Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology, Part D journal homepage: www.elsevier.com/locate/cbpd

Biochemical and proteomic characterisation of haemolymph serum reveals the origin of the alkali-labile phosphate (ALP) in mussel (Mytilus galloprovincialis) Caterina Oliveri a, Lorena Peric c, Susanna Sforzini a, Mohammed Banni a,b, Aldo Viarengo a, Maria Cavaletto a, Francesco Marsano a,⁎ a b c

Department of Science and Technological Innovation, University of Piemonte Orientale “Amedeo Avogadro”, Viale Teresa Michel, 11, 15121 Alessandria, Italy Laboratory of Biochemistry and Environmental Toxicology, ISA Chott-Mariem, 4042 Sousse, Tunisia Ruder Bošković Instititute, Center for Marine Research, G. Paliaga 5, 52210 Rovinj, Croatia

a r t i c l e

i n f o

Article history: Received 27 May 2014 Received in revised form 16 July 2014 Accepted 17 July 2014 Available online 24 July 2014 Keywords: Haemolymph serum Mytilus galloprovincialis Alkali-labile phosphate (ALP) Vitellogenin (Vtg)-like proteins Proteomics

a b s t r a c t Mollusc haemolymph proteins are known to play several important physiological roles in the immune system, heavy metal transport and the tissue distribution of lipophilic compounds. In this study, we analysed acetone-extracted proteins from mussel haemolymph by one- and two-dimensional gel electrophoresis. The proteins were identified by comparing mass spectrometry data with the invertebrate EST database, allowing us to establish the mussel haemolymph serum proteome. Extrapallial protein (EP) precursor represents the most abundant serum protein; astacin and CuZn superoxide dismutase were also detected. Slight contamination from muscle proteins, due to the sampling method, was also found. No differences were observed in the profiles obtained for male and female serum proteins. One aspect of interest was the previously reported finding that alkali-labile phosphate (ALP) from haemolymph serum may be representative of vitellogenin (vtg)-like protein content in the circulatory fluid of molluscs. In our analysis of mussel haemolymph serum, vitellogenin-like proteins were never found. To confirm these data, a typical methyl-tert-butyl-ether (MTBE) extraction, which is specific for vtg-like proteins, was performed, and the results of the electrophoretic analyses were compared with those obtained by acetonic precipitation. The results showed that the electrophoretic profiles are similar and that vtg-like proteins cannot be identified. Moreover, the main phosphoprotein present in female and male extracts is EP protein precursor. In addition, agarose gel electrophoresis demonstrates that high-molecular-weight forms of vtg-like proteins are not detectable. © 2014 Elsevier Inc. All rights reserved.

1. Introduction Bivalve haemolymph cells (i.e., haemocytes) play an important role in the internal immune defence, and it is hypothesised that these cells can be involved in numerous other processes including wound-healing, shell repair, transport and the intracellular digestion of nutrients (Cheng, 1981, 1984; Fisher, 1986; Wootton and Pipe, 2003; Hong et al., 2006; Hégaret et al., 2011). The proteins present in the haemolymph serum in mussels are also known to be involved in the immune response and to bind a variety of metals (Koutsogiannaki and Kaloyianni, 2010; Itoh et al., 2011; Seufi et al., 2012; Xue et al., 2012). In recent years, although technical advances in proteomics (i.e., twodimensional gel electrophoresis and mass spectrometry) and its application in the field of comparative biochemistry, physiology and environmental research have increased significantly, few data have ⁎ Corresponding author. Tel.: +39 131360229; fax: +39 131360243. E-mail address: [email protected] (F. Marsano).

http://dx.doi.org/10.1016/j.cbd.2014.07.003 1744-117X/© 2014 Elsevier Inc. All rights reserved.

been produced regarding the Mytilus sp. haemolymph serum proteome (Riffeser and Hock, 2002). An aspect of great interest was the proposal that in bivalves the alkali-labile phosphate (ALP) from haemolymph serum proteins can be representative of the vitellogenin (vtg)-like protein content (Blaise et al., 1999). This method, based on the determination of labile phosphates released by proteins after alkali hydrolysis, has been used in plasma and tissues as an indicator of endocrine disruption due to the effects of the presence of xenoestrogens in the environment. Despite genomic research aimed at identifying specific genes associated with maturation and oestrogen exposure in Mytilus edulis (Ciocan et al., 2011), there are still no definitive data regarding the identification of vitellogenin-like proteins in mollusc serum. The aim of the present study is to analyse the haemolymph serum proteome of Mytilus galloprovincialis and to identify the main expressed protein(s). Another important aspect of this research, given the use of phosphorylatable proteins as environmental biomarkers for the assessment of endocrine disruptor chemicals (EDCs), is to identify such protein(s) in the serum. One- and two-

30

C. Oliveri et al. / Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

dimensional electrophoresis, western blotting followed by in situ digestion and mass spectrometric analysis were performed both in male and female individuals. The results provide a snapshot of the most abundant proteins present in the haemolymph; however, vitellogenin-like proteins cannot be identified, despite the use of a specific methyl-tert-butyl-ether (MTBE) extraction method that was previously proposed to enhance the extraction of vtg-like proteins (Blaise et al., 1999). Moreover, the data obtained from a specific stain for phospho-proteins and subsequent identification by mass spectrometry show that almost all the phosphate is bound to different forms of extrapallial protein (EP) with different apparent molecular weights. EP is the major haemolymph serum protein identified in the marine mussel (Yin et al., 2005); a database search revealed its high homology with heavy metal binding protein (Buck et al., 2003), Keystonein protein (Ferrier et al., 2014), and putative C1q domain containing protein MgC1q6 (Gerdol et al., 2011).

2. Material and methods 2.1. Mussel cell-free haemolymph collection M. galloprovincialis individuals of 4–5 cm in length, obtained from a mussel farm in Arborea, Sardinia (Italy), were used for the analysis. Sex was determined via microscopic examination of the gonadal tissue at the same time the stage of the reproductive cycle was evaluated and all the mussels were in developing gonad phase 2 or 3 according to Hillman (1993). Haemolymph was sampled from the posterior adductor muscle using a 1 mL sterile syringe with a 21 G (0.5 mm inner diameter), 1 1/2″ needle. Serum was obtained after centrifugation at 500 g for 10 min at 4 °C, the supernatant was immediately frozen in liquid nitrogen, and stored at − 80 °C.

2.2. Protein precipitation Haemolymph serum proteins were precipitated overnight with 5 volumes of ice-cold acetone (80%) and then centrifuged at 14,000 g for 30 min. The pellets were washed twice with ice-cold acetone (80%). Alternatively, according to the method of Blaise et al. (1999), 1.5 mL of haemolymph was mixed with 500 μL of MTBE (Sigma-Aldrich) and kept at R.T. for 15 min. A volume of 400 μL of the organic (upper) phase was dried in a speed vacuum (Concentrator 5301, Eppendorf) and utilised in the subsequent experiments. Total protein concentrations were estimated by the Bradford (1976) method. 2.3. SDS-polyacrylamide gel electrophoresis SDS-PAGE was performed according to Laemmli (1970), with 12%resolving gels (pH 8.8), 4%-stacking gels (pH 6.8), and a Tris/glycine running buffer (pH 8.3). Proteins were resuspended in 20 mM Tris– HCl pH 7, with 4% CHAPS, 20 μg of protein was mixed 1:1 with 2 × SDS-PAGE sample buffer (66% v/v glycerol, 6% SDS, 20 mM Tris–HCl pH 7, 20 mM DTT, 4 mM EDTA, traces of bromophenol blue), and heated at 95 °C for 2 min before loading. A prestained protein ladder (low range, Bio-Rad Laboratories S.r.l.) was used as a marker to estimate the molecular weights of the proteins. After electrophoresis, proteins were fixed (40% ethanol, 10% acetic acid) for 60 min and stained with colloidal Coomassie blue (Neuhoff et al., 1988) or with Pro-Q® Diamond phosphoprotein gel stain (Molecular Probes) according to the manual. 2.4. Agarose gel electrophoresis Agarose gel electrophoresis (3%) in resolving buffer containing 500 mM Tris base (pH 8.5, 160 mM boric acid, 1 M urea) was performed on a horizontal apparatus (Bio-Rad Laboratories S.r.l.). Protein samples (20 μg) were loaded after mixing with an equal volume of 2× sample

Fig. 1. SDS-PAGE (12%) of serum proteins from female (♀) and male (♂) mussels. A total of 20 μg of protein, obtained by either MTBE or acetone precipitation, was loaded. The first lane (marker) shows the protein ladder with MW (kDa) values indicated. The numbers (1–9) correspond to the bands analysed. The identification details are in Table 1.

C. Oliveri et al. / Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

31

Table 1 Proteins identified from one-dimensional gel bands by ESI-MS/MS Q-TOF and Mascot MS/MS Ions Search. Band no.

Protein name

1 2 3

Not identified Not identified EP protein precursor or C1q domain containing protein MgC1q6 CuZn Superoxide dismutase

4

5

6

7

8

9

EP protein precursor or C1q domain containing protein MgC1q6 EP protein precursor or C1q domain containing protein MgC1q6

EP protein precursor or C1q domain containing protein MgC1q6

EP protein precursor or C1q domain containing protein MgC1q6

EP protein precursor or C1q domain containing protein MgC1q6

Organism

GeneInfo identifier

No. of matched peptides

Peptide sequences

Mytilus edulis and M. galloprovincialis

gi34304719 gi325504311

2

K.HLHEEVEYFK.S K.AEFDLTSLNADLEK.F

Mytilus galloprovincialis

gi402122771

4

K.HEGHTGLHIHEVKP.– K.MTDETISLHGENSLIGR.S + oxidation (M) R.SIAIHEGPDDLGMGGDAGSLK.G + oxidation (M) K.HDDEVHAHCEVFPNSNVADGDDIR.G + carbamidomethyl (C) K.HLHEEVEYFK.S K.AEFDLTSLNADLEK.F K.HEIDELHQEIK.H K.HLHEEVEYFK.S K.HEIDELHQEIK.H K.AEFDLTSLNADLEK.F K.EIHDVENHTEHNK.H K.FVAPEEGFFYFSVTICTK.R + carbamidomethyl (C) K.SHHVAFSAELTHPIENIAAEEIAHFDK.V + deamidated (NQ) K.HLHEEVEYFK.S K.HEIDELHQEIK.H K.AEFDLTSLNADLEK.F K.EIHDVENHTEHNK.H K.FVAPEEGFFYFSVTICTK.R + carbamidomethyl (C) K.SHHVAFSAELTHPIENIAAEEIAHFDK.V + deamidated (NQ) K.HLHEEVEYFK.S K.HEIDELHQEIK.H K.AEFDLTSLNADLEK.F K.FVAPEEGFFYFSVTICTK.R + carbamidomethyl (C) K.SHHVAFSAELTHPIENIAAEEIAHFDK.V + deamidated (NQ) K.HLHEEVEYFK.S K.AEFDLTSLNADLEK.F K.SHHVAFSAELTHPIENIAAEEIAHFDK.V + deamidated (NQ)

Mytilus edulis and M. galloprovincialis

gi34304719 gi325504311

3

Mytilus edulis and M. galloprovincialis

gi34304719 gi325504311

6

Mytilus edulis and M. galloprovincialis

Mytilus edulis and M. galloprovincialis

Mytilus edulis and M. galloprovincialis

gi34304719 gi325504311

gi34304719 gi325504311

gi34304719 gi325504311

buffer (126 mM Tris–HCl, 33% glycerol, 4% SDS, trace amount of BBF). The running buffer contained 90 mM Tris base with pH 8.5, 90 mM boric acid, and 0.1% SDS. The proteins were separated at 50 V until the prestained marker reached the end of the gel. The gel was stained with Coomassie blue stain solution (40% methanol, 10% acetic acid, 0.25% Coomassie brilliant blue G250) for approximately 1 h, and destained in a solution containing 20% methanol, and 5% acetic acid with gentle shaking until the protein bands became visible. 2.5. 2-D gel electrophoresis For the first dimension, 7-cm IPG strips (Bio-Rad Laboratories S.r.l.), pH 4.0–7.0, were pre-hydrated for 7 h in 5 M urea, 2 M thiourea, 4% CHAPS, 1% IPG buffer, 20 mM DTT, and trace BBF. Isoelectric focusing (IEF) was performed at 30 V for 7 h and then 300 V for 4 h, followed by a gradient of 300–1000 V in 30 min, 1000–5000 V in 90 min, and then kept at 5000 V for 30 min. The proteins in the strips were reduced for 10 min in 1% DTT in equilibration buffer (50 mM Tris–HCl with pH 8.8; 5 M urea; 2 M thiourea; 30% glycerol; 2% SDS; trace BBF), and then alkylated for 10 min in 2.5% iodoacetamide in the same buffer. The proteins were then separated in a second dimension according to their apparent molecular mass on

6

5

3

Mascot individual ion scores

Threshold for significant identification of individual peptides (p b 0.05)

62 70

60

137 80

58

72 270 70 110 98 105 109 128 118 206

60

58

79 98 87 76 93 86

58

59 70 62 75 80

58

90 62 71 60

58

a 12% acrylamide gel, pH 8.8. After electrophoresis, the gels were fixed, and stained as above for SDS-PAGE. 2.6. Western blotting After SDS-PAGE and 2-DE, the samples were transferred onto polyvinyl difluoride (PVDF) membranes by electroblotting at 50 V for 1 h according to the instruction manual (Bio-Rad Laboratories S.r.l.). The membranes were blocked with a 5% BSA solution in TBST at 4 °C for 1 h, and incubated with a 1:500-diluted rabbit anti-EP protein precursor antibody (Cambridge Research Biochemical) in TBST at 4 °C overnight. Then, the membranes were washed four times with TBST and incubated with a 1:1000-diluted HRP-conjugated goat anti-rabbit antibody (BioRad Laboratories S.r.l.) in TBST at R.T. for 90 min. After washing four times with TBST, the protein bands on the membrane were identified using an ECL detection kit (GE Healthcare Life Science). 2.7. In-gel digestion The gel bands were excised, cut into small pieces (approximately 1 mm2), and in-gel digested at 37 °C overnight using trypsin (25 ng/ μL trypsin proteomic grade, Roche Diagnostics S.p.A., Milan, Italy) in

32

C. Oliveri et al. / Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

Fig. 2. Agarose gel electrophoresis to separate high-molecular-weight proteins. Marker, and low-range protein marker (Bio-Rad) with MW (kDa) values are indicated. BSA, 5 μg of BSA standard. IgY from chicken egg yolk, 5 μg. Lane 1, 20 μg of mussel serum protein extract. Lane 2, the same sample as in lane 1 heated for 5 min at 95 °C.

25 mM ammonium hydrogen carbonate (Shevchenko et al., 2007). Briefly, the gel pieces were destained in acetonitrile, reduced using 10 mM DTT in 100 mM ammonium bicarbonate for 30 min at R.T., and alkylated with 100 mM iodoacetamide in 100 mM ammonium bicarbonate for 30 min at R.T. in the dark. The peptide samples were extracted after digestion by sonicating for 10 min at R.T., then vacuum dried and stored at −20 °C until nano-HPLC ESI-Q-TOF MS/MS analysis. 2.8. Protein identification by nano-HPLC–MS/MS analysis and database search Nano-HPLC ESI-Q-TOF MS/MS analysis (Ultimate 3000, Dionex coupled with Q-Star XL, AB Sciex) was performed as previously described (Marsano et al., 2010). MS/MS data from the protein samples were searched as a Mascot generic file against all entries in the public NCBI nr database (version 20140228; 37332560 sequences) (http://www.ncbi.nlm.nih.gov/)

Fig. 4. Immunolocalisation of HRG on PVDF after 12% SDS-PAGE separation of serum proteins from male (lanes 1, 2) and female (lanes 3, 4) individuals performed on both MTBEextracted and acetone-precipitated protein samples.

using the online Mascot (www.matrixscience.com) search engine. Carbamidomethylation of cysteine residues, oxidation of methionine, deamidation of asparagine and glutamine were set as possible variable modifications; trypsin was selected as the protease. One missed trypsin cleavage site was allowed, and the peptide MS and MS/MS tolerances were set to 50 ppm and 0.3 Da, respectively. No taxonomy filter was used in the first search to exclude possible contamination. If the search returned no protein candidates or hits with a low number of peptides after excluding the contaminant peptides (trypsin, keratins, etc.), another search was performed against the Invertebrate EST database to identify the sequence, and names were then assigned using a Basic Local Alignment Search Tool (BlastP) (Altschul et al., 1997). 3. Results 3.1. Biochemical characterisation of mussel haemolymph by gel electrophoresis and mass spectrometry

Fig. 3. Phosphoprotein detection in SDS polyacrylamide gels by Pro-Q Diamond. Phosphate in the serum is mainly localised in the band at approximately 35 kDa, and no difference can be observed with respect to the extraction method (acetone or MTBE) or specimen gender (male or female). Marker P is the phosphate marker used as a positive control. Marker LR is the low-range protein marker (Bio-Rad) used as a negative control.

One-dimensional SDS gels of mussel haemolymph serum analysed using the “detect bands” tool embedded in Quantity One (ver. 4.6.1 build 055, Bio-Rad Laboratories) revealed nine bands (Fig. 1). The main bands have apparent masses of 60, 39, 37 and 35 kDa (Fig. 1, bands 4, 6, 7 and 8) and five fainter bands were also visible at approximately 100, 87, 80 and 30 kDa (Fig. 1, bands 1–3, 5, 9). Bands 1–3 are very weak and cannot be identified by mass spectrometry. In order to clarify whether the method suggested by Blaise et al. (1999) may lead to changes in the mussel proteome through the specific extraction of lipoproteins, the MTBE-extracted serum haemolymph fraction was compared on the same gel with the acetonic preparation (Fig. 1). Qualitative differences in the protein profiles cannot be observed with the different extraction methods; however, we noticed a difference in the total amount of protein extracted, which is substantially lower in the MTBE-extracted samples. The results of mass spectrometric protein identification are reported in Table 1. The main protein bands can be identified as EP protein precursor, or C1q domain containing protein MgC1q6, from the same set

C. Oliveri et al. / Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

33

Fig. 5. Two-dimensional gel electrophoresis, with a pH gradient of 4–7, of Mytilus galloprovincialis serum haemolymph after acetonic precipitation from female (A) and male (B) individuals, stained using Coomassie blue. The numbers correspond to the proteins identified in Table 2.

of peptides (Fig. 1, bands 3, 5–9), and CuZn superoxide dismutase (Fig. 1, band 4). A multiple sequence alignment (Supplemental file 1) confirmed that EP protein precursor, putative C1q domain containing protein MgC1q6,

Keystonein and heavy metal binding protein are all the same protein, which accounts for 60–70% of the protein content in the haemolymph. The same amounts of serum proteins (20 μg) from both male and female mussels (acetonic precipitation and MTBE extraction) were

Table 2 Proteins identified from two-dimensional gel spots by ESI-MS/MS Q-TOF and Mascot MS/MS Ions Search. Spot no.

Protein name

Organism

GeneInfo identifier of matched sequence(s)

No. of matched peptides

Peptide sequence

1

EP protein precursor or C1q domain containing protein MgC1q6

Mytilus edulis and M. galloprovincialis

gi34304719 gi325504311

6

2

Actin

Mytilus galloprovincialis

gi5114428

3

3

Mg_Nor01_53P03 (astacin by BlastP)

Mytilus galloprovincialis

gi238638946 accession FL495425 (invertebrate EST)

7

K.HLHEEVEYFK.S K.HEIDELHQEIK.H K.AEFDLTSLNADLEK.F K.EIHDVENHTEHNK.H K.FVAPEEGFFYFSVTICTK.R + carbamidomethyl (C) K.SHHVAFSAELTHPIENIAAEEIAHFDK.V R.HQGVMVGMGQK.D R.VAPEEHPVLLTEAPLNPK.A K.DLYANTVLSGGTTMFPGIADR.M + oxidation (M) K.VSLSISDVHNLK.R R.LLSQTIAAGEYCVR.F + carbamidomethyl (C) R.DSFVNVMYDNIKDEYK.Q + oxidation (M) R.YTNAFTFTQPTVLVVEAIVGR.S + deamidated (NQ) R.NDGNDVLLEEMEGNQGNEWHR.Y + oxidation (M) R.GPTPTTMTGPNSDYSSGIGYYLLAEAR.T + oxidation (M) R.RGPTPTTMTGPNSDYSSGIGYYLLAEAR.T + oxidation (M) K.MTDETISLHGENSLIGR.S + oxidation (M) R.SIAIHEGPDDLGMGGDAGSLK.G + oxidation (M) R.LEDELLTEK.E K.LAITEVDLER.A K.ISMLEEDIMK.S + 2 oxidation (M) R.KLAITEVDLER.A K.EVDRLEDELLTEK.E K.QIAEHEQEIQSLTR.K K.AISDELDATFAELAGY.– K.VIDLEEQLTVVGANIK.T K.VTDLQSELENAQK.E R.LTGELEDLGIDVER.A K.NLAEEIHELTEQLSEGGR.S K.EELQAALEEAESALEQEEAK.V K.QIQFVNDIVDFLDVGSDETR.V K.ELLADIFFVLDQSSSIK.T K.TLHFVTGVIDYLQVSPSETQVGALK.F

4

CuZn superoxide dismutase

Mytilus galloprovincialis

gi402122771

2

5

Tropomyosin

Mytilus galloprovincialis

gi6647862 gi6647863

8

6

Catchin protein or myosin heavy chain

Mytilus galloprovincialis

gi6682323 gi6682319

4

7

Mg_Nor01_57K02 Mg_Nor01_33L06 Collagen (by BlastP)

Mytilus galloprovincialis Crassostrea gigas Bathymodiolus thermophilus

gi238645034 accession FL501143 gi405961288 gi262233854

X

Not identified

Mascot individual ion scores

62 78 80 95 63 79 45 119 90 32 86 52 50

Threshold for significant identification of individual peptide (p b 0.05) 59

58

59

25 43 65 95 77

57

47 76 97 112 63 39 82 138 84 54 92 90 94 97 94

57

59

59

34

C. Oliveri et al. / Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

Fig. 6. Two-dimensional gel electrophoresis of MTBE-extracted Mytilus galloprovincialis serum proteins from pooled female haemolymph samples. The numbers correspond to the proteins identified in Table 2.

compared by SDS-PAGE to investigate possible sex-dependent variations, but no differences can be observed (Fig. 1). Agarose gel electrophoresis was performed to assess the presence of larger proteins in the haemolymph, but high-molecular-mass proteins cannot be observed (Fig. 2, lanes 1 and 2). Serum proteins from both males and females, precipitated by either acetone or MTBE, were stained with Pro-Q Diamond dye (Molecular Probes) after separation on SDS-PAGE gels. A phosphorylated band was observed in each sample (Fig. 3) and was identified as EP protein precursor that in fact has several putative phosphorylation sites (Supplemental file 2), which has no homology with the sequence of vitellogenin from M. edulis (Supplemental file 3). 3.2. Immunostaining of EP protein precursor Western blot analyses were performed using a custom polyclonal anti-EP antibody (Cambridge Research Biochemicals Limited, UK). The analyses recognised the EP protein at approximately 35 kDa in both male and female organisms after acetonic precipitation or MTBE extraction (Fig. 4). 3.3. Two-dimensional gel electrophoresis Two-dimensional electrophoresis (2-DE) was performed for both male and female samples using pooled haemolymph from four individual mussels (Fig. 5). No significant differences between the two sexes can be detected in the serum extracts. The main pI spot cluster (Fig. 5, spot number 1) was identified as EP protein precursor (Table 2). As with

the one-dimensional gel electrophoresis, we also identified a cluster corresponding to CuZn SOD (Fig. 5, spot number 4). Most likely due to the increased resolving power of 2-DE, we also identified a metalloprotease, astacin (Fig. 5, spots numbered 3). Moreover, actin, tropomyosin, catchin or myosin heavy chain and collagen (Fig. 5, spot numbers 2, 5, 6 and 7) were also identified, most likely as a result of the contamination of the haemolymph by muscle proteins during the puncturing of the abductor muscle with the syringe. The same protein profile was also obtained with methyl-t-butyl-ether (MTBE)-extracted samples as shown in Fig. 6. To further verify the nature of the phosphoproteins extracted by methyl-t-butyl-ether (MTBE), the MTBE-enriched serum haemolymph fraction (Blaise et al., 1999) was also analysed by 2-DE and stained with Pro-Q Diamond dye (Fig. 7). The results indicate that the MTBEextracted protein samples do not differ from those obtained by acetone precipitation (Fig. 5). In fact, we obtained similar phospho-protein profiles, which were influenced by neither the extraction method (Fig. 7A, B, acetone; Fig. 7C, MTBE) nor the sex of the individual organism (Fig. 7A, female; Fig. 7B, male). 4. Discussion Both the cellular and serum components of mussel haemolymph have been studied for various aims. Here, we present the first characterisation of the proteomic component of M. galloprovincialis haemolymph serum. In the past, serious drawbacks stemmed from attempts to identify mussel proteins from mass spectrometric data, and only a few proteins could be identified with statistical confidence (Apraiz et al., 2006, 2009). Because of the incomplete sequencing of the mussel genome, the proteins in this organism can be identified mainly through de novo sequencing. In the present work, tandem mass spectrometry combined with the possibility offered by Mascot to search against the invertebrate expressed sequence tag (EST) database allowed us to identify mussel proteins with good confidence from both in one-dimensional gel electrophoresis bands and 2-DE protein spots (Table 2). The data obtained allowed the characterisation of the serum proteome, in particular the identification of the main protein species. The 2-DE proteomic maps of M. galloprovincialis serum allow us to resolve proteins never identified previously by this technique, though a few (e.g., actin, tropomyosin, catchin or myosin heavy chain, and collagen) can be ascribed to contamination due to the method of haemolymph collection, which involved puncturing the adductor muscle, that should be taken into account in research involving haemolymph. The role of the other proteins identified in the haemolymph serum, such as EP protein precursor, astacin and SOD, has to be studied further.

Fig. 7. Two-dimensional analysis (pH 4–7) of Mytilus galloprovincialis serum after acetonic precipitation from female (A) and male (B) individuals. (C) 2-DE of a female sample obtained after MTBE extraction. All gels were stained with Pro-Q Diamond dye. The marker is the PeppermintStick™ phosphoprotein molecular weight standard.

C. Oliveri et al. / Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

Astacin-like metalloproteases are ubiquitous in the animal kingdom, but their functions are unclear (Möhrlen et al., 2006). The astacin family of metalloendopeptidases was recognised in the 1990s, and more than 20 members of this family have been identified in species ranging from hydra to humans (Dumermuth et al., 1991; Bond and Beynon, 1995). The proposed functions of these proteases include the activation of growth factors, the degradation of polypeptides, and the processing of extracellular proteins (Dumermuth et al., 1991; Bond and Beynon, 1995). Astacin-like metalloproteases have also been identified in the digestive fluid of the araneid spider Argiope aurantia (Foradori et al., 2006) and in haemocytes of the pearl oyster Pinctada fucata (Xiong et al., 2006). In mussel haemolymph, a metalloprotease with gelatin-specific protease activity was characterised (Mannello et al., 2001). Superoxide dismutase was previously immunolocalised in blood cells from the mussel M. edulis (Pipe et al., 1993; Monari et al., 2007), and the effects of temperature on SOD were studied in both haemocyte lysates and cell-free haemolymph from Chamelea galina (Monari et al., 2007). EP protein precursor (Hattan et al., 2001; Yin et al., 2005) exhibits extensive amino acid sequence similarity with three other proteins identified in the haemolymph (Supplemental file 1), including serum protein band 1 (SPB1), identified as the most abundant humoral protein of the haemolymph of M. edulis (Renwrantz and Werner, 2008) and heavy metal binding protein (HIP) identified as protein carrier of divalent cations in plasma (Schneeweiss et al., 2002). EP protein also has sequence homology to the putative C1q domain containing protein (Gestal et al., 2010; Gerdol et al., 2011), which is considered to be a pattern recognition protein (PRP) that binds to a broad range of pathogenassociated molecular patterns (PAMPs) on bacteria, viruses, and parasites, thereby enhancing pathogen phagocytosis (Medzhitov and Janeway, 2002; Bohlson et al., 2007). Recently, a new sequence from mussel has been deposited in the database as the Keystonein protein (Ferrier et al., 2014), which also shows high homology with EP protein (Supplemental file 1). Moreover, EP protein precursor is a glycoprotein and has recently been shown to contain a peculiar structure of fucosylated N-glycans (Zhou et al., 2013). Although a sequence for histidine rich glycoprotein (HRG) (Abebe et al., 2007) – another protein previously identified as the most abundant in mussel serum – has yet to be deposited in a public database, we propose to refer to the main serum protein as HRG, because it seems the most correct definition for a protein whose localisation and functions are as yet unknown but which is undoubtedly histidine rich. This abundant and complex protein evidently has many different functions in the haemolymph. Previous results demonstrated that HRG can bind Ca, Cd, Hg, Mg, Ni, Pd, and Zn. The finding that a single mussel plasma protein may be responsible for binding all these metals raises important questions about how these different metals are subsequently transferred from HRG to different tissues of the mussel (Devoid et al., 2007). Probably, the highly phosphorylatable nature of HRG (Supplemental file 2) has a role in the modulation of the metal binding strengths. In fact, it was previously hypothesised that phosphorylation could influence the protein–metal interactions (Lu et al., 2011) but this aspect should be further studied. Given that a common biomarker used to assess the presence of endocrine disruptors in the environment in mussel haemolymph is based on the alkali-labile phosphate assay, we also characterised the haemolymph proteome after specific extraction using MTBE, as proposed by Blaise et al. (1999). This assay has previously been applied in fish (Brink et al., 2012; Nyina-wamwiza et al., 2012), gastropods (Gagnaire et al., 2009), crabs and other crustaceans (Martin-Diaz et al., 2004) but also in bivalves (Gagné et al., 2001; Blaise et al., 2003; Matozzo and Marin, 2005; Baussant et al., 2011; Farcy et al., 2011; Gagné et al., 2011; Saavedra et al., 2012). The data obtained show that the main phosphorylated protein in Mytilus serum was EP, which is known to be highly phosphorylatable. Furthermore, the alignment of the EP and vitellogenin sequences from

35

M. edulis revealed no homology between these proteins (Supplemental file 3). Considering the high apparent molecular weight of the native form of vitellogenin in fishes (Pan et al., 2012) and in freshwater mussels (Won et al., 2005), we performed agarose gel electrophoresis to resolve high-molecular-weight proteins in the haemolymph serum, but no proteins, and in particular no vtg-like proteins, can be detected. Furthermore, during one-dimensional electrophoresis, no proteins were found at the top of the gels, thus confirming the absence of proteins with high molecular weights. Comparing the acetone extract and the MTBE phase partition via SDS page revealed no appreciable differences when loading the same amount of protein per lane; the only difference observed was the lower amount of total protein per volume obtained via MTBE extraction. Similarly, there were no differences between the male and female haemolymph proteomes. The phosphate in all the samples (acetone or MTBE extraction, male or female) can be detected at the same amount and position associated with HRG. To better characterise the haemolymph proteome, we performed 2-DE to better resolve the protein spots. This sensitive approach, despite allowing the identification of more proteins with respect to SDS PAGE, did not shown any qualitative differences among the samples. Unfortunately, the SDS PAGE and 2-DE methods used in this study do not allow for the identification of low-molecular-mass proteins below 10 kDa (defensin, mytilin, myticin, etc.) which, however, have already been studied in mussel for the implication in immune responses on both levels of proteomics and genomics (Charlet et al., 1996; Li et al., 2009; Venier et al., 2011; Gerdol et al., 2012). For high- and middle-abundance proteins, our study provides the first comprehensive picture of the composition of the haemolymph proteome in M. galloprovincialis. These data demonstrate that there are no vitellogenin-like phosphorylatable proteins in the haemolymph serum; therefore, great care should be taken in using the ALP method in mussels. The identified proteins in this work may represent a starting point for comparative studies of the proteins that characterise different physiological or stress conditions in mussel haemolymph serum using proteomic or biochemical approaches. Acknowledgements This work was partially funded by Theme 6 of the EC seventh framework programme through the Marine Ecosystem Evolution in a Changing Environment (MEECE No. 212085) Collaborative Project. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.cbd.2014.07.003. References Abebe, A.T., Devoid, S.J., Sugumaran, M., Etter, R., Robinson, W.E., 2007. Identification and quantification of histidine-rich glycoprotein (HRG) in the blood plasma of six marine bivalves. Comp. Biochem. Physiol. B 147, 74–81. Apraiz, I., Mi, J., Cristobal, S., 2006. Identification of proteomic signatures of exposure to marine pollutants in mussels (Mytilus edulis). Mol. Cell. Proteomics 5, 1274–1285. Apraiz, I., Cajaraville, M.P., Cristobal, S., 2009. Peroxisomal proteomics: biomonitoring in mussels after the Prestige's oil spill. Mar. Pollut. Bull. 58, 1815–1826. Baussant, T., Ortiz-Zarragoitia, M., Cajaraville, M.P., Bechmann, R.K., Taban, I.C., Sanni, S., 2011. Effects of chronic exposure to dispersed oil on selected reproductive processes in adult blue mussels (Mytilus edulis) and the consequences for the early life stages of their larvae. Mar. Pollut. Bull. 62, 1437–1445. Blaise, C.,Gagné, F.,Pellerin, J.,Hansen, P., 1999. Determination of vitellogenin-like properties in Mya arenaria hemolymph (Saguenay Fjord, Canada): a potential biomarker for endocrine disruption. Environ. Toxicol. 14, 455–465. Blaise, C.,Gagne, F.,Pellerin, J., 2003. Bivalve population status and biomarker responses in Mya arenaria clams (Saguenay Fjord, Quebec, Canada). Fresenius Environ. Bull. 12, 956–960.

36

C. Oliveri et al. / Comparative Biochemistry and Physiology, Part D 11 (2014) 29–36

Bohlson, S.S., Fraser, D.A., Tenner, A.J., 2007. Complement proteins C1q and MBL are pattern recognition molecules that signal immediate and long-term protective immune functions. Mol. Immunol. 44, 33–43. Bond, J.S., Beynon, R.J., 1995. The astacin family of metalloendopeptidases. Protein Sci. 4, 1247–1261. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248–254. Brink, K.,van Vuren, J.,Bornman, R., 2012. Responses of laboratory exposed catfish (Clarias gariepinus) to environmentally relevant concentrations of p, p′-DDT. Environ. Toxicol. Pharmacol. 34, 919–925. Buck, F.,Schmalmack, W.,Luedeking, A.,Jarck, F.,Behlich, L.K.,Schneeweiss, H., 2003. Cloning of the HIP Protein From Mytilus edulis. http://www.uniprot.org/uniprot/Q708T0. Charlet, M., Chernysh, S., Philippei, H., Hetru, C., Hoffmann, J.A., Bulet, P., 1996. Innate immunity. Isolation of several cysteine-rich antimicrobial peptides from the blood of a mollusc, Mytilus edulis. J. Biol. Chem. 271, 21808–21813. Cheng, T.C., 1981. Invertebrate Blood Cells. In: Rowley, A.F. (Ed.), Academic Press, London, pp. 231–300. Cheng, T.C., 1984. Invertebrate Blood: Cells and Serum Factors. Springer-Verlag, Berlin. Ciocan, C.M.,Cubero-Leon, E.,Minier, C.,Rotchell, J.M., 2011. Identification of reproductionspecific genes associated with maturation and estrogen exposure in a marine bivalve Mytilus edulis. PLoS One 6, e22326. Devoid, S.J.,Etter, R.,Sugumaran, M.,Wallace, G.T.,Robinson, W.E., 2007. Histidine-rich glycoprotein from the hemolymph of the marine mussel Mytilus edulis L. binds class A, class B, and borderline metals. Environ. Toxicol. Chem. 26, 872–877. Dumermuth, E., Sterchi, E.E., Jiang, W., Wolz, R.L., Bond, J.S., Flannery, A.V., Beynon, R.J., 1991. The astacin family of metalloendopeptidases. J. Biol. Chem. 266, 21381–21385. Farcy, E., Gagnè, F., Martel, L., Fortier, M., Trépanier, S., Brousseau, P., Fournier, M., 2011. Short-term physiological effects of a xenobiotic mixture on the fresh water mussel Elliptio complanata exposed to municipal effluents. Environ. Res. 111, 1096–1106. Ferrier, G.A.,Zimmer, R.K.,Zimmer, C.A.,Kim, S.J.,Loo, J.A., 2014. Keystone Species and Molecules of Keystone Significance. http://www.ncbi.nlm.nih.gov/protein/ AFY13479.1. Fisher, W.S., 1986. Immunity in Invertebrates. In: Brehelin, M. (Ed.), Springer-Verlag, Berlin, pp. 25–35. Foradori, M.J.,Tillinghast, E.K.,Smith, J.S.,Townley, M.A.,Mooney, R.E., 2006. Astacin family metallopeptidases and serine peptidase inhibitors in spider digestive fluid. Comp. Biochem. Physiol. B 143, 257–268. Gagnaire, B., Gagné, F., André, C., Blaise, C., Abbaci, K., Budzinski, H., Dévier, M.-H., Garric, J., 2009. Development of biomarkers of stress related to endocrine disruption in gastropods: alkali-labile phosphates, protein-bound lipids and vitellogenin-like proteins. Aquat. Toxicol. 92, 155–167. Gagné, F., Blaise, C., Lachance, B., Sunahara, G.I., Sabik, H., 2001. Evidence of coprostanol estrogenicity to the freshwater mussel Elliptio complanata. Environ. Pollut. 115, 97–106. Gagné, F., Bouchard, B., André, C., Farcy, E., Fournier, M., 2011. Evidence of feminization in wild Elliptio complanata mussels in the receiving waters downstream of a municipal effluent outfall. Comp. Biochem. Physiol. C 153, 99–106. Gerdol, M., Manfrin, C., De Moro, G., Figueras, A., Novoa, B., Venier, P., Pallavicini, A., 2011. The C1q domain containing proteins of the Mediterranean mussel Mytilus galloprovincialis: a widespread and diverse family of immune-related molecules. Dev. Comp. Immunol. 35, 635–643. Gerdol, M., De Moro, G., Manfrin, C., Venier, P., Pallavicini, A., 2012. Big defensins and mytimacins, new AMP families of the Mediterranean mussel Mytilus galloprovincialis. Dev. Comp. Immunol. 36, 390–399. http://dx.doi.org/10.1016/j.dci. 2011.08.003. Gestal, C.,Pallavicini, A.,Venier, P.,Novoa, B.,Figueras, A., 2010 Sep. Dev Comp Immunol. 34 (9), 926–934. http://dx.doi.org/10.1016/j.dci.2010.02.012. PMID:20219531. Hattan, S.J., Laue, T.M., Chasteen, N.D., 2001. Purification and characterization of a novel calcium-binding protein from the extrapallial fluid of the mollusc, Mytilus edulis. J. Biol. Chem. 276, 4461–4468. Hégaret, H., Da Silva, P.M., Wikfors, G.H., Haberkorn, H., Shumway, S.E., Soudant, P., 2011. In vitro interactions between several species of harmful algae and haemocytes of bivalve molluscs. Cell Biol. Toxicol. 27, 249–266. Hillman, R.E., 1993. Gonadal index and histopathology for the East and West coasts used in the National Status and Trends Mussel Watch Project. Sampling and Analytical Methods of the National Status and Trends Program National Benthic Surveillance and Mussel Watch Projects 1984–1992. Comprehensive Descriptions of Complementary Measurements, vol. II. NOAA/NOS/ORCA, Silver Spring, MD, pp. 45–53 (NOAA Tech. Mem. NOS ORCA 71). Hong, X.T.,Xiang, L.X.,Shao, J.Z., 2006. The immunostimulating effect of bacterial genomic DNA on the innate immune responses of bivalve mussel, Hyriopsis cumingii Lea. Fish Shellfish Immunol. 21, 357–364. Itoh, N.,Xue, Q.-G.,Schey, K.L.,Li, Y.,Cooper, R.K.,La Peyre, J.F., 2011. Characterization of the major plasma protein of the eastern oyster, Crassostrea virginica, and a proposed role in host defense. Comp. Biochem. Physiol. B 158, 9–22. Koutsogiannaki, S., Kaloyianni, M., 2010. Signaling molecules involved in immune responses in mussels. Invertebr. Surviv. J. 7, 11–21. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685.

Li, Cheng-Hua,Zhao, Jian-Min,Song, Lin-Sheng, 2009. A review of advances in research on marine molluscan antimicrobial peptides and their potential application in aquaculture. Molluscan Res. 29, 17–26. Lu, Y., Prudent, M., Fauvet, B., Lashuel, H.A., Girault, H.H., 2011. Phosphorylation of αsynuclein at Y125 and S129 alters its metal binding properties: implications for understanding the role of α-synuclein in the pathogenesis of Parkinson's disease and related disorders. ACS Chem. Neurosci. 2, 667–675. Mannello, F., Canesi, L., Gazzanelli, G., Gallo, G., 2001. Biochemical properties of metalloproteinases from the hemolymph of the mussel Mytilus galloprovincialis Lam. Comp. Biochem. Physiol. B 128, 507–515. Marsano, F.,Boatti, L.,Ranzato, E.,Cavaletto, M.,Magnelli, V.,Dondero, F.,Viarengo, A., 2010. Effects of mercury on Dictyostelium discoideum: proteomics reveals the molecular mechanisms of physiological adaptation and toxicity. J. Proteome Res. 9, 2839–2854. Martin-Diaz, M.L., Sales, D., Del Valls Casillas, T.A., 2004. Influence of salinity in hemolymph vitellogenin of the shore crab Carcinus maenas, to be used as a biomarker of contamination. Bull. Environ. Contam. Toxicol. 73, 870–877. Matozzo, V.,Marin, M.G., 2005. Can 4-nonylphenol induce vitellogenin-like proteins in the clam Tapes philippinarum? Environ. Res. 97, 43–49. Medzhitov, R.,Janeway, C.A., 2002. Decoding the patterns of self and nonself by the innate immune system. Science 296, 298–300. Möhrlen, F., Maniura, M., Plickert, G., Frohme, M., Frank, U., 2006. Evolution of astacin-like metalloproteases in animals and their function in development. Evol. Dev. 8, 223–231. Monari, M., Matozzo, V., Foschi, J., Cattani, O., Serrazanetti, M.G.G.P.,Marin, 2007. Effects of high temperatures on functional responses of haemocytes in the clam Chamelea gallina. Fish Shellfish Immunol. 22, 98–114. Neuhoff, V.,Arold, N.,Taube, D.,Ehrhardt, W., 1988. Improved staining of proteins in polyacrylamide gels including isoelectric focusing gels with clear background at nanogram sensitivity using Coomassie Brilliant Blue G-250 and R-250. Electrophoresis 9, 255–262. Nyina-wamwiza, L.,Defreyne, P.,Ngendahayo, L.,Milla, S., Mandiki, S.,Kestemont, P., 2012. Effects of partial or total fish meal replacement by agricultural by-product diets on gonad maturation, sex steroids and vitellogenin dynamics of African catfish (Clarias gariepinus). Fish Physiol. Biochem. 38, 1287–1298. Pan, Z., Tian, H., Wang, W., Wang, J., Ru, S., 2012. Identification, purification, and immunoassay of stone flounder (Kareius bicolouratus) vitellogenin. J. Korean Soc. Appl. Biol. Chem. 55, 219–227. Pipe, R.K.,Porte, C.,Livingstone, D.R., 1993. Antioxidant enzymes associated with the blood cells and haemolymph of the mussel Mytilus edulis. Fish Shellfish Immunol. 3, 221–233. Renwrantz, L., Werner, I., 2008. Origin of a metal-binding protein in serum of Mytilus edulis. J. Molluscan Stud. 74, 11–17. Riffeser, M.,Hock, B., 2002. Vitellogenin levels in mussel hemolymph—a suitable biomarker for the exposure to estrogens? Comp. Biochem. Physiol. C Toxicol. Pharmacol. 132, 75–84. Saavedra, L.,Leonardi, M.,Morin, V., Quiñones, R.A., 2012. Induction of vitellogenin-like lipoproteins in the mussel Aulacomya ater under exposure to 17beta-estradiol. Rev. Biol. Mar. Oceanogr. 47, 429–438. Schneeweiss, H., Schmalmack, W., Luedeking, A., Jarck, F., Behlich, L.-K., Buck, F., 2002. Heavy metal binding protein HIP. Swill-Prot DatabaseSwiss Institute of Bioinformatics, Geneva, Switzerland (Accession number P83425). Seufi, A.M.,Galal, F.H.,Hafez, E.E., 2012. Characterization of multisugar-binding C-type lectin (Spli Lec) from a bacterial-challenged cotton leafworm, Spodoptera littoralis. PLoS One 7, e42795. Shevchenko, A.,Tomas, H.,Havli&sbreve, J., Olsen, J.V., Mann, M.,, 2007. In-gel digestion for mass spectrometric characterization of proteins and proteomes. Nat. Protoc. 1, 2856–2860. Venier, P.,Varotto, L.,Rosani, U.,Millino, C.,Celegato, B.,Bernante, F.,Lanfranchi, G.,Novoa, B., Roch, P., Figueras, A., Pallavicini, A., 2011. Insights into the innate immunity of the Mediterranean mussel Mytilus galloprovincialis. BMC Genomics 12, 69. Won, S.-J.,Novillo, A., Custodia, N.,Rie, M.T.,Fitzgerald, K., Osada, M., Callard, I.P., 2005. The freshwater mussel (Elliptio complanata) as a sentinel species: vitellogenin and steroid receptors. Integr. Comp. Biol. 45, 72–80. Wootton, E.,Pipe, R., 2003. Structural and functional characterisation of the blood cells of the bivalve mollusc, Scrobicularia plana. Fish Shellfish Immunol. 15, 249–262. Xiong, X., Chen, L., Li, Y., Xie, L., Zhang, R., 2006. Pf-ALMP, a novel astacin-like metalloproteinase with cysteine arrays, is abundant in hemocytes of pearl oyster Pinctada fucata. Biochim. Biophys. Acta Gene Struct. Expr. 1759, 526–534. Xue, Q.,Gauthier, J.,Schey, K.,Li, Y.,Cooper, R.,Anderson, R.,La Peyre, J., 2012. Identification of a novel metal binding protein, segon, in plasma of the eastern oyster, Crassostrea virginica. Comp. Biochem. Physiol. B 163, 74–85. Yin, Y., Huang, J., Paine, M.L., Reinhold, V.N., Chasteen, N.D., 2005. Structural characterization of the major extrapallial fluid protein of the mollusc Mytilus edulis: implications for function. Biochemistry 44, 10720–10731. Zhou, H., Hanneman, A.J., Chasteen, N.D., Reinhold, V.N., 2013. Anomalous N glycan structures with an internal fucose branched to GlcA and GlcN residues isolated from a mollusk shell-forming fluid. J. Proteome Res. 12, 4547–4555.