down to 12 °C and varied time of expression, however these measures resulted in only a limited success and only a very small fraction (much less than 5%) of the protein, usually not visible upon Coomassie staining, was present in a soluble fraction (data not shown). This confirms the data of the pioneering works on StAR [5, 17], where even under mild conditions a major part of the protein was insoluble that required denaturation and refolding and suggested significant problems with intrinsic solubility of bacterially expressed StAR. In cases when expressed protein is not misfolded and completely aggregated and is present in both soluble and insoluble fractions, ArcticExpress strain of E. coli constitutively expressing chaperones Cpn60 and Cpn10 from psychrophilic bacterium Oleispira antarctica [27] was reported to be especially helpful and can dramatically improve target protein solubility [28]. When we transformed pNIC-Bsa4-STAR.66-274 plasmid into ArcticExpress(DE3) cells, we observed only slightly improved solubility but the majority of the protein expressed at 8 °C was still in insoluble fraction (Fig. 1A). His-tagged unfused StARD1 protein was purified from both soluble, minor fraction and also from inclusion bodies. Purification of StARD1 from soluble fraction was carried out in one step using a HisTrap column, however, in this case only very small amounts of protein could be obtained (less than 1 mg per liter of bacterial culture). Furthermore, despite ArcticExpress cells allowed to obtain more of soluble protein, it was required to separate StARD1 from the constitutively expressed strain-specific chaperone Cpn60 (Fig. 1B) tending to co-purify with His-tagged proteins [29]. In addition, even in spite of those small quantities of soluble StARD1, the protein could not survive dialysis showing significant propensity for aggregation/precipitation. Similar problems were reported previously, when it was concluded that only a very small amount of soluble StARD1 could be obtained in E. coli and only at significantly decreased growth temperatures [17]. Then we tried to obtain StARD1 from inclusion bodies by using denaturation in 8 M urea and the following refolding. This purification scheme was more successful and allowed us to obtain milligram quantities of electrophoretically homogenous StARD1, however as judged from the presence of lots of protein in the flowthrough (Fig. 2) and of flakes of precipitate in the eluate, there were doubts about correctness of its folding and this preparation was very difficult to handle because of precipitation. Design, expression and purification of His.MBP-StARD1 fusion protein 8
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
As previously authors also noticed problems with solubility of StARD1 [17, 30], first, we tried to reassess and optimize the boundaries of StAR construct used in our work. According to a multiple sequence alignment of different human START proteins (not shown), the most conservative part containing functional START domain is roughly between 66–69 and 274 residues (human StAR1 numbering). Since the first 62 N-terminal residues contain mainly a leader sequence nonessential for activity [6] and since, on contrary, the last C-terminal proteasesensitive 10 residues were reported to be important for protein functionality [6], the shortest optimal construct appeared to be between residues 66–69 and 285. On the other hand, sequencebased prediction of protein solubility of StAR using, for instance, CamSol web server [31] showed that residues 62–69 almost equally contribute to the predicted protein solubility indicating no substantial bias in selecting the N-terminal boundary. Second, we wanted to improve solubility of StARD1 by adding some fusion partner at its N-terminus while having the opportunity to cleave it off. Among all considered fusion partners, Maltose-Binding Protein (MBP) was shown to cause the most pronounced effect on target's solubility and expression levels [23, 32, 33]. Since the latter may at least partially be due to the protecting and stabilizing action of MBP tag, its utilization could be even more advantageous. Thus, to maximize solubility in bacterial expression system we cloned the StAR sequence (residues 66–285) into the H-MBP3C plasmid and obtained a vector in which N-terminal His-tagged MBP was connected to the downstream StARD1.66–285 sequence through an Asn-rich linker followed by a site for 3C protease (Fig. 3A). As a result, after digestion by 3C protease there would be four extra residues at the N-terminus of StARD1 (GPGS-) (Fig. 3A), i.e. small electroneutral addition which would unlikely affect protein properties significantly. The lack of the His-tag could be advantageous in future studies dealing with interaction of StAR with other proteins. His.MBP-StARD1 fusion protein was readily expressed in ArcticExpress(DE3) cells at high levels and in a completely soluble form, however the strain-specific Cpn60 chaperone protein (~65 kDa) was also expressed at almost the same levels (Fig. 3B). A ~69 kDa IPTGinducible protein band having apparent molecular mass similar to the expected value for His.MBP-StARD1 fusion (68.5 kDa) was excised and verified by mass spectrometry which confirmed the identity of the fusion protein by detecting peptides belonging to its different regions (Table 1). The fusion could be effectively isolated by the first IMAC step (Fig. 4A), easily cleaved by 3C protease, and the second IMAC step allowed to obtain in a flowthrough highly pure StARD1 protein (Fig. 4B), without the necessity of subsequent removal of high imidazole concentrations by dialysis. Under these conditions, His-tagged MBP and 3C protease remained bound to the resin (Fig. 4B). Although a priori ArcticExpress(DE3) cells seemed to be the most promising strain because of the presence of additional chaperone system [27] and decreased growth temperatures, which more likely would have resulted in a well-folded and soluble protein, prolonged expression, typical for this strain, sometimes led to an appearance of other bands of higher mobility (data not shown) suggesting partial degradation of StARD1 by bacterial proteases. At the same time, the presence of comparable amounts of Cpn60 chaperonin in the lysate was not a problem because it could easily be separated from the fusion during the first IMAC step (note the presence of all Cpn60 in the first flowthrough on Fig. 4A). This indirectly indicates that these proteins were not tightly bound and that the target protein did not require chaperones to be soluble. Fortunately, it turned out that even at 37 °C in C41(DE3) cells, His.MBP-StARD1 fusion protein could be expressed in a completely soluble form even in the absence of Cpn60 (Fig. 3B) and shorter expression times allowed to prevent protein degradation in cells. In other respects the results obtained with the two strains were essentially similar and either strain allowed us to obtain highly pure fusion protein already after the first IMAC step (Fig. 4A). Therefore, we conclude that MBP indeed significantly improved solubility of StARD1, probably because of the natural properties of the former, i.e., its pI, net charge, solubility, etc. [32]. Treatment of His.MBP-StARD1 fusion protein with 3C protease resulted in complete and specific proteolysis and the following IMAC step allowed to obtain almost homogenous StARD1 9
ACCEPTED MANUSCRIPT
SC
RI PT
preparation (Fig. 4B) with no visible signs of protein precipitation. At the same time we could not exclude that the preparation contained soluble aggregates and therefore performed sizeexclusion chromatography on a Superdex200 column to further purify the protein. As expected, on the elution profile we observed small peak at around void volume ("Peak 1" on Fig. 5), most likely corresponding to aggregated StARD1, and also a slightly larger and broader peak, presumably corresponding to oligomeric forms of STARD1 ("Peak 2" on Fig. 5). However, the major protein was eluted as a symmetric peak at ~92 min (fractions 85-97) and most likely corresponded to StARD1 monomer ("Peak 3" on Fig. 5). The presence of different peaks on sizeexclusion chromatography profile of soluble and already almost homogenous StARD1 underlines the tendency of this protein for oligomerization/aggregation and necessitates utilization of size-exclusion chromatography as a final step of StAR purification. To avoid concentration and dialysis stages, fractions with the highest StARD1 content (fractions 85–97) were combined and the resultant preparation of StARD1 was further characterized. The yield per 1 L of bacterial culture was about 20–25 mg of pure and soluble StARD1 that appears to be the most successful among the protocols used in our work.
AC C
EP
TE D
M AN U
Properties of StARD1 protein obtained from the MBP fusion To characterize the protein obtained from MBP fusion we used various spectroscopy methods. According to the data of far-UV CD spectroscopy, the spectrum of StARD1 had one maximum at 192 nm and two negative maxima at 208 and 222 nm with the most pronounced one at 208 nm (Fig. 6A). Accurate estimation of secondary structure elements using DichroWeb server [25] yielded the values matching the recently published ones [34], at the same time being in a good agreement with the data of X-ray crystallography for StARD1 and relative StARD6 protein (see Table 2). The overall shape of the spectrum (Fig. 6A) was also very similar to that reported recently [34, 35] and suggested correct protein conformation. At the same time, there are many inconsistencies in reported data on the CD spectra and estimation of secondary structures in StAR (e.g., compare data of [5, 14, 16, 17, 34, 36]) and this probably reflects not only differences in the methods of this estimation or errors in protein concentration measurements but also differences in protein properties, most likely arising from differences in isolation protocols and oligomeric status of the protein. It is worthwhile to mention, that the latter has rarely been checked in the previous reports. Having four tryptophan residues, upon excitation at 295 nm StARD1 demonstrated relatively bright fluorescence and its spectrum at pH 7.1 had a maximum at 337 nm and a barely visible shoulder at around 350 nm (Fig. 6B). At the same time, at pH 3.7 the spectrum of StARD1 became significantly (by ~6–7 nm) red-shifted and had noticeably decreased amplitude (Fig. 6B), together indicating transition of Trp residues to a more solvent-exposed position upon acidification. This is in line with the hypothesis that StAR displays characteristics of a molten globule and undergoes structural rearrangements in a pH-dependent manner [5, 14, 37]. It should be noted that some reported spectra for StARD1 at pH 7.0–8.5 significantly differ from one another. For example, spectra of intrinsic fluorescence of StARD1 and its mutants obtained from a pTWIN1 vector with a help of intein fusion and self-splicing, had a maximum at around 370 nm [14]. At the same time the corresponding maximum in case of the refolded StARD1 was around 340 nm [5, 34], which is rather similar to our results. This again suggests that different purification schemes may result in some differences in protein properties and underlines the necessity of standardization of StAR purification. Cholesterol binding is a prerequisite for the biological activity of StAR [36]. In order to validate this activity in case of the StARD1 obtained from our MBP fusion, we used a fluorescent cholesterol analogue, 20NP [18]. This compound possesses covalently attached to the 20th carbon of the sterol moiety [18] the 7-nitrobenz-2-oxa-1,3-diazol-4-yl (NBD) fluorescent group which is known to react on the polarity of the environment by increasing the quantum yield of fluorescence and by the blue spectral shift upon translocation to the hydrophobic/masked regions. These features could be used to monitor labeled steroid’s interaction with cognate 10
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
proteins as in the case of commercially available NBD-cholesterols, for example, 22-NBDcholesterol [14, 38]. It should be noted that the fluorescent cholesterol analogue 20NP, used in our study, is a slightly smaller homologue of commonly used 22-NBD-cholesterol, differing only by a lack of additional methylene group in the NBD-bearing steroid backbone, and therefore could also be used to probe StARD1/cholesterol interaction. Moreover, despite the presence of bulky fluorescent NBD group, the NBD-labeled cholesterols were shown to be almost equally good substrates to measure StAR binding as the radiolabeled cholesterol [39]. The fluorescence of free 20NP dissolved in buffer was characterized by a very small amplitude and the maximum of the spectrum excited at 460 nm was around 545–550 nm (Fig. 6C). At the same time pre-incubation of 20NP with StARD1 at pH 7.1 resulted in a remarkable rise of its fluorescence and an appearance of a large peak at 535 nm and therefore reflected StARD1-steroid interaction (Fig. 6C). By contrast, when pH was lowered down to 3.7, we failed to observe the corresponding increase of fluorescence of 20NP, which in this case was only slightly larger than that of free 20NP in buffer (Fig. 6C). This is in accordance with the data of [36] indicating gradual reduction in cholesterol binding upon lowering pH from 7.4 to 2.5 and the hypothesis that reduced pH causes conformational changes in StAR associated with releasing of bound steroid [5, 14, 36]. Likewise, hydrophilic coiled-coil protein α-tropomyosin, preincubated with 20NP, did not show such characteristic spectral changes and the spectrum in this case was indistinguishable from that of buffer (Fig. 6C), implying that the amplitude of the 20NP spectrum at ~535 nm and the position of its maximum can indeed be used as a good indicator of the protein-steroid interactions [38, 40]. Interestingly, incubation of StARD1 with 20NP at pH 7.1 also led to a significant decrease of the intrinsic StARD1 fluorescence (excited at 295 nm) indicating the occurrence of FRET from Trp residues of StARD1 to the bound 20NP (Fig. 6D) which indirectly confirms the protein-steroid interaction and is also in accordance with the data reported earlier [40]. Noteworthy, the possibility of close localization of the Trp241 residue in the active site of StARD1 and the NBD-group of 20NP upon corresponding protein/ligand interactions has been confirmed in silico using docking simulation (data not shown). To determine binding parameters for the StARD1/cholesterol analogue interaction, we titrated two different concentrations of StARD1 by increasing amounts of 20NP while monitoring 20NP fluorescence as a function of total concentration of 20NP added (Fig. 7). Titration of StARD1 at low fixed concentration of 90 nM allowed us to use a stoichiometric representation [41] where linear extrapolation of the two parts of the binding curve allows to assess the stoichiometry of binding (Fig. 7A). As expected from the data of literature [14, 30, 36, 39], the most likely stoichiometry of binding under conditions used was ~1:1 (Fig. 7A). Titration of StARD1 at larger fixed concentration of 0.52 µM by 20NP also allowed us to estimate the affinity of interaction with the apparent KD equal to 23 ± 5 nM (Fig. 7B). In this case the saturation was observed at total 20NP concentrations above ~0.5 µM indirectly supporting the 1:1 binding. These data indicate that StARD1 protein obtained using our strategy binds fluorescently labeled cholesterol with a high affinity which is in accordance with the results of literature obtained using different methods [36, 40]. Therefore, the protein displays properties characteristic to the native StAR, thus suggesting its correct folding and functionality. Conclusion
Despite many successful and interesting studies concerning steroid acute regulatory protein (StAR), the detailed mechanism of its functioning still is far from being completely clear. Moreover, due to the different, poorly standardized purification schemes, sometimes the reported properties of StAR may significantly differ. This seems to be further complicated because of the nature of this protein, particularly, its limited intrinsic solubility and therefore the tendency to aggregation [17, 30]. In support of this, we found extremely difficult to produce sizeable quantities of soluble StAR using available genetic constructs, with the most successful approach based on expression of the protein in insoluble form, followed by denaturation/refolding 11
ACCEPTED MANUSCRIPT
procedure. However, this procedure is costly, time-consuming and very difficult to standardize to meet, for instance, high demands of biotechnology. Therefore, our newly-developed strategy reported in this work and allowing to produce large amounts of completely soluble, aggregatesfree and native StAR with a help of its fusion to the cleavable His-tagged MBP, seems rather perspective in future biotechnology and drug discovery applications and could foster further physical and structural investigation of this biomedically important protein. Characterized properties of StAR protein obtained in our work are similar to the reported ones and support the relevance of our isolation protocol.
AC C
EP
TE D
M AN U
SC
RI PT
Acknowledgements. This investigation was supported by Russian Foundation for Basic Science (grant 14-04-00146a to N.N.S.) and the Program “Molecular and Cell Biology” of the Russian Academy of Sciences (to D.I.L. and N.N.S.). N.N.S. thanks Prof. H. Bose (Mercer Health Sciences Center, School of Medicine, Memorial University Medical Center, Savannah, GA, USA) and Dr. H. Schuler (Structural Biology, Dept. of Medical Biochemistry and Biophysics (MBB), Karolinska Institutet, Stockholm, Sweden) for the provided plasmids of StAR.
12
ACCEPTED MANUSCRIPT References
AC C
EP
TE D
M AN U
SC
RI PT
[1] B.J. Clark, J. Wells, S.R. King, D.M. Stocco, The purification, cloning, and expression of a novel luteinizing hormone-induced mitochondrial protein in MA-10 mouse Leydig tumor cells. Characterization of the steroidogenic acute regulatory protein (StAR). J Biol Chem 269 (1994) 28314-28322. [2] D.M. Stocco, StAR protein and the regulation of steroid hormone biosynthesis. Annu Rev Physiol 63 (2001) 193-213. [3] D. Lin, T. Sugawara, J.F. Strauss, 3rd, B.J. Clark, D.M. Stocco, P. Saenger, A. Rogol, W.L. Miller, Role of steroidogenic acute regulatory protein in adrenal and gonadal steroidogenesis. Science 267 (1995) 1828-1831. [4] C.B. Kallen, J.T. Billheimer, S.A. Summers, S.E. Stayrook, M. Lewis, J.F. Strauss, 3rd, Steroidogenic acute regulatory protein (StAR) is a sterol transfer protein. J Biol Chem 273 (1998) 26285-26288. [5] H.S. Bose, R.M. Whittal, M.A. Baldwin, W.L. Miller, The active form of the steroidogenic acute regulatory protein, StAR, appears to be a molten globule. Proc Natl Acad Sci U S A 96 (1999) 7250-7255. [6] F. Arakane, T. Sugawara, H. Nishino, Z. Liu, J.A. Holt, D. Pain, D.M. Stocco, W.L. Miller, J.F. Strauss, 3rd, Steroidogenic acute regulatory protein (StAR) retains activity in the absence of its mitochondrial import sequence: implications for the mechanism of StAR action. Proc Natl Acad Sci U S A 93 (1996) 13731-13736. [7] L.F. Epstein, N.R. Orme-Johnson, Regulation of steroid hormone biosynthesis. Identification of precursors of a phosphoprotein targeted to the mitochondrion in stimulated rat adrenal cortex cells. J Biol Chem 266 (1991) 19739-19745. [8] Z. Granot, E. Silverman, R. Friedlander, N. Melamed-Book, S. Eimerl, R. Timberg, K.H. Hales, D.B. Hales, D.M. Stocco, J. Orly, The life cycle of the steroidogenic acute regulatory (StAR) protein: from transcription through proteolysis. Endocr Res 28 (2002) 375-386. [9] F. Arakane, S.R. King, Y. Du, C.B. Kallen, L.P. Walsh, H. Watari, D.M. Stocco, J.F. Strauss, 3rd, Phosphorylation of steroidogenic acute regulatory protein (StAR) modulates its steroidogenic activity. J Biol Chem 272 (1997) 32656-32662. [10] F. Alpy, C. Tomasetto, Give lipids a START: the StAR-related lipid transfer (START) domain in mammals. J Cell Sci 118 (2005) 2791-2801. [11] J.F. Strauss, 3rd, T. Kishida, L.K. Christenson, T. Fujimoto, H. Hiroi, START domain proteins and the intracellular trafficking of cholesterol in steroidogenic cells. Mol Cell Endocrinol 202 (2003) 59-65. [12] F. Arakane, C.B. Kallen, H. Watari, J.A. Foster, N.B. Sepuri, D. Pain, S.E. Stayrook, M. Lewis, G.L. Gerton, J.F. Strauss, 3rd, The mechanism of action of steroidogenic acute regulatory protein (StAR). StAR acts on the outside of mitochondria to stimulate steroidogenesis. J Biol Chem 273 (1998) 16339-16345. [13] H.S. Bose, V.R. Lingappa, W.L. Miller, The steroidogenic acute regulatory protein, StAR, works only at the outer mitochondrial membrane. Endocr Res 28 (2002) 295-308. [14] B.Y. Baker, D.C. Yaworsky, W.L. Miller, A pH-dependent molten globule transition is required for activity of the steroidogenic acute regulatory protein, StAR. J Biol Chem 280 (2005) 41753-41760. [15] E. Barbar, J.G. Lehoux, P. Lavigne, Toward the NMR structure of StAR. Mol Cell Endocrinol 300 (2009) 89-93. [16] A. Roostaee, E. Barbar, P. Lavigne, J.G. LeHoux, The mechanism of specific binding of free cholesterol by the steroidogenic acute regulatory protein: evidence for a role of the Cterminal alpha-helix in the gating of the binding site. Biosci Rep 29 (2009) 89-101.
13
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[17] H.S. Bose, M.A. Baldwin, W.L. Miller, Incorrect folding of steroidogenic acute regulatory protein (StAR) in congenital lipoid adrenal hyperplasia. Biochemistry 37 (1998) 9768-9775. [18] Y.V. Faletrov, E.V. Rudaya, H.V. Hlushko, I.P. Edimecheva, V.M. Shkumatov, Revealing of novel affine interactions of NBD-labeled fluorescent steroids with bacterial steroid–converting oxidoreductases using molecular docking, FEBS J Paris, France, 2014, p. 637. [19] M. Adamczyk, J. Grote, A straightforward synthesis of 3 alpha- and 3 betaaminodigoxigenin. Steroids 61 (1996) 562-564. [20] I.F. Craig, D.P. Via, W.W. Mantulin, H.J. Pownall, A.M. Gotto, Jr., L.C. Smith, Low density lipoproteins reconstituted with steroids containing the nitrobenzoxadiazole fluorophore. J Lipid Res 22 (1981) 687-696. [21] E.O. Porta, P.B. Carvalho, M.A. Avery, B.L. Tekwani, G.R. Labadie, Click chemistry decoration of amino sterols as promising strategy to developed new leishmanicidal drugs. Steroids 79 (2014) 28-36. [22] A. Alexandrov, K. Dutta, S.M. Pascal, MBP fusion protein with a viral protease cleavage site: one-step cleavage/purification of insoluble proteins. Biotechniques 30 (2001) 1194-1198. [23] P. Sun, J.E. Tropea, D.S. Waugh, Enhancing the solubility of recombinant proteins in Escherichia coli by using hexahistidine-tagged maltose-binding protein as a fusion partner. Methods Mol Biol 705 (2011) 259-274. [24] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-685. [25] L. Whitmore, B.A. Wallace, DICHROWEB, an online server for protein secondary structure analyses from circular dichroism spectroscopic data. Nucleic Acids Res 32 (2004) W668-673. [26] M.V. Medvedeva, T.L. Bushueva, V.P. Shirinsky, T.J. Lukas, D.M. Watterson, N.B. Gusev, Interaction of smooth muscle caldesmon with calmodulin mutants. FEBS Lett 360 (1995) 89-92. [27] M. Ferrer, H. Lunsdorf, T.N. Chernikova, M. Yakimov, K.N. Timmis, P.N. Golyshin, Functional consequences of single:double ring transitions in chaperonins: life in the cold. Mol Microbiol 53 (2004) 167-182. [28] D. Hartinger, S. Heinl, H.E. Schwartz, R. Grabherr, G. Schatzmayr, D. Haltrich, W.D. Moll, Enhancement of solubility in Escherichia coli and purification of an aminotransferase from Sphingopyxis sp. MTA144 for deamination of hydrolyzed fumonisin B(1). Microb Cell Fact 9 (2010) 62. [29] L. Belval, A. Marquette, P. Mestre, M.C. Piron, G. Demangeat, D. Merdinoglu, J.F. Chich, A fast and simple method to eliminate Cpn60 from functional recombinant proteins produced by E. coli Arctic Express. Protein Expr Purif 109 (2015) 29-34. [30] Y. Tsujishita, J.H. Hurley, Structure and lipid transport mechanism of a StAR-related domain. Nat Struct Biol 7 (2000) 408-414. [31] P. Sormanni, F.A. Aprile, M. Vendruscolo, The CamSol method of rational design of protein mutants with enhanced solubility. J Mol Biol 427 (2015) 478-490. [32] S. Raran-Kurussi, D.S. Waugh, The ability to enhance the solubility of its fusion partners is an intrinsic property of maltose-binding protein but their folding is either spontaneous or chaperone-mediated. PLoS One 7 (2012) e49589. [33] R.B. Kapust, D.S. Waugh, Escherichia coli maltose-binding protein is uncommonly effective at promoting the solubility of polypeptides to which it is fused. Protein Sci 8 (1999) 1668-1674. [34] H.S. Bose, R.M. Whittal, M. Bose, D. Debnath, Hydrophobic core of the steroidogenic acute regulatory protein for cholesterol transport. Biochemistry 48 (2009) 1198-1209.
14
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
[35] H.S. Bose, R.M. Whittal, D. Debnath, M. Bose, Steroidogenic acute regulatory protein has a more open conformation than the independently folded smaller subdomains. Biochemistry 48 (2009) 11630-11639. [36] A. Roostaee, E. Barbar, J.G. Lehoux, P. Lavigne, Cholesterol binding is a prerequisite for the activity of the steroidogenic acute regulatory protein (StAR). Biochem J 412 (2008) 553-562. [37] R.C. Tuckey, H.S. Bose, I. Czerwionka, W.L. Miller, Molten globule structure and steroidogenic activity of N-218 MLN64 in human placental mitochondria. Endocrinology 145 (2004) 1700-1707. [38] J. Reitz, K. Gehrig-Burger, J.F. Strauss, 3rd, G. Gimpl, Cholesterol interaction with the related steroidogenic acute regulatory lipid-transfer (START) domains of StAR (STARD1) and MLN64 (STARD3). FEBS J 275 (2008) 1790-1802. [39] B.Y. Baker, R.F. Epand, R.M. Epand, W.L. Miller, Cholesterol binding does not predict activity of the steroidogenic acute regulatory protein, StAR. J Biol Chem 282 (2007) 1022310232. [40] A.D. Petrescu, A.M. Gallegos, Y. Okamura, J.F. Strauss, 3rd, F. Schroeder, Steroidogenic acute regulatory protein binds cholesterol and modulates mitochondrial membrane sterol domain dynamics. J Biol Chem 276 (2001) 36970-36982. [41] S. Rajagopalan, A.M. Jaulent, M. Wells, D.B. Veprintsev, A.R. Fersht, 14-3-3 activation of DNA binding of p53 by enhancing its association into tetramers. Nucleic Acids Res 36 (2008) 5983-5991.
15
ACCEPTED MANUSCRIPT Tables Table 1. Identification of His.MBP-StARD1 fusion protein by in-gel trypsinolysis and the following tandem mass-spectrometry. Mass 1125.6 Da
Residues belong to StAR
..PQVAATGDGPDIIFWAHDR..
2470.3 Da
MBP
..AQTNSSSNNNNNNNNNNLGIEGR..
2459.6/2599.3 Da
MBP+Linker
RI PT
Sequenced peptide ..AAEAAGNLVGPR..
MBP+Linker
..VTVEHPDKLEEKFPQVAATGDGPDIIFWAHDR..
3617.9 Da
MBP
..DKPLGAVALK..
1011.6 Da
MBP
1130.5 Da
MBP
SC
..QTVDEALKDAQTNSSSNNNNNNNNNNLGIEGR.. 3459.6 Da
M AN U
..YENGKYDIK..
AC C
EP
TE D
Table 2. Estimation of secondary structure elements in the StARD1 obtained from MBP fusion by CONTIN and CDSSTR algorithms using DichroWeb server [25]. Secondary CONTINa (set3*) CDSSTRb (set3*) 3P0L PDB entry 2MOU PDB structure type (StARD1) entry (StARD6) α-helices 22.4 % 22 % 19 % 22 % β-strands 24 % 26 % 31 % 34 % Turns 22.4 % 22 % 50 % 44 % Unordered 31.3 % 30 % * - using reference set 3 of proteins optimized for the range of 185–240 nm. a - Goodness-of-fit (NRMSD) = 0.043 b - Goodness-of-fit (NRMSD) = 0.015.
16
ACCEPTED MANUSCRIPT Legends to Figures
RI PT
Fig. 1. Obtaining of His-tagged unfused StARD1 protein from soluble fraction. A. Solubility of StARD1 expressed in BL21(DE3), C41(DE3)-RIL or ArcticExpress(DE3) cells at different temperature (indicated). Samples of the total lysate (T) containing overexpressed protein was divided into soluble (S) or insoluble (P) fractions analyzed by SDS-gel electrophoresis. StARD1 position is indicated by arrows and rectangles. B. Nickel affinity chromatography of soluble fraction of StARD1 expressed in ArcticExpress(DE3) cells. "L" refers to the sample loaded on the column, "FL" stands for the unbound material (flowthrough at 10 mM imidazole). Fractions are numbered and imidazole gradient is shown as a dashed line. Positions of Cpn60 and StARD1 are indicated. Combined fractions are marked by bold font.
M AN U
SC
Fig. 2. Purification of the refolded His-tagged unfused StARD1 protein. StARD1 protein was expressed in C41(DE3)-RIL cells, refolded from inclusion bodies ("P") and then subjected to nickel affinity chromatography (see Materials and methods for details). "L" refers to the sample loaded on the column, "FL" stands for the unbound material (flowthrough at 20 mM imidazole). Fractions are numbered and imidazole gradient is shown as a dashed line. Position of StARD1 is indicated. Combined fractions are marked by bold font.
TE D
Fig. 3. Design and expression of StARD1 fusion with MBP. A. Schematic representation of the His.MBP-StARD1 fusion construct obtained in this work on the basis of the vector H-MBP-3C [22]. Mature MBP sequence (malE) devoid of a signal peptide had a hexahistidine tag immediately attached to the N-terminus and was connected to the StARD1 sequence by the indicated linker, where IEGR designates optional site for Factor Xa cleavage. The main site for 3C cleavage utilized in this study is also shown. B. Test expression of the fusion protein in ArcticExpress(DE3) or C41(DE3) cells at indicated temperature. Samples of the total lysate (T) containing overexpressed protein was divided into soluble (S) and insoluble (P) fractions analyzed by SDS-gel electrophoresis. Non-induced sample is marked by "-I". Positions of Cpn60 and His.MBP-StARD1 ("FUSION") are indicated by arrows. Molecular masses of protein markers are shown in kDa.
AC C
EP
Fig. 4. Obtaining of StARD1 protein from its fusion with MBP. A. Purification of the fusion expressed in ArcticExpress(DE3) cells by nickel affinity chromatography using a linear gradient of imidazole (dashed line). "T" refers to the total lysate, "E1" and "E2" – to the first and the second soluble extractions, "L" – to the combined "E1" plus "E2" sample loaded on the column, "fl1", "fl2" and "fl3" – stand for the material washed at 10, 20 and 50 mM of imidazole, respectively. Positions of Cpn60 and His.MBP-StARD1 ("FUSION") are indicated by arrows. Combined fractions are marked by bold font. B. Cleavage of the fusion (first lane) by 3C protease (second lane) and separation of released StARD1 protein (third lane) from His.MBP and 3C (last lane). Molecular masses of protein markers are indicated in kDa to the left. Fig. 5. Size-exclusion chromatography of StARD1 obtained from MBP fusion. Positions of aggregates (peak 1), oligomers (peak 2) and individual StARD1 (peak 3) as well as positions of StARD1 and molecular mass markers (in kDa) on the gel are indicated. Void volume approximately corresponds to fraction 42. "L" refers to the sample loaded on the column. Dashed rectangle indicates the fraction range collected. Chromatography was performed on a Superdex 200 26/60 HiLoad column (GE Healthcare) equilibrated with a 20 mM Tris-HCl buffer (pH 7.6) containing 300 mM NaCl, 0.1 mM EDTA, 0.1 mM PMSF and 2 mM ME at room temperature and a flow rate of 2.6 mL min-1. 17
ACCEPTED MANUSCRIPT
RI PT
Fig. 6. Properties of StARD1 from MBP fusion investigated by spectroscopy methods. A. Averaged far-UV circular dichroism (CD) spectrum recorded for StARD1 protein at a concentration of 0.6 mg mL-1 at pH 7.2 and 20 °C (solid line) and its reconstruction using deconvolution by CDSSTR method with a help of DichroWeb server (dashed line) (normalized rmsd = 0.015). B. Intrinsic fluorescence spectra of StARD1 (0.5 µM) recorded at pH 7.1 (solid line) and pH 3.7 (dashed line) at 37 °C. Positions of maxima are indicated. C. Fluorescence spectra of NBD-cholesterol (20NP) in a buffer or in the presence of StARD1 (1.0 µM) at pH 7.1, pH 3.7 or in the presence of α-tropomyosin (α-Tm; 2.0 µM) at 37 °C. Positions of characteristic maxima of 20NP fluorescence are indicated. D. Intrinsic fluorescence spectra of StARD1 (0.5 µM) at pH 7.1 in the absence (solid line) or in the presence (dashed line) of 20NP at 37 °C.
AC C
EP
TE D
M AN U
SC
Fig. 7. Estimation of binding parameters from titration of StARD1 protein by fluorescent cholesterol analogue 20NP. A. Stoichiometric titration of low (90 nM) concentration of StARD1 by 20NP. Dashed line corresponds to 1:1 stoichiometry of binding. B. Titration of 0.52 µM of StARD1 protein by 20NP and approximation of the binding curve against total ligand concentration with a quadratic equation [26]. Intensity of fluorescence was registered at 37 °C and pH 7.1 using a Cary Eclipse fluorescence spectrophotometer (Varian Inc.) at 535 nm upon excitation at 460 nm (slits 5 nm) (see details in Materials and methods).
18
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Figures
19
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
20
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
21
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
22
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
23
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
24
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
25
