Immunohistochemical expression of E–cadherin in different tissues of the teleost fish Scophthalmus maximus

1 Immunohistochemical expression of E–cadherin in different 1 tissues of the teleost fish Scophthalmus maximus 2 Paolo Ronzaa* ; Antonio Villamarínb; Lucía Méndezc; Belén G. Pardod; Roberto 3 Bermúdeza; María Isabel Quirogaa 4 aDepartamento de Anatomía, Producción Animal y Ciencias Clínicas Veterinaria, 5 Facultad de Veterinaria, Universidade de Santiago de Compostela, 27002 Lugo, Spain. 6 bDepartamento de Bioquímica y Biología Molecular, Facultad de Veterinaria, 7 Universidade de Santiago de Compostela, 27002 Lugo, Spain. 8 cInstituto de Investigaciones Marinas, Consejo Superior de Investigaciones Científicas 9 (IIM-CSIC), 36208 Vigo, Spain. 10 dDepartamento de Zoología, Genética y Antropología Física, Facultad de Veterinaria, 11 Universidade de Santiago de Compostela, 27002 Lugo, Spain. 12 *Corresponding author 13 E-mail: pa[email protected]; Phone number: +34 982822306 14 Abstract 15 E–cadherin is an evolutionary conserved protein, whose main role as the principal 16 component of adherens junctions is supporting epithelial cell–cell adhesion. It is an 17 essential molecule for the maintenance of the epithelial barrier function and the analysis 18 of its immunohistochemical expression is a valuable resource in morphopathological, 19 ontogenetic and pathogenesis studies in mammals. As well, there is an increasing 20 understanding of the importance of E-cadherin in the physiology of the immune system 21 Accepted manuscript © 2018 Elsevier B.V. This manuscript version is made available under the CC-BY-NC-ND 4.0 license (http://creativecommons.org/licenses/by-nc-nd/4.0) 2 and the development of the immune response. Mucosal health is a primary issue in 22 aquaculture research; nevertheless, there is a lack of immunohistochemical studies of cell 23 junction proteins in fish species. In this work, an immunohistochemical technique was 24 optimized in Bouinand formalin-fixed paraffin-embedded tissues of turbot 25 Scophthalmus maximus, employing a commercial antibody raised against human E-26 cadherin. The specificity of the antibody in recognizing the molecule in this teleost 27 species was tested by western blot and mass spectrometry-based proteomic analyses. The 28 assays showed a good specificity and indicated that the antibody recognizes the well 29 conserved cytoplasmic domain of the protein. Immunohistochemistry showed the 30 localisation of E-cadherin at cell-cell contact in the epithelia of the different organs, 31 between the hepatocytes and the pancreatic acinar cells, as well as in the reticulo-32 epithelial stroma of the thymus. Also, the immunoreaction was observed in the cells 33 constituting the melano-macrophage centres in the spleen and kidney. No 34 immunostaining was detected, as expected, only in the heart and brain. No significant 35 difference was noticed between the two fixative used for collecting the tissues samples. 36 This is the first description of E-cadherin immunohistochemical expression in several 37 tissues of a teleost. The immunohistochemical technique represents a useful tool to be 38 used in the different areas of fish health research. 39 Keyword: Immunohistochemistry; cadherin-1; cell junctions; epithelial barriers; mucosal 40 health 41 42 Accepted manuscript 3 1. Introduction 43 Cell–cell adhesion is a fundamental structural feature of multicellular organisms, 44 mediated by a set of specialized membrane structures termed intercellular junctions. 45 Particularly, the ability of epithelial cells to organize into monolayered sheets is a 46 prerequisite for multicellularity, thereby providing tissue integrity, barrier function, and 47 tissue polarity in metazoan organisms (Bruser and Bogdan, 2017). Over the course of 48 morphological evolution, metazoan animals have diversified the architecture of their cell–49 cell junctions, which includes tight junctions, adherens junctions (AJ), and desmosomes. 50 Among them, AJs are the only detected throughout the metazoan phyla, whereas other 51 junctional types show restricted phylogenetic distributions (Oda and Takeichi, 2011). 52 Thus, AJs could be considered the universal adhesion machinery for the generation and 53 maintenance of multicellular animal bodies. E-cadherin, a calcium–dependent 54 transmembrane protein, is the structural and functional core of AJs, mediating trans–55 homophilic interactions between neighbouring cells. Its adhesive functions are mediated 56 by the extracellular cadherin repeat domains, while the highly conserved intracellular 57 domains form a complex with catenins linking E–cadherin to the actin cytoskeleton, being 58 so involved in junctional maintenance, dynamics, and plasticity of epithelial tissues 59 (Bruser and Bogdan, 2017; Oda and Takeichi, 2011; van Roy and Berx, 2008). 60 In teleost, the mucosal surfaces (skin, gills and gastrointestinal tract) constitute the first 61 line of defence against pathogens invasion and carry out multiple physiological processes, 62 such as osmoregulation, waste excretion and nutrient adsorption. For fish, even more than 63 for terrestrial species, mucosal barriers and their health are of primary importance to face 64 the continue interactions with the aquatic microbiota (Peatman and Beck, 2015). Almost 65 all of the most urgent areas of aquaculture research require a deeper understanding of 66 Accepted manuscript 4 mucosal barriers, from pathology issues to vaccine delivery, nutrition, nutraceutics and 67 microflora modulation (Peatman et al., 2015). The importance of E–cadherin as 68 diagnostic marker and the main role it plays in human diseases, from several cancer types 69 to different skin and gastrointestinal conditions, have been widely reported (Baniak et al., 70 2016; Christou et al., 2017; Maretzky et al., 2008; Singhai et al., 2011; Sugihara, 2016; 71 von Zeidler et al., 2014; Wu et al., 2007; Zbar et al., 2004). In fish pathology, E–cadherin 72 has recently been identified as a key molecule for resistance to infectious pancreatic 73 necrosis (IPN) in Atlantic salmon (Moen et al., 2015), and another work on the same 74 species reported modulated intestinal E–cadherin gene expression in response to an 75 experimental dietary treatment affecting intestinal fluid permeability (Hu et al., 2016). 76 The investigation of E–cadherin immunohistochemical expression is a primary focus in 77 morphological studies, including ontogenetic and morphopathological characterization of 78 the different organs, as well as an irreplaceable complement for studies based on gene 79 expression analysis (Bondow et al., 2012; Fuertes et al., 2013; Gassler et al., 2001; 80 Kuwahara et al., 2001; Sakamoto et al., 2008; Schneider et al., 2010). However, no 81 comprehensive study of E–cadherin immunolocalisation in teleost tissues has been 82 addressed and there is still a scarce knowledge of the physiological distribution of this 83 protein and its change under pathological conditions. In this study, an 84 immunohistochemical technique based on a commercial antibody was optimized in 85 Bouinand formalin-fixed paraffin–embedded tissues from turbot (Scophthalmus 86 maximus), an economically–important marine species, and the distribution of E–cadherin 87 in different tissues of healthy specimens analysed. 88 2. Materials and methods 89 2.1. Fish and sampling procedures 90 Accepted manuscript 5 For this study, 10 adult turbot (2,127 ± 182.2 g mean weight) were employed. Fish were 91 euthanized by overexposure to tricaine methane sulfonate (MS222, Sigma–Aldrich, 92 Denmark) and necropsied. For histological examination and immunohistochemistry, 93 tissues samples from kidney, spleen, thymus, digestive tract, liver, pancreas, heart, gills, 94 brain and skin were collected. Samples were fixed in Bouin’s fluid at 4 ºC or in formalin 95 during 24 hours and then paraffin–embedded. Histological analysis was performed on 96 H&E and toluidine blue–stained sections. For western blot analysis, samples (1 g) of 97 anterior and posterior intestine were collected and extensively washed with ice cold PBS 98 containing 1 mM of phenylmethylsulfonyl fluoride (PMSF) before storage at –80ºC. 99 All experimental protocols were approved by the Institutional Animal Care and Use 100 Committee of the University of Santiago de Compostela (Spain). 101 2.2. Western blotting and E-cadherin sequences analysis 102 Western blot analysis was performed to assess the specificity of the primary antibody on 103 protein extracts from anterior and posterior intestine of turbot. Protein extracts from 104 human skin and intestine (ileum) were used as positive control. Tissues were 105 homogenized in ice–cold homogenization buffer (50 mM Tris–HCl, 150 mM NaCl, 5mM 106 EDTA, 1 mM dithiothreitol, 1 mM PMSF, 1 mM benzamidine hydrochloride hydrate, 5 107 µg/mL pepstatin A, 5 µg/mL aprotinin, 5 µg/mL leupeptin, 1% triton X–100, 0,5 % 108 sodium deoxycholate, 0,1 % sodium dodecyl sulfate pH 7.4). Homogenates were 109 centrifuged at 35,000 g for 15 minutes at 4 ºC, and supernatants filtered through gauze 110 and frozen in aliquots at –80º C until required. Protein concentrations were determined 111 by Bradford method according to manufacturer’s instruction (Bio–Rad, Hercules, CA, 112 USA). 113 Accepted manuscript 6 Samples of protein extracts were mixed with 1/4 vol of 5X SDS sample buffer (250 mM 114 Tris–HCl, 8 % SDS, 40% glycerol, 20% β–mercaptoethanol, pH 6.8) and denatured by 115 heating at 37 ºC for 20 min. Approximately 30 µg of total protein per tissue were resolved 116 by SDS–PAGE using 7,5% polyacrylamide slabs–gel and proteins were transferred to 117 polyvinylidene difluoride (PVDF) membranes (Immobilon–P; Millipore, Bedford, MA, 118 USA) by applying a 275 mA current for 3 h at 4 ºC. Sodium dodecyl sulphate (SDS) was 119 added (0,025%) to the transference buffer in order to facilitate the transfer of high 120 molecular weight proteins. After blocking for 2 h and 30 min at room temperature with 121 5% non–fat dry milk in TTBS (20 mM Tris–HCl buffer, pH 7.5, 0.15 M NaCl, 0.1% 122 Tween 20), membranes were incubated overnight at 4º C with the anti–human E–cadherin 123 primary antibody (mouse monoclonal antibody, clone NCH–38, M3612, Dako, Denmark) 124 diluted 1:500 with TTBS. The blots were then incubated for 1 h at room temperature with 125 anti–mouse secondary antibody conjugated to horseradish peroxidase (Sigma–Aldrich, St 126 Louis, MO, USA) diluted 1:50,000 with TTBS and, finally, developed with the 127 chemiluminiscent HRP substrate (Millipore) and exposed to X–ray film (Curix RP2 Plus; 128 Agfa–Gevaert, Mortsel, Belgium) for a few seconds. Membranes were washed 5 times 129 for 10 min with TTBS between subsequent steps. Negative control was performed by 130 substituting primary antibody with TTBS. In order to estimate the apparent molecular 131 weight of detectable bands, a sample of molecular weight standards (Unstained Broad 132 Range SDS-PAGE Standards, Bio-Rad) was also applied on the gel. The membranes were 133 finally stained with a Ponceau-S solution to mark the positions of the molecular weight 134 standards. 135 Additionally, for all tested samples SDS–PAGE and gel staining with Coomassie Brilliant 136 Blue R (Sigma–Aldrich) were carried out. 137 Accepted manuscript 7 The amino acid sequence of turbot E–cadherin was inferred using ExPASy Translate tool 138 (http://web.expasy.org/translate/) from the nucleotide sequence (GenBank accession 139 number: MG137250) identified in a previous study (Robledo et al., 2014), and its 140 theoretical molecular weight estimated by ExPASy Compute pI/Mw tool 141 (http://web.expasy.org/compute_pi/). For comparison purposes, the amino acid sequence 142 alignments were performed by using the BLASTP 2.6.1 online program (Altschul et al., 143 2005; Altschul et al., 1997). 144 2.3. In–gel digestion and protein identification by nano–HPLC–ESI–IT–MS/MS 145 Based on western blot results, the labelled bands of 110 kDa y 38 kDa obtained in turbot 146 were manually excised from SDS–PAGE Coomassie-stained gels. Proteins were in–gel 147 reduced, alkylated and digested with trypsin as previously described (Jensen et al., 1999). 148 Briefly, protein bands were three times washed with water and dehydrated with 149 acetonitrile. Gel plugs were then dehydrated in a vacuum centrifuge and finally 150 rehydrated with a 0.5 μM solution of sequencing grade bovine trypsin (Promega, 151 Madison, WI) in 25 mM ammonium bicarbonate buffer, pH 8.0, for at least 40 min on 152 ice. After the rehydration step, samples were digested overnight at 37 °C and peptides 153 were then cleaned–up by using ZipTip C18 according to manufacturer’s instructions. The 154 peptide mixture was dried in vacuum centrifugation, and dissolved again in 15 μL of 1% 155 formic acid prior to mass analysis. 156 Trypsin–digested proteins were analysed using a Dionex UltiMate 3000 RSLCnano 157 system (Thermo Fisher Scientific) coupled to a LTQ Velos–Pro mass spectrometer 158 (Thermo Fisher Scientific). The separation of the peptides was done on an Acclaim 159 PepMap100 Nano Trap Column, C18, 5 μm, 100 Å, 100 μm × 1 cm (Thermo Fisher 160 Scientific) coupled to an RP column Acclaim PepMap RSLC 75 μm × 150 mm, C18, 2 161 Accepted manuscript 8 μm, 100 Å (Thermo Fisher Scientific). Mobile phases A and B were respectively 0.1% 162 formic acid in water and in 100% acetonitrile. A 90 min linear gradient from 5% to 35% 163 B, at a flow rate of 300 nL/min, was used. For ionization, a spray voltage of 2.10 kV and 164 a capillary temperature of 200 °C were used. Peptide detection was performed by using 165 survey scans from 350 to 1600 Da (2 μscans), followed by MS/MS scans (2 μscans) of 166 the six more intense peaks using an isolation width of 1 Da and a normalised collision 167 energy of 35%. Singly charged ions were excluded from MS/MS fragmentation and a 168 dynamic exclusion enabled, with repeat count set to 2 and exclusion duration of 30 s. 169 Protein identification was performed using the database searching function of the PEAKS 170 7 software (Ma et al., 2003; Zhang et al., 2012) (Bioinformatics Solutions Inc, Waterloo, 171 Ontario, Canada), to compare experimental MS/MS spectra against reference MS/MS 172 spectra from the UniProtKB/TrEMBL database (release 2017_04; 84827567 sequence 173 entries), which also included their respective decoy sequences. The following limitations 174 were used for the searches: tryptic cleavage, up to 2 missed cleavage sites, and tolerances 175 ±1.0 Da for precursor ions and ±0.8 Da for MS/MS fragments ions. The FDR was kept 176 below 1% and only proteins with at least one unique peptide were considered. When 177 protein identification failed, the PEAKS algorithm was used to perform de novo 178 sequencing for each input spectrum. The same parameters (mass error tolerance and 179 PTMs) specified for database search are also used for de novo sequencing. Average Local 180 confidence (ALC) for each de novo sequence was set at ≥60%. 181 2.4. Immunohistochemistry 182 Thin sections (3 µm thick) were obtained from all the sampled organs. The sections were 183 placed on slides treated with silane to improve section adherence and dried overnight at 184 37 ºC. After deparaffination with xylene (two 5–min washes) and hydration through a 185 graded ethanol series, slides were incubated with a peroxidase–blocking solution (Dako, 186 Accepted manuscript 9 Denmark) for 30 minutes to quench endogenous peroxidase activity. Antigen retrieval 187 was performed using high pH antigen retrieval buffer (Dako, Denmark), following the 188 manufacturer instruction. The optimal working dilution for the monoclonal antibody anti–189 human E–cadherin was determined to be 1:50 with an incubation time of 2 h at room 190 temperature. Then, slides were incubated during 30 min with horseradish peroxidase 191 (HRP)–labelled polymer conjugated to rabbit secondary antibody and peroxidase reaction 192 was developed using a diaminobenzidine–positive chromogen (EnVision+ System–HRP 193 kit, K4007, Dako). All incubations were performed in a humid chamber at room 194 temperature, and the sections were washed three times for 5 min in 0.1 M phosphate 195 buffered saline containing 0.05% Tween–20 between all subsequent steps. Sections were 196 finally counterstained with haematoxylin, dehydrated and coverslipped with DePeX 197 mounting medium for microscopic observation. Tissue sections from formalin-fixed 198 paraffin-embedded human intestine were used as positive controls. In sections included 199 as negative controls, primary antibody was replaced with either PBS or antibody diluent. 200 3. Results 201 3.1. Histological analysis 202 The histological evaluation of the sampled tissues did not show any significant 203 pathological alteration or presence of pathogenic agents. 204 3.2. Western blotting 205 In human tissues, an expected signal corresponding to a molecular weight of 206 approximately 97 kDa was detected in both the skin and intestinal extracts. Furthermore, 207 an additional intensely labelled band of about 38 kDa was observed for the skin. In turbot, 208 the band corresponding to the mature form of E–cadherin reached approximately 110 209 Accepted manuscript 16 Kuwahara, M., Hatoko, M., Tada, H., Tanaka, A., 2001. E-cadherin expression in wound 376 healing of mouse skin. J. Cutan. Pathol. 28, 191-199. 377 Lim, Y.-S., Lee, H.C., Lee, H.-S., 2007. Switch of cadherin expression from Eto N-type 378 during the activation of rat hepatic stellate cells. Histochem. Cell Biol. 127, 149-379 160. 380 Ma, B., Zhang, K., Hendrie, C., Liang, C., Li, M., Doherty-Kirby, A., Lajoie, G., 2003. 381 PEAKS: powerful software for peptide de novo sequencing by tandem mass 382 spectrometry. Rapid Commun. Mass Spectrom. 17, 2337-2342. 383 Malhao, F., Urbatzka, R., Navas, J.M., Cruzeiro, C., Monteiro, R.A., Rocha, E., 2013. 384 Cytological, immunocytochemical, ultrastructural and growth characterization of 385 the rainbow trout liver cell line RTL-W1. Tissue Cell. 45, 159-174. 386 Marambaud, P., Shioi, J., Serban, G., Georgakopoulos, A., Sarner, S., Nagy, V., Baki, L., 387 Wen, P., Efthimiopoulos, S., Shao, Z., Wisniewski, T., Robakis, N.K., 2002. A 388 presenilin-1/gamma-secretase cleavage releases the E-cadherin intracellular 389 domain and regulates disassembly of adherens junctions. EMBO J. 21, 1948-390 1956. 391 Maretzky, T., Scholz, F., Koten, B., Proksch, E., Saftig, P., Reiss, K., 2008. ADAM10-392 mediated E-cadherin release is regulated by proinflammatory cytokines and 393 modulates keratinocyte cohesion in eczematous dermatitis. J. Invest. Dermatol. 394 128, 1737-1746. 395 Maretzky, T., Reiss, K., Ludwig, A., Buchholz, J., Scholz, F., Proksch, E., de Strooper, 396 B., Hartmann, D., Saftig, P., 2005. ADAM10 mediates E-cadherin shedding and 397 regulates epithelial cell-cell adhesion, migration, and beta-catenin translocation. 398 Proc. Natl. Acad. Sci. U. S. A. 102, 9182-9187. 399 Moen, T., Torgersen, J., Santi, N., Davidson, W.S., Baranski, M., Ødegård, J., Kjøglum, 400 S., Velle, B., Kent, M., Lubieniecki, K.P., Isdal, E., Lien, S., 2015. Epithelial 401 cadherin determines resistance to infectious pancreatic necrosis virus in Atlantic 402 salmon. Genetics. 200, 1313. 403 Oda, H., Takeichi, M., 2011. Evolution: structural and functional diversity of cadherin at 404 the adherens junction. J. Cell Biol. 193, 1137-1146. 405 Padrós, F., Crespo, S., 1996. Ontogeny of the lymphoid organs in the turbot Scophthalmus 406 maximus: a light and electron microscope study. Aquaculture. 144, 1-16. 407 Peatman, E., Beck, B.H., 2015. 1 - Why mucosal health?, Mucosal Health in Aquaculture. 408 Academic Press, San Diego, pp. 1-2. 409 Peatman, E., Lange, M., Zhao, H., Beck, B.H., 2015. Physiology and immunology of 410 mucosal barriers in catfish (Ictalurus spp.). Tissue Barriers. 3, e1068907. 411 Robledo, D., Ronza, P., Harrison, P.W., Losada, A.P., Bermúdez, R., Pardo, B.G., 412 Redondo, M.J., Sitjà-Bobadilla, A., Quiroga, M.I., Martínez, P., 2014. RNA-seq 413 analysis reveals significant transcriptome changes in turbot (Scophthalmus 414 maximus) suffering severe enteromyxosis. BMC Genomics. 15, 1149. 415 Sakamoto, A., Murata, K., Suzuki, H., Yatabe, M., Kikuchi, M., 2008. 416 Immunohistochemical observation of co-expression of Eand N-cadherins in rat 417 organogenesis. Acta Histochem. Cytochem. 41, 143-147. 418 Schneider, M.R., Dahlhoff, M., Horst, D., Hirschi, B., Trülzsch, K., Müller-Höcker, J., 419 Vogelmann, R., Allgäuer, M., Gerhard, M., Steininger, S., Wolf, E., Kolligs, F.T., 420 2010. A key role for E-cadherin in intestinal homeostasis and Paneth cell 421 maturation. PLoS One. 5, e14325. 422 Singhai, R., Patil, V.W., Jaiswal, S.R., Patil, S.D., Tayade, M.B., Patil, A.V., 2011. E-423 Cadherin as a diagnostic biomarker in breast cancer. N. Am. J. Med. Sci. 3, 227-424 233. 425 Accepted manuscript 17 Steinel, N.C., Bolnick, D.I., 2017. Melanomacrophage centers as a histological indicator 426 of immune function in fish and other poikilotherms. Front Immunol. 8, 827. 427 Sugihara, T., 2016. Loss of adherens junction protein E-cadherin is a biomarker of high-428 grade histology and poor prognosis in endometrial cancer. Ann. Clin. Lab. Res. 4, 429 1. 430 Tsuchiya, B., Sato, Y., Kameya, T., Okayasu, I., Mukai, K., 2006. Differential expression 431 of N-cadherin and E-cadherin in normal human tissues. Arch. Histol. Cytol. 69, 432 135-145. 433 Van den Bossche, J., Van Ginderachter, J.A., 2013. E-cadherin: from epithelial glue to 434 immunological regulator. Eur. J. Immunol. 43, 34-37. 435 van Roy, F., Berx, G., 2008. The cell-cell adhesion molecule E-cadherin. Cell Mol. Life 436 Sci. 65, 3756-3788. 437 Vigliano, F.A., Losada, A.P., Castello, M., Bermúdez, R., Quiroga, M.I., 2011. 438 Morphological and immunohistochemical characterisation of the thymus in 439 juvenile turbot (Psetta maxima, L.). Cell Tissue Res. 346, 407-416. 440 von Zeidler, S.V., de Souza Botelho, T., Mendonça, E.F., Batista, A.C., 2014. E-cadherin 441 as a potential biomarker of malignant transformation in oral leukoplakia: a 442 retrospective cohort study. BMC Cancer. 14, 972. 443 Wu, S., Rhee, K.J., Zhang, M., Franco, A., Sears, C.L., 2007. Bacteroides fragilis toxin 444 stimulates intestinal epithelial cell shedding and gamma-secretase-dependent E-445 cadherin cleavage. J. Cell Sci. 120, 1944-1952. 446 Zbar, A.P., Simopoulos, C., Karayiannakis, A.J., 2004. Cadherins: an integral role in 447 inflammatory bowel disease and mucosal restitution. J. Gastroenterol. 39, 413-448 421. 449 Zhang, J., Xin, L., Shan, B., Chen, W., Xie, M., Yuen, D., Zhang, W., Zhang, Z., Lajoie, 450 G.A., Ma, B., 2012. PEAKS DB: De novo sequencing assisted database search 451 for sensitive and accurate peptide identification. Mol. Cell. Proteomics. 11. 452 453 Accepted manuscript 18 Table 1. NanoHPLC-ESI-IT-MS/MS analysis of the 110 kDa and 38 kDa labeled 454 bands found by western blot with an anti-E-cadherin antibody in Scophthalmus 455 maximus. 456 Band ID Presence of E-cadherin Protein description Uniprot Code - 10lgP #Peptides #Unique #Sequences SC (%) MWt (kDa) 110 kDa Band YES Cadherin-1 OS= Fundulus heteroclitus PE=4 SV=1 A0A146 XRX5 47,63 1 1 R.PANPD ELGNFID DNLK.A 2 83.6 38 kDa Band YES Cadherin-1 OS= Fundulus heteroclitus PE=4 SV=1 A0A146 XRX5 37,07 1 1 R.PANPD ELGNFID DNLK.A 2 83.6 110 kDa Band YES E–cadherin protein sequence of S. maximus - - 1 1 D.NQGLS QDNTVQ TK.V* 1.3 110 Different parameters supporting the successful protein identification of E-cadherin by MS are 457 indicated: PEAKS protein score (–10lgP), number of matching total and unique peptides (FDR < 458 1%), peptide sequences, % sequence coverage (SC) and theoretical protein mass (MWt). 459 *Obtained by de novo sequencing using PEAKS algorithm and alignment with E–cadherin protein 460 sequence of S. maximus using the BLASTP 2.6.1 (Score: 28.9 bits(63); Expect: 1e-06; Method: 461 Compositional matrix adjust; Identities: 13/13(100%); Gaps: 0/13(0%)). 462 Accepted manuscript 19 463 Fig 1. Western blot analysis of E-cadherin expression in protein extracts from 464 human and turbot (Scophthalmus maximus) tissues. S = human skin; I = human ileum; 465 AI = turbot anterior intestine; PI = turbot posterior intestine. At the right, the position of 466 the molecular weight standards is indicated. 467 468 Fig 2. Comparisons between human and turbot E-cadherin amino acid sequences. 469 A) Output of BLASTp alignment of human (P12830) and turbot (Scophthalmus maximus) 470 E-cadherin. B) Comparison between the amino acid sequences of the C-terminal fragment 471 Accepted manuscript 20 known in human as CTF1 and its counterpart in turbot. The amino acids highlighted in 472 red corresponds to the unique peptide sequence found by nano–LC–MS/MS analysis of 473 the excised bands from Western blot in turbot tissues that matched with the E–cadherin 474 sequence of the teleost Fundulus heteroclitus. Conservative amino acid changes are 475 shown by a "+" sign between the aligned residues. 476 477 Fig 3. E-cadherin immunolocalisation at cell-cell contacts in the lining epithelia of 478 turbot skin (A), gills (B), intestine (C) and stomach (D). Note the immunoreaction of 479 the gastric pits (3D, arrows), while this is not present in the underlying gastric glands. 480 Bars = 100 µm (Inset 3A= 50 µm). 481 Accepted manuscript 21 482 Fig 4. E-cadherin immunohistochemical expression in turbot (Scophthalmus 483 maximus) kidney (A), spleen (B) and thymus (C). A) The immunoreaction was 484 observed in the epithelia of the renal ducts and tubules, as well as in the epithelium of the 485 Bowman’s capsule (inset). Bar = 200 µm; Inset’s bar = 25 µm. B) Immunostaining of the 486 splenic melano-macrophage centres. Bar = 100 µm. C) Immunolocalisation of E-cadherin 487 in thymus stroma, showing a stronger staining intensity in the inner (I) than in the outer 488 (O) part of the organ. Note also the immunoreaction of the epithelium separating the 489 thymus from the gill chamber (arrowheads). Bar = 100 µm. 490 Accepted manuscript 22 491 Fig 5. E-cadherin immunohistochemistry in turbot (Scophthalmus maximus) liver 492 (A) and pancreas (B). A) Membranous staining pattern of the hepatocytes and in the 493 epithelium of a biliar duct (Inset). Bars = 50 µm. B) Immunostaining of E-cadherin at 494 cell-cell contacts between acinar cells, while this is not noticed in a pancreatic islet of 495 Langerhans (star). Bar = 100 µm. 496 497 Accepted manuscript