A Mn(II)–Mn(II) center in human prolidase

Share Embed


Descrição do Produto

Biochimica et Biophysica Acta 1834 (2013) 197–204

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbapap

A Mn(II)–Mn(II) center in human prolidase Roberta Besio a, Maria Camilla Baratto b, Roberta Gioia a, Enrico Monzani c, Stefania Nicolis c, Lucia Cucca c, Antonella Profumo c, Luigi Casella c, Riccardo Basosi b, Ruggero Tenni a, Antonio Rossi a, Antonella Forlino a,⁎ a b c

Department of Molecular Medicine, Section of Biochemistry, University of Pavia, via Taramelli 3 b-27100 Pavia, Italy Department of Chemistry, University of Siena, via A. De Gasperi 2‐53100 Siena, Italy Department of Chemistry, University of Pavia, via Taramelli 12‐2710 Pavia, Italy

a r t i c l e

i n f o

Article history: Received 6 August 2012 Received in revised form 10 September 2012 Accepted 12 September 2012 Available online 19 September 2012 Keywords: Prolidase Metallo-enzyme Mn(II)–Mn(II) cluster Apoenzyme

a b s t r a c t Human prolidase, the enzyme responsible for the hydrolysis of the Xaa-Pro/Hyp peptide bonds, is a key player in the recycling of imino acids during the final stage of protein catabolism and extracellular matrix remodeling. Its metal active site composition corresponding to the maximal catalytic activity is still unknown, although prolidase function is of increasing interest due to the link with carcinogenesis and mutations in prolidase gene cause a severe connective tissue disorder. Here, using EPR and ICP-MS on human recombinant prolidase produced in Escherichia coli (hRecProl), the Mn(II) ion organized in a dinuclear Mn(II)–Mn(II) center was identified as the protein cofactor. Furthermore, thermal denaturation, CD/fluorescence spectroscopy and limited proteolysis revealed that the Mn(II) is required for the proper protein folding and that a protein conformational modification is needed in the transition from apo- to Mn(II)loaded-enzyme. The collected data provided a better knowledge of the human holo-prolidase and, although limited to the recombinant enzyme, the exact identity and organization of the metal cofactor as well as the conformational change required for activity were proven. © 2012 Elsevier B.V. All rights reserved.

1. Introduction Human prolidase (peptidase D, EC 3.4.13.9) is responsible for the hydrolysis of dipeptides containing proline or hydroxyproline residues at the C-terminal end, playing an unique role in imino acids recycling during the final stage of protein catabolism and in particular in collagen turnover [1,2]. The human enzyme is a cytosolic homodimer of 123 kDa. Together with methionine aminopeptidase (MetAP, EC 3.4.1 1. la) and aminopeptidase P (APPro, EC 3.4.1 1.9) it belongs to the “pita-bread” enzyme family, characterized by the same type of internal two fold structural symmetry resembling a “pita-bread” fold with a dimetal(II) core in the C-terminal region [3,4]. The majority of prolidase proteins from various organisms previously studied are metalloenzymes requiring for activity divalent cations such as Zn(II), Mn(II) or Co(II) in their active site [5,6]. Although it is known that in in vitro assay human prolidase necessitates Mn(II) and reducing conditions for activity [7], the metal composition of the catalytic center is still unknown. In fact using human recombinant prolidase we previously demonstrated by ICP-MS and by X-ray absorption spectroscopy (XAS) that in solution two different metals, Mn(II) and Zn(II), could be

Abbreviations: ECM, extracellular matrix; ICP-MS, inductively coupled plasma mass spectrometry; XANES, X-ray absorption near edge structure; DTT, dithiothreitol; GSH, reduced glutathione ⁎ Corresponding author. Tel.: +39 0382 987235; fax: +39 0382 423108. E-mail address: [email protected] (A. Forlino). 1570-9639/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbapap.2012.09.008

simultaneously present in the enzyme. Based on activity measurements the prolidase containing 1 Mn(II) and 3 Zn(II) ions was still 25% active [8]. The discovery of Zn(II) ions in the active site of prolidase in humans (hRecProl), in the archeae Pyrococcus furiosus [9] and Pyrococcus horikoshii [10], both sharing 28% identity with human enzyme, leaves a puzzle on the nature of the cofactor and on the metal active site composition of the active human enzyme. Prolidase is linked to the metabolism of many biologically important molecules and it has a central role in the extracellular matrix (ECM) remodeling, being essential for certain aspects of normal physiology including embryonic development, cell migration, wound healing and tissue resorption [1,11–13]. Recently, proteomic studies have shown increased proline and hydroxyproline consumptions in patients with metastatic prostate tumors [14] and an increase in prolidase activity has been detected in various types of tumor, such as breast [15], endometrial [16] and ovarian cancers [17]; lung carcinoma and adenocarcinoma, for which it was indicated as sensitive marker [18]. The higher activity and expression of prolidase detected in tumor cells suggested its role in cancer development and its potential use as an endogenous target for selective activation of proline pro-drugs [19]. Moreover, mutations in the prolidase gene cause prolidase deficiency (PD: OMIM 170100), a severe recessive genetic disease characterized mainly by intractable skin ulcers, mental retardation and respiratory infections, and for which no therapy is available. Considering the broad fields of relevance of prolidase related to human health, the modulation of its activity could be a promising target for therapeutic approaches and a way to act on the ECM remodeling, but this will require a deep understanding of the enzyme active

198

R. Besio et al. / Biochimica et Biophysica Acta 1834 (2013) 197–204

site. To satisfy this demand we used the human recombinant protein produced in Escherichia coli (hRecProl) and, following the generation of its apo form (apo-hRecProl), we clarified by EPR and ICP-MS spectroscopy the nature and the stoichiometry of the metal ions in the active site. Thermal denaturation profiles, CD spectra, fluorescence analysis and limited proteolysis patterns elucidated the role of the cofactor on both prolidase structure and stability. 2. Materials and methods 2.1. Expression and purification of human recombinant prolidase Human recombinant prolidase (hRecProl) was obtained in prokaryotic host (Escherichia coli) and the identity of the enzyme was confirmed by western blot with antibody against human prolidase (kindly provided by Dr. Phang JM) as previously described [20]. The purified protein was extensively dialyzed against 10 mM Tris–HCl pH 7.8, 0.57 mM DTT and 300 mM NaCl, aliquoted and stored at −80 °C. 2.2. Prolidase activity assay Prolidase activity was determined as described in Besio et al. [7]. Briefly the reaction was performed in 50 mM Tris–HCl, pH 7.8 (50 μl) using 0.05 μg of hRecProl. An activation step before substrate incubation was performed, consisting in the addition of 1 mM MnCl2, 0.75 mM GSH, at 50 °C for 20 min. After the activation step, the dipeptide substrate (100 mM Gly-Pro, Promega) was added to the mixture, the reaction was incubated at 50 °C for 20 min and finally stopped with 0.45 M trichloroacetic acid. The amount of proline released was measured in the supernatant using Chinard's ninhydrin reagent [21]. Spectrophotometric measurements were performed at 515 nm. hRecProl activity was expressed as mmol of proline released per min per mg of protein. All measurements were performed in triplicate using a Jasco V-550 UV/VIS spectrophotometer (Jasco-Europe). Proline in 5 mM HCl was used as standard. 2.3. Apo-hRecProl generation and metal reconstitution studies Several attempts were performed to generate the apoprotein and prolidase activity was used as a first indicator of the effectiveness of the approaches. In the first attempts, hRecProl (0.28 mg/ml), as described in [8], was dialyzed against 10 mM EDTA, 50 mM Tris–HCl, pH 7.8 for 48 h at 4 °C or supplemented with 10 mM EDTA or EGTA for 72 h at 4 °C or 1 mM potassium cyanide for 1 h at 50 °C was added. Activity without the incubation step with MnCl2 and GSH was evaluated and compared to that of the hRecProl enzyme. In the attempt to denature the protein, an incubation with 3.5 M guanidine hydrochloride was carried out, followed by a 24 h dialysis against 50 mM Tris– HCl, pH 7.8, 0.1 mM EDTA performed to remove the metals. In a further attempt, to facilitate the release of the metal cofactor, hRecProl was dialyzed against 1 mM EDTA, 50 mM sodium acetate at different pH (3.2, 4, 5) overnight at 4 °C. After these metal removal procedures, the protein was dialyzed against 50 mM Tris–HCl, pH 7.8 overnight at 4 °C to remove chelating or denaturating agents and activity was determined prior and following the activation step. Finally, hRecProl (0.28 mg/ml) was incubated with 0.7 M trisodium citrate and 0.01 M disodium tetraborate for 48 h at 4 °C and then dialyzed against 20 mM trisodium citrate, 0.3 mM disodium tetraborate for 72 h at 4 °C to remove the salt excess and activity was determined again prior and following the activation step. After the treatments, protein recovery was determined by RC DC Protein Assay (Biorad). When the apoprotein was generated a reconstitution step with different metals was performed by incubating the enzyme with 1 mM MnCl2, ZnCl2, CoCl2, CaCl2 or MgCl2 (Sigma Aldrich) for 20 min at 50 °C and the activity was again analyzed. All the buffers used for the apoprotein generation were Chelex100-treated and the glassware was washed with nitric acid to remove any metals.

2.4. EPR measurement EPR measurements (CW X-band at 9 GHz) were carried out with a Bruker Elexsys E500 series using the Bruker ER4122 SHQE cavity and the low temperature spectra were recorded with an Oxford ESR900 helium continuous flow cryostat. The analyzed samples, apo and native prolidase, were concentrated to 12 mg/ml in 10 mM Tris–HCl, pH 7.8. The apo-hRecProl was analyzed also after the incubation with 0.75 mM GSH and 1 mM MnCl2 for 20 min at 50 °C. For a better resolution of the spectra, thanks to the addition of 20% glycerol and 30% acetonitrile, hRecProl was concentrated to 22 mg/ml and analyzed after an 8 h dialysis against 10 mM Tris–HCl, pH 7.8. All shown spectra were recorded at 9.39 GHz, 100 kHz modulation frequency, 1 mT modulation amplitude, and 20 mW microwave power. 2.5. ICP-MS measurement The hRecProl (3.2× 10−4 μmol), after a 48 h dialysis at 4 °C against Chelex100-treated 50 mM Tris–HCl, pH 7.8, 300 mM NaCl, and the apo-hRecProl (7.8 × 10−4 μmol), obtained as described in Section 2.3, were analyzed by ICP-MS to measure the number of tightly bound divalent metal ions. The measurements were performed on a Perkin Elmer Mod ELAN DRC-e instrument, following the standard procedures suggested by the manufacturer. 2.6. Thermal stability analysis with thermofluor technology Thermofluor experiments were carried out with the real time PCR instrument Mx3000P (Stratagene). In a final volume of 20 μl the recombinant protein (5 μM) was mixed with the fluorescent dye SYPRO Orange (Sigma Aldrich) in a Thermo-Fast 96-well PCR plate (VWR International). The plate was heated at a rate of 1 °C/min from 25 °C to 95 °C, and fluorescence was measured in 1 °C increments. Fluorescence was filtered through custom interference excitation (492 nm) and emission (568 nm) filters. The primary data (relative fluorescence intensity versus temperature) were fit to standard equations describing protein thermal stability, as described previously (Sigma Plot) [22]. Background correction was performed to avoid the interference of the SYPRO Orange fluorescence. 2.7. CD spectroscopy Far-UV (190–260 nm) and near-UV (250–320 nm) circular dichroism (CD) measurements were performed at 20 °C in 0.1 and 1.0 cm pathlength quartz cuvettes, respectively. CD spectra were recorded on a Jasco J-720 spectropolarimeter at a scan rate of 50 nm/min with a 1 nm spectral band width and collecting points every 0.2 nm. All the spectroscopic measurements were performed in 50 mM Tris–HCl pH 7.8. Far- and near-UV measurements were performed on protein preparations concentrated to 0.2 mg/ml and 0.8 mg/ml respectively. Far-UV spectra were recorded also in the presence of 0.57 mM DTT and after a 20 min incubation at 50 °C with 1 mM MnCl2 and 0.57 mM DTT. Twenty scans were averaged for each spectrum and the contribution from the buffer was subtracted in each case. The results were expressed as the mean residue ellipticity. The secondary structure content was estimated from the CD spectra using the deconvolution algorithms CONTIN [23], CDSSTR [24] and SELCON3 [25] with the data set 4 at the DICHROWEB server [26,27]. 2.8. Fluorescence spectroscopy Fluorescence spectra were measured with a Jasco FP-6500 spectrofluorimeter at room temperature, using a quartz cell with a path length of 0.2 cm. Measurements were performed on protein preparation concentrated 0.2 mg/ml. Emission spectra were recorded in the

R. Besio et al. / Biochimica et Biophysica Acta 1834 (2013) 197–204

range of 290–400 nm at the excitation wavelengths of 270, 280, and 295 nm with a scanning speed of 500 nm/min. 2.9. Limited proteolysis Apo-hRecProl and hRecProl (10 μg) were digested in a reaction volume of 100 μl with 250 μg/ml α-chymotrypsin (Cooper Biomedical) in 50 mM Tris–HCl, pH 7.8, at 37 °C or with 0.064 U/ml papain (Sigma Aldrich) in 0.1 M sodium acetate, pH 5.6, 5 mM cysteine, 5 mM EDTA at 4 °C or with 120 μg/ml trypsin (Sigma Aldrich) in 1 mM HCl at 37 °C. After 0, 5, 30, and 60 min the reactions were stopped by adding Laemmli sample buffer (60 mM Tris–Cl, pH 6.8, 2% SDS, 10% glycerol, 0.1 M DTT, 0.01% bromophenol blue) and by immediately heating at 90 °C. The proteolytic patterns were analyzed by 10% SDS-PAGE under reducing conditions with the Mini-Protean 3 system (Bio-Rad). Gels were stained with Coomassie blue, acquired by Versadoc 3000 (Biorad) and band intensity was evaluated by QuantityOne software (Biorad). For the densitometric analysis the baseline for each sample was the density measured at time 0. 2.10. Mn(II) binding The Mn(II) binding was evaluated through a combined use of thermofluor technology and prolidase activity assay. The apoprotein, 5 μM (20 μl), was incubated with 1 mM MnCl2, 0.75 mM GSH and with the fluorescent dye SYPRO Orange (Sigma Aldrich) in a Thermo-Fast 96-well PCR plate (VWR International). The plate was treated as described in Section 2.6 using a temperature range from 25 °C to 55 °C. At 30, 37 and 50 °C prolidase activity was determined. After the thermal treatment the protein was returned to 37 °C for 2 h and activity was again measured. 3. Results 3.1. Apo-hRecProl generation and reconstitution studies To investigate the prolidase metal active site composition and the relation between prolidase structure and function, a recombinant human prolidase (hRecProl) was generated in E. coli [20]. Firstly we attempted to produce the apoprotein taking advantage of the most common approaches described in literature: the use of chelating agents, the protonation of the metal ligand residues and the protein denaturation [28–30]. In order to evaluate the success in obtaining the apo form, poorly active with respect to the holoenzyme, we measured the prolidase function comparing it to the one of the hRecProl previously characterized as containing 1 Mn(II) ion and 3 Zn(II) ions and still partly active [8]. No decrease in activity was detected following EDTA dialysis and EDTA/EGTA or potassium cyanide incubation (Table 1). The treatments with acidic pH and guanidine hydrochloride revealed a reduced/ undetectable activity and this result was considered a first evidence of the successful generation of the apoprotein. Unfortunately, following the reconstitution step with MnCl2 very low activity was obtained, both conditions caused an irreversible protein precipitation (Table 1). Finally, the hRecProl was incubated with a high concentration of sodium ions, as described in Materials and methods. After the incubation, the protein was poorly active (0.05± 0.01 mmol pro/mg vs 2.93 ± 0.32 mmol pro/mg for hRecProl) and its function was restored only following the incubation with MnCl2, whereas incubation with ZnCl2, CoCl2, CaCl2 and MgCl2 did not produce any effect (Fig. 1). This was also true by performing the substrate incubation at 37 °C, demonstrating that the absence of activity was not caused by a poorly binding of the metal due to protein instability (Supplementary Table S1). No loss of the protein was detected (Table 1), making this the elective procedure for apoenzyme generation. The absence of metals in the apoprotein preparations was demonstrated by ICP-MS. The analysis was performed

199

Table 1 Approaches for the apoprotein generation. Loss of activity after the treatment and the subsequent activity rescue after the incubation step with 1 mM MnCl2 and 0.75 mM GSH for 20 min at 50 °C was considered a first indicator of the effectiveness of the approaches. Treatment

None EDTA

EGTA

Potassium cyanide Acidic pH

Guanidine hydrochloride Trisodium citrate Disodium tetraborate

Activity (mmol pro/mg)

Activity (mmol pro/mg) MnCl2 incubation

Protein recovery (%)

10 mM 20 mM 50 mM 100 mM 10 mM 20 mM 50 mM 0.6 M

2.93 ± 0.32 2.94 ± 0.11 3.67 ± 0.36 4.3 ± 0.18 3.52 ± 0.39 2.93 ± 0.2 3.06 ± 0.17 3.46 ± 0.2 2.29 ± 0.23

9.8 ± 0.96 np np np np np np np 2.35 ± 0.24

100 100 100 100 100 100 100 100 np

3.2 4 5 3.5 M

0.76 ± 0.06 0.76 ± 0.03 0.74 ± 0.07 nd

0.29 ± 0.06 0.47 ± 0.06 0.59 ± 0.03 0.53 ± 0.09

17 21 24 29

0.7 M

nd

9.6 ± 0.8

100

10 mM

np: not performed; nd: not detectable.

on the recombinant enzyme both before and after the metal removal and extensive dialysis. The presence of a tightly bound Mn(II) ion and of 3 Zn(II) ions in hRecProl confirmed our previous data [8], while the absence of metals in the apo-hRecProl proved the success of our attempt (Table 2). To our knowledge these results obtained with the apoenzyme, demonstrated for the first time that Mn(II) ion is the cofactor of human recombinant prolidase since all the previous available data pointed to Mn(II) as cofactor only based on an increase of the basal enzymatic activity following MnCl2 incubation [6,7,31]. The presence of Mn(II) and Zn(II) traces detected in the apoenzyme preparations could be due to a slight contamination of the buffers used for the dialysis or, more likely, to the non complete removal of the bound metals from all the protein molecules.

Fig. 1. Human recombinant prolidase activity dependence on divalent ions. Prolidase activity was measured in: hRecProl as described in [7]; hRecProl after the incubation with 1 mM MnCl2 [Mn(II)-hRecProl]; apo-hRecProl; apo-hRecProl after the incubation with 1 mM MnCl2 [Mn(II)-apo-hRecProl], with 1 mM ZnCl2 [Zn(II)-apo-hRecProl], with 1 mM CoCl2 [Co(II)-apo-hRecProl], with 1 mM CaCl2 [Ca(II)-apo-hRecProl], with 1 mM MgCl2 [Mg(II)-apo-hRecProl]; Mn(II)-apo-hRecProl after an 8 h dialysis against 10 mM Tris–HCl, pH 7.8 to remove the free metal excess, as analyzed by EPR spectroscopy [EPR-hRecProl].

200

R. Besio et al. / Biochimica et Biophysica Acta 1834 (2013) 197–204

Table 2 ICP-MS quantitation of the Mn(II) and Zn(II) ions in hRecProl and apo-hRecProl. The analysis was performed after a 48 h dialysis against metal free buffer.

hRecProl mol dimer:mol metal apo-hRecProl mol dimer:mol metal

Protein ×10−4 μmol

Mn(II) ×10−4 μmol

Zn(II) ×10−4 μmol

3.2

2.8 ± 0.3 ~1:1 0.8 ± 0.3 ~1:0.1

9.0 ± 0.9 ~1:3 5.2 ± 0.3 ~1:0.7

7.8

3.2. EPR fingerprints of the metal active site EPR spectroscopy at cryogenic temperature is a suitable tool to characterize the prosthetic group of metalloenzymes [32–35]. CW EPR spectra of prolidase were recorded at 9 GHz and at different temperatures (10 K, 25 K, 50 K) with the attempt to discriminate the Mn(II)–Mn(II) center contribution. The spectra at 25 K were chosen as the optimal condition in terms of spectra resolution. The apo-hRecProl in the absence of added cofactor did not give rise, as expected, to any Mn(II) signal further supporting its nature (Fig. 2A). When the apoprotein was incubated at 50 °C for 20 min with GSH and with 1 mM MnCl2, spectra revealed the six-line pattern typical of mononuclear Mn(II) (Fig. 2A). The signal of the exchange coupled Mn(II)–Mn(II) center at 275 mT was weak and not well resolved into the expected eleven structured components. To confirm the presence of the dinuclear center, an amplification of the protein signal was attempted using higher protein concentration, possible only through the addition of the stabilizing agents such as glycerol and acetonitrile. Concentrated hRecProl was activated with GSH and MnCl2 at 50 °C for 20 min and dialyzed to avoid the interference of excess of free Mn(II) ions. The X-band EPR spectrum, beyond the six-line pattern typical of mononuclear Mn(II) sites with a g = 2.03 and a hyperfine coupling constant of 9 mT, revealed, in the region 250–300 mT, the characteristic signal of an exchange-coupled Mn(II)–Mn(II) center [36–38]: eleven lines centered around g =2.4 with relative intensities of approximately 1:2:3:4:5:6:5:4:3:2:1 and a hyperfine coupling of 4.4 mT (Fig. 2B, C). The observed interaction was consistent with the event that the two Mn(II) sites are bound to each other sufficiently close to have a direct orbital interaction. The possibility that metal centers bridged by a common ligand favoring exchange cannot be ruled out. The coupling of two S=5/2 Mn(II) ions gives rise to a series of new states with a total spin S=0,1,2,3,4,5 each of them with a degeneracy of (2S+1). In the case of ferromagnetic coupling the S = 5 spin state lies lowest in energy, whereas with the antiferromagnetic coupling the diamagnetic S= 0 spin state is the one with the lowest energy [38]. With strong exchange conditions the exchange coupling energy is much larger than the electronic Zeeman energy (|J|≫ gβB) and the zero-field splitting (|J|≫ D) and therefore the dimer can be regarded as a ladder of isolated spin manifolds. The separation between the spin manifolds is larger than the microwave energy at X-band and no transitions are expected between them [39]. Thus, in the analyzed hRecProl, spin coupling between the two Mn(II) ions was detected revealing the presence of a dinuclear Mn(II)–Mn(II) center at least in one of the two active sites of the homodimer. The narrow signal at 156 mT with a g = 4.3 derives from an orthorhombic Fe(III) ion, commonly present in biological systems as impurity [38,39].

Fig. 2. X-band EPR spectra of apo-hRecProl and Mn(II)loaded-hRecProl. (A) 25 K EPR spectra of apo-hRecProl (grey line) and of hRecProl incubated with Mn(II) ions (black line). The apo-hRecProl spectrum did not reveal any Mn(II) signal. The hRecProl incubated with 1 mM MnCl2, 0.75 mM GSH at 50 °C for 20 min revealed the six-line pattern typical of the mononuclear magnetically isolated Mn(II) center. (B) Amplification of the EPR signal of the exchange coupled Mn(II)–Mn(II) center in hRecProl using an higher protein concentration. X-band EPR spectrum was measured on hRecProl after the activation step followed by dialysis. The arrow highlights the eleven lines characteristic of the Mn(II)–Mn(II) coupling. (C) Expansion of the eleven components is shown in the 250– 300 mT region. Frequency 9.39 GHz; central field 330 mT; scan range 650 mT; 20 mW microwave power; modulation amplitude 1 mT; gain 60 dB; temperature 25 K.

exposure to the solvent and thus to the binding of the fluorescence dye SYPRO Orange. The melting temperature of the apoenzyme (50 °C) was significantly lower than the one measured for hRecProl (57 °C), indicating an effect of the Mn(II) binding on prolidase stability (Fig. 3A).

3.3. Metal effect on hRecProl thermal stability 3.4. Metal effect on hRecProl secondary structure To further elucidate the effect of Mn(II) ions on prolidase structure and stability, thermal denaturation profiles of apo-hRecProl and hRecProl were determined with thermofluor technology [40]. The denaturation profiles derived from changes in the fluorescence emission due to protein

The secondary structure of apo-hRecProl was analyzed by CD in the far-UV region. The spectrum revealed two minima, at 205–210 and 215–220 nm respectively (Fig. 3B). CD spectra deconvolution, using

R. Besio et al. / Biochimica et Biophysica Acta 1834 (2013) 197–204

three independent algorithms, allowed the evaluation of the relative content of secondary structure elements: 23% α helix, 28% β-sheet and 23% coil structure. Spectra were recorded also in the presence of the reducing agent DTT and of the Mn(II) cofactor to reproduce the optimal functional status. The incubation did not produce any structural change (Fig. 3B), suggesting that the absence of metals in the apoprotein was not affecting its secondary structure. The detected reduction of the signal intensity likely indicated a small loss of the protein amount.

201

respectively. The proteolytic patterns, shown by electrophoretic analysis, revealed that the apoenzyme was more sensitive to all three tested enzymes. In particular a densitometric analysis revealed after 30 min of chymotrypsin digestion a reduction of the full length apo-hRecProl with respect to hRecProl (21% vs 67%) a completely different tryptic peptide pattern was detected at all time points and finally after just 5 min of papain digestion all apoprolidase disappeared (Fig. 4). 3.6. Conformational transition from apo- to holoenzyme

3.5. Metal effect on hRecProl tridimensional structure The near-UV CD spectra of the protein did not reveal any absorption (data not shown) indicating a great mobility of the aromatic residues. Thus, fluorescence spectroscopy was used to evaluate the solvent accessibility in the hRecProl and in its apo-form. All fluorescence spectra revealed in the apoenzyme a marked decreased intensity of the maximum of fluorescence emission, indicating the movement of aromatic residues to a more polar environment and thus a change in tridimensional structure (Fig. 3C). In order to further support the presence of protein conformational changes induced by the presence of the metal, a limited proteolysis was performed with three different enzymes: α-chymotrypsin, papain and trypsin, specific for peptide amide bonds where the carboxyl side is an aromatic, a basic/leucine/glycine, or a lysine/arginine residue,

The apo-hRecProl was incubated in presence of Mn(II) ions and SYPRO Orange in order to follow, through thermofluor technology, the conformational changes of the protein during the phase of metal loading. Activity measurements were performed at selected temperature points. At increased temperatures the fluorescence trend revealed as expected a more open arrangement of the protein, but the enzyme showed a low activity both at 30 °C (1.9 mmol pro/mg) and at 37 °C (2.5 mmol pro/mg), only at 50 °C the activity reached 9.8 mmol pro/mg. At the end of the 50 °C incubation the protein was returned to 37 °C for 2 h and the 86% activity (8.3± 0.81 mmol pro/mg) was detected demonstrating that the increase in activity at 50 °C was primarily due to the metal loading rather than to the high temperature. These data further supported that conformational changes are required in the transition apo-hRecProl to Mn(II)loaded-hRecProl (Fig. 5).

Fig. 3. Effect of Mn(II) ions on prolidase thermal stability, secondary and tertiary structure. (A) Denaturation profiles of hRecProl and apo-hRecProl. The solvatochromic dye SYPRO Orange was used as an indicator of protein unfolding (fluorescence excitation λ = 492 nm; fluorescence emission λ = 568 nm). (B) CD spectra in the far-UV spectrum region of apo-hRecProl (__) and of apo-hRecProl in the presence of 0.75 mM DTT (·····) and of both 0.75 mM DTT and 1 mM MnCl2 (−−−). (C) Fluorescence spectra of hRecProl (__) and apo-hRecProl (−− −−) exiting tryptophan residues. The excitation wavelength was set at 295 nm; the monitoring emission from 305 to 400 nm. Spectra recorded at the wavelengths specific for phenylalanine and tyrosine were similar and thus not reported.

202

R. Besio et al. / Biochimica et Biophysica Acta 1834 (2013) 197–204

Fig. 4. Limited proteolysis of hRecProl and apo-hRecProl. The proteolytic digestions were performed with α-chymotrypsin, papain and trypsin. At the indicated time points the reactions were stopped and the proteins were analyzed by SDS-PAGE under reducing conditions. Gels were stained with Coomassie blue. In lines §, *, and † only the proteolytic enzymes chymotrypsin, trypsin and papain were loaded, respectively.

4. Discussion Since human prolidase is involved in many biological processes and it is correlated to many forms of tumors and to a severe genetic disease, whose molecular mechanism is still under investigation, the characterization of its catalytic site is an urgent requirement. The metal

Fig. 5. From apo-hRecProl to Mn(II)loaded-hRecProl: follow up the Mn(II) binding. The Mn(II) loading of apo-hRecProl was followed by thermofluor tecnology (•) and by measuring the prolidase activity (bars) at the selected temperatures. The solvatochromic dye SYPRO Orange was used as an indicator of protein conformational change (fluorescence excitation λ=492 nm; fluorescence emission λ=568 nm).

coordinating amino acids were determined for P. furiosus prolidase by site directed mutagenesis [9] and are highly conserved among different species (Fig. S1). Furthermore, for three of the metal binding residues in human prolidase (D276, E412 and E452), the substitution with a different amino acid was causative for the recessive connective tissue disorder prolidase deficiency [41–43]. Prolidase proteins extracted from different organisms need divalent cations such as Zn(II), Mn(II) or Co(II) for their catalytic activity [5,6], but for the human enzyme the metal composition of the active form of the enzyme is still unknown [8]. Interestingly, quantum mechanical (QM) studies have previously investigated in silico the presence of metal cations such as Zn(II), Co(II) and Mn(II) in the active site of the human enzyme, but without revealing a clear dependence of the catalytic activity on a specific ion. A better performance of the cobalt-containing cluster was detected looking at the potential energy profiles [44]. Our previous data on the human prolidase activity of the recombinant enzyme, expressed in CHO cells and prokaryotic host, and of the endogenous enzyme, from fibroblasts, revealed an increase in activity following Mn(II) incubation, while less than 30% activity was reported in presence of other cations such as Zn(II) and Co(II) [20]. Zn(II) ions were detected in the active site of the partially active enzyme (hRecProl), as analyzed by XAS [8], but the presence of Zn(II) ions was detected also in the inactive forms of prolidase from other organisms: X-ray crystallography of prolidase from P. furiosus (PDB ID: 1PV9) and from P. horikoshii (PDB ID: 1WY2), both sharing 28% identity with human prolidase, revealed Zn(II) ions in the binding site, although their activity was reported to be Co(II) dependent [4,45]. To determine the metal active site composition of the active enzyme, the recombinant human prolidase was used and its apo form was

R. Besio et al. / Biochimica et Biophysica Acta 1834 (2013) 197–204

generated. The metal removal, through the use of chelating agents, protein denaturation and protonation of the metal ligand residues, was unsuccessful or biased by a low yield of protein recovery (Table 1). Interestingly, the dialysis against EDTA, a procedure successfully used to remove the metal from P. furiosus recombinant prolidase (PfRecProl) [9], failed on our recombinant protein. This finding suggested that the tightly bound metal ion in the human enzyme could be less accessible to the solvent or more tightly bound with respect to Co(II) in the archaea protein. For both enzymes the affinity constant for this strongly bound divalent ion was not measurable but, interestingly, the Kb for the less tightly bound metal was greater in hRecProl respect with PfRecProl [7,9]. Thus, a similar behavior could be hypothesized for the more tightly bound ion, explaining the different responses to the incubation with the chelating agents. The catalytically inactive apoenzyme was finally obtained by the protein incubation with sodium ions and absence of metals was proven by ICP-MS and EPR measurements (Table 2, Fig. 2A). The apo-hRecProl recovered its maximal activity, comparable to MnCl2 incubated hRecProl, following MnCl2 and GSH incubation, whereas no increase in activity was detected following ZnCl2, CoCl2, CaCl2 and MgCl2 incubations (Fig. 1). Thanks to the apoenzyme generation these data unequivocally demonstrated the hRecProl specific requirement for the Mn(II) ions. We also investigated the effect of the Mn(II) bound in the active site on the enzyme structure and stability. The apoenzyme had a lower thermal stability (Fig. 3A) as well as a different susceptibility to proteolytic digestion (Fig. 4) and a different solvent exposure of aromatic residues (Fig. 3C), indicating the crucial role of the metals in the protein folding maintenance. Furthermore, the follow up of the Mn(II) binding confirmed the requirement for a protein conformational modification during the transition from the apo to the active holoenzyme (Fig. 5). According to our previous XAS results, the catalytic activity of the partially active hRecProl was associated to a tightly bound Mn(II) ion, but with the addition of free Mn(II) an increase in prolidase activity was detected [8]. This finding suggested that one or more Mn(II) ions, in addition to that observed by XAS [8], could be weakly bound in the active site. The XAS measurements could be performed only after extensive dialysis that could indeed be responsible for the loss of the less tightly bound metals. With the EPR spectroscopy, that can be used in the presence of a limited amount of free metals in solution, we demonstrated the presence of inter-Mn(II) spin coupling associated with maximum prolidase activity, indicating that at least one of the two metal clusters in the active site was constituted by 2 coupled Mn(II) ions (Fig. 2B,C). The mechanism of prolidase catalysis is still largely unknown. Mock and Liu, using picolinylproline as substrate analogue, and Lowther and Matthews, by analogy with the “pita-bread” enzyme family, proposed a reaction mechanism. In their model metal ions activate a coordinated water molecule, by lowering the pKa for proton ionization, generating the hydroxide group that performs a nucleophilic attack to the peptide bond [6,46–48]. Therefore the increased activity of hRecProl, resulting from the addition of a second Mn(II), may be explained by the likely lower pKa of the water molecule bound as a bridge between the two coupled metal ions. We were not able to determine whether the other site was occupied by two Mn(II) ions as well, due to the interference of the free Mn(II) signal and to the impossibility of obtaining the high protein concentration required for this type of measurement. However, it is reasonable to hypothesize that the second active site also is comprised of two Mn(II) ions based on the homology with other peptidases belonging to the same family. As originally identified by sequence analysis [49], prolidase shares a common pita bread fold with aminopeptidase P (APP) and methionine aminopeptidase (MetAP), the active site structures are almost identical, with two divalent metal ions bound to the same set of ligands [50,51] and they are presumed to share a common mechanism of catalysis [6].

203

XAS data of a partially active APP from E. coli showed the presence of only one Mn(II) ion in the active site [52]. On the other hand, in the EPR spectrum at liquid-He temperature of Mn(II)loaded-APP an exchange-coupled dinuclear Mn(II) center was detected, although without any information about which monomer was linked to [52]. The X-ray crystal structure (PDB ID: 1WL9) confirmed the presence of 2 Mn(II) ions, in both the monomers, indicating that the Mn(II) ion in the second position of the dinuclear site was weakly bound [36]. The Mn(II)–Mn(II)loaded-APP retained activity in the crystal unlike the Zn(II)–Zn(II)loaded-APP that was catalytically inactive [53]. Furthermore, the methionine aminopeptidase (MetAP) from P. furiosus showed a crystal structure (PDB ID: 1XGM) containing 2 Co(II) ions in both the monomers confirming its nature of homodimer for the active site metal composition as well as for the amino acid sequence [51]. In conclusion, our findings shed new light on the catalytic site of human recombinant prolidase, clarifying the nature and organization of the metal cofactor as well as the requirement for a conformation change for the apo-hRecProl to holo-hRecProl transition. Author contribution A. Forlino designed the study; R. Besio, M.C. Baratto, R. Gioia, and L. Cucca performed the experiments; A. Forlino, R. Besio, E. Monzani, A. Profumo, R. Tenni, A. Rossi, S. Nicolis, R. Basosi, and L. Casella analyzed the data; and A. Forlino and R. Besio wrote the paper. Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.bbapap.2012.09.008. Acknowledgements This work was supported by PRIN 2008 (2008XA48SC), by Progetto Regione Lombardia (cod. SAL/45) “Dalla scienza dei materiali alla medicina molecolare” to A.F. and by CARIPLO 2011–0270 to R.T. References [1] A. Lupi, R. Tenni, A. Rossi, G. Cetta, A. Forlino, Human prolidase and prolidase deficiency: an overview on the characterization of the enzyme involved in proline recycling and on the effects of its mutations, Amino Acids 35 (2008) 739–752. [2] J.M. Phang, W. Liu, O. Zabirnyk, Proline metabolism and microenvironmental stress, Annu. Rev. Nutr. 30 (2010) 441–463. [3] S.L. Roderick, B.W. Matthews, Structure of the cobalt-dependent methionine aminopeptidase from Escherichia coli: a new type of proteolytic enzyme, Biochemistry 32 (1993) 3907–3912. [4] M.J. Maher, M. Ghosh, A.M. Grunden, A.L. Menon, M.W. Adams, H.C. Freeman, J.M. Guss, Structure of the prolidase from Pyrococcus furiosus, Biochemistry 43 (2004) 2771–2783. [5] D.E. Wilcox, Binuclear metallohydrolases, Chem. Rev. 96 (1996) 2435–2458. [6] W.T. Lowther, B.W. Matthews, Metalloaminopeptidases: common functional themes in disparate structural surroundings, Chem. Rev. 102 (2002) 4581–4608. [7] R. Besio, E. Monzani, R. Gioia, S. Nicolis, A. Rossi, L. Casella, A. Forlino, Improved prolidase activity assay allowed enzyme kinetic characterization and faster prolidase deficiency diagnosis, Clin. Chim. Acta 412 (2011) 1814–1820. [8] R. Besio, S. Alleva, A. Forlino, A. Lupi, C. Meneghini, V. Minicozzi, A. Profumo, F. Stellato, R. Tenni, S. Morante, Identifying the structure of the active sites of human recombinant prolidase, Eur. Biophys. J. 39 (2010) 935–945. [9] X. Du, S. Tove, K. Kast-Hutcheson, A.M. Grunden, Characterization of the dinuclear metal center of Pyrococcus furiosus prolidase by analysis of targeted mutants, FEBS Lett. 579 (2005) 6140–6146. [10] C.M. Theriot, S.R. Tove, A.M. Grunden, Characterization of two proline dipeptidases (prolidases) from the hyperthermophilic archaeon Pyrococcus horikoshii, Appl. Microbiol. Biotechnol. 86 (2009) 177–188. [11] C.M. Nelson, M.J. Bissell, Of extracellular matrix, scaffolds, and signaling: tissue architecture regulates development, homeostasis, and cancer, Annu. Rev. Cell Dev. Biol. 22 (2006) 287–309. [12] T.D. Tlsty, L.M. Coussens, Tumor stroma and regulation of cancer development, Annu. Rev. Pathol. 1 (2006) 119–150. [13] C. Yan, D.D. Boyd, Regulation of matrix metalloproteinase gene expression, J. Cell. Physiol. 211 (2007) 19–26. [14] G. Catchpole, A. Platzer, C. Weikert, C. Kempkensteffen, M. Johannsen, H. Krause, K. Jung, K. Miller, L. Willmitzer, J. Selbig, S. Weikert, Metabolic profiling reveals key metabolic features of renal cell carcinoma, J. Cell Mol. 15 (2009) 109–118.

204

R. Besio et al. / Biochimica et Biophysica Acta 1834 (2013) 197–204

[15] M. Cechowska-Pasko, J. Palka, M.Z. Wojtukiewicz, Enhanced prolidase activity and decreased collagen content in breast cancer tissue, Int. J. Exp. Pathol. 87 (2006) 289–296. [16] D.T. Arioz, H. Camuzcuoglu, H. Toy, S. Kurt, H. Celik, N. Aksoy, Serum prolidase activity and oxidative status in patients with stage I endometrial cancer, Int. J. Gynecol. Cancer 19 (2009) 1244–1247. [17] H. Camuzcuoglu, D.T. Arioz, H. Toy, S. Kurt, H. Celik, N. Aksoy, Assessment of preoperative serum prolidase activity in epithelial ovarian cancer, Eur. J. Obstet. Gynecol. Reprod. Biol. 147 (2009) 97–100. [18] E. Karna, A. Surazynski, J. Palka, Collagen metabolism disturbances are accompanied by an increase in prolidase activity in lung carcinoma planoepitheliale, Int. J. Exp. Pathol. 81 (2000) 341–347. [19] S. Mittal, X. Song, B.S. Vig, C.P. Landowski, I. Kim, J.M. Hilfinger, G.L. Amidon, Prolidase, a potential enzyme target for melanoma: design of proline-containing dipeptide-like prodrugs, Mol. Pharm. 2 (2005) 37–46. [20] A. Lupi, S. Della Torre, E. Campari, R. Tenni, G. Cetta, A. Rossi, A. Forlino, Human recombinant prolidase from eukaryotic and prokaryotic sources. Expression, purification, characterization and long-term stability studies, FEBS J. 273 (2006) 5466–5478. [21] F.P. Chinard, Photometric estimation of proline and ornithine, J. Biol. Chem. 199 (1952) 91–95. [22] T.M. Mezzasalma, J.K. Kranz, W. Chan, G.T. Struble, C. Schalk-Hihi, I.C. Deckman, B.A. Springer, M.J. Todd, Enhancing recombinant protein quality and yield by protein stability profiling, J. Biomol. Screen. 12 (2007) 418–428. [23] S.W. Provencher, J. Glockner, Estimation of globular protein secondary structure from circular dichroism, Biochemistry 20 (1981) 33–37. [24] W.C. Johnson, Analyzing protein circular dichroism spectra for accurate secondary structures, Proteins 35 (1999) 307–312. [25] N. Sreerama, R.W. Woody, Estimation of protein secondary structure from circular dichroism spectra: comparison of CONTIN, SELCON, and CDSSTR methods with an expanded reference set, Anal. Biochem. 287 (2000) 252–260. [26] A. Lobley, L. Whitmore, B.A. Wallace, DICHROWEB: an interactive website for the analysis of protein secondary structure from circular dichroism spectra, Bioinformatics 18 (2002) 211–212. [27] 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–W673. [28] N. Carvajal, V. Lopez, M. Salas, E. Uribe, P. Herrera, J. Cerpa, Manganese is essential for catalytic activity of Escherichia coli agmatinase, Biochem. Biophys. Res. Commun. 258 (1999) 808–811. [29] E.W. Moomaw, A. Angerhofer, P. Moussatche, A. Ozarowski, I. Garcia-Rubio, N.G. Richards, Metal dependence of oxalate decarboxylase activity, Biochemistry 48 (2009) 6116–6125. [30] P. Vatsyayan, P. Goswami, Acidic pH conditions induce dissociation of the haem from the protein and destabilise the catalase isolated from Aspergillus terreus, Biotechnol. Lett. 33 (2011) 347–351. [31] D.F. Cunningham, B. O'Connor, Proline specific peptidases, Biochim. Biophys. Acta 1343 (1997) 160–186. [32] M.A. Aceves, M.C. Baratto, R. Basosi, R. Vazquez-Duhalt, R. Pogni, Spectroscopic characterization of a manganese-lignin peroxidase hybrid isozyme produced by Bjerkandera adusta in the absence of manganese: evidence of a protein centred radical by hydrogen peroxide, J. Mol. Catal. B Enzym. 16 (2001) 159–167. [33] J. Verdin, R. Pogni, A. Baeza, M.C. Baratto, R. Basosi, R. Vazquez-Duhalt, Mechanism of versatile peroxidase inactivation by Ca(2+) depletion, Biophys. Chem. 121 (2006) 163–170. [34] B. Valderrama, H. Garcia-Arellano, S. Giansanti, M.C. Baratto, R. Pogni, R. VazquezDuhalt, Oxidative stabilization of iso-1-cytochrome c by redox-inspired protein engineering, FASEB J. 20 (2006) 1233–1235. [35] S.V. Khangulov, P.J. Pessiki, V.V. Barynin, D.E. Ash, G.C. Dismukes, Determination of the metal ion separation and energies of the three lowest electronic states of dimanganese

[36]

[37]

[38]

[39] [40]

[41]

[42]

[43] [44] [45]

[46] [47] [48]

[49]

[50]

[51]

[52]

[53]

(II, II) complexes and enzymes: catalase and liver arginase, Biochemistry 34 (1995) 2015–2025. M.C. Wilce, C.S. Bond, N.E. Dixon, H.C. Freeman, J.M. Guss, P.E. Lilley, J.A. Wilce, Structure and mechanism of a proline-specific aminopeptidase from Escherichia coli, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) 3472–3477. S.V. Khangulov, T.M. Sossong Jr., D.E. Ash, G.C. Dismukes, L-arginine binding to liver arginase requires proton transfer to gateway residue His141 and coordination of the guanidinium group to the dimanganese(II, II) center, Biochemistry 37 (1998) 8539–8550. A.J. Copik, B.P. Nocek, S.I. Swierczek, S. Ruebush, S.B. Jang, L. Meng, V.M. D'Souza, J.W. Peters, B. Bennett, R.C. Holz, EPR and X-ray crystallographic characterization of the product-bound form of the MnII-loaded methionyl aminopeptidase from Pyrococcus furiosus, Biochemistry 44 (2005) 121–129. A.P. Golombek, M.P. Hendrich, Quantitative analysis of dinuclear manganese(II) EPR spectra, J. Magn. Reson. 165 (2003) 33–48. U.B. Ericsson, B.M. Hallberg, G.T. Detitta, N. Dekker, P. Nordlund, Thermofluorbased high-throughput stability optimization of proteins for structural studies, Anal. Biochem. 357 (2006) 289–298. A. Tanoue, F. Endo, A. Kitano, I. Matsuda, A single nucleotide change in the prolidase gene in fibroblasts from two patients with polypeptide positive prolidase deficiency. Expression of the mutant enzyme in NIH 3T3 cells, J. Clin. Invest. 86 (1990) 351–355. A. Lupi, A. Rossi, E. Campari, F. Pecora, A.M. Lund, N.H. Elcioglu, M. Gultepe, M. Di Rocco, G. Cetta, A. Forlino, Molecular characterisation of six patients with prolidase deficiency: identification of the first small duplication in the prolidase gene and of a mutation generating symptomatic and asymptomatic outcomes within the same family, J. Med. Genet. 43 (2006) e58. P. Ledoux, C.R. Scriver, P. Hechtman, Expression and molecular analysis of mutations in prolidase deficiency, Am. J. Hum. Genet. 59 (1996) 1035–1039. M.E. Alberto, M. Leopoldini, N. Russo, Can human prolidase enzyme use different metals for full catalytic activity? Inorg Chem 50 (2011) 3394–3403. J. Jeyakanthan, K. Takada, M. Sawano, K. Ogasahara, H. Mizutani, N. Kunishima, S. Yokoyama, K. Yutani, Crystal structural and functional analysis of the putative dipeptidase from Pyrococcus horikoshii OT3, J. Biophys. 2009 (2009) 434038. W.L. Mock, P.C. Green, Mechanism and inhibition of prolidase, J. Biol. Chem. 265 (1990) 19606–19610. W.L. Mock, P.C. Green, K.D. Boyer, Specificity and pH dependence for acylproline cleavage by prolidase, J. Biol. Chem. 265 (1990) 19600–19605. W.L. Mock, Y. Liu, Hydrolysis of picolinylprolines by prolidase. A general mechanism for the dual-metal ion containing aminopeptidases, J. Biol. Chem. 270 (1995) 18437–18446. J.F. Bazan, L.H. Weaver, S.L. Roderick, R. Huber, B.W. Matthews, Sequence and structure comparison suggest that methionine aminopeptidase, prolidase, aminopeptidase P, and creatinase share a common fold, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 2473–2477. X. Li, Z. Lou, W. Zhou, M. Ma, Y. Cao, Y. Geng, M. Bartlam, X.C. Zhang, Z. Rao, Structure of human cytosolic X-prolyl aminopeptidase: a double Mn(II)-dependent dimeric enzyme with a novel three-domain subunit, J. Biol. Chem. 283 (2008) 22858–22866. T.H. Tahirov, H. Oki, T. Tsukihara, K. Ogasahara, K. Yutani, K. Ogata, Y. Izu, S. Tsunasawa, I. Kato, Crystal structure of methionine aminopeptidase from hyperthermophile, Pyrococcus furiosus, J. Mol. Biol. 284 (1998) 101–124. L. Zhang, M.J. Crossley, N.E. Dixon, P.J. Ellis, M.L. Fisher, G.F. King, P.E. Lilley, D. MacLachlan, R.J. Pace, H.C. Freeman, Spectroscopic identification of a dinuclear metal centre in manganese (II)-activated aminopeptidase P from Escherichia coli: implications for human prolidase, J. Biol. Inorg. Chem. 3 (1998) 470–483. S.C. Graham, C.S. Bond, H.C. Freeman, J.M. Guss, Structural and functional implications of metal ion selection in aminopeptidase P, a metalloprotease with a dinuclear metal center, Biochemistry 44 (2005) 13820–13836.

Lihat lebih banyak...

Comentários

Copyright © 2017 DADOSPDF Inc.