Expression, purification, phosphorylation and characterization of recombinant human statherin

Protein Expression and Purification 69 (2010) 219–225

Contents lists available at ScienceDirect

Protein Expression and Purification journal homepage: www.elsevier.com/locate/yprep

Expression, purification, phosphorylation and characterization of recombinant human statherin Barbara Manconi a, Tiziana Cabras a, Alberto Vitali b, Chiara Fanali c, Antonella Fiorita d, Rosanna Inzitari c, Massimo Castagnola b,c,e, Irene Messana a, Maria Teresa Sanna a,* a

Dipartimento di Scienze Applicate ai Biosistemi, Sezione di Biochimica e Biologia Molecolare, Università di Cagliari, Cagliari, Italy Istituto per la Chimica del Riconoscimento Molecolare, Consiglio Nazionale delle Ricerche (C.N.R.), Roma, Italy Istituto di Biochimica e Biochimica Clinica, Facoltà di Medicina, Università Cattolica, Roma, Italy d Istituto di Clinica Otorinolaringoiatrica, Facoltà di Medicina, Università Cattolica, Roma, Italy e Istituto Scientifico Internazionale ‘‘Paolo VI”, Facoltà di Medicina, Università Cattolica, Roma, Italy b c

a r t i c l e

i n f o

Article history: Received 24 July 2009 and in revised form 29 July 2009 Available online 3 August 2009 Keywords: Intein ESI–MS Golgi-casein kinase ATR-FTIR

a b s t r a c t This work reports the successful recombinant expression of human statherin in Escherichia coli, its purification and in vitro phosphorylation. Human statherin is a 43-residue peptide, secreted by parotid and submandibular glands and phosphorylated on serine 2 and 3. The codon-optimized statherin gene was synthesized and cloned into commercial pTYB11 plasmid to allow expression of statherin as a fusion protein with intein containing a chitin-binding domain. The plasmid was transformed into E. coli strains and cultured in Luria–Bertani medium, which gave productivity of soluble statherin fusion protein of up to 47 mg per liter of cell culture, while 112 mg of fusion protein were in the form of inclusion bodies. No significant refolded target protein was obtained from inclusion bodies. The amount of r-h-statherin purified by RP–HPLC corresponded to 0.6 mg per liter of cell culture. Attenuated total reflection-Fourier transform infrared spectroscopy experiments performed on human statherin isolated from saliva and r-h-statherin assessed the correct folding of the recombinant peptide. Recombinant statherin was transformed into the diphosphorylated biologically active form by in vitro phosphorylation using the Golgienriched fraction of pig parotid gland containing the Golgi-casein kinase. Ó 2009 Elsevier Inc. All rights reserved.

Introduction Human statherin (hSTAT)1 is a 43-residue peptide, secreted by parotid and submandibular glands and phosphorylated on serine 2 and 3 in its mature form. One of the primary functions of this multifunctional molecule is connected to its great affinity for calcium phosphate minerals [1] reflecting the ability to inhibit precipitation and crystal growth of hydroxyapatite from supersaturated solutions of calcium phosphate [2]. Moreover, statherin and its C-terminal fragments inhibit the growth of anaerobic bacteria isolated from the oral cavity [3], contributing to the protection of oral mucosal and hard tissues. The physiological significance of statherin for the

* Corresponding author. Address: Dipartimento di Scienze Applicate ai Biosistemi, Università di Cagliari, Cittadella Universitaria Monserrato, 09042 Monserrato, Cagliari, Italy. Fax: +39 070 6754523. E-mail address: [email protected] (M.T. Sanna). 1 Abbreviations used: ATR-FTIR, attenuated total reflection-Fourier transform infrared; CBD, chitin-binding domain; DTT, 1,4-dithiotreithol; ESI–MS, electrospray ionization–mass spectrometry; hSTAT, human statherin; IB, inclusion body; IPTG, isopropyl-beta-D-thiogalactopyranoside; LB, Luria–Bertani; r-hSTAT, recombinant hSTAT; TFA, 2,2,2-trifluoroacetic acid; XIC, extracted ion current. 1046-5928/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2009.07.015

oral cavity health has stimulated widespread interest in its purification and in the characterization and analysis of the structural features responsible for its biological properties. However, the low abundance of hSTAT in saliva (about 5 lmol/L) was always a serious hindrance for the achievement of these aims. The length of this peptide connected to the presence of phosphorylated residues near to the N-terminus precludes an economical production via synthetic methods. Direct expression of small polypeptides in Escherichia coli is often unsuccessful, because short peptides are often subject to degradation by proteases in the host cells. This heterogeneity resulted in a difficult purification of the target peptide from their degraded fragments and a significant yield decrease [4]. Several approaches have been undertaken in order to overcome these drawbacks. One of them is based on the fusion of the target peptide to a carrier protein named ‘‘fusion partner” [5,6]. On this regard, intein-mediated self-cleavage system has been recently developed as a powerful tool for protein expression, purification, ligation, and amidation [7–12]. Target peptides can be synthesized in fusion to inteins that have been genetically engineered to achieve cleavage of the peptide bond either at N- or C-sequence termini without the use of specific protease or chemical reagents [8–13].

220

B. Manconi et al. / Protein Expression and Purification 69 (2010) 219–225

Recombinant human statherin (r-hSTAT) retaining the N-terminal methionine has been expressed by Gilbert and Stayton [14] using pCYB1 intein-mediate expression system. In this study, r-hSTAT with an N-terminus identical to that of native hSTAT has been obtained using the plasmid pTYB11, which allowed expression of the target peptide fused to the C-terminus of a modified intein from Saccharomyces cerevisiae (VMA1) containing a chitin-binding domain (CBD) affinity tag. Polynucleotide sequence used for r-hSTAT expression has been optimized according to the code of K12 strain, resulting in a noticeable enhancement of the expression yield. Furthermore, by using Golgi-casein kinase enriched-fraction of pig parotid gland we were able to phosphorylate serine residues at position 2 and 3, thus obtaining the functional mature form of the human peptide. Materials and methods Materials The modified hSTAT gene was purchased from EZBiolab (Dolan Way, IN, USA). Plasmid pTYB11 and chitin beads were purchased from New England Biolabs (Beverly, MA, USA). All enzymes, restriction endonucleases and DNA markers were from Fermentas (Ont., Canada). Plasmid purification kit and gel extraction kit were from Sigma–Aldrich (St. Louis, MO, USA). Sequencing primers were synthesized by Invitrogen (Paisley, UK). Competent BL21 (DE3) E. coli cells were from Novagen (Madison, WI, USA). Agarose and all other chemicals were of analytical grade. Construction of plasmid pTYB11–hSTAT and expression of r-hSTAT fusion protein The hSTAT sequence used for protein expression, designed on the base of the nucleotide sequence reported in GenBank (accession no. AK311812), was adapted to E. coli codon usage preference and engineered in order to insert SapI and PstI recognition sites at the 50 and 30 termini, respectively. The synthetic hSTAT gene and the expression vector pTYB11 were both cleaved with SapI and PstI restriction endonucleases. After 1% agarose gel electrophoresis purification, hSTAT fragment and pTYB11 were ligated with T4 DNA ligase to yield the expression plasmid pTYB11–hSTAT, used to transform competent BL21 (DE3) E. coli cells. Colony PCR was used to screen the transformants for the presence and correct insertion of the hSTAT insert, positive clones were selected by DNA sequencing using the intein forward primer 50 -CGCGGATTTTATTTCGAGTT-30 . Positive clones were inoculated into 5 mL of Luria–Bertani (LB) medium in the presence of 100 lg/mL ampicillin and cultured overnight at 37 °C with shaking. 0.5 mL of the culture was added to 50 mL of LB broth containing 100 lg/mL ampicillin and shacked at 37 °C at 250 rpm until the A600nm reached 0.6. Several concentrations (0.5, 0.3, 0.2 mM) of isopropyl-beta-D-thiogalactopyranoside (IPTG) at different induction temperatures (37, 30, 25 °C) and induction times (3, 5, 8 h, overnight) were tested for protein expression optimization. Cells from 1 L of culture were harvested by centrifugation at 5000g at 4 °C for 15 min and the pellet (2.8 g) was sonicated on ice (10 rounds of 15 s sonication with 30 s intervals) in 20 mL of 20 mM Tris–HCl pH 8.0, 0.5 M NaCl, 1 mM EDTA (column buffer), containing 1 tablet of a cocktail of protease inhibitors – mini Complete EDTA-free, from Roche Diagnostics. Cell debris were removed by centrifugation at 20,000g for 30 min at 4 °C and the clarified supernatant was loaded onto a column packed with 10 mL of chitin beads equilibrated with column buffer at a flow rate of 0.5 mL/min; 40 lL of the flow-through were collected for SDS–PAGE analysis. Twenty volumes of 20 mM

Tris–HCl pH 8.5, 1.2 M NaCl, 1 mM EDTA (washing buffer) were used to wash away nonspecific protein binding, then the column was quickly flushed with three volumes of 20 mM Tris–HCl pH 8.0, 0.5 M NaCl, 1 mM EDTA, 50 mM DTT (cleavage buffer) and incubated at 22 °C for 40 h. r-hSTAT was eluted using the cleavage buffer in the absence of 1,4-dithiotreithol (DTT). The eluted fractions were dialyzed (benzoylated dialysis tubing, 1.5 kDa molecular weight cut-off, Sigma–Aldrich) overnight against distilled water at room temperature, lyophilized and further purified by RP–HPLC.

Purification of r-hSTAT by RP–HPLC r-hSTAT was purified by RP–HPLC on a Beckman Gold 125S apparatus equipped with a diode array detector (Palo Alto, CA, USA) using a Hypersil BDS-C18, 100 mm  4 mm, 3 lm particle diameter (Hewlett–Packard, Palo Alto, CA, USA). The following solutions were utilized: (eluent A) 0.2% (v/v) aqueous 2,2,2-trifluoroacetic acid (TFA) and (eluent B) 0.2% (v/v) TFA in acetonitrile– water 80/20. Proteins were eluted using a linear gradient from 0 to 85% of B in 50 min, at a flow rate of 1 mL/min. Retention time was compared with that of the naturally occurring nonphosphorylated statherin in human saliva.

SDS–PAGE analysis Expression of the fusion protein, intein self-cleavage and elution from column were monitored by 15% SDS–PAGE [15]. Quantification of the bands of interest was done by computer-assisted scanning analysis [16].

Isolation of the Golgi-enriched fraction from pig salivary gland Isolation of the different subcellular fractions of pig parotid gland was essentially performed according to Morrè [17]. Parotid glands were obtained from Sus scrofa animals present in the animal house of the Catholic University, during surgery treatment. The animals were treated according to the ethical rules approved by the Ethical Committees of the Catholic University, which are in agreement with those accepted by the European Community. Parotid glands were trimmed free of fatty tissue, minced and re-suspended in cold 50 mM Tris–maleate pH 6.75, 1 mM MgCl2, 5 mM 2-mercaptoethanol, 0.5 M sucrose, 1% dextran (homogenization buffer) at 2 mL per gram. Homogenization was performed for 40–60 s at 10,000 rpm using an Ultra Turrax T25 (IKA, Staufen, Germany) and the homogenate was centrifuged at 1000g for 5 min. The precipitate (nuclei and unbroken cells) was discarded, while the supernatant (post-nuclear fraction) was diluted with an equal volume of the homogenization buffer and centrifuged at 10,000g for 20 min at 4 °C. The resulting pellet was re-suspended in homogenization buffer to give a final concentration of 0.5 g/mL, overlaid on 5 mL of 50 mM Tris–maleate pH 6.75, 1 mM MgCl2, 1.2 M sucrose and centrifuged at 100,000g for 60 min at 4 °C using a SW55Ti rotor (Beckman). The Golgi-enriched fraction was collected at the 0.5/1.2 M sucrose interface, diluted with an equal volume of 50 mM Tris–maleate pH 7.0, 1 mM MgCl2, 0.25 M sucrose (pelleting buffer) and sedimented by centrifugation at 14,000g for 15 min. Protein kinase activity was solubilized from the Golgi-enriched fraction with 0.7% Triton X-100 in the presence of 5 mM Tris–HCl pH 7.5, 1 mM EGTA, 5 mM MgCl2 and 1 mM 2mercaptoethanol. After 1 h of stirring at 4 °C, the insoluble material was discarded by centrifugation at 37,000g for 15 min, and the supernatant was used for the enzymatic phosphorylation of rhSTAT.

B. Manconi et al. / Protein Expression and Purification 69 (2010) 219–225

Phosphorylation of r-hSTAT The extract of the Golgi-enriched fraction (5 lg) was incubated with 33 lM r-hSTAT for 120 min at 37 °C in 30 lL of 50 mM MopsKOH pH 6.6, 15 mM MgCl2, 15 mM MnCl2, 0.15 M NaCl, 5 mM ATP. The enzymatic reaction was stopped by adding 0.2% TFA in 1:1 v/v ratio. Incorporation of phosphate into the substrate was evaluated by HPLC–electrospray ionization–mass spectrometry (ESI–MS) analysis. RP–HPLC–ESI–MS analysis The RP–HPLC–ESI–MS apparatus was a Surveyor HPLC system (ThermoFisher, San Jose, CA, USA) connected by a T splitter to a photodiode array detector and to a LCQ Deca XP Plus mass spectrometer (ThermoFisher). The chromatographic column was a Vydac (Hesperia, CA, USA) C8 with 5 lm particle diameter (column dimensions 150 mm  2.1 mm). The following solutions were utilized: (eluent A) 0.056% (v/v) aqueous TFA and (eluent B) 0.05% (v/v) TFA in acetonitrile–water 80/20. Proteins were eluted using a linear gradient from 0 to 55% of B for 40 min, at a flow rate of 0.30 mL/min. The T splitter permitted 0.20 mL/min to flow toward the diode array detector and 0.10 mL/ min to flow toward the ESI source. Mass spectra were collected every 3 ms in the positive ion mode. The MS spray voltage was 4.50 kV, and the capillary temperature was 220 °C. Protein quantification Bradford method was applied in order to determine lysate total protein concentration and r-hSTAT concentration; bovine serum albumin (0.1–1.2 mg/mL) was assayed to plot standard curves. Relative abundances of r-hSTAT and its phosphorylated derivatives were calculated by the eXtracted Ion Current (XIC) peak areas, revealed by searching the specific multiply charged ions of each form in the total ion current profile. The following m/z values (±0.5) were used for the search: 1306.1 [M + 4H]4+, 1741.2 [M + 3H]3+, nonphosphorylated r-hSTAT; 1325.9 [M + 4H]4+, 1767.6 [M + 3H]3+, monophosphorylated r-hSTAT; 1345.9 [M + 4H]4+, 1794.2 [M + 3H]3+, diphosphorylated r-hSTAT.

221

were used to determine the contribution of each secondary structure motif. Data analysis Deconvolution of the averaged ESI–MS spectra was automatically performed by using either the Bioworks Browser software provided with the LCQ Deca XP instrument or MagTran 1.0 software [19]. Experimental mass values were compared with average theoretical mass values available at the Swiss-Prot data bank (http://us.expasy.org/tools) with the accession no. P02808, using the PeptideMass program. Results and discussion Cloning and expression of hSTAT gene Since statherin is a small 43-amino acid peptide, we artificially synthesized the hSTAT gene according to the preference codon usage of K12 strain of E. coli to ensure efficient recombinant protein expression. The hSTAT fragment, without the Met start codon, was inserted immediately downstream of the intein splice site through SapI and PstI sites of the pTYB11 vector. The SapI site of pTYB11 places the N-terminal amino acid of the target protein immediately adjacent to the cleavage site (Asn) of the intein tag allowing its expression with no extra vector-derived residues. DNA sequencing analysis confirmed correct insertion of the hSTAT gene fragment into pTYB11 vector. The engineered strain E. coli BL21(DE3)/pTYB11–hSTAT was able to express the fusion protein (CBD–intein–r-hSTAT) in the form of both inclusion bodies (IBs) and soluble proteins in different percentages depending on the experimental conditions applied. To optimize the expression of soluble protein form, we tested different induction conditions including IPTG concentration, temperature and time. The fusion protein was exclusively in the insoluble form when the expression was performed at 37 °C regardless of IPTG concentration and time expression. The maximal amount of soluble proteins was obtained at the induction temperature of 25 °C and 0.2 mM IPTG. Under these conditions the total soluble and insoluble proteins determined by the Bradford method were

Secondary structure analysis Attenuated total reflection-Fourier transform infrared (ATRFTIR) spectra of r-hSTAT and hSTAT were recorded on a Spectrum One spectrophotometer equipped with an ATR accessory with a ZnSe reflection element (Perkin–Elmer, Waltham, MA, USA). Spectra were recorded after 25 scans at a 1 cm1 of resolution. hSTAT was purified as reported [18]. The samples were firstly cleared from salts and other contaminants using C-18 Zip-Tips (Millipore, Billerica, MA, USA) following the manufacturer indications. Before peptide analysis, an open beam background spectrum of clean crystal was recorded. During measurements performed at 25 °C, the crystal was continually flushed with nitrogen to eliminate residual water vapours. Usually 1 lL of 1 mg/mL (w/v) CH3CN/ H2O (1:1, v/v) solutions of r-hSTAT and hSTAT was employed for any measurements. Peak Fit 4.12 (Sea Solve Software, Inc., San Jose, CA, USA) program was used to obtain the second derivative from ATR-FTIR spectra. A 21% smoothing process, employing the Savitzky–Golay algorithm, was performed and the resulting peaks were used as a reference for the subsequent peak fitting analysis performed with the same program. A linear baseline was employed and Gaussian peaks were produced after an iterative adjustment of data until the SSE statistical parameter was under 1  104, indicating a good fitting analysis. The resulting peak areas of Amide I

Fig. 1. SDS–PAGE analysis: Lane 1, low molecular weight standards (BioRad); lane 2, uninduced cells; lane 3, induced cells; lane 4, soluble lysate from induced cells (arrow indicates CBD–intein–r-hSTAT); lane 5, lysate pellet from induced cells; lane 6, chitin column flow-through; lane 7, chitin column flow-through following cleavage and elution (arrow indicates r-hSTAT); lane 8, stripped affinity beads after DTT cleavage and statherin elution (arrow indicates intein remaining in the chitin column after elution of r-hSTAT).

222

B. Manconi et al. / Protein Expression and Purification 69 (2010) 219–225

Table 1 Purification of r-hSTAT from 1 L E. coli cell lysatea.

a

Purification step

Total proteins (mg)

CBD–intein–r-hSTAT purity (%)

Total cell extract Supernatant Chitin column and intein-mediated cleavage C18 RP–HPLC

224 190 1.7

23% 25%

0.6

Protein concentration was determined by Bradford method.

r-hSTAT purity (%)

70%

95%

190 and 249 mg/L, respectively. The soluble and insoluble recombinant CBD–intein–r-hSTAT accounted for 25% and 45% of the total protein, respectively, as determined by computer-assisted scanning of SDS–PAGE gel (Fig. 1), with no differences between 5 and 8 h induction time, while the overnight expression led to loss of the soluble fusion protein. In order to solubilize IBs and increase yield of recombinant peptide, we tested different denaturants, combination of denaturant and salt, and addictives [20]. We succeeded to solubilize IBs using 6 M guanidine hydrochloride-containing buffer, and to maintain soluble the fusion protein during the following dialysis steps by

Fig. 2. RP–HPLC–ESI–MS analysis of r-hSTAT. (A) Total Ion Current profile in the elution range 5.02–21.98 min; (B) average ESI–MS spectrum in the elution range 16.9– 17.6 min; (C) deconvoluted spectrum showing the experimental average molecular mass values of r-hSTAT.

223

B. Manconi et al. / Protein Expression and Purification 69 (2010) 219–225

adding L-Arg to buffers. However, the fusion protein almost completely precipitated when dialyzed against column buffer probably due to the presence of unfolded protein. Therefore, r-hSTAT was expressed under the experimental condition that ensured the highest yield of soluble protein (0.2 M IPTG, 25 °C for 5 h induction time), leaving out the IBs. Fusion protein purification After ultrasonication on ice of the re-suspended pellet of cells, the clarified lysate was loaded onto the chitin affinity column at a slow speed in order to increase the binding of fusion protein. Nonspecific proteins were washed away with twenty volumes of washing buffer. Due to the strong binding ability of chitin-binding domain, high flow rate (2 mL/min) and high salt concentration (1.0 M NaCl) washing procedure were used to reduce nonspecific binding of other E. coli proteins. No CBD–intein–r-hSTAT was detected in the flow-through (Fig. 1). Intein self-cleavage was induced by adding three volumes of cleavage buffer which led to the release of r-hSTAT from the fusion protein. SDS–PAGE analysis showed that other proteins with higher molecular weight were present besides r-hSTAT (Fig. 1), while the N-terminal cleavage product (N-extein, 1.6 kDa), part of the maltose-binding protein generated by the DTT-mediated self-spicing event, was not detected. In a previous study, it was shown the effect of the N-terminal residue of a target protein on the intein C-terminal cleavage; most of the 20 amino acid residues at the N-terminal allowed > 50% cleavage at 23 °C after 16 h. The cleavage at 4 °C was most efficient for Met, Ala and Gln, but less efficient for other residues such as Val, Ile, Asp, Glu, Lys, Arg and His [9]. After DTT cleavage and statherin elution a small aliquot of affinity beads was washed with SDS–PAGE loading buffer (stripping) and supernatant was analyzed by 15% SDS–PAGE. Comparison of the intensity of the CBD–intein–r-hSTAT band with the CBD–intein band showed that CBD–intein–r-hSTAT exhibited 90% cleavage (Fig. 1). The percentage of cleavage obtained is in good agreement with the presence of a Glu residue at the N-terminus of hSTAT. Fractions containing the r-hSTAT were dialyzed to eliminate DTT and lyophilized. Total amount of eluted proteins determined by the Bradford method was 1.7 mg per liter of cell culture (Table 1); r-hSTAT accounted for over 70% of the total, as determined by computer-assisted scanning of the 15% SDS–PAGE gel (Fig. 1). rhSTAT was further purified by preparative RP–HPLC and purity assessed by HPLC–ESI–MS (Fig. 2). The typical yield is 0.6 mg per liter of cell culture. The final amount is lower than expected, probably due to the sticky behavior of this peptide. Indeed, comparable losses occurred during purification of statherin from human saliva (unpublished results). Physico-chemical characterization of recombinant statherin HPLC–ESI–MS analysis of r-hSTAT is reported in Fig. 2. Retention time of recombinant peptide was identical to that of nonphosphorylated statherin characterized in human saliva [21,22]. Deconvolution of the ESI–MS spectrum allowed determining the average molecular mass of 5219.6 Da, consistent with the theoretical value of the human nonphosphorylated peptide and with the absence of any additional vector-derived amino-acid. Small amounts of the nonphosphorylated forms of statherin DesTF42-43 (experimental average mass: 4971.2 Da) were also detected, probably generated by proteolytic activity present in E. coli cells. It is interesting to outline that the corresponding truncated form of statherin is usually detectable in human saliva [21,22].

Fig. 3. ATR-FTIR spectroscopic analysis of r-hSTAT (grey line) compared with a sample of hSTAT (black line) purified from whole saliva. The arrow indicates the position of the signal of phosphate groups.

Secondary structure analysis In order to assess the secondary structure and the correct folding of r-hSTAT, an ATR-FTIR spectroscopy analysis was performed comparing the recombinant product with a sample of hSTAT purified from whole saliva. As depicted in Fig. 3, the spectra of the two peptides obtained in the range between 1800 and 1000 cm1 are almost identical, especially in the fingerprint signal of the Amide I centered for both the peptides at 1655 cm1. The difference between the two spectra is found in the strong signal around 1100 cm1 in hSTAT, typical of phosphates groups [23]. These data suggest that the correct folding of hSTAT is obtained, in vivo, before complete post-translational diphosphorylation. The Amide I peak value centered at 1655 cm1 is in general indicative of a prevalent a-helical structural architecture. A deeper estimation of the secondary structure of r-hSTAT and hSTAT was obtained performing a peak deconvolution of Amide I band. The resulting components (reported in Table 2) confirm a prevalence of alpha helix (signal at 1657 cm1, 25%) and random coil (signal at 1644 cm1, 22%) motifs followed by b-turn motifs (signals at 1674 cm1, 18% and 1689 cm1, 9%). This is in a good agreement with the computational prevision obtained by Goobes and coll. [24] based on NMR data of hydroxyapatite-bound statherin, where they observed both the helical and coiled structures. Phosphorylation of recombinant statherin Experiments measuring statherin’s ability to control crystallization in vitro demonstrated that the N-terminus of statherin is the most important fragment involved in binding to mineral surfaces and in inhibiting secondary precipitation of calcium phosphate salts [25]. It has been shown that the majority of secreted salivary proteins and peptides that adsorb onto enamel powder and hydroxyapatite beads are phosphorylated [26]. The active form of hSTAT is diphosphorylated at two serine residues (positions 2 and 3) that fulfill the requirements of the consensus sequence

Table 2 Amide I peak fitting analysis of hSTAT and r-hSTAT. Centroid (cm1)

Peak

Area % hSTAT

r-hSTAT

hSTAT

r-hSTAT

1 2 3 4 5 6

8.2 18.5 9 16.5 25 22.6

8.4 18.2 9.1 16.2 25.4 22.5

1614 1674 1689 1629 1657 1644

1614 1672 1683 1629 1657 1643

224

B. Manconi et al. / Protein Expression and Purification 69 (2010) 219–225

SXE/pS of the Golgi-casein kinase [27]. This kinase has been first recognized as responsible for the in vivo phosphorylation of casein in mammary glands [28] and it is denoted by the acronym G-CK. Its

activity has been also revealed in rat in many other secretory organs, notably liver, brain, and to a lesser extent, kidney and spleen [29]. The G-CK is unique among all known kinases and recognizes

Fig. 4. RP–HPLC–ESI–MS analysis of r-hSTAT incubated with Golgi-casein kinase enriched fraction for 120 min at 37 °C. (A) Total Ion Current profile in the elution range 24.67–32.73 min; (B) extracted ion current peak of nonphosphorylated r-hSTAT; (C) extracted ion current peak of monophosphorylated r-hSTAT; (D) extracted ion current peak of diphosphorylated r-hSTAT; (E) average ESI–MS spectrum collected in the elution range 28.4–29.4 min; (F) deconvoluted spectrum showing the experimental average molecular mass values of r-hSTAT and its two mono and diphosphoryated derivatives.

B. Manconi et al. / Protein Expression and Purification 69 (2010) 219–225

motifs different from the consensus of either casein kinase-1 or casein kinase-2 [30–32]. The phosphorylation process of salivary proteins has been investigated using Golgi-casein kinase enriched-fraction of sublingual [33] and parotid [34] glands on fragments of primate salivary peptides. Despite the recurrent efforts of several laboratories, the G-CK has not yet been purified to homogeneity [35]. Recently, it has been shown that several pig salivary peptides are phosphorylated at a level of serine residues [36] that fulfill the requirements of G-CK enzyme [either SXE/pS or SXQXX(D/E)3] [32]. r-hSTAT, expressed in E. coli, is obviously not phosphorylated. Thus, in order to obtain the biologically active diphosphorylated peptide, in vitro phosphorylation assays have been developed using fractions of pig parotid gland homogenates. Kinase activity was present in the Golgi-enriched fraction of pig parotid gland homogenate, as demonstrated by the ability to diphosphorylate the recombinant peptide. Fig. 4 shows the RP– HPLC–ESI–MS total ion current profile of recombinant statherin incubated with the extract of the Golgi-enriched fraction (5 lg) for 120 min at 37 °C. XIC search was used to evidence the presence of phosphorylated derivatives of r-h-STAT, and the area of the XIC peaks to evaluate the relative amount of the nonphosphorylated (60%), the monophosphorylated (27%) and the diphosphorylated forms (13%). Conclusions The intein-fused expression and purification strategy applicable for large-scale production of the r-hSTAT is here reported. This approach has clear advantages. Firstly, intein fused to r-hSTAT allowed the purification of r-hSTAT with no additional residues at the N-terminus, resulting in the properly folded peptide. Secondly, the high expression levels of the soluble peptide and simple purification procedures decreased the loss of expressed protein and resulted in high yields. The self-cleavage properties of intein avoided using chemical reagents or proteases, thus simplifying the purification procedure and resulting in increase of safety and lower cost. Finally, we showed that recombinant statherin can be transformed into the diphosphorylated biologically active form using the G-CK like enzyme activity of pig parotid gland. Acknowledgments We acknowledge the financial support of Università di Cagliari, Università Cattolica in Rome, Italian National Research Council (CNR), Regione Sardegna, International Scientific Institute ‘‘Paolo VI” (ISI), MIUR, FILAS project n. AR2/2007/0000051. References [1] P.A. Raj, M. Johnsson, M.J. Levine, G.H. Nancollas, Salivary statherin. Dependence on sequence, charge, hydrogen bonding potency, and helical conformation for adsorption to hydroxyapatite and inhibition of mineralization, J. Biol. Chem. 267 (1992) 5968–5976. [2] D.H. Schlesinger, D.I. Hay, Complete covalent structure of statherin, a tyrosinerich acidic peptide which inhibits calcium phosphate precipitation from human parotid saliva, J. Biol. Chem. 252 (1977) 1689–1695. [3] B. Kochanska, A. Kedzia, W. Kamysz, Z. Mackiewicz, G. Kupryszewski, The effect of statherin and its shortened analogues on anaerobic bacteria isolated from the oral cavity, Acta Microbiol. Pol. 49 (2000) 243–251. [4] S. Gottesman, Genetics of proteolysis in Escherichia coli, Annu. Rev. Genet. 23 (1989) 163–198. [5] J. Nilsson, S. Stahl, J. Lundeberg, M. Uhlen, P.A. Nygren, Affinity fusion strategies for detection, purification, and immobilization of recombinant proteins, Protein Expr. Purif. 11 (1997) 1–16. [6] E.R. Lavallie, J.M. McCoy, Gene fusion expression systems in Escherichia coli, Curr. Opin. Biotechnol. 6 (1995) 501–506. [7] E.J. Cantor, S. Chong, Intein-mediated rapid purification of Cre recombinase, Protein Expr. Purif. 22 (2001) 135–140.

225

[8] P.A. Nygren, S. Stahl, M. Uhlen, Engineering proteins to facilitate bioprocessing, Trends Biotechnol. 12 (1994) 184–188. [9] S. Chong, G.E. Montello, A. Zhang, E.J. Cantor, W. Liao, M. Xu, J. Benner, Utilizing the C-terminal cleavage activity of a protein splicing element to purify recombinant proteins in a single chromatographic step, Nucleic Acids Res. 26 (1998) 5109–5115. [10] I.R. Cottinghan, A. Millar, E. Emslie, A. Colman, A.E. Schnieke, C.A. McKee, Method for the amidation of recombinant peptides expressed as intein fusion proteins in Escherichia coli, Nat. Biotechnol. 19 (2001) 974–977. [11] S. Mathys, T.C. Evans, I.C. Chute, H. Wu, S. Chong, J. Benner, X. Liu, M. Xu, Characterization of a self-splicing mini-intein and its conversion into autocatalytic N- and C-terminal cleavage elements: facile production of protein building blocks for protein ligation, Gene 231 (1999) 1–13. [12] M. Xu, H. Paulus, S. Chong, Fusions to self-splicing inteins for protein purification, Methods Enzymol. 362 (2000) 376–418. [13] C. Morassutti, D.F. Amicis, B. Skerlavaj, M. Zanetti, S. Marchetti, Production of a recombinant antimicrobial peptide in transgenic plants using a modified VMA intein expression system, FEBS Lett. 519 (2002) 141–146. [14] M. Gilbert, P.S. Stayton, Expression and characterization of human salivary statherin from Escherichia coli using two different fusion constructs, Protein Expr. Purif. 16 (1999) 243–250. [15] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. [16] M.D. Abramoff, P.J. Magelhaes, S.J. Ram, Image Processing with ImageJ, Biophotonics Int. 11 (2004) 36–42. [17] J. Morrè, Isolation of Golgi apparatus, Methods Enzymol. 22 (1971) 130–148. [18] T. Cabras, R. Inzitari, C. Fanali, E. Scarano, M. Patamia, M.T. Sanna, E. Pisano, B. Giardina, M. Castagnola, I. Messana, HPLC–MS characterization of cyclostatherin Q-37, a specific cyclization product of human salivary statherin generated by transglutaminase 2, J. Sep. Sci. 29 (2006) 2600–2608. [19] Z. Zhang, A.G. Marshall, A universal algorithm for fast and automated charge state deconvolution of electrospray mass-to-charge ratio spectra, J. Am. Soc. Mass. Spectrom. 9 (1998) 225–233. [20] R. Rudolph, H. Lilie, In vitro folding of inclusion body proteins, FASEB J. 10 (1996) 49–57. [21] R. Inzitari, T. Cabras, D.V. Rossetti, C. Fanali, A. Vitali, M. Pellegrini, G. Paludetti, A. Manni, B. Giardina, I. Messana, M. Castagnola, Detection in human saliva of different statherin and P–B fragments and derivatives, Proteomics 23 (2006) 6370–6379. [22] T. Cabras, E. Pisano, R. Boi, A. Olianas, B. Manconi, R. Inzitari, C. Fanali, B. Giardina, M. Castagnola, I. Messana, Age-dependent modifications of the human salivary secretory protein complex, J. Proteome Res. (2009) Jul 10. (Epub ahead of print) [23] G. Socrates, Infrared Characteristic Group Frequencies, Wiley, Chichester, England, 1994, pp. 24–25. [24] G. Goobes, R. Goobes, O. Schueler-Furman, D. Baker, P.S. Stayton, G.P. Drobny, Folding of the C-terminal bacterial binding domain in statherin upon adsorption onto hydroxyapatite crystals, Proc. Natl. Acad. Sci. USA 103 (2006) 16083–16088. [25] S.S. Schwartz, D.I. Hay, S. Schlukebier, Inhibition of calcium phosphate precipitation by human salivary statherin: structure–activity relationships, Calcif. Tissue Int. 50 (1992) 511–517. [26] J.L. Jensen, M.S. Lamkin, F.G. Oppenheim, Adsorption of human salivary proteins to hydroxyapatite: a comparison between whole saliva and glandular salivary secretions, J. Dent. Res. 71 (1992) 1569–1576. [27] P.A. Price, J.S. Rice, M.K. Williamson, Conserved phosphorylation of serines in the Ser-X-Glu/Ser(P) sequences of the vitamin K-dependent matrix Gla protein from shark, lamb, rat, cow, and human, Protein Sci. 3 (1994) 822–830. [28] E.W. Bimgam, H.M. Farrell, J.J. Basch, Phosphorylation of casein, J. Biol. Chem. 247 (1972) 8193–8194. [29] M. Lasa, O. Marin, L.A. Pinna, Rat Liver Golgi apparatus contains a protein kinase similar to the casein kinase of lactating mammary gland, Eur. J. Biochem. 243 (1997) 719–725. [30] M. Lasa-Benito, O. Marin, F. Meggio, L.A. Pinna, Golgi apparatus mammary gland casein kinase: monitoring by a specific peptide substrate and definition of specificity determinants, FEBS Lett. 382 (1996) 149–152. [31] L.A. Pinna, M. Ruzzene, How do protein kinases recognize their substrates? Biochim. Biophys. Acta 1314 (1996) 191–225. [32] A.M. Brunati, O. Marin, A. Bisinella, A. Salviati, L.A. Pinna, Novel consensus sequence for the Golgi apparatus casein kinase, revealed using proline-rich protein-1 (PRP1)-derived peptide substrates, Biochem. J. 351 (2000) 765–768. [33] Y. Nam, G. Madapallimattam, L. Drzymala, A. Bennick, Characterization of human sublingual-gland protein kinase by phosphorylation of a peptide related to secreted proteins, Arch. Oral Biol. 42 (1997) 527–537. [34] M.S. Lamkin, P. Lindhe, Purification of kinase activity from primate parotid glands, J. Dent. Res. 80 (2001) 1890–1894. [35] E. Tibaldi, G. Arrigoni, A.M. Brunati, P. James, L.A. Pinna, Analysis of a subproteome which co-purifies with and is phosphorylated by the Golgi casein kinase, Cell. Mol. Life Sci. 63 (2006) 378–379. [36] M. Patamia, I. Messana, R. Petruzzelli, A. Vitali, R. Inzitari, T. Cabras, C. Fanali, E. Scarano, A. Contucci, A. Galtieri, M. Castagnola, Two proline-rich peptides from pig (Sus scrofa) salivary glands generated by pre-secretory pathway underlying the action of a proteinase cleaving ProAla bonds, Peptides 26 (2005) 1550– 1559.