Characterization and kinetics studies of water buffalo (Bubalus bubalis) myoglobin

Comparative Biochemistry and Physiology, Part B 145 (2006) 230 – 238 www.elsevier.com/locate/cbpb

Characterization and kinetics studies of water buffalo (Bubalus bubalis) myoglobin Roberta Dosi a,b , Antimo Di Maro a,b , Angela Chambery a , Giovanni Colonna b,c , Susan Costantini c,d , Giuseppe Geraci e , Augusto Parente a,b,⁎ b

a Department of Life Sciences, Second University of Naples, Via Vivaldi 43, I-81100 Caserta, Italy Laboratorio Integrato per la Qualità e la Sicurezza degli Alimenti, Centro Regionale di Competenza “Produzioni Agroalimentari”, Second University of Naples, Italy c Department of Biochemistry and Biophysics and CRISCEB, (Research Center of Computational and Biotechnological Sciences), Second University of Naples, Via Costantinopoli 16, 80138 Naples, Italy d Institute of Food Science, CNR, Via Roma 52 A/C, 83100 Avellino, Italy e Department of Biological Sciences, Via Mezzocannone 8, University of Naples Federico II, Italy

Received 20 March 2006; received in revised form 18 July 2006; accepted 20 July 2006 Available online 25 July 2006

Abstract The colour of buffalo (Bubalus bubalis L.) meat is darker than bovine meat. Since meat colour depends on the concentration of myoglobin (Mb) and its oxidation state, we have determined the main structural and functional properties of buffalo Mb. Buffalo Mb was purified from longissimus dorsi muscles and its molecular mass determined by ESI Q-TOF mass spectrometry. The molecular mass 17,034.50 was 86.20 Da higher than the bovine Mb. This was confirmed by analysing its primary structure, using a combined approach based on Edman degradation and MALDI-TOF mass spectrometry. Comparing the amino acid sequences of both Mbs, we found three amino acid differences out of 153 amino acid residues. One is a conservative substitution (Dbov141Ebuf), and the other two (Abov19Tbuf and Abov117Dbuf) are nonconservative. These amino acid substitutions are unlikely to cause structural changes because they are located far from the heme binding pocket, as revealed by the 3D structure of buffalo Mb elaborated by homology modelling. Stability analyses show no difference with the bovine Mb for helix E and only minor differences in the stability values for helices A and G. Moreover, autoxidation rates of purified buffalo and bovine myoglobins at 37 °C, pH 7.2, were almost identical, 0.052 ± 0.001 h− 1 and 0.054 ± 0.002 h− 1, respectively, as were their oxygen-binding Kd values, 3.7 ± 0.1 μM and 3.5 ± 0.1 μM, respectively. The percent of MetMb values were almost identical. The results presented here suggest that the darker buffalo meat depends on factors other than the oxidation rate of its Mb, as, for example, the Mb content (0.393 ± 0.005 g/100 g of tissue) and consequently MetMb, which are almost twice as high as bovine meat (Mb: 0.209 ± 0.003 g/100 g of tissue). © 2006 Elsevier Inc. All rights reserved. Keywords: Autoxidation; Bubalus bubalis; Buffalo; Edman degradation; Homology modelling; Mass spectrometry; Meat colour; Myoglobin

1. Introduction Myoglobin (Mb) is the most important determinant for meat colour. In living animals, there is an equilibrium between the

⁎ Corresponding author. Dipartimento di Scienze della Vita, Seconda Università degli Studi di Napoli, Via Vivaldi, 43, I-81100 Caserta, Italy. Tel.: +39 0823 274583; fax: +39 0823 274571. E-mail address: [email protected] (A. Parente). 1096-4959/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.cbpb.2006.07.006

purplish-red Mb form (deoxyMb) and the cherry-red form (oxyMb). During meat storage, these two reduced Mb forms readily become oxidized to the brownish-red metMb (Faustman and Cassens, 1990). Over the last 20 years it has been shown that there is up to a 12-fold difference in the rate at which Mb oxidizes in postmortem mammalian muscles, depending on the species and muscle type (Livingston et al., 1986; Trout and Gutzke, 1995). For example, Livingston and Brown (1981) reported that fish Mbs are 2.5 times more susceptible to oxidation than mammalian

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

Mbs, due to a cysteinyl residue at position 9 (Cys 9). Similarly, the autoxidation rate of beluga whale (Delphinapterus leucas) pure oxymyoglobin is more rapid than that reported for terrestrial Mbs (Stewart et al., 2004). African elephant Mb shows an unusual autoxidation behaviour, probably due to the glutamine at position 64, in place of the distal histidine, which is conserved in almost all Mbs (Tada et al., 1998). Furthermore, research results also indicate that there are differences in the oxidation rate of Mbs isolated from different muscle types (Bembers and Satterlee, 1975; Foucat et al., 1994). The colour of fresh meat is an important property, influencing its market value and the consumer's purchase decision (Hood and Riordan, 1973; Liu et al., 1996). Recently, water buffalo meat, derived from male animals, is being introduced in the fresh meat market of Southern Italy, as an alternative to the bovine one. Compared to the latter, water buffalo meat is presented as having higher nutritional properties, although there are conflicting reports (Syed Ziauddin et al., 1993; Cutrignelli, 1996; Infascelli, 1996; Infascelli et al., 2004; Spanghero et al., 2004). Furthermore, until now Bovine Spongiform Encephalopathy (BSE) has never been observed in water buffalo. However, it is generally accepted that dry-aged buffalo meat becomes dark faster than bovine meat, discouraging consumer purchase. The aim of this study is to investigate whether the reasons for the faster darkening process of buffalo meat might depend on structural differences between buffalo and bovine Mbs. Therefore, we have: i) purified buffalo Mb; ii) determined its average amount in meat; iii) determined its primary structure and main structural properties, and iv) studied Kd values and autoxidation kinetics to establish the possible occurrence of correlation with the amino acid substitutions. 2. Materials and methods 2.1. Enzymes and chemicals HPLC-grade solvents and reagents were obtained from Carlo Erba (Milan, Italy). Endoproteinase Glu-C and trypsin (sequencing grade) were purchased from Sigma (Milan, Italy). Cyanogen bromide was obtained from Fluka (Milan, Italy). Reagents for automated Edman degradation were supplied by Applera (Monza, Italy). The following buffers were used: buffer A, 25 mM Na/P, pH 7.2, containing 0.35 M NaCl; buffer B, 10 mM Tris· Cl, pH 8,6; buffer C, 10 mM Tris· Cl containing 50 mM NaCl; buffer D, 25 mM phosphate buffer, pH 7.2, containing 50 μM EDTA. 2.2. Animal samples Myoglobin was purified from the Italian water buffalo (Bubalus bubalis) skeletal muscle (Longissimus dorsi), provided by “Cooperativa La Baronia”, Pontelatone, Caserta (Italy). The muscle was quickly withdrawn from the butchered animal and kept frozen at − 20 °C until use. Bovine Mb, which is not commercially available, was purified from the same type of muscle, obtained from a local butcher.

231

2.3. Extraction and purification procedure Both buffalo and bovine Mbs were purified essentially using the procedure reported by Balestrieri et al. (1973), with modifications. Frozen muscle (20 g), thawed overnight at 4 °C, was partially freed of gross fat and connective tissue, cut into small pieces, placed in Milli Q water (1:1; w:v) and homogenized in a Waring Blender homogenizer (Waring Products, Torrington, CT, USA). The homogenate was then centrifuged at 22,000 ×g for 40 min at 4 °C, in a rotor JA-14 (AVANT J-25 centrifuge; Beckman Coulter, Fullerton, CA (USA)). The supernatant fluid (about 25 mL) was filtered through Mirachlot paper (Calbiochem, San Diego, USA) to remove fatty material. The clear filtrate was then dialyzed, using a Spectra/Por® membrane (molecular mass cutoff: 6000–8000) for 24 h against three successive changes of Milli Q water (2 L). The white precipitate formed during the dialysis was removed by centrifugation for 15 min at 15,800 ×g. Supernatant (20 mL; protein concentration ≅ 10 mg/mL) was gelfiltered on a Sephacryl S-100 column (ϕ 2.6 × 115 cm, Amersham Pharmacia Biotech, Milan, Italy), equilibrated with buffer A. The column was then eluted with the same buffer at a flow rate of 35 mL/h, collecting 5 mL fractions. The eluate was monitored at 280 and 409 nm. The Mb-containing fractions were concentrated using an ultrafiltration cell (Amicon Inc., Beverly, USA; membrane molecular mass cut-off: 10,000) and then dialyzed against 2 L of buffer B (three changes). The dialyzed protein solution was centrifuged at 15,800 ×g for 15 min, at 4 °C. The supernatant was absorbed onto a DEAE-Sepharose column (ϕ 3 × 8 cm, BioRad, Milan, Italy), equilibrated with buffer B and eluted with a gradient made up by buffer B and buffer C (1 L final volume), at a flow rate of 1.5 mL/min. The eluate absorbance was measured at 280 and 409 nm and the sample purity checked by SDS-PAGE (Laemmli and Favre, 1973). For determination of both bovine and buffalo tissue Mb content, analytical separations have been used (Dosi, 2005). Mb concentration was determined at the isosbestic point of 527 nm using a E1%1cm coefficient of 2.7. The Mb amount is expressed in g/100 g of tissue. Samples were analyzed in triplicate for each Mb source and the mean values presented. Determination of percent values of MetMb in the purified preparations of buffalo and bovine samples, collected after the same time from slaughtering, were performed spectroscopically according to Krzywicki (1982). 2.4. Autoxidation rate measurement The autoxidation of MbO2 to metMb was monitored by recording, over time, the changes of the absorption spectrum in the range 500–700 nm at measured time intervals, and estimating the absorbance decrease at 582 nm. Spectra were collected every 15 min for a time sufficient to characterize the autoxidation process (5 h in total). Triplicate experiments were performed, each with freshly prepared MbO2. Absorption spectra in the visible were recorded on a Hewlett Packard UV-visible spectrophotometer (Model 8453), equipped with a thermostatically controlled (within ±0.1 °C) cell holder, set at 37 °C.

232

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

MbO2 was prepared by reduction of pure Mb in buffer D by addition of sodium dithionite (1 mg/mL final concentration), and immediate desalting through a small Sephadex G-25 column (9.1 mL), equilibrated and eluted with the same buffer. The Mb solution was immediately placed in the spectrophotometer cuvette and absorbance measurements started. Absorbance data were analyzed by plotting the log values of the % decreasing oxyMb concentration vs. time. The apparent first-order rate constants were determined from this type of graphic analysis. Data were statistically analyzed using the ANOVA test. Linear interactions were tested for significance ( p b 0.05). 2.5. Determination of buffalo myoglobin primary structure 2.5.1. Preparation of apomyoglobin For structural studies, reversed-phase HPLC on a C4 column (150 × 4.6 mm, Alltech, Italy) was used to separate apo-Mb from the heme group, applying a linear gradient of 0.1% TFA and acetonitrile containing 0.1% TFA, from 5 to 65%, over 60 min, at a flow-rate of 1 mL/min, monitoring the absorbance both at 214 and 280 nm. 2.5.2. N-terminal sequence determination Automated Edman degradation was performed on the first 41 amino acid residues of the HPLC purified buffalo apoMb, on an Applied Biosystems Procise sequencer, Model 491 (Parente et al., 1993). 2.5.3. Mass spectrometry The relative molecular mass (Mr) of whole myoglobin was determined by a Q-TOF Micro mass spectrometer (Waters, Milford, MA USA) equipped with a CapLC system. The capillary source voltage and the cone voltage were set at 3000 V and 43 V, respectively. The source temperature was kept at 80 °C and nitrogen was used as drying gas (flow rate about 50 L/h). The sample was brought up with a mixture of acetonitrile, 0.2%

formic acid in water (50:50, v:v) to a concentration of 1 pmol/μL and infused into the system at a flow rate of 5 μL/min. The acquisition and deconvolution of data were performed on a Mass Lynx Windows NT PC data system. Mass spectra of peptides obtained by chemical and enzymatic digestion were analyzed with a MALDI-TOF micro LR (Waters Micromass Co., Manchester, UK). Briefly, RP-HPLC eluted peptides were concentrated and analyzed. Prior to the acquisition of spectra, 1 μL of each peptide solution was mixed with 1 μL of saturated α-cyano-4-hydroxycinnamic acid matrix solution [10 mg/mL in ethanol:water (1:1; v:v), containing 0.1% trifluoroacetic acid] and a droplet of the resulting mixture (1 μL) placed on the mass spectrometer's sample target. The droplet was dried at room temperature. Once the liquid was completely evaporated, the sample was loaded into the mass spectrometer and analyzed. The instrument was externally calibrated using a tryptic alcohol dehydrogenase digest (Waters, Milford, MA USA) in reflectron mode. A mass accuracy near to the nominal (50 ppm) was achieved for each standard. All spectra were processed and analyzed using the MassLynx 4.0 software (Waters, Milford, MA USA). 2.5.4. Protein digestions Buffalo and bovine apo-Mbs (100 μg) were digested with trypsin (Sigma, Milan, Italy), in 0.1 M Tris·Cl, pH 8.5, containing 20 mM CaCl2 and 10% acetonitrile. The enzyme was added in three steps with a final enzyme-to-substrate ratio of 1:50 (w:w) and the reaction was carried out at 37 °C for 24 h. Glu-C (Sigma, Milan, Italy) hydrolysis of buffalo apo-Mb was performed in 0.05 M phosphate buffer, containing 0.5 M Gdn·Cl and 1% CH3CN. Chemical digestion with cyanogen bromide (CNBr, Fluka—Milan, Italy) was carried out as described (Gross, 1967). 2.5.5. Peptide separations Separation of tryptic, endoproteinase Glu-C and cyanogen bromide peptides by RP-HPLC was obtained on a Waters instrument, using a C18 column (Alltech; 0.46 × 25 cm; 5 μm

Fig. 1. Deconvoluted ESI-QTOF mass spectrum of buffalo apo-myoglobin and, (insert) SDS-PAGE of purified buffalo muscle Mb (lane 2) compared to protein molecular mass standards (lane 1).

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

particle size), at a flow-rate of 1 mL/min. Insoluble material after digestions, was separated by centrifugation, solubilized with formic acid (50% final concentration) and analyzed similarly. Peptide elution was obtained using linear gradients of 0.1% TFA and acetonitrile containing 0.1% TFA, from 5 to 55%, over 150 min. Peptides elution was monitored at 214 nm. 2.5.6. Peptide recognition Signals recorded in the mass spectra were associated with the corresponding peptides on the basis of the expected molecular mass calculated by using a suitable computer program (Peptide Tools, Hewlett-Packard). When necessary, Edman degradation steps were performed on the HPLC purified peptides, in order to confirm the assignment. 2.6. Protein modelling The three-dimensional models of bovine and buffalo myoglobins were created following the homology modelling

233

strategy already used with success by our group (Facchiano et al., 2001; Scapigliati et al., 2004; Costantini et al., 2005). The searches for sequence similarity within databases were performed with the BLAST program (Altschul et al., 1990). These searches have evidenced that sequence identities higher than 85% exist between both bovine and buffalo Mbs and the corresponding proteins in horse and pig, so that the homology modelling strategy can be applied with success by using the experimental models of horse (PDB code: 1WLA) and pig Mbs (PDB code: 1MWD) as template (Maurus et al., 1997; Krzywda et al., 1998). The alignment of the protein sequences was made with CLUSTALW program (Thompson et al., 1994) The program MODELLER (Sali and Blundell, 1993) implemented in the Quanta molecular simulation package (Accelrys, San Diego, CA) was used to build 10 full-atom models by setting 4.0 Å as RMS deviation among the structures of the templates and fully optimized models, with multiple cycles of refinement with conjugate gradient minimization and molecular dynamics with simulated annealing. To select the best model among them,

Fig. 2. RP-HPLC separation of tryptic peptides obtained from buffalo (a) and bovine (b) Mbs. Arrows indicate peptide peaks containing amino acid substitutions.

234

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

Table 1 Amino acid sequences and molecular weights determined by MALDI-TOF mass spectrometry of tryptic (T), endoproteinase Glu-C (G) and cyanogen bromide (CB) peptides used to assemble the amino acid sequence of buffalo Mb Peptide

MW

N-terminal – T-1

T-5 T-7 T-9 T-10 T-12

1304.321 1006.103 721.544 940.301 1270.238 1530.107 1622.040 748.505 1393.464 1770.227 3427.916

G-1 G-3 G-4 G-5 G-6 G-7 G-13 CB-3

696.247 1421.782 1622.855 1666.271 1322.526 1282.669 2002.402 8308.166

T-2 T-3 T-4

Sequence

Note a

GLSDGEWQLV LNAWGKVETD 1 → 41 VAGHGQEVLI RLFTGHPETL E KG HHEAEVK 79 → 87 HLA ESHANK 88 → 96 HKIP VK 97 → 102 N EMAAQYK 140 → 147 LFTGHPETL EK 32 → 42 HP SDFGADAQAA MSK 119 → 133 VETD VAGHGQEVLI R 17 → 31 ALELFR 134 → 139 HGNTVLT ALGGILK 64 → 77 GLSDGEWQLV LNAWGK 1 → 16 YLEFISDA IIHVLHDKHP 103 → 133 SDFGADAQAA MSK VKHLA E 86 → 91 KFDKFKHLK TE 42 → 52 KFDKFKHLK TEAE 42 → 54 SHANKHKIP VKYLE 92 → 105 MAAQYKVLG FHG 142 → 153 VLI RLFTGHPE 28 → 38 GLSDGEWQLV LNAWGKVE 1 → 18 KASED LKKHGNTVLT ALGGILKKKG 56 → 131 hheaevkhla eshankhkip vkylefisda iihvlhdkhp sdfgadaqaa m

The N-terminal sequence is also shown. Lower case letters indicate amino acid residues of peptide CB3 not sequenced. a Peptide sequence position.

their stereochemical quality was evaluated with the program PROCHECK (Laskowski et al., 1993), in particular by comparing the Ramachandran plots and the related statistical

analysis. Search for structural classification was performed in CATH (Orengo et al., 1997; Pearl et al., 2000) database. The models were analyzed with DSSP program to assign secondary structure (Kabsch and Sander, 1983). In the obtained models the docking of heme into its binding pocket was performed using as starting reference the positions of the hemes in the crystallographic templates. The “Protein–Protein Interaction Server” and the program NACCESS (Hubbard et al., 1991) were used to identify the amino acid residues in the Mb heme binding pocket. Model figures were drawn with InsightII package (Accelrys, San Diego, CA). The total contribution of helices to protein stability was evaluated by considering five terms according to Munoz and Serrano (1995) and Facchiano et al. (1998). 3. Results 3.1. Myoglobin purification and quantitation The purification procedure described in Materials and methods yielded reproducibly purified preparations of Mbs from the skeletal muscle of both water buffalo and bovine. The insert of Fig. 1 shows the SDS-PAGE of buffalo purified Mb with no detectable contaminating protein bands. The amount of Mb in samples from both animal sources was 0.393 ± 0.005 g/100 g of tissue for buffalo and 0.209 ± 0.003 g/ 100 g of tissue for bovine. These values are consistent with those reported for mammalian skeletal muscles (Fox and Condon, 1981). Percent MetMb values in the purified buffalo and bovine samples, after the same time from slaughtering, were 57% and 47%, respectively.

Fig. 3. Amino acid sequence of B. bubalis Mb, including the peptides obtained by treatment of Mb with trypsin, Glu-C endoproteinase and cyanogen bromide, compared to Bos taurus Mb. Substitutions are underlined.

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

235

Table 2 Summary of amino acid identities and similarities among Mb sequences

Horse Pig Bovine Buffalo

Horse

Pig

Bovine

Buffalo

100

90 (93) 100

88 (93) 88 (91) 100

86 (91) 86 (89) 98 (98) 100

Numbers indicate the percentage identities, in parenthesis are indicated the percentage similarities.

3.2. Primary structure of buffalo myoglobin The primary structure of buffalo myoglobin was obtained using the strategy described by Di Maro et al. (2001). Initially, the molecular weights of both buffalo and bovine apo-Mbs, obtained from RP-HPLC, were determined by ESI-QTOF analysis, giving the values of 17,034.50 and 16,948.50, respectively. The transformed mass spectrum of buffalo Mb is shown in Fig. 1. The bovine Mb molecular mass was similar, within the experimental error, to that calculated from the gene sequence (16,946 Da; Han et al., 1970). The mass difference of 86.20 Da between buffalo and bovine Mbs indicated that they might have different primary structure. This possibility has been confirmed by automated Edman degradation analysis of the sequence of the first 41 amino acid residues, which, when compared with the sequence of bovine Mb, showed one amino acid difference at position 19 (Abov19Tbuf). To map the complete buffalo Mb sequence, both buffalo and bovine Mbs were digested with trypsin and the resulting peptides, separated by RP-HPLC (Fig. 2), were analyzed by MALDI-TOF mass spectrometry (Table 1). An overlay of the RP-HPLC separation patterns showed that most tryptic peptides from the two proteins had the same elution times and also identical masses. Therefore, they were considered to have identical amino acid sequence (unless the isobaric Leu/Ile residues). A mass difference of 1558 Da was found between tryptic peptides T12 (buffalo) and

Fig. 5. 3D molecular model of buffalo Mb. Backbone ribbon and the secondary structure topology are shown: black cylinders represent alpha helices. Heme is reported in stick and in gray. Three different amino acids in two sequences are indicated in cpk and in gray.

T12′ (bovine) and Edman degradation of T12′ revealed that its sequence covers positions 103–118 in the bovine Mb sequence. Peptide T12 results from a miss-cleavage at Lys118. Indeed, it has an additional stretch of amino acids up to lysyl 133 (Table 1), showing the same sequence up to residue 116 and a substitution (Abov117Dbuf) at position 117. The difference in the elution time between T12 and T12′ is likely related both to their difference in length and to the substitution found. Moreover, Edman degradation of buffalo Mb tryptic peptide T3 revealed that its sequence covers position 140–147 and this allowed us to find a third substitution at position 141 (Dbov141Ebuf). Finally, to map the entire sequence, it was necessary to cleave buffalo Mb with endoproteinase Glu-C and CNBr. The two sets

Fig. 4. Multiple sequence alignment of horse, pig, bovine and buffalo Mb sequences. Stars (★) indicate amino acids conserved within the reported sequences. Amino acids in bold indicate eight α-helices and amino acids underlined indicate a short 310 helix in pig Mb. The three amino acid difference between bovine and buffalo Mbs are evidenced with arrows.

236

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

Table 3 RMS deviations (expressed in Å) between pairs backbones of myoglobin models

Horse Pig Bovine Buffalo

Horse

Pig

Bovine

Buffalo

0

0.97 0

0.9 0.36 0

0.87 0.39 0.17 0

of peptides, separated by RP-HPLC, were analyzed by MALDITOF mass spectrometry (Table 1). On the basis of their molecular masses, we confirmed that the sequence differences between buffalo and bovine Mbs are only the three found. This conclusion was further confirmed by the sequences of peptides G-4, G-6 and CB-3 (up to residue 80), determined by Edman degradation. The complete sequence of buffalo Mb, compared to the bovine one, and the strategy employed are reported in Fig. 3. 3.3. Modelling of bovine and buffalo myoglobins The sequences of bovine (derived from the gene sequence) and buffalo (this paper) Mbs have been analyzed by computer programs in order to find similar sequences in databases and perform structural predictions to obtain a theoretical 3D model of these proteins. The BLAST (Altschul et al., 1990) search has resulted in two experimental three-dimensional models of myoglobin from horse and pig that could be considered as template for modelling both bovine and buffalo sequences. The BLAST pairwise alignments between each target and the horse/ pig sequences returned amino acid identities higher of 86% (see Table 2). In these conditions, a homology modelling strategy can be applied with very good results (Rodriguez et al., 1988; Westhead and Thornton, 1998; Martin et al., 1997; Tramontano, 1998), because the alignment between the new sequences and Table 4 Helix stability contributions (expressed in kJ/mol) a in four myoglobin models concerning total structures and each one of the eight helices

Total chains ΔΔGH bond ΔΔGintrinsic ΔΔGnonH ΔΔGdipole ΔΔGSD ΔΔGtotal Helices ΔΔGtotal (Helix A) ΔΔGtotal (Helix B) ΔΔGtotal (Helix C) ΔΔGtotal (Helix D) ΔΔGtotal (Helix E) ΔΔGtotal (Helix F) ΔΔGtotal (Helix G) ΔΔGtotal (Helix H)

Horse

Pig

Bovine

Buffalo

− 340.61 488.63 − 13.48 − 13.27 − 31.77 89.50

− 340.66 485.37 − 12.98 − 13.14 − 40.65 77.98

−360.08 492.90 − 13.48 − 11.93 − 35.46 71.96

− 353.63 489.55 − 13.48 − 10.93 − 35.96 75.60

10.21 15.20 16.62 − 0.79 17.83 6.24 10.88 − 4.31

13.48 15.20 17.43 − 0.79 17.83 6.24 14.73 − 8.54

10.51 14.53 17.46 − 0.67 16.95 16.74 12.85 1.13

9.96 13.48 18.06 − 1.47 17.83 16.70 3.43 − 0.063

a Five terms contribute to the total free energy of helices and to the overall stability of the protein. The most stabilizing effect corresponds to the lower value of this free energy.

the reference models can be obtained with no gaps. The alignment of four protein sequences was made with CLUSTALW program (Thompson et al., 1994) (Fig. 4). On the basis of this sequence alignment, we created ten structural models of bovine Mb, as well as of buffalo Mb, by using the Modeller program (Sali and Blundell, 1993) and the experimental horse and pig models as templates. The backbones of the ten models overlap properly and the most reliable structure was chosen by evaluating the stereochemical quality of the models using the PROCHECK package (Laskowski et al., 1993). In these obtained models, the heme (bound to His 93) was inserted in its binding pocket using as reference the experimental structures of horse and pig proteins. Fig. 5 shows the representation of the modelled protein structure for buffalo, with secondary structure topology. Eight alpha helices (indicated with black cylinders), involving about 76–77% of the sequence, have been assigned by DSSP program and define the global structure as globin-like with orthogonal bundle architecture. Therefore, the models can be considered as a “mainly α” fold, in agreement to the structural classification reported by CATH database (Orengo et al., 1997; Pearl et al., 2000) for the reference structures of horse and pig myoglobin. Three different amino acids in two sequences (Abov19Tbuf, Abov117Dbuf and Dbov141Ebuf) are located on the surface and far from the heme binding pocket, so that they probably do not cause evident structural changes. The Mb models were compared by structural superimposition, obtaining RMSD values (Table 3) and in terms of secondary structures. The superposition of backbones of bovine and buffalo models with those of horse and pig (ones) gives RMSD values of about 0.9 and 0.36 Å, respectively. These values, compared to the 0.97-Angstrom value, obtained by superimposition of horse and pig myoglobins, suggest that our target sequences are well compatible with experimental template structures and that the models obtained by homology are more similar to pig structure. Moreover, RMSD value of 0.17 Å between the bovine and buffalo Mb models indicates that two conformations are very much alike. The comparison of

Fig. 6. Time dependent spectral changes in the autoxidation of buffalo MbO2. Isosbestic points are at 527 and 595 nm. The final spectrum is that of the acidic met form. Protein concentration: 0.07 mg/mL.

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

secondary structures, assigned by DSSP program in all four models and reported in Fig. 4, shows that helices B, F, G and H are completely conserved among species, but few external residues can result added or excluded in the helices A, C, D and E. In detail, horse Mb compared with other models has one less amino acid in helices A and D, and two in helix E. Moreover, helix C of bovine Mb includes six amino acids in comparison to four present in other models. This could probably depend on the short 310 helix present in pig Mb between the helices C and D. The Mb amino acids at the interface with the heme, identified as reported in Methods, are well conserved in various organisms except few substitutions that do not cause evident functional differences (data not shown). Finally, we analyzed the stability of helical regions according to the method of Munoz and Serrano (1995) and Facchiano et al. (1998). As shown in Table 4, the calculation of the energetic contribution of alpha helices to the global energy of the structure indicates that the helices in horse model are more destabilized in comparison to those in other three species and that the total free energy of helices (ΔΔGtotal) in the bovine myoglobin is less than −1 kcal/mol lower than that in buffalo. A detailed analysis of the global energy shows no differences for the B, C, D and E helices among four species but few differences can be evidenced for the others. In particular, helix F is more stabilized in bovine and buffalo myoglobins with respect to other models. Analysis of helices A, G and H, each one bearing one of the three different amino acids found between bovine and buffalo myoglobins, Abov19Tbuf, Abov117Dbuf and Dbov141Ebuf, showed that while helix H in buffalo is more stabilized helices A and G are destabilized. 3.4. Kinetic studies The results of the determinations of the autoxidation rate constant are shown in Fig. 6, which reports an example of the spectral changes during the autoxidation reaction of buffalo MbO2 in 25 mM phosphate buffer (pH 7.2), at 37 °C. The spectra changed with a set of isosbestic points at 527 and 595 nm, and the final spectrum was that of the acidic met form, obtained by sodium nitrite treatment. The first order rate constants (k) of buffalo and bovine MbO2 autoxidation, obtained by plotting the decreasing of MbO2 concentration vs. time, were almost identical, with values 0.052 ± 0.001 h− 1 and 0.054 ± 0.002 h− 1, respectively. 4. Discussion The aim of the present study was to determine buffalo Mb structural characteristics, to compare these with those of bovine Mb and possibly to relate the differences of buffalo Mb to its autoxidation behaviour. The combined use of Mass Spectrometry and automated Edman degradation for the analysis of peptides obtained from trypsin, endoproteinase Glu-C and cyanogen bromide digestions, allowed us to obtain the entire sequence of B. bubalis myoglobin (also providing the correct overlapping of the fragments obtained with the cleavage methods). The amino acid residues not positioned through measurements of the molecular mass of the pertinent peptide, were obtained by Edman degradation. Finally,

237

the agreement between the experimental mass of the native protein (17,034.5 Da) and that calculated on the basis of the amino acid sequence (17,034.46 Da) is a further confirmation of the correctness of the analysis. The amino acid sequence of buffalo Mb, compared with the bovine one, shows three amino acid substitutions (1.96%) out of 153 amino acid residues, one of which is conservative Dbov141Ebuf and two non-conservative, Abov19Tbuf and Abov117Dbuf. These substitutions account well for the difference of 86 Da between the two Mbs. The three-dimensional structures of bovine and buffalo Mbs, obtained by homology modelling are very much alike and, also, the Mb amino acids at the interface with the heme are well conserved in both organisms. The three amino acid substitutions between buffalo and bovine Mbs are located on the surface of the protein and far from the heme binding pocket and do not cause appreciable structural changes. It appears unlikely that they may be responsible of changes in molecular properties. Nevertheless, helices A and G in buffalo Mb are more destabilized. This last result may be explained considering that, in buffalo, helix A, in position 19, has a threonyl amino acid residue, that is a destabilizing β-branched residue (Abov19Tbuf). In fact, it is known that an overall greater stability of helices in proteins is related to a combination of stabilizing factors, the most common being the low content of β-branched residues. Moreover, the presence of a negatively charged residue (Abov117Dbuf) at the C-terminus of helix G, certainly induces a destabilization of the helix dipole (Richardson and Richardson, 1988). Buffalo and bovine MbO2, at pH 7.2 and 37 °C, are equally resistant to autoxidation. The O2 dissociation curves of purified bovine and buffalo Mbs, in identical experimental conditions at pH 7.2, were almost superimposable, providing similar Kd values (3.7 ± 0.1 μM and 3.5 ± 0.1 μM, respectively). These observations indicate that the structural architecture of the heme pockets in the two proteins is similar, with similar functional properties. Therefore, Mb autoxidation rate does not influence buffalo meat colour. This might depend on the higher Mb content found, likely together with other factors (Faustman and Cassens, 1990). Acknowledgments This work was supported by Regione Campania, Italy (Centro Regionale di Competenza “Produzioni Agroalimentari”, in the framework of the Project line C: “Carne di bufalo” (P.O.R. 2000–2006, Misura 3.16)) and by funds from the Second University of Naples. We thank Dr. Angelo M. Facchiano of Istituto di Scienze dell’Alimentazione, CNR, Avellino (Italy), for useful suggestions on computational analysis. R. Dosi thanks the Camera di Commercio of Caserta (Italy) for a fellowship. References Altschul, S.F., Gish, W., Miller, W., Myers, E., Lipman, D.J., 1990. Best local alignment search tool. J. Mol. Biol. 215, 403–410. Balestrieri, C., Colonna, G., Irace, G., 1973. The skeletal muscle myoglobin of water buffalo (Bos bubalus L.). Comp. Biochem. Physiol. B 46, 667–672.

238

R. Dosi et al. / Comparative Biochemistry and Physiology, Part B 145 (2006) 230–238

Bembers, M., Satterlee, L.D., 1975. Physico-chemical characterization of normal and PSE porcine muscle myoglobins. J. Food Sci. 40, 40. Costantini, S., Colonna, G., Rossi, M., Facchiano, A.M., 2005. Modelling of HLA-DQ2 and simulations of its interaction with gluten peptides to explain molecular recognition in celiac disease. J. Mol. Graph. Model. 23, 419–431. Cutrignelli, M.I., 1996. Proceedings of XXXI International Symposium of Zootechny, Milano, Italy. Di Maro, A., Ferranti, P., Mastronicola, M., Polito, L., Bolognesi, A., Stirpe, F., Malorni, A., Parente, A., 2001. Reliable sequence determination of ribosome inactivating proteins by combining electrospray mass spectrometry and Edman degradation. J. Mass Spectrom. 36, 38–46. Dosi, R., 2005. Caratterizzazione strutturale, chimico-fisica e cinetica della mioglobina di bufalo (Bubalus bubalis). PhD Thesis. Facchiano, A.M., Colonna, G., Ragone, R., 1998. Helix stabilizing factors and stabilization of thermophilic proteins: an X-ray based study. Protein Eng. 11, 753–760. Facchiano, A.M., Stiuso, P., Chiusano, M.L., Caraglia, M., Giuberti, G., Marra, M., Abruzzese, A., Colonna, G., 2001. Homology modelling of the human eukaryotic initiation factor 5A (eIF-5A). Protein Eng. 14, 881–890. Faustman, C., Cassens, R.G., 1990. The biochemical basis for discoloration in fresh meat: a review. J. Muscle Foods 1, 217–243. Foucat, L., Renerre, M., Gatellier, P., Anton, M., 1994. 1H NMR study of bovine myoglobin autoxidation. Influence of muscle type and time post-mortem. Int. J. Food Sci. Technol. 29, 1–8. Fox, P.F., Condon, J.J., 1981. Food Proteins. Applied Science Publishers, London. Gross, E., 1967. The cyanogen bromide reaction. Methods Enzymol. 11, 238–257. Han, K., Dautrevaux, M., Chaila, X., Biserte, G., 1970. The covalent structure of beef heart myoglobin. Eur. J. Biochem. 16, 465–471. Hood, D.E., Riordan, E.B., 1973. Discolouration in pre-packaged beef measurement by reflectance spectrophotometery and shopper discrimination. J. Food Technol. 8, 333–343. Hubbard, S.J., Campbell, S.F., Thornton, J.M., 1991. Molecular recognition. Conformational analysis of limited proteolytic sites and serine proteinase protein inhibitors. J. Mol. Biol. 220, 507–530. Infascelli, F., 1996. Proceedings of XXXI International Symposium of Zootechny, Milano, Italy. Infascelli, F., Gigli, S., Campanile, G., 2004. Buffalo meat production: performance infra vitam and quality of meat. Vet. Res. Commun. 1 (suppl.28), 143–148. Kabsch, W., Sander, C., 1983. Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features. Biopolymers 22, 2577–2637. Krzywda, S., Murshudov, G.N., Brzozowski, A.M., Jaskolski, M., Scott, E.E., Klizas, S.A., Gibson, Q.H., Olson, J.S., Wilkinson, A.J., 1998. Stabilizing bound O2 in myoglobin by valine68 (E11) to asparagine substitution. Biochemistry 37, 15896–15907. Krzywicki, K., 1982. The determination of haem pigments in meat. Meat Sci. 7, 29–36. Laemmli, U.K., Favre, M., 1973. Maturation of the head of the bacteriophage T4. J. Mol. Biol. 80, 575–589. Laskowski, R.A., MacArthur, M.W., Moss, D.S., Thornton, J.M., 1993. PROCHECK: a program to check the stereochemical quality of protein structures. J. Appl. Crystallogr. 26, 283–291. Liu, Q., Scheller, K.K., Arp, S.C., Schaefer, D.M., Williams, S.N., 1996. Titration of fresh meat color stability and malondialdehyde development with Holstein steers fed vitamin E-supplemented diets. J. Anim. Sci. 74, 117–126. Livingston, D.J., Brown, W.D., 1981. The chemistry of myoglobin and its reactions. Food Technol. 35, 244–252. Livingston, D.J., Watts, D.A., Brown, W.D., 1986. Myoglobin interspecies structural differences. Effects on autoxidation and oxygenation. Arch. Biochem. Biophys. 249, 106–115.

Martin, A.C.R., MacArthur, M.W., Thornton, J.M., 1997. Assessment of comparative modelling in CASP2. Proteins Suppl. 1, 14–28. Maurus, R., Overall, C.M., Bogumil, R., Luo, Y., Mauk, A.G., Smith, M., Brayer, G.D., 1997. A myoglobin variant with a polar substitution in a conserved hydrophobic cluster in the heme binding pocket. Biochim. Biophys. Acta 1341, 1–13. Munoz, V., Serrano, L., 1995. Elucidating the folding problem of helical peptides using empirical parameters. II. Helix macrodipole effects and rational modification of the helical content of natural peptides. J. Mol. Biol. 245, 275–296. Orengo, C.A., Michie, A.D., Jones, S., Jones, D.T., Swindells, M.B., Thornton, J.M., 1997. CATH — a hierarchic classification of protein domain structures. Structure 5, 1093–1108. Parente, A., Verde, C., Malorni, A., Montecucchi, P.C., Aniello, F., Geraci, G., 1993. Amino-acid sequence of the cooperative dimeric myoglobin from the radular muscles of the marine gastropod Nassa mutabilis. Biochim. Biophys. Acta 1162, 1–9. Pearl, F.M.G., Lee, D., Bray, J.E., Sillitoe, I., Todd, A.E., Harrison, A.P., Thornton, J.M., Orengo, C.A., 2000. Assigning genomic sequences to CATH. Nucleic Acids Res. 28, 277–282. Richardson, J.S., Richardson, D.C., 1988. Amino acid preferences for specific locations at the ends of alpha helices. Science 240, 1648–1652. Rodriguez, R., Chinea, G., Lopez, N., Pons, T., Vriend, G., 1988. Homology modelling, model and software evaluation: three related resources. Bioinformatics 14, 523–528. Sali, A., Blundell, T.L., 1993. Comparative protein modelling by satisfaction of spatial restraints. J. Mol. Biol. 234, 779–815. Scapigliati, G., Costantini, S., Colonna, G., Facchiano, A., Buonocore, F., Bossú, P., Holland, J.W., Secombes, C.J., 2004. Modelling of fish interleukin 1 and its receptor. Dev. Comp. Immunol. 28, 429–441. Spanghero, M., Gracco, L., Valusso, R., Piasentier, E., 2004. In vivo performance, slaughtering traits and meat quality of bovine (Italian Simmental) and buffalo (Italian Mediterranean) bulls. Livest. Prod. Sci. 91, 129–141. Stewart, J.M., Blakely, J.A., Karpowicz, P.A., Kalanxhi, E., Thatcher, B.J., Martin, B.M., 2004. Unusually weak oxygen binding, physical properties, partial sequence, autoxidation rate and a potential phosphorylation site of beluga whale (Delphinapterus leucas) myoglobin. Comp. Biochem. Physiol., B 137, 401–412. Syed Ziauddin, K., Mahendrakar, N.S., Rao, D.N., Ramesh, B.S., Amla, B.L., 1993. Observations on some chemical and physical characteristics of buffalo meat. Meat Sci. 37, 103–105. Tada, T., Watanabe, Y., Matsuoka, A., Ikeda-Saito, M., Imai, K., Ni-hei, Y., Shikama, K., 1998. African elephant myoglobin with an unusual autoxidation behavior: comparison with the H64Q mutant of sperm whale myoglobin. Biochim. Biophys. Acta 1387, 165–176. Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. Clustal W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22, 4673–4680. Tramontano, A., 1998. Homology modelling with low sequence identity. Methods 14, 293–300. Trout, G.R., Gutzke, D., 1995. The relationship between the discolouration rate during refrigerated display and the autoxidation rate of purified myoglobin: the effects of muscle type and species. Proceedings of 41st International Congress of Meat Science and Technology, San Antonio, TX, p. 404. Westhead, D.R., Thornton, J.M., 1998. Protein structure prediction. Curr. Opin. Biotechnol. 9, 383–389.