Identification of multiple isoforms of the cAMP-dependent protein kinase catalytic subunit in the bivalve mollusc Mytilus galloprovincialis ´ ´ Jose R Bardales1, Ulf Hellman2 and J A Villamarın1 ´ ´ Departamento de Bioquımica e Bioloxıa Molecular, Facultade de Veterinaria, Universidade de Santiago de Compostela, Lugo, Spain Ludwig Institute for Cancer Research, Uppsala, Sweden Keywords cAMP-dependent protein kinase; catalytic subunit; C-subunit isoforms; MALDI-TOF ⁄ TOF MS; Mytilus Correspondence ´ J A Villamarın, Departamento de ´ ´ Bioquımica e Bioloxıa Molecular, Facultade de Veterinaria, Universidade de Santiago de Compostela, Campus de Lugo, 27002 Lugo, Spain Fax: +34 82 252 195 Tel: +34 82 285 900 E-mail: antonio.villamarin@usc.es (Received 28 March 2008, revised July 2008, accepted 10 July 2008) doi:10.1111/j.1742-4658.2008.06591.x Several isoforms of the cAMP-dependent protein kinase catalytic subunit (C-subunit) were separated from the posterior adductor muscle and the mantle tissues of the sea mussel Mytilus galloprovincialis by cation exchange chromatography, and identified by: (a) protein kinase activity; (b) antibody recognition; and (c) peptide mass fingerprinting Some of the isozymes seemed to be tissue-specific, and all them were phosphorylated at serine and threonine residues and showed slight but significant differences in their apparent molecular mass values, which ranged from 41.3 to 44.5 kDa The results from the MS analysis suggest that at least some of the mussel C-subunit isoforms arise as a result of alternative splicing events Furthermore, several peptide sequences from mussel C-subunits, determined by de novo sequencing, showed a high degree of homology with the mammalian Ca-isoform, and contained some structural motifs that are essential for catalytic function On the other hand, no significant differences were observed in the kinetic parameters of C-subunit isoforms, determined by using synthetic peptides as substrate and inhibitor However, the C-subunit isoforms separated from the mantle tissue differed in their ability to phosphorylate in vitro some proteins present in a mantle extract The cAMP-dependent protein kinase (PKA; EC 2.7.11.11) plays a crucial role in the regulation of several physiological processes, as it is the main mediator of the effects of cAMP in eukaryotic organisms Inactive PKA is a tetrameric holoenzyme composed of two functionally distinct subunits: a dimeric regulatory subunit (R-subunit) and two monomeric catalytic subunits (C-subunits) The main function of the R-subunit is to inhibit the phosphotransferase activity of the C-subunit The transitory increase of cAMP levels inside the cell, induced by an extracellular signal, and the binding of cyclic nucleotide to R-subunits cause the dissociation of C-subunits which, once free, can phosphorylate protein substrates, mainly in the cytoplasm, but also in the nucleus [1,2] It has been widely reported that PKA is involved in the regulation of some physiological events that specifically occur in bivalve molluscs as a consequence of environmental adaptation For example, the relaxation of mollusc ‘catch’ muscles, induced by serotonin, occurs through the PKA-mediated phosphorylation of twitchin, a high molecular mass protein present in the thick filaments [3,4] The mollusc ‘catch’ muscles, such as the posterior adductor muscle (PAM), are specialized muscles that can sustain high tension for very long periods with low energy expenditure [5] On the Abbreviations CAF-PSD, chemically assisted fragmentation–post-source decay; C-subunit, catalytic subunit of cAMP-dependent protein kinase; PAM, posterior adductor muscle; PKA, cAMP-dependent protein kinase; PKI(5–24), protein kinase inhibitor peptide; PMF, peptide mass fingerprinting; PTM, post-translational modification; R-subunit, regulatory subunit of cAMP-dependent protein kinase FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS 4479 Mussel C-subunit isoforms J R Bardales et al other hand, phosphofructokinase from the sea mussel Mytilus galloprovincialis, unlike that from mammals, was clearly activated when phosphorylated by PKA at a serine residue [6]; moreover, the enzyme activity changed seasonally in parallel with its phosphorylation degree [7] These and other results reported by other authors [8] suggest that PKA activation contributes to the regulation of carbohydrate metabolism during bivalve gametogenic development, through the reversible phosphorylation of key regulatory enzymes Finally, various authors have argued that PKA-mediated protein phosphorylation could be responsible for metabolic rate depression, a strategy that bivalve molluscs use to survive during the long periods of aerial exposure causing environmental hypoxia [9,10] Therefore, to understand the biochemical basis of these molluscan regulatory events, the diverse forms of PKA in these organisms must be defined Over the last few years, we have identified and purified two different isoforms of the PKA R-subunit from the sea mussel M galloprovincialis, which were named Rmyt1 and Rmyt2 [11–13] Interestingly, both isoforms have identical apparent molecular masses of 54 kDa, but they differ in: (a) their isoelectric point; (b) their biochemical properties; (c) their antigenicity; and (d) their tissue distribution [12–14] According to its physicochemical and biochemical properties, a partial amino acid sequence from Rmyt1 showed a clear homology with the type I R-subunits from both mammalian and invertebrate sources [13]; likewise, Rmyt2 was shown to be homologous to the type II R-subunits from the same species [14] The purpose of the work described in this article was to investigate the possible existence of different isoforms of the PKA C-subunit in the sea mussel M galloprovincialis Results Separation of different isoforms of the C-subunit In order to demonstrate the presence of different isoforms of the PKA C-subunit in mussels, the protein was partially purified from the PAM and the mantle tissues of the mollusc, and then subjected to cation exchange chromatography on a Mono-S column Figure 1A shows the elution profile corresponding to the PAM C-subunit The application of a salt gradient resulted in separation of four absorbance peaks Three of them – labelled peak I, peak II and peak III – showed protein kinase activity; they eluted at 0.13, 0.16 and 0.25 m NaCl, respectively SDS ⁄ PAGE analysis and Coomassie staining revealed the presence of a 4480 A B C Fig Separation and identification of PKA C-subunit isoforms from mussel PAM (A) Elution profile of C-subunit from a Mono-S HR ⁄ column A sample (2 mL, 1.5 mg of protein) of C-subunit purified from PAM as described in Experimental procedures was applied to the column and eluted with a linear salt gradient Fractions of 0.5 mL were collected and assayed for protein kinase activity Three distinct peaks associated with protein kinase activity were separated: I, II and III Aliquots of fractions were also analysed by (B) Coomassie-stained SDS ⁄ PAGE, and (C) western blotting with an antibody against the human Ca-isoform protein with apparent molecular mass  40 kDa in the fractions corresponding to peak I, peak II and peak III (Fig 1B) This protein band was recognized by an antibody raised against the human Ca-isoform of the C-subunit (Fig 1C) Therefore, peak I, peak II and peak III correspond to three different isoforms of the C-subunit, which we named C1, C2 and C3, respectively On the other hand, fraction 18, corresponding to the first absorbance peak, without protein kinase activity, contained an unidentified protein < 30 kDa, and fractions 21–23 also contained an unidentified high molecular mass protein (Fig 1B) None of these proteins was recognized by the Ca-isoform antibody in the western blot analysis (Fig 1C) Figure 2A shows a representative elution pattern of the C-subunit preparation obtained from the mantle tissue Two absorbance peaks, associated with protein kinase activity, were separated; these eluted at 0.19 and 0.25 m NaCl, and were labelled peak I and peak II, respectively Coomassie staining of an SDS ⁄ PAGE gel revealed that fractions corresponding to peak I FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS J R Bardales et al Mussel C-subunit isoforms A A B B C C Fig Separation and identification of PKA C-subunit isoforms from mussel mantle tissue (A) Elution profile of C-subunit from a Mono-S HR ⁄ column A sample (2 mL, 1.8 mg of protein) of C-subunit purified from mantle tissue as described in Experimental procedures was applied to the column and eluted with a linear salt gradient Fractions of 0.5 mL were collected and assayed for protein kinase activity Two distinct peaks associated with protein kinase activity were separated: I and II Aliquots of fractions were also analysed by (B) Coomassie-stained SDS ⁄ PAGE, and (C) western blotting with an antibody against the human Ca-isoform contained only a  40 kDa protein, whereas those corresponding to peak II contained two different protein bands of  41 and  43 kDa (Fig 2B) The three protein bands showed reactivity with the human Ca-isoform antibody in the western blot analysis (Fig 2C) In summary, three different isoforms of C-subunit were separated from the mantle tissue preparation: the isoform named C4, which corresponded to peak I of the Mono-S chromatogram, and the isoforms named C5 and C6, which coeluted together at peak II C4 was 3–4-fold more abundant than C5 and C6 together Fig Characterization of mussel C-subunit isoforms (A) Determination of molecular mass by SDS ⁄ PAGE Samples (1–2 lg of protein) were subjected to 10% SDS ⁄ PAGE in a 16 · 16 cm polyacrylamide gel which was Coomassie stained Lane 1: molecular mass standards Lane 2: sample of bovine heart C-subunit, fraction CB [17] Lanes 3–5: samples of fractions 20, 22 and 29 of Fig 1A chromatogram, respectively Lanes and 7: fractions 24 and 28 of Fig 2A chromatogram, respectively The apparent molecular mass of mussel C-subunit isoforms was estimated from the positions of molecular mass standards and bovine C-subunit (B, C) Western blot analysis of mussel C-subunit isoforms Samples (140– 300 ng of protein) of the same fractions were subjected to 10% SDS ⁄ PAGE, and C-subunit was detected by western blotting using monoclonal antibodies against phosphoserine (B) or phosphothreonine (C) tissues The values of the apparent molecular mass ranged between 41.3 kDa for C4 and 44.5 kDa for C6 All the mussel isoforms were slightly heavier than the bovine C-subunit used as a control (lane 2), whose molecular mass, determined by MS, was exactly 40 855.7 Da [15] On the other hand, samples of purified mussel C1–C6 were probed with both phosphoserine and phosphothreonine antibodies, and they were all serine and threonine phosphorylated, as shown in Fig 3B,C, respectively Moreover, incubation of C-subunit isoforms with MgATP did not change their mobility on SDS ⁄ PAGE (not shown) Structural analysis of C-subunit isoforms Characterization of C-subunit isoforms Samples of purified C1–C6 were analysed by SDS ⁄ PAGE, using a 16 · 16 cm polyacrylamide gel As shown in Fig 3A, slight but significant differences were observed in the migration behaviour among mussel isozymes Only C3 (from PAM) and C5 (from mantle) have identical apparent mobilities, which suggests that they could be the same isoform present in both In order to determine possible structural differences among mussel C-subunit isoforms, samples of purified C1–C6 proteins were subjected to ‘in-gel’ tryptic digestion, and peptide mixtures were analysed by MALDI-TOF MS The corresponding peptide mass fingerprinting (PMF) spectra are shown in Fig Furthermore, a sample of C-subunit purified from bovine heart (fraction CB), consisting mainly of the FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS 4481 2833.537 2621.455 2556.224 2213.147 2283.220 2109.072 2152.230 1917.072 2833.535 2621.471 2556.215 2283.218 2109.054 2166.932 2212.978 C3 2849.241 2621.197 2555.979 2308.088 2212.749 2108.832 2151.991 1970.785 2833.334 2575.086 2621.275 2386.962 2212.806 * 2308.159 1970.816 2152.041 C4 2092.888 1854.821 1854.857 * 1768.624 1661.745 1713.717 1582.658 1494.737 1419.650 1253.594 1172.633 1014.523 1059.513 959.439 0.2 804.210 0.4 734.385 0.6 859.331 0.8 1916.898 1661.717 1713.684 1605.705 1494.720 1419.625 1253.577 1338.627 1172.619 1118.434 1014.508 958.924 895.386 774.288 713.389 x10 1.0 1916.857 842.446 1.5 0.5 x10 C5 2849.328 2621.281 2308.166 2108.888 2152.047 2198.787 1970.830 1916.904 1854.867 1661.757 1713.724 1605.744 1494.752 1419.657 1253.606 0.5 1172.646 774.292 1.0 1014.535 1065.993 1.5 959.450 842.465 2.0 895.402 x10 C6 2849.451 2591.142 2623.34 * 2212.893 2108.941 2152.088 1970.869 1864.886 1916.952 1661.810 1713.780 1605.786 1494.794 895.434 1014.569 1066.030 0.2 774.322 0.4 713.435 0.6 1419.706 0.8 1253.636 842.491 1.0 1172.682 Intens [a.u.] 1970.975 x10 1.0 Intens [a.u.] C2 1917.058 1855.016 1745.860 1661.897 1713.870 1605.878 1494.882 1419.783 1338.783 1172.754 1230.64 1253.712 1040.614 959.070 0.25 1118.577 * 859.451 804.333 1.00 0.50 Intens [a.u.] 1970.996 1855.033 1661.911 1713.884 1494.892 1582.823 1338.785 1172.758 1.25 0.75 Intens [a.u.] C1 x10 734.496 Intens [a.u.] 0.25 1419.797 0.50 1253.716 0.75 1138.649 1.00 * 966.537 1014.650 1059.640 1.25 734.499 791.44 804.339 Intens [a.u.] x105 J R Bardales et al 859.453 Mussel C-subunit isoforms 0.0 750 1000 1250 1500 1750 2000 2250 2500 2750 m/z Fig Peptide mass fingerprints of mussel C-subunit isoforms after tryptic digestion The asterisk(s) indicate the peak(s) observed only in the spectrum from a particular isoform Arrowheads indicate the peak at 1059.5 Da, common to C1 and C4, and arrows indicate the peak at 1605.7 Da, common to the remaining isoforms: C2, C3, C5 and C6 Ca-isoform [15,16], was also digested and analysed A detailed analysis of data allowed us to draw the following conclusions (a) Eight peptide masses were found to be common to the bovine Ca-isoform and all mussel C-subunit isozymes; these are (in Da): 734.5, 744.5, 759.4, 895.5, 1138.6, 1419.8, 1661.9, and 1917.1 The partial sequences with theoretical masses identical 4482 to those measured are marked with dashed lines in the whole sequence of the bovine Ca-isoform in Fig 5A Furthermore, two of these common peptides, which yielded peaks at m ⁄ z 744.5 and 895.5, were sequenced de novo by chemically assisted fragmentation–postsource decay (CAF-PSD) (Table 1: peptides and 2), and exactly matched the sequences of the bovine FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS J R Bardales et al Mussel C-subunit isoforms Fig Comparison of amino acid sequences from mussel C-subunit isoforms (in bold) with homologous regions of (A) bovine Ca-isoform (UniProtKB P00517), and (B) Aplysia C-subunit (UniProtKB Q16958) Identical residues are in black boxes Asterisks indicate residues playing a key role in the catalytic function (see text) Dashed lines show the partial sequences of the bovine Ca-isoform corresponding to the eight measured m ⁄ z peaks that were common to bovine and all mussel C-subunit isoforms Table Peptides from Mytilus C-subunit isoforms identified by de novo sequencing Peptide Measured m ⁄ z (Da) 744.47 895.47 10 11 12 13 859.44 870.51 884.48 1014.64 1016.65 1040.61 1172.75 1338.80 1494.89 1910.82 1605.88 a Present in spectra from Bovine Ca, C1–C6 Bovine Ca, C1–C6 C1–C6 C1–C6 C1–C6 C1–C6 C1–C6 C1–C6 C1–C6 C1–C6 C1–C6 C1–C6 C2, C3, C5, C6 De novo peptide sequence [I ⁄ L][I ⁄ L]DKQK T[I ⁄ L]GTGSFGR FSEPHSR VF[I ⁄ L]VQHK VTDFGFAK KVDAPF[I ⁄ L]PK S[I ⁄ L][I ⁄ L]QVD[I ⁄ L]TK VTDFGFAKR S[I ⁄ L][I ⁄ L]QVD[I ⁄ L]TKR [I ⁄ L]KQVEHT[I ⁄ L]NEK [I ⁄ L]KQVEHT[I ⁄ L]NEKR GPGDASNFDDYEEEP[I ⁄ L]R KGDVPMNVKE(x)Ka x Dmass = 360.3 Da Ca-isoform corresponding to amino acids 73–78 and 48–56, respectively (Fig 5A) (b) Most peptide masses were common to all mussel isoforms, C1–C6 Several of these peptides were also sequenced de novo (Table 1: peptides 3–12), showing high amino acid sequence identities to the bovine Ca-isoform (Fig 5A) (c) With regard to the C1, C2, C4 and C6 spectra, there was at least one peptide mass that was unique for each isoform, being absent in the remainder This was particularly true for m ⁄ z peaks labelled with asterisks in the spectra of Fig 5: 791.4 (only in C1); 1230.6 (only in C2); 1768.6 and 2386.9 (only in C4); and 2623.3 (only in C6) (d) There was one peak at 1059.5 Da observed only in the C1 and C4 spectra, whereas another peak at 1605.7 Da was found in the spectra of the remaining isoforms: C2, C3, C5 and C6 Interestingly, an incomplete sequence derived from this last peak (Table 1: peptide 13) matches a sequence lying at the N-terminus of a C-subunit (N1-isoform) from the mollusc Aplysia (Fig 5B) This result indicates that mussel C1 and C4 differ from C2, C3, C5 and C6 at the N-terminal region (e) When spectra from C3 and C5 were compared, no significant difference was observed, which suggests that both C3 and C5 are the same C-subunit isoforms present in the PAM and the mantle tissue respectively Kinetic characterization of C-subunit isoforms and protein phosphorylation In order to determine possible functional differences among mussel C-subunit isoforms, the kinetic parameters were determined for each purified isozyme It should be noted that C5 and C6 coeluted from the Mono-S column, and therefore, samples containing a mix of both isoforms were used in the kinetic experiments No significant differences among C-subunit isoforms regarding the values of apparent Km for Kemptide and Vmax were observed Furthermore, all the mussel isozymes were inhibited by the protein kinase inhibitor peptide [PKI(5–24)] with similar I50 values (Table 2) The ability of mussel C-subunit isoforms to phosphorylate proteins in vitro was also investigated Thus, FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS 4483 Mussel C-subunit isoforms J R Bardales et al previously demonstrated [3,4] The intermediate protein band was identified as actin by PMF and de novo sequence analysis (not shown) The protein band with the lowest molecular mass could not be identified by PMF; a correct sequence of 14 amino acids (RESEFQSGDLWEVR) was then obtained by de novo sequencing, although no clear identity could be drawn from the databases For the mantle extract (Fig 6B), the patterns of proteins phosphorylated by C4 and the mixture of C5 and C6 were also apparently similar, although densitometric analysis of the autoradiograph showed some protein bands, marked by asterisks, that seemed to be phosphorylated by C4 but not by the mix of C5 and C6 Thus, it is possible that mantle isoforms have different abilities to phosphorylate some proteins of mantle tissue Table Kinetic parameters of mussel C-subunit isoforms The apparent Km for Kemptide and Vmax values were determined at 0.2 mM ATP The I50 values for PKI(5–24) were determined at 100 lM Kemptide and 0.2 mM ATP The data are expressed as means ± SE of three independent experiments Isoform Km (lM) Vmax (nmol PỈmin)1Ỉlg)1) I50 (nM) C1 C2 C3 C4 C5 + C6 15.8 24.7 14.7 18.6 12.5 6.6 4.9 4.3 3.6 4.0 8.9 7.8 9.5 6.9 7.2 ± ± ± ± ± 8.0 9.5 5.3 2.4 5.2 ± ± ± ± ± 1.9 2.2 1.3 0.5 1.3 ± ± ± ± ± 1.7 3.3 2.1 2.9 2.7 C1, C2 and C3 (purified from PAM) were individually incubated, in the presence of labelled ATP, with aliquots of a PAM extract In the same way, C4 and the mixture of C5 and C6 were incubated with samples of a mantle tissue extract Densitometric analysis of autoradiographs corresponding to the PAM samples showed identical protein phosphorylation patterns for C1, C2 and C3 (Fig 6A) Three protein bands, marked with arrows in Fig 6A, were mainly phosphorylated by each C-subunit isoform The protein with the highest molecular mass ( 600 kDa) was identified as twitchin, whose PKA-mediated phosphorylation had been C1 C2 C3 – + A – + – + Discussion In this article, we describe the separation and identification of several catalytically active isoforms of the PKA C-subunit from the sea mussel M galloprovincialis The isozymes named C1, C2 and C3 were isolated from the PAM tissue, whereas C4, C5 and C6 were separated from the mantle tissue However, it C4 C5+C6 – + B – + twitchin actin PAM extract mantle extract ? 0.4 0.4 C1 C2 C3 * C4 C5+C6 A A * 0.2 * 0.2 0.0 0.0 0.5 Rf 4484 1.0 0.0 0.0 0.5 Rf 1.0 Fig In vitro phosphorylation of proteins from mussel extracts by C-subunit isoforms Aliquots of a crude extract from PAM, 100 lg of protein (A), and from mantle tissue, 120 lg of protein (B), were individually incubated with MgCl2 and [32P]ATP[cP] in the absence and presence of each C-subunit isoform isolated from the corresponding tissue (5 unitsỈmg)1 protein) Samples of C1, C2 and C3 were from fractions 20, 22 and 29 of the Fig 1A chromatogram, respectively, and samples of C4 and C5 + C6 were from fractions 24 and 28 of the Fig 2A chromatogram, respectively At 20 min, all the reactions were stopped by adding SDS sample buffer and boiling for Samples were then analysed by 10% SDS ⁄ PAGE, and the gel was Coomassie stained, destained, dried and exposed for autoradiography at )80 °C ?, unidentified protein The lower figures show the densitometric analysis of the autoradiographs FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS J R Bardales et al seems highly likely that C3 and C5 are the same isoform present in both tissues, as they showed identical apparent molecular masses, were eluted from a MonoS column at the same salt concentration, indicating similar pI values, and yielded near-identical PMF results In essence, mussel C-subunits could be: (a) encoded by various different genes; (b) generated by alternative splicing from a single gene; or (c) produced by posttranslational modifications (PTMs) Several authors have reported that the purified C-subunits from different mammalian species can be separated into two fractions, called CA and CB, by means of cation exchange chromatography [17,18] CA arises from CB, as a result of the in vivo deamidation of the Asn2 residue, and therefore the only difference between CA and CB was the presence of aspartic acid or asparagine, respectively, at position of their sequences [15] Unlike mammalian CA and CB, all the mussel C-subunits showed significant differences in their molecular masses, as revealed by SDS ⁄ PAGE mobility This finding rules out the possibility that some of them are produced by a similar PTM to that generating mammalian CA and CB, despite the fact that they were also separated by cation exchange chromatography On the other hand, all the mussel C-subunits are phosphorylated at serine and threonine residue(s), and they could not be interconverted by treatment with MgATP, which suggests that the differences were not due to autophosphorylation Finally, the comparison of PMF results from tryptic digests showed that, with the exception of C3 and C5, there was at least one peptide mass that was unique for each mussel C-subunit Therefore, taken together, these results clearly indicated that mussel C-subunits are not generated as a consequence of the PTMs typical of the PKA C-subunit, but rather they differ in their amino acid sequences In most mammalian species, two principal genes for the C-subunit have been identified and termed Ca and Cb [19,20]; additionally, the human genome contains a third gene encoding the Cc-isoform, which appears to be expressed only in testis [21] Among invertebrates, the nematode Caenorhabditis elegans also has two genes for the PKA C-subunit: the kin-1 gene, with potential to generate several C-subunit isoforms by alternative splicing, and the F47F2.1b gene, encoding a catalytic subunit-like protein [22,23] Other invertebrate species, such as the fruit fly Drosophila melanogaster [24], the mollusc Aplysia californica [25], the honeybee Apis mellifera [26] and the tick Amblyomma americanum [27], seem to have a single gene encoding the C-subunit Our results from MS analysis Mussel C-subunit isoforms revealed that almost all tryptic peptide masses were common to all C-subunit isoforms, and only a few m ⁄ z peaks were specific for a particular isoform, which indicates that amino acid differences are not scattered over the whole sequences, but rather limited to a particular region of the proteins On the other hand, the presence of a peak at 1605.8 Da was observed in the spectra of C2, C3 ⁄ C5 and C6 that was absent in those of C1 and C4; moreover, a partial amino acid sequence derived from this peak matches a sequence located at the N-terminal region of an alternatively spliced C-subunit isoform from the mollusc Aplysia Thus, taken together, these results indicate that C2, C3 ⁄ C5 and C6 differ from C1 and C4 at the N-termini; that is, both sets of isoforms are likely to be encoded by two alternative first exons Interestingly, C1 and C4 also had a common peptide (m ⁄ z peak 1059.5 Da), absent in the remaining isoforms, which would be the equivalent to that of 1605.8 Da, although, unfortunately, its sequence could not be determined In conclusion, structural data strongly suggest that at least some of the C-subunits identified in mussel arise as a result of differential splicing events involving various forms of the first exon, as has been widely reported for C-subunits from both mammalian and invertebrate sources [22,23,25, 26,28–32] Sequence alignments of tryptic peptides from mussel C-subunit isoforms with the bovine Ca-isoform showed a degree of sequence identity near to 90%, which confirms that the PKA C-subunit is a highly conserved protein As expected, mussel sequences contain some structural motifs, conserved throughout the protein kinase family, that are crucial for Mg2+ and ATP binding [2] For example: (a) the glycine-rich loop or nucleotide positioning motif (GxGxxG), which is particularly important for positioning the phosphates of ATP; (b) the glutamic acid residue occupying position 91 in the bovine Ca-isoform, which suitably positions Lys72, which, in turn, binds to the a-phosphate and b-phosphate of ATP; and (c) the Mg2+ positioning loop or DGF motif, with the aspartic acid residue chelating the primary Mg2+ ion that bridges the b-phosphate and c-phosphate of ATP Various authors have proposed that the functional significance of C-subunit diversity could be related to the different ability of C-subunit isoforms to phosphorylate cellular proteins, and ⁄ or to interact with partner proteins that determine the subcellular distribution of PKA activity [23,33,34] In mussel, the C-subunit isoforms isolated from the PAM tissue displayed an identical pattern of protein phosphorylation; however, the C-subunit isoforms from the FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS 4485 Mussel C-subunit isoforms J R Bardales et al mantle tissue showed minor but reproducible differences in this pattern, despite the fact that they phosphorylated a synthetic peptide substrate with similar apparent affinity Specifically, certain proteins from a mantle tissue extract were phosphorylated in vitro by C4, the main C-subunit isoform present in that tissue, but not by C5 or C6 Therefore, this finding suggests that some of the mussel C-subunit isoforms differ in their ability to phosphorylate cellular proteins, as has also been reported for Aplysia C-subunit isoforms [33] In summary, in this work we demonstrate the presence of several structurally different isoforms of the PKA C-subunit in mussel tissues In principle, the combination of these catalytically active C-subunits with the two types of R-subunit previously identified (Rmyt1 and Rmyt2 [11–13]) could potentially generate multiple PKA holoenzymes In order to establish the functional differences among these PKA isoforms, it would now be interesting to investigate the ability of C-subunits to interact with partner proteins, including Rmyt1 and Rmy2, and to examine the cellular distribution of both R-subunit and C-subunit isoforms in the mussel tissues Experimental procedures Molluscs Sea mussels of the species M galloprovincialis Lmk were collected from a sea farm located at the Rı´ a de Betanzos (Galicia, north-west Spain) Molluscs were placed in tanks containing seawater and transported to the laboratory Tissues were dissected out and immediately frozen at )20 °C until use Mussel extracts Mantle tissue was homogenized : (m ⁄ v) in ice-cold buffer A (pH 7.0) [55 mm potassium phosphate, mm EDTA, mm dithiothreitol, mm phenylmethanesulfonyl fluoride, mgỈL)1 leupeptin and mgỈL)1 pepstatin A (SigmaAldrich Quı´ mica, Madrid, Spain)], using a Potter-Elvehjem homogenizer PAM tissue was homogenized : (m ⁄ v) in ice-cold buffer B (pH 7.0) (30 mm potassium phosphate, mm EDTA, mm dithiothreitol, mm phenylmethanesulfonyl fluoride, mgỈL)1 leupeptin and mgỈL)1 pepstatin A), using a blade homogenizer (VirTis Tempest IQ2; SP Industries, Warminster, PA, USA) The homogenates were centrifuged at 35 000 g for 30 at °C in a refrigerated centrifuge (Beckman Coulter, Fullerton, CA, USA), and the supernatants, once filtered through glass wool, constituted the crude extracts 4486 Separation of C-subunit isoforms First, C-subunit was purified from PAM and mantle tissues as described previously [35,36] Briefly, the procedure is based on the binding of PKA, through its R-subunit, to DEAE–cellulose, and the specific elution of the C-subunit by addition of cAMP, which causes the dissociation of holoenzyme The crude extract obtained from each tissue was mixed with DEAE–cellulose (DE52; Whatman International, Maidstone, UK) at 30 mL gel per gram of protein After h of gentle stirring, the gel was allowed to settle – to allow the supernatant containing unbound proteins to be discarded – and then packed into a chromatographic column Next, the gel was extensively washed with the homogenization buffer, and then C-subunit was specifically eluted with the same buffer containing 0.12 mm cAMP (Sigma-Aldrich Quı´ mica) The fractions showing protein kinase activity were pooled and concentrated to  mL by ultrafiltration through a PM-30 membrane (Millipore, Bedford, MA, USA) This procedure allows enzymatic preparations containing mainly C-subunit together with minor contaminant proteins to be obtained The separation of C-subunit isoforms was performed by means of cation exchange chromatography on a Mono-S HR ⁄ FPLC column (GE Healthcare Bioscience, Uppsala, Sweden) Samples (2 mL) of the enzymatic preparations obtained from the PAM and the mantle tissues were applied to the column, previously equilibrated with buffer C (pH 6.8) (45 mm potassium phosphate, mm dithiothreitol) The column was then washed with buffer C to eliminate most contaminant proteins, and C-subunit isoforms were eluted by applying a continuous NaCl gradient (0–0.4 m in buffer C) The collected fractions of 0.5 mL were assayed for protein kinase activity, and also analysed by SDS ⁄ PAGE and western blotting C-subunit from bovine heart was purified following the procedure of Pepperkok et al [16], and purified enzyme was separated into fractions CA and CB by cation exchange chromatography on a Mono-S HR ⁄ column (GE Healthcare Bioscience) [16] Assay of C-subunit activity and determination of kinetic parameters C-subunit activity was assayed using the synthetic peptide Kemptide (Sigma-Aldrich Quı´ mica) as substrate In a total volume of 50 lL, the assay contained 50 mm Tris ⁄ HCl (pH 7.0), mm dithiothreitol, mm MgCl2, 0.2 mm [32P]ATP[cP] ( 100 c.p.m.Ỉpmol)1) (Hartmann Analytic, Braunschweig, Germany), and a sample, suitably diluted, containing C-subunit The reactions were started by addition of 100 lm Kemptide In the kinetic experiments, the concentrations of Kemptide ranged from to 150 lm and the concentrations of PKI(5–24) (Sigma-Aldrich Quı´ mica) FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS J R Bardales et al ranged from to 200 nm After 10 at 25 °C, the reactions were stopped by addition of 10 lL of 300 mm phosphoric acid Next, 30 lL of the mixture was spotted onto a phosphocellulose disc paper, and the discs were: (a) washed three times with 75 mm phosphoric acid and gently shaken to remove free ATP; (b) dried under a lamp; and (c) counted with mL of scintillation liquid Ecoscint H (National Diagnostics, Hessle, UK) in a scintillation counter One activity unit was defined as the quantity of enzyme that transfers nmol of phosphate to Kemptide per Experimental data describing the dependence of protein kinase activity on Kemptide concentrations were fitted to the Michaelis–Menten equation, and the values of the Michaelis–Menten (Km) and maximum velocity (Vmax) constants were determined from the plots ⁄ Vo versus ⁄ [S], where Vo is the initial rate at a given substrate concentration [S] The I0.5 for PKI(5–24) (concentration of peptide that reduces enzyme activity by 50%) was determined from plots of Vo versus [PKI(5–24)] at saturating concentrations of Kemptide and ATP Phosphorylation of mussel proteins by purified C-subunit isoforms Aliquots of the crude extract from PAM (100 lg of protein) or from mantle tissue (120 lg of protein) were individually incubated at 25 °C with each tissue-purified C-subunit isoform (5 unitsỈmg)1 protein) in the presence of mm MgCl2 and 0.2 mm [32P]ATP[cP] ( 500 c.p.m.Ỉpmol)1) At 20 min, reactions were stopped by adding a one-quarter volume of SDS sample buffer [250 mm Tris ⁄ HCl (pH 6.8), 8% (m ⁄ v) SDS, 20% (v ⁄ v) 2-mercaptoethanol, 40% (v ⁄ v) glycerol] and boiled for Samples were then analysed by 10% SDS ⁄ PAGE, and the gel was stained with Coomassie Brilliant Blue R (Sigma-Aldrich Quı´ mica), destained, dried, and exposed for autoradiography at )80 °C Densitometric evaluation of the autoradiographs was carried out using the versadoc imaging system (Bio-Rad Laboratories, Hercules, CA, USA) SDS/PAGE and western blotting SDS ⁄ PAGE was carried out according to Laemmli [37], using 10% polyacrylamide slab-gels of size 16 · 16 cm (Protean II xi cell) or 8.2 · 6.2 cm (Mini Protean cell) (Bio-Rad Laboratories) For performance of western blot analysis, the proteins were transferred to a poly(vinylidene difluoride) membrane (Immobilon-P; Millipore) by applying a 400 mA current for h at °C After blocking for h at room temperature with 5% nonfat dry milk in 20 mm Tris ⁄ HCl with Tween-20 (Tris ⁄ HCl, pH 7.5, 0.15 m NaCl, 0.1% Tween-20), membranes were washed with Tris ⁄ HCl with Tween-20 and then incubated overnight at °C with the primary antibodies: (a) polyclonal antibody against human Ca-isoform (sc903; Santa Cruz Biotechnology, Mussel C-subunit isoforms Santa Cruz, CA, USA) diluted : 2500 in Tris ⁄ HCl with Tween-20; (b) monoclonal antibody against phosphoserine (P3430; Sigma-Aldrich Quı´ mica) diluted : 2000 in Tris ⁄ HCl with Tween-20; or (c) monoclonal antibody against phosphothreonine (P6623; Sigma-Aldrich Quı´ mica) diluted : 1500 in Tris ⁄ HCl with Tween-20 After washing with Tris ⁄ HCl with Tween-20, the blots were incubated for h at room temperature with secondary antibodies (anti-rabbit IgG or anti-mouse IgG, diluted : 50 000 and : 25 000 in Tris ⁄ HCl with Tween-20, respectively) conjugated to horseradish peroxidase (Sigma-Aldrich Quı´ mica) Next, the blots were: (a) extensively washed; (b) developed with the chemiluminiscent horseradish peroxidase substrate (Millipore); and (c) exposed to X-ray film (Curix RP2 Plus; Agfa-Gevaert, Mortsel, Belgium) for a few seconds MS Samples of mussel C-subunit isoforms and bovine C-subunit (fraction CB [16,17]) were first reduced by adding a one-quarter volume of SDS sample buffer supplemented with dithiothreitol to a final concentration of 10 mm, and then alkylated with 20 mm iodoacetamide (Sigma-Aldrich Quı´ mica) for 30 in the dark Next, proteins were separated by SDS ⁄ PAGE and subjected to ‘in-gel’ digestion procedure with modified trypsin, sequence grade (Promega, Madison, WI, USA) [38] Digested samples were analysed using a MALDI-TOF ⁄ TOF Ultraflex instrument (Bruker Daltonics, Bremen, Germany) in reflector mode to obtain PMF spectra Sequences of tryptic peptides from C-subunits were determined using the CAF-PSD approach [39] After the peptide mixture was sulfonated by the CAF reagent (GE Healthcare Bioscience), PMF was performed again and peptide masses that had increased their masses by 136 or 204 Da were searched for The former represent tryptic peptides with a C-terminal arginine, and the latter those with a C-terminal lysine, where the e-amino groups of lysine residues were blocked with 2-methoxy-4,5-dihydro-1H-imidizole (Lys Tag 4H; Agilent Technologies, Santa Clara, CA, USA) in order to prevent lysine from being sulfonated The MALDI instrument was switched to PSD mode and the ion selector was adjusted for the appropriate mass PSD analysis was performed according to the manufacturer’s instructions, and the generated spectra, mainly consisting of a clean y-ion series, were interpreted manually A sequence homology search for the mussel C-subunit isoforms was conducted using the ‘short, nearly exact matches’ option of blast [40] Acknowledgements This work was supported by grant PGIDIT02RMA26101PR from the Autonomous Government of Galicia (Xunta de Galicia) FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS 4487 Mussel C-subunit isoforms J R Bardales et al References Taylor SS, Buechler JA & Yonemoto W (1990) cAMPdependent protein kinase: framework for a diverse family of regulatory enzymes Annu Rev Biochem 59, 971–1005 Smith CM, Radzio-Andzelm E, Madhusudan, Akamine P & Taylor SS (1999) The catalytic subunit of cAMPdependent protein kinase: prototype for an extended network of communication Prog Biophys Mol Biol 71, 313–341 Siegman MJ, Funabara D, Kinoshita S, Watabe S, Hartshorne D & Butler TM (1998) Phosphorylation of a twitchin-related protein controls catch and calcium sensitivity of force production in invertebrate smooth muscle Proc Natl Acad Sci USA 95, 5383– 5388 Yamada A, Yoshio M, Kojima H & Oiwa K (2001) An in vitro assay reveals essential protein components for the catch state of invertebrate smooth muscle Proc Natl Acad Sci USA 98, 6635–6640 Ishii N, Simpson AWM & Ashley CC (1989) Free calcium at rest during catch in single smooth muscle cells Science 243, 1367–1368 ´ Fernandez M, Cao J, Vega FV, Hellman U, Wernstedt C & Villamarı´ n JA (1997) cAMP-dependent phosphorylation activates phosphofructokinase from mantle tissue of the mollusc Mytilus galloprovincialis Identification of the phosphorylated site Biochem Mol Biol Int 43, 173–181 ´ Fernandez M, Cao J & Villamarı´ n JA (1998) In vivo phosphorylation of phosphofructokinase from the bivalve mollusk Mytilus galloprovincialis Arch Biochem Biophys 353, 251–256 ´ ´ ´ Sanjuan-Serrano F, Fernandez-Gonzalez M, Sanchez´ Lopez JL & Garcı´ a-Martı´ n LO (1995) Molecular mechanism of the control of glycogenolysis by calcium ions and cyclic AMP in the mantle of Mytilus galloprovincialis Lmk Comp Biochem Physiol B 110, 577–582 Storey KB & Storey JM (1990) Facultative metabolic rate depression: molecular regulation and biochemical adaptation in anaerobiosis, hibernation and aestivation Q Rev Biol 65, 145–174 10 Storey KB & Storey JM (2004) Metabolic rate depression in animals: transcriptional and translational controls Biol Rev 79, 207–233 11 Cao J, Ramos-Martı´ nez JI & Villamarı´ n JA (1995) Characterization of a cAMP-binding protein from the bivalve mollusc Mytilus galloprovincialis Eur J Biochem 232, 664–670 12 Rodrı´ guez JL, Barcia R, Ramos-Martı´ nez JI & Villamarı´ n JA (1998) Purification of a novel isoform of the regulatory subunit of cAMP-dependent protein kinase from the bivalve mollusk Mytilus galloprovincialis Arch Biochem Biophys 359, 57–62 4488 13 Dı´ az-Enrich MJ, Ibarguren I, Hellman U & Villamarı´ n JA (2003) Characterization of a type I regulatory subunit of cAMP-dependent protein kinase from the bivalve mollusk Mytilus galloprovincialis Arch Biochem Physiol 416, 119–127 14 Bardales JR, Hellman U & Villamarı´ n JA (2007) CK2mediated phosphorylation of type II regulatory subunit of cAMP-dependent protein kinase from the mollusc Mytilus galloprovincialis Arch Biochem Physiol 461, 130–137 15 Jedrzejewski PT, Girod A, Tholey A, Konig N, Thullă ner S, Kinzel V & Bossemeyer D (1998) A conserved deamidation site at Asn in the catalytic subunit of mammalian cAMP-dependent protein kinase detected by capillary LC-MS and tandem mass spectrometry Protein Sci 7, 457–469 16 Pepperkok R, Hotz-Wagenblatt A, Konig N, Girod A ă & Bossemeyer D (2000) Intracellular distribution of mammalian protein kinase A catalytic subunit altered by conserved Asn2 deamidation J Cell Biol 148, 715– 726 17 Kinzel V, Hotz A, Konig N, Gagelmann M, Pyerin W, ă Reed J, Kubler D, Hofmann F, Obst C, Gensheimer ă HP et al (1987) Chromatographic separation of two heterogeneous forms of the catalytic subunit of cyclic AMP-dependent protein kinase holoenzyme type I and type II from striated muscle of different mammalian species Arch Biochem Biophys 253, 341–349 18 Herberg FW, Bell S & Taylor SS (1993) Expression of the catalytic subunit of cAMP-dependent protein kinase in Escherichia coli: multiple isozymes reflect different phosphorylation states Protein Eng 6, 771–777 19 Uhler MD, Carmichael DF, Lee DC, Chrivia JC, Krebs EG & McKnight GS (1986) Isolation of cDNA clones coding for the catalytic subunit of mouse cAMP-dependent protein kinase Proc Natl Acad Sci USA 83, 1300– 1304 20 Uhler MD, Chrivia JC & McKnight GS (1986) Evidence for a second isoform of the catalytic subunit of cAMP-dependent protein kinase J Biol Chem 261, 15360–15363 21 Beebe SJ, Oyen O, Sandberg M, Froysa A, Hansson V & Jahnsen T (1990) Molecular cloning of a tissuespecific protein kinase (Cc) from human testis – representing a third isoform for the catalytic subunit of cAMP-dependent protein kinase Mol Endocrinol 4, 465–475 22 Gross RE, Bagchi S, Lu X & Rubin CS (1990) Cloning, characterization and expression of the gene for the catalytic subunit of cAMP-dependent protein kinase in Caenorhabditis elegans J Biol Chem 265, 6896–6907 23 Tabish M, Clegg RA, Rees HH & Fischer MJ (1999) Organization and alternative splicing of the Caenorhabditis elegans cAMP-dependent protein kinase catalytic-subunit gene (kin-1) Biochem J 339, 209–216 FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS J R Bardales et al 24 Kalderon D & Rubin GM (1998) Isolation and characterization of Drosophila cAMP-dependent protein kinase genes Genes Dev 2, 1539–1556 25 Beushausen S, Bergold P, Sturner S, Elste A, Roytenberg V, Schwartz H & Bayley H (1988) Two catalytic subunits of cAMP-dependent protein kinase generated by alternative RNA splicing are expressed in Aplysia neurons Neuron 1, 853–864 26 Eisenhardt D, Fiala A, Brau P, Rosenboom H, Kress H, Ebert PR & Menzel R (2001) Cloning of a catalytic subunit of cAMP-dependent protein kinase from the honeybee (Apis mellifera) and its localization in the brain Insect Mol Biol 10, 173–181 27 Palmer MJ, McSwain JL, Spatz MD, Tucker JS, Essenberg RC & Sauer JR (1999) Molecular cloning of cAMP-dependent protein kinase catalytic subunit isoforms from the lone star tick, Amblyomma americanum (L) Insect Biochem Mol Biol 29, 43–51 28 Beushausen S, Lee E, Walker B & Bayley H (1992) Catalytic subunits of Aplysia neuronal cAMP-dependent protein kinase with two different N termini Proc Natl Acad Sci USA 89, 1641–1645 29 Tabish M, Clegg RA, Turner CP, Jonczy J, Rees HH & Fisher MJ (2006) Molecular characterization of cAMPdependent protein kinase (PK-A) catalytic subunit isoforms in the male tick, Amblyomma hebraeum Mol Biochem Parasitol 150, 330–339 30 Desseyn JL, Burton KA & Mcknight GS (2000) Expression of a nonmyristylated variation of the catalytic subunit of protein kinase A during male germ-cell development Proc Natl Acad Sci USA 97, 6433–6438 ´ 31 Reinton N, Orstavik S, Haugen TB, Jahnsen T, Tasken T & Skalhegg BS (2000) A novel isoform of human cyclic 3¢,5¢-adenosine monophosphate-dependent protein kinase, Ca-s, localizes to sperm midpiece Biol Reprod 63, 607–611 32 Orstavik S, Reinton N, Frengen E, Langeland BT, Jahnsen T & Skalhegg BS (2001) Identification of novel splice variants of the human catalytic subunit Cb of Mussel C-subunit isoforms 33 34 35 36 37 38 39 40 cAMP-dependent protein kinase Eur J Biochem 268, 5066–5073 Panchal RG, Cheyley S & Bayley H (1994) Differential phosphorylation of neuronal substrates by catalytic subunits of Aplysia cAMP-dependent protein kinase with alternative N termini J Biol Chem 269, 23722– 23730 Bowen LC, Bicknell AV, Tabish M, Clegg RA, Rees HH & Fischer MJ (2006) Expression of multiple isoforms of the cAMP-dependent protein kinase (PK-A) catalytic subunit in the nematode Caenorhabditis elegans Cell Signal 18, 2230–2237 ´ Bejar P & Villamarı´ n JA (2006) Catalytic subunit of cAMP-dependent protein kinase from a match muscle of the bivalve mollusk Mytilus galloprovincialis: purification, characterization, and phosphorylation of muscle proteins Arch Biochem Biophys 450, 133–140 ´ ´ Cao J, Fernandez M, Vazquez-Illanes MD, RamosMartinez JI & Villamarı´ n JA (1995) Purification and characterization of the catalytic subunit of cAMPdependent protein kinase from the bivalve mollusc Mytilus galloprovincialis Comp Biochem Physiol B 111, 453–462 Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4 Nature 277, 680–685 ´ Hellman U, Wernstedt C, Gomez J & Heldin CH (1995) Improvement of an ‘In-Gel’ digestion procedure for the micropreparation of internal protein fragments for amino acid sequencing Anal Biochem 224, 451–455 Hellman U & Bhikhabhai R (2002) Easy amino acid sequencing of sulfonated peptides using post-source decay on a matrix-assisted laser desorption ⁄ ionization time-of-flight mass spectrometer equipped with a variable voltage reflector Mass Spectrom 16, 1851–1859 Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W & Lipman DJ (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs Nucleic Acids Res 25, 3389–3402 FEBS Journal 275 (2008) 4479–4489 ª 2008 The Authors Journal compilation ª 2008 FEBS 4489 ... mussel M galloprovincialis Results Separation of different isoforms of the C -subunit In order to demonstrate the presence of different isoforms of the PKA C -subunit in mussels, the protein was... isoforms [33] In summary, in this work we demonstrate the presence of several structurally different isoforms of the PKA C -subunit in mussel tissues In principle, the combination of these catalytically... tissuespecific protein kinase (Cc) from human testis – representing a third isoform for the catalytic subunit of cAMP-dependent protein kinase Mol Endocrinol 4, 465–475 22 Gross RE, Bagchi S, Lu X & Rubin