ORIGINAL RESEARCH published: 04 February 2021 doi: 10.3389/fimmu.2020.613824 Skin Multi-Omics-Based Interactome Analysis: Integrating the Tissue and Mucus Exuded Layer for a Comprehensive Understanding of the Teleost Mucosa Functionality Edited by: as Model of Study Samuel A. M. Martin, University of Aberdeen, Felipe E. Reyes-López 1,2,3*†, Antoni Ibarz 4*†, Borja Ordóñez-Grande 4, Eva Vallejos-Vidal 5, United Kingdom Karl B. Andree 6, Joan Carles Balasch 1, Laura Fernández-Alacid 4, Ignasi Sanahuja 4, Reviewed by: Sergio Sánchez-Nuño 4, Joana P. Firmino 6,7, Leonardo Pavez 8, Javier Polo 9, Lluis Tort 1 James Watkin Wynne, and Enric Gisbert 6 CSIRO, Australia Monica Fengsrud Brinchmann, 1Departament de Biologia Cel·lular, Fisiologia i Immunologia, Universitat de Autònoma de Barcelona (UAB), Bellatera, Spain, Nord University, Norway 2Facultad de Medicina Veterinaria y Agronomía, Universidad de Las Américas, Providencia, Chile, 3 Consorcio Tecnológico *Correspondence: de Sanidad Acuícola, Ictio Biotechnologies S.A., Santiago, Chile, 4 Departament de Biologia Cel·lular, Fisiologia i Felipe E. Reyes-López Immunologia, Universitat de Barcelona (UB), Barcelona, Spain, 5 Centro de Biotecnologı´a Acuı´cola, Facultad de Quı´mica y Felipe.Reyes@uab.cat Biologı´a, Universidad de Santiago de Chile, Edificio de Investigación Eduardo Morales, Santiago, Chile, 6 IRTA-SCR, Antoni Ibarz Aquaculture Program, Sant Carles de la Rápita, Spain, 7 PhD Program in Aquaculture, Universitat Autònoma de Barcelona, tibarz@ub.edu Bellaterra, Spain, 8 Instituto de Ciencias Naturales, Universidad de las Américas, Santiago, Chile, 9 APC Europe SL, † Granollers, Spain These authors have contributed equally to this work From a general structural perspective, a mucosal tissue is constituted by two main Specialty section: matrices: the tissue and the secreted mucus. Jointly, they fulfill a wide range of functions This article was submitted to Comparative Immunology, including the protection of the epithelial layer. In this study, we simultaneously analyzed the a section of the journal epithelial tissue and the secreted mucus response using a holistic interactome-based Frontiers in Immunology multi-omics approach. The effect of the gilthead sea bream (Sparus aurata) skin mucosa Received: 03 October 2020 Accepted: 24 December 2020 to a dietary inclusion of spray-dried porcine plasma (SDPP) was evaluated. The epithelial Published: 04 February 2021 skin microarrays-based transcriptome data showed 194 differentially expressed genes, Citation: meanwhile the exuded mucus proteome analysis 35 differentially synthesized proteins. Reyes-López FE, Ibarz A, Separately, the skin transcripteractome revealed an expression profile that favored Ordóñez-Grande B, Vallejos-Vidal E, Andree KB, Balasch JC, biological mechanisms associated to gene expression, biogenesis, vesicle function, Fernández-Alacid L, Sanahuja I, protein transport and localization to the membrane. Mucus proteome showed an Sánchez-Nuño S, Firmino JP, Pavez L, Polo J, Tort L and Gisbert E (2021) enhanced protective role with putatively higher antioxidant and antimicrobial properties. Skin Multi-Omics-Based Interactome The integrated skin mucosa multi-interactome analysis evidenced the interrelationship and Analysis: Integrating the Tissue and synergy between the metabolism and the exuded mucus functions improving specifically Mucus Exuded Layer for a Comprehensive Understanding of the the tissue development, innate defenses, and environment recognition. Histologically, the Teleost Mucosa Functionality as skin increased in thickness and in number of mucous cells. A positive impact on animal Model of Study. Front. Immunol. 11:613824. performance, growth and feed efficiency was also registered. Collectively, the results doi: 10.3389/fimmu.2020.613824 suggest an intimate crosstalk between skin tissue and its exuded mucus in response to Frontiers in Immunology | www.frontiersin.org 1 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis the nutritional stimulus (SDPP supplementation) that favors the stimulation of cell protein turnover and the activation of the exudation machinery in the skin mucosa. Thus, the multi- omics-based interactome analysis provides a comprehensive understanding of the biological context of response that takes place in a mucosal tissue. In perspective, this strategy is applicable for evaluating the effect of any experimental variable on any mucosal tissue functionality, including the benefits this assessment may provide on the study of the mammalian mucosa. Keywords: microarrays, multi-omics analyses, functional networks, integrative analysis, immune response, histology, skin mucosa, functional diet INTRODUCTION the mucosa itself, including epithelia and the overlying mucus layer. Numerous studies have described the benefits of an Skin is a stratified squamous epithelial surface strategically adequate diet or the dietary additives to enhance human and located at the interface with the external environment, where animal condition and welfare, with special attention on gut it has evolved to detect, integrate and respond to a diverse health. However, efforts to intensify animal production of range of stimuli from the environment, including stressors and valuable species can lead to increased stress, limited growth aggressions (1–3). Skin function is a crucial component for performance and poor welfare; thus, the research for organism survival, acting like physical barrier as the outermost nutritional strategies focused on “functional feeds” is a priority organ and requiring precise calibration of its responses with a task (19–22). high degree of local autonomy (4, 5). Described as the body’s Spray-dried porcine plasma (SDPP) is an abattoir by-product largest organ, the vertebrate integument is a conserved organized obtained from animal blood after exclusion of cells, and structure consisting of the epidermis, dermis, and hypodermis (6, subjected to concentration and spray drying (23). It has been 7). Nonetheless, the skin of terrestrial and aquatic vertebrates has widely used as a safe and high-quality feed ingredient for acquired differential specific adaptations because of its livestock, especially at the time of weaning because this relationship with the environment. Whereas mammalian skin ingredient promotes feed intake, somatic growth and reduces acquired dead keratinized cell layers, hair follicles, sweat glands stress, as well as morbidity and mortality (23–26). Furthermore, but loosed mucus production capacity (8), the teleost skin several proteins with distinct functions have been found in SDPP presents mucous glands, which produce antifungal and such as immunoglobulins, albumin, growth factors and antibacterial substances, and it also serves as an osmotic barrier biologically active peptides, which mediated anti-inflammatory (9). According to these characteristics, teleost mucosal surfaces may effects (27–29). Regarding aquatic species, several studies have closely resemble type I mucosal surfaces of mammals, represented reported that SDPP enhances growth in rainbow trout by the intestine, the respiratory tract and the uterus, exerting similar (Oncorhynchus mykiss) (30), gilthead seabream (Sparus aurata) physiological (10) and immunological functions (11, 12). For (31) and Nile tilapia (Oreochromis niloticus) (32). These results instance, from an evolutionary point of view, regardless of their may be attributed to its high digestibility, the improvement of phylogenetic origin and tissue localization, mucosal feed intake and feed efficiency, as well as its content in growth immunoglobulins operate under the guidance of primordially promoting factors. Recently, the dietary inclusion of SDPP has conserved principles from fish to mammals (13). Moreover, the shown to enhance innate immunity and antioxidant enzyme systematic exploration of fish skin models have been proposed as activities in gilthead seabream (31). These results are in biologically, clinically and technologically relevant, opening agreement with gained evidence in higher vertebrates, which interesting new opportunities for dermatological research (14). hypothesizes that SDPP protects the organism via the immune Regarding fish skin, it is well assumed that this interface tissue system or directly acting against pathogens (27). also acts as a multifunctional organ, playing roles in protection, Multi-omics approaches pursue the integration of different communication, sensory perception, locomotion, respiration, ion biological entities to understand their interrelation and the regulation, excretion, and thermal regulation (15). functioning of larger systems, and serve to identify new Research on mucosal tissues mainly tackled unique and biomarkers in specific tissues (33–35). In general, all evolved immune mechanisms of defense in mammals as well experimental ‘omics’ approaches can be considered to share as in fish. The epithelial structure, including impermeable tight some major features in contrast to traditional procedures. junctions, goblet cells distribution and density, or the different However, single omics analysis does not always provide enough mucosa-associated lymphoid tissues (MALTs) are of major information to understand the behavior and responsiveness of a interest to understand mucosal properties in mammals (16, 17) cellular system. Therefore, a combination of multiple omics as well as in fish species (13, 15, 18). The maintenance of these analyses, the so-called multi-omics approach, is required to mucosal tissues in healthy conditions is complex and relies on a acquire a precise picture of living organisms (36). ‘Omics’ are delicate balance between the diet, the commensal microbiota and high-throughput, holistic, top-down methodologies and data- Frontiers in Immunology | www.frontiersin.org 2 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis driven, which also attempt to understand the cell metabolism like TABLE 1 | Ingredient list and proximate composition on dry weight basis (%) of an ‘integrated system’, rather than as mere collections of different experimental diets. parts by using information of the relationships between many Ingredients Diets measured molecular species (36). Several popular ‘omics’ platforms Ingredients (%) Control SDPP in tissue biological systems include transcriptomics, which measures mRNA transcript levels; proteomics, the set of proteins Fishmeal LT 70 36.90 33.35 expressed by an organism, tissue, or cell; metabolomics, which Fishmeal 60 12.50 12.50 determines abundance of small cellular metabolites; and CPSP 90 4.00 4.00 Squid meal 6.00 6.00 interactomics, which integrates the whole set of molecular Appetein GS 3.00 interactions in cells. To date, information regarding cellular Wheat Gluten 7.60 7.60 metabolism has been acquired through application of individual Soybean meal 48 (micronized) 7.00 7.00 ‘omics’ approaches (37), but scarce studies tackled on multi-omics Wheat meal 7.70 7.70 data to analyze tissue functionality and none on mucosal tissues, Pea starch 4.50 4.80 Fish oil 11.20 11.45 including epidermal/epithelial cells and its mucus exudation. Vitamin and Mineral Premix PV01 1.00 1.00 The present study proposes a multi-omic analysis for the Choline chloride 0.10 0.10 sustained dietary supplementation impacts of a functional diet Soy lecithin 0.50 0.50 including 3% of SDPP on fish skin mucosa. The fish model Binder (guar gum) 1.00 1.00 selected was the gilthead sea bream due to its well-known Total 100.00 100.00 physiology and its high economical value. To propose an Proximate composition basis CTRL SDPP Crude protein (%) 51.09 51.11 adequate description of the mucosal skin response, we combined Crude fat (%) 17.17 17.16 several molecular biological disciplines that measure the entirety Fiber (%) 0.51 0.51 of biomolecules differentially expressed by means of skin Ash (%) 11.75 11.38 transcriptomic analyses and the modification on mucus layer Gross Energy (MJ/kg)* 20.56 20.69 exudation by the analysis of the mucus proteome. Histological Amino acid composition (as feed basis) analyses for skin structure and mucous cell density, growth Histidine 1.05 1.08 performance and feed efficiency parameters were also Isoleucine 2.00 2.02 Leucine 3.76 3.85 performed. Collectively, the results suggest the whole-mucosal Lysine 3.52 3.60 interactome as a useful strategy for representing the beneficial Threonine 2.35 2.40 effects of functional diets. Prospectively, this methodology arises as Tryptophan 0.54 0.56 a promising alternative for the applicability of treatments upon the Valine 2.32 2.39 Methionine + Cysteine 1.82 1.84 mammalian type I mucosal surfaces. Phenylalanine + Tyrosine 4.32 4.42 Taurine 0.16 0.15 *Gross energy content was estimated by using the following: total carbohydrate = 17.2 J/ MATERIALS AND METHODS kg, fat = 39.5 J/kg and protein = 23.5 J/kg. Diets Spain) and transported by road to IRTA-Sant Carles de la Rapita To assay how a functional diet would benefit skin mucosa research facilities (Sant Carles de la Ràpita, Spain), where they functionality two diets were formulated as follows: a control were acclimated in two 2000-L tanks for two weeks. After their diet (Diet C), equivalent to commercial diet containing 51% acclimation, all fish were anesthetized (tricaine methanesulfonate crude protein, 17% crude fat and 20.6 MJ/kg gross energy that [MS-222], 150 mg/L) and individually weighed for initial body fulfill the nutritional requirements of juvenile sea bream. Based weight (BWi) on this basal formulation, another diet named Diet SDPP was and measured for standard length (SLi) to the nearest 0.1 g manufactured where Fishmeal LT70 was substituted by 3% SDPP and 1 mm, respectively, and then distributed into eight 500-L (Apetein GS, APC Europe SL, Granollers, Spain; Table 1). Diets cyclindroconical tanks at a density of 50 fish per tank (4 tanks/ were manufactured by Sparos Lda (Portugal). In brief, main replicates per diet). ingredients were ground (<250 mm) in a micropulverizer Fish (BWi = 10.6 ± 0.1 g, n = 400, mean ± standard deviation, hammer mill (Hosokawa Micron). Powder ingredients and oils SD) were fed for 95 days with both experimental diets by means were then mixed according to the target formulation in a paddle of automatic feeders (ARVO-TEC T Drum 2000; Arvotec, mixer (RM90; Mainca). All diets were manufactured by Huutokosk, Finland) at the rate of 2.5% of the stocked temperature-controlled extrusion (pellet sizes: 0.8 and 1.5 mm) biomass, which approached apparent satiation. Feed ration was by means of a low-shear extruder (P55; Italplast). Upon evenly distributed in seven meals per day from 08:00 to 18:00 h. extrusion, all feed batches were dried in a convection oven (OP Fish were regularly sampled at a monthly basis in order to 750-UF; LTE Scientific) for 4 h at 45°C. evaluate their growth in BW and adjust the feeding ratio to Fish and Experimental Design stocked biomass. During the trial, water temperature and pH Gilthead seabream fry (average body size 9.5 g) were obtained (pH meter 507; Crison Instruments, Barcelona, Spain), salinity from a commercial hatchery (Piscimar SL, Andromeda Group, (MASTER-20T; ATAGO Co., Ltd., Tokyo, Japan), and dissolved Frontiers in Immunology | www.frontiersin.org 3 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis oxygen (OXI330; Crison Instruments) were 22.1 ± 0.4°C, 7.0 ± Transcriptional Analysis 0.01, 36 mg/L, and 7.2 ± 0.3 mg/L (mean ± SD), respectively. Total RNA was extracted individually from fish skin using Water flow rate in experimental tanks was maintained at QIAGEN RNeasy® Mini Kit following the manufacturer’s approximately 9.0–10.1 liter/min via a recirculation system recommendations. The total RNA concentration was (IRTAmar®; Spain) that maintained adequate water quality quantified using a NanoDrop ND-2000 (Thermo Scientific) (total ammonia and nitrite were ≤0.15 and 0.6 mg/L, and RNA integrity and quality checked with the Experion respectively) through UV, biological, and mechanical filtration. (Automated Electrophoresis Station, Bio-Rad) using the Photoperiod followed natural changes according to the season of Experion Standard Sens RNA chip (Bio-Rad). Only the the year (November to February; 40°37′41″ N). samples with an RNA integrity number (RIN) > 8.0 were At the end of the trial, fish were anaesthetized as previously chosen for further analysis. Transcriptional analysis was described; mucus was gently scrapped off from the skin surface carried out using the AquaGenomic Sparus aurata (n = 15 per diet) using a sterile glass slide avoiding blood, urine oligonucleotide microarray (SAQ) platform (39). The complete and feces during collection and transferred into 2-ml Eppendorf information on this platform and our data is available through tubes and stored at -80 °C as is described in Ferná ndez-Alacid the public repository Gene Expression Omnibus (GEO) et al. (38). All fish in experimental tanks were measured for final (accession numbers GPL13442 and GSE162501, respectively) at BW and SL as indicated. Then, 10 fish per tank were sacrificed the United States National Center for Biotechnology Information with an overdose of MS-222 (200 mg/L) for tissue sampling (NCBI). A transcriptomic analysis was conducted to determine purposes. Skin samples (1 cm2 of anterior dorsal body region) differences at the expression level between control and SDPP were dissected and immediately transferred into RNAlater groups at the end of feeding trial (95 days). For each (Ambion®), fixed overnight and then frozen at -80°C until experimental group (control; SDPP group) total RNA samples further RNA extraction. Fish growth and feed utilization from were pooled (n = 3 pools each group; n = 4 fish each pool, n = 1 experimental groups was evaluated by means of the following fish taken at random from each tank for each pool) using the indices: Fulton’s condition factor (K) = (BW/SL3) × 100; specific same final concentration (133 ng/µL each pool). One-color growth rate in BW (SGR, %) = ((ln BWf −ln BWi) × 100)/time (d) microarray was carried out according to the manufacturer’s and feed conversion ratio (FCR, g/g) = FI/(Bf −Bi), where FI was protocols. Briefly, 200 ng of total RNA was reversed the total feed intake during the experimental period (g) and, Bi transcribed along with spike-in (Agilent One-Color RNA and Bf were the initial and final biomass (g), respectively. Spike-In kit, Agilent Technologies, United States). The solution All animal experimental procedures were conducted in was then used as template for Cyanine-3 (Cy3) labeled cRNA compliance with the experimental research protocol approved synthesis and amplification with the Quick Amp Labeling kit. by the Committee of Ethics and Animal Experimentation of the cRNA samples were purified using the RNeasy micro kit Institut de Recerca i Tecnologia Agroalimentàries and in (Qiagen) according to manufacturer’s instructions. Dye accordance with the Guidelines of the European Union incorporation and cRNA yield were checked with the Council (86/609/EU) for the use of laboratory animals. NanoDrop ND-2000 Spectrophotometer. 1.5 mg of Cy3- labeled cRNA with specific activity of >6.0 pmol Cy3/mg cRNA was then fragmented at 60°C for 30 min, and the Histological Organization of the Skin samples were then mixed with hybridization buffer and For histological purposes, 1 cm2 of the skin from the dorsal hybridized to the array (ID 025603, Agilent Technologies) at anterior region of the body from 16 fish per dietary treatment 65°C for 17 h, using the Gene expression hybridization kit. was dissected (n= 4 fish per tank) and fixed in 4% buffered Washes were conducted as recommended by the manufacturer, formaldehyde (pH = 7.4), dehydrated in a graded series of using gene expression wash buffers and a stabilization and drying ethanol, cleared with xylene, embedded in paraffin and cut solution (Agilent Technologies). Microarray slides were scanned in serial sections (3–5 mm thick). Sections were stained with with Agilent Technologies Scanner model G2505B. Spot hematoxylin-esosin (HE) for general histological descriptions, intensities and other quality control features were extracted whereas slides were stained with Periodic Acid Schiff for with Agilent’s Feature Extraction software version 10.4.0.0 goblet cell identification (neutral mucins produced by mucous (Agilent Technologies). Quality reports were checked for each cells stain in magenta). All sections were observed under a array. The extracted raw data were imported and analyzed with light microscope (Leica DM LB; Leica Microsystems) and GeneSpring (version 14.5 GX software, Agilent Technologies). photographed (Olympus DP70 Digital Camera; Olympus The 75% percentile normalization was used to standardize the Imaging Europa GmbH). Digital images (600 dpi) were arrays for comparisons, and data were filtered by expression. The processed and analyzed using an image analysis software differential expressed genes (DEGs) were obtained from a gene- package (ANALYSIS; Soft Imaging Systems GmbH). level differential expression analysis. Expression values with a p- Measurements of the thickness of different skin regions value < 0.05 were considered statistically significant. The DEGs (epidermis and dermis), as well as mucous cell number (full were grouped according to its fold-change value (p-value < 0.05) and empty) were based on the analysis of three to five randomly and represented using the GraphPad software v7.0 for Windows. chosen fields from the skin per fish. The number of epidermal The Principal Component Analysis (PCA) was carried out using mucous cells was expressed over a length of 100 mm. GeneSpring software (Agilent), four eigenvectors were calculated Frontiers in Immunology | www.frontiersin.org 4 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis to describe the aggrupation of the control and SDPP groups in a bottom of an Ettan DALT II system (Ammersham Biosciences, 3D plot. The gene expression values (log2-expression ratios) Stockholm, Sweden). Proteins were fixed for 1 h in methanol: were represented by a hierarchical clustering heatmap analysis acetic acid 40:10 and stained overnight using colloidal Coomassie using MeV software (v4.0), with Pearson distance and average blue G-250. Gel staining was removed by consecutive washing linkage as it was described before (40). steps with distilled water until the best visualization was achieved. Gene Ontology and Pathway Enrichment Gel Image Analysis and Protein Digestion Transcriptome Analysis Coomassie blue stained gels were scanned in a calibrated In order to classify the DEGs (both up- and down-regulated) Imagescanner (Bio-Rad, Spain) and digital images captured according to its functional annotation, genes were imported into using Quantity-One software (Bio-Rad). The images were the web-tool Protein ANalysis Through Evolutionary saved as uncompressed TIFF files. Gel images were analyzed Relationships (PANTHER) classification system (version 13.0) using the software package ImageMaster 2D, version 6.01 (GE (41). This web resource allows understanding the biological Healthcare, Spain). Proteins were detected using the automated meaning behind large list of DEGs based on their GO routine of ImageMaster 2.0 software, combined with manual classification. For the enrichment analysis, the major over- editing when necessary to remove artefacts. The background was represented GO were chosen according to a p-value < 0.05 removed and normalized volumes were calculated as follows: the criteria (biological processes; cellular component). The volume of each protein spot was divided by the total volume of biological interpretation of the DEGs obtained was all the protein spots included in the analysis. Normalized protein complemented using the free access databases GeneCards spot values were used to select the 300 most abundant proteins in (www.genecards.org) (42) and UniProt (www.uniprot.org) (43). each condition to be further analyzed for their differential expression. The obtained protein spots with differential Proteomic Analysis of Exuded Mucus expression, henceforth differential expressed spots (DESs) were Protein Extraction and Two-Dimensional manually cut from the gel and in-gel tryptic digestion was Electrophoresis Separation performed in an InvestigatorTM Progest (Genomic Solution) Mucus samples for two-dimensional electrophoresis (2-D automatic protein digestion system as it was detailed for fish PAGE) protocols were solubilized in equal volume of ice-cold mucus samples in Sanahuja and Ibarz (44). lysis buffer (7 M urea; 2 M thiourea, 2% w/v CHAPS and 1% protease inhibitor mixture) and centrifuged at 20,000 g for LC-MS/MS Analysis and Database Search 15 min at 4°C, whereas the resultant supernatant was aliquoted Dried-down peptide mixtures were analyzed in a nanoAcquity avoiding pellet resuspension and surface lipid layer. The liquid chromatographer (Waters) coupled to a LTQ-Orbitrap supernatants obtained were submitted to a clean-up procedure Velos (Thermo Scientific) mass spectrometer. Tryptic digests (ReadyPrep 2-d clean-up kit, Bio-Rad) in order to enhance were resuspended in 1% FA solution and an aliquot was injected protein extraction as described in Sanahuja and Ibarz (44). The for chromatographic separation. Peptides were trapped on a proteome map of soluble epidermal mucus proteins was Symmetry C18TM trap column (5 µm 180 µm x 20mm, Waters), obtained by 2D-electrophoresis. Protein concentration was and separated using a C18 reverse phase capillary column determined by Bradford assay with bovine serum albumin as (ACQUITY UPLC M-Class Peptide BEH column; 130 Å, standard (Bio-Rad). 1.7µm, 75 µm x 250mm, Waters). The gradient used for the Pools of three mucus samples were prepared in order to obtain elution of the peptides was 1 to 40% B in 20 min, followed by 450 µg of protein dissolved in 450 µL of rehydration buffer gradient from 40 to 60% during 5 min (A:0.1% FA; B: 100% containing 7M urea, 2M thiourea, 2% w/v CHAPS, and 0.5% v/ CAN, 0.1% FA), with a flow rate of 250 nl/min. Eluted peptides v IPG buffer, 80 mM DTT and 0.002% bromophenol blue. were subjected to electrospray ionization in an emitter needle Five samples of skin mucus protein extract from each dietary (PicoTipTM, New Objective) with an applied voltage of 2,000 V. condition (Control and SDPP diets) were loaded onto 24 cm, pH Peptide masses (m/z 300–1,700) were analyzed in data 3–10 NL IPG strips (GE Healthcare, Madrid, Spain). Isoelectric- dependent mode where a full Scan MS was acquired in the focusing was performed using an IPGhor instrument (Amersham Orbitrap with a resolution of 60,000 FWHM at 400 m/z. Up to Biosciences), following the manufacturer’s instructions (active the 10th most abundant peptides (minimum intensity of 500 rehydratation at 50 V for 12 h followed by linear gradient from counts) were selected from each MS scan and then fragmented in 500 to 8000 V until 48,000 V/h). The focused strips were the linear ion trap using CID (38% normalized collision energy) equilibrated in two steps as follows: 15 min with equilibration with helium as the collision gas. The scan time settings were: Full buffer I (65 mM DTT, 50 mM Tris-HCl, 6 M urea, 30% glycerol, MS: 250 ms (1 microscan) and MSn: 120 ms. Generated.raw data 2% SDS, bromophenol blue) and then, 15 min with equilibration files were collected with Thermo Xcalibur (v.2.2). buffer II (135 mM iodoacetamide, 50 mM Tris-HCl, 6 M urea, Files obtained from mass spectrometry analyses were used to 30% glycerol, 2% SDS, bromophenol blue). Equilibrated strips search against the public database Uniprot Actinopterygii (v.23/ were set directly onto 12.5% polyacrylamide gels, sealed with 0.5% 3/17). A database containing common laboratory contaminant w/v agarose, and separated at a constant voltage of 50 V for 30 min proteins was added to this database. The software used as followed by 200 V for about 6 h, until the blue dye reached the Thermo Proteome Discoverer (v1.4.1.14) with Sequest HT as Frontiers in Immunology | www.frontiersin.org 5 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis the search engine. The following search parameters were applied: those more stringent statistically significant GO obtained from the two missed cleavage sites as well as fixed and variable enrichment analysis were assigned to each one of the nodes modifications; carbamidomethyl of cysteine and oxidation of represented in the functional network. The nodes classified in methionine, respectively. Peptide tolerance was 10 ppm and 0.6 different clusters according to their functionality were represented Da for MS and MS/MS spectra, respectively. Both target and by integrative ameboid graphics using Adobe Photoshop decoy databases were searched in order to obtain a false version CC2018. discovery rate (FDR), and thus, estimate the number of incorrect peptide-spectrum matches that exceeded a given Statistical Analysis threshold. The results were filtered so only proteins identified Data on growth performance, feed conversion and skin with at least two high confidence (FDR >1%) peptides were morpohometrics were compared by means of a t-test. included in the lists. To sort the search results, proteins were Regarding, microarray data, an unpaired t-test) was conducted ranked by a first criterion of the higher Score together with and a using the GeneSpring software GX 14.5 to detect DEGs (p < 0.05) second criterion of the higher number of Sequence Coverage and between the control and SDPP groups. Mucus differential Peptides matched. The principal component analysis (PCA) was synthesized proteins (DESs) that were found to vary in their used to check the quality of the data from each replicate and abundance between the control and SDPP diets were analyzed identify the subsets of samples that are associated with the two for significance using a t-test. The Shapiro-Wilk test was first different groups under study. The protein intensity values (log2- used to ensure the normal distribution of the data, while the expression ratios) were represented by a hierarchical clustering uniformity of the variances was determined by Levene’s test. The heatmap analysis using MeV software (v4.0), with Pearson DEGs and DESs fold-change graphs were represented with distance and average linkage. GraphPad software version 7.0. The PCA analyses for DEGs and DESs were obtained from GeneSpring software version 14.5 Functional Network Analyses: Interactomes (Agilent Technologies) and Analyse-it Software versió n The complete map of interactions that can occur in a living organism 5.4, respectively. (interactome) was obtained from the DEGs obtained in the microarrays-based transcriptomics analysis (transcripteractome), Ethics Statement from the DESs (proteinteractome) and the functional integrated Animal experimental procedures were conducted in compliance network for the transcriptomic and proteomic outcomes both with the research protocol approved by the IRTA’s Committee of together (multinteractomics). For this purpose, the Search Tool Ethics and Animal Experimentation and in accordance with the for the Retrieval of Interacting Genes (STRING) public repository Guidelines of the European Union Council (86/609/EU) for the version 10.0 (https://string-db.org) was used (45). Protein-protein use of laboratory animals. interaction (PPI) network for the differentially expressed genes was conducted with a high-confidence interaction score (0.9). The mechanisms of response in which DEGs, DESs and both omics RESULTS outcome together are involved, were obtained from a comparative analysis based on Homo sapiens as a reference organism in order Profiling of the Skin Transcriptome and to extract the maximum information currently available. Thus, an the Mucus Proteome orthologue H. sapiens Entrez Gene ID was assigned based on Transcriptional analysis of the skin was carried out using the S. sequence homology. Briefly, we selected the best tBlastX (NCBI) aurata oligonucleotide custom microarray and results are shown hit between the entire set of DEGs and DESs query sequence for S. in Figure 1. A total of 194 DEGs were found in the skin of fish fed aurata and the human transcriptome database. We only consider the SDPP diet (Figure 1A). Among them, 121 DEGs were those matches with at least E value ≤ 1E-10. The Uniprot (43) and annotated, whereas 73 were unknown genes (Table S1). Genecards databases (42) were used to confirm match of the gene Among them, 93 DEGs were up-regulated and 101 down- acronym tag between both species. The detailed list of human regulated. Results from the PCA divided the dataset into three orthologues is available on Table S9. Gene ontology (GO) principal components and revealed a clear differential gene pathway enrichment analysis (biological processes; cellular expression pattern among the control and SDPP groups component) was also performed for the DEGs, DESs, and both (Figure 1B). When representing DEGs from both experimental entities together (DEGs+DESs) by STRING using a Fisher’s exact groups using a hierarchical clustering heatmap, we found a clear test followed by a correction for multiple testing (46, 47). A p- differential expression profile with a clear grouping of DEGs of value < 0.05 was considered as significant. The GO terms obtained fish fed with the control diet (Figure 1C; top half of the panel) were then identified in the ancestor GO chart using the QuickGO compared to those fed the SDPP diet (Figure 1C; bottom half). web-tool (https://www.ebi.ac.uk/QuickGO/) (48). The GO chart High-resolution 2D maps of epidermal mucus proteomes for each GO obtained from the enrichment analysis were then were obtained for each individual sample by a combination of mapped in a single chart in order to identify those less redundant broad range, 3–10NL IPG strips with large format SDS gels. In GO terms and thus propose more stringent GO terms associated brief, 950 protein spots were detected in the mucus proteome of to the mechanism of response for the SDPP-fed fish. The potential all samples after 2DE-gel staining. In primary matched sets, a interaction between the GO found into the functional networks representative master gel was obtained for the Control diet was estimated based on an integrative cluster analysis. To do it, (Figure S1) and the 300 spots with higher normalized intensity Frontiers in Immunology | www.frontiersin.org 6 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis FIGURE 1 | Skin mucus microrrays-based transcriptomic analysis for gilthead sea bream fed with SDPP supplemented diet. (A) Number of total differential expressed genes (DEGs). The green (upregulation) and red (downregulation) color scheme indicates the gene modulation according to its fold-change magnitude interval. (B) Principal component analysis (PCA). Control (red spheres) and SDPP groups (yellow spheres) are represented. (C) Hierarchical clustering heatmap representing the 194 DEGs. The normalized intensity values (log2) obtained for each microarray analyzed for control (C1, C2, and C3) and SDPP group (S1, S2, and S3) are shown. were further analyzed for their differently synthesis between both the central core interaction network (31 up-regulated DGEs; 16 experimental groups. A total of 35 proteins, whose abundance down-regulated DEGs), while eight DEGs (seven up-regulated were significantly changed, were considered as DESs since they DEGs and one down-regulated DEGs) were more isolated. On accomplished the criterion over the 2-fold spot intensity the other hand, 35 DEGs (13 up-regulated DGEs; 22 down- difference. Importantly, none of the significant proteins that regulated DEGs) showed no interaction with the main functional changed matched with the identified DEGs. A total of 33 network obtained (Figure S2). The gene ontology (GO) pathway proteins were up-regulated (Figure 2A). The PCA analysis for enrichment analysis revealed that the dietary inclusion of SDPP DESs suggested a differential protein synthesis pattern between induced sustained changes in several biological processes the skin mucus of gilthead seabream fed the control and SDPP including “RNA splicing” (GO.0008380) and “mRNA diets (Figure 2B). This result was confirmed by the hierarchical processing” (GO.0006397), “ribonucleoprotein complex clustering heatmap, showing a clear differential synthesis profile biogenesis” (GO.0022613) and “ribosome biogenesis” between both experimental groups; thus, grouping the DESs (GO.0042254), “intracellular protein transport” (GO.0006886), profile of the skin mucus proteome from fish fed with the control and “protein localization to membrane” (GO.0072657) (Figure diet (Figure 2C; top half of the panel) compared to that of fish S2). Biological processes linked to “membrane budding” fed the SDPP diet (Figure 2C; bottom half). (GO.0006900) and “cellular catabolic process” (GO.0044248) were also detected. The full list of the GO biological process Transcripteractome: The Skin enrichment analysis is on Table S2. The complexity of the Transcriptomics Functional Network interaction between the biological processes was assessed A functional network analysis was carried out based on the DEGs through an integrative cluster analysis (Figure 3) where the obtained from the skin tissue transcriptional analysis from fish sharing nodes of DEGs were considered as evidences of fed with the SDPP diet. From the 121 annotated DEGs, a interacting network within biological processes. functional association was registered between 90 DEGs, (74.4% In order to determine the functional network upon cellular of the total annotated DEGs) with the generation of 143 edges structures, compartments, or macromolecular complexes, DEGs between them (Figure S2). In particular, 47 DEGs belonged to were classified according to their GO cellular component (Figure Frontiers in Immunology | www.frontiersin.org 7 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis FIGURE 2 | Skin mucus 2-Dimensional electrophoresis separation-based proteomic analysis for gilthead sea bream fed with SDPP supplemented diet. (A) Number of total differential expressed spots (DESs). The green (upregulation) and red (downregulation) color scheme indicates the protein modulation according to its fold- change magnitude interval. (B) Principal component analysis (PCA). Control (red spheres) and SDPP groups (yellow spheres) are represented. (C) Hierarchical clustering heatmap representing the 35 DESs. The results obtained for each 2-D electrophoresis separation for control (C0, C2, C3, and C4) and SDPP group (S0, S1, S2, S3, and S4) are shown. S3; Table S3). According to the GO terms for biological process identification and protein code by UniProtKB. From the total enrichment analysis, the cellular component enrichment DESs, only two of them were down-regulated meanwhile the determined the presence of macromolecular complexes other 33 showed a clear higher relative abundance on the mucus associated with splicing (GO.0097525 “spliceosomal snRNP of fish fed the SDPP diet compared to the control group (Figure complex”; GO.0005681 “spliceosomal complex”) and S4). According to their function in epidermal mucus, DESs were biogenesis (GO.0030529 “ribonucleoprotein complex”), classified as structural-, metabolic- or defense-related proteins whereas GO terms associated with vesicle compartment (Figure S4). Among the DESs with protective-related roles, two (GO.0030662 “coated vesicle membrane”; GO.0030135 “coated groups of up-regulated proteins were found: proteins with vesicle”) were also identified. Importantly, GO terms related to chaperone activity and proteins with enzymatic defensive cell junction (GO.0070161 “anchoring junction”; GO.0005925 activities including proteasomal and esterase activity. Four “focal adhesion”) were found, suggesting the effect of SDPP upon identified proteins were associated with cell redox activity: the classical cellular structures for the maintenance epithelial protein disulfide-isomerase related to protein disulfide bonds tissue integrity. formation, and three enzymes related to glutathione The relationship between the cellular component GO terms biosynthesis, which participate in the synthesis of cysteine, was evaluated by an integrative cluster analysis. According to the glutamate and serine, respectively. Together with glutathione data obtained for biological processes GO terms, the biosynthesis, a miscellaneous group of metabolic proteins and transcripteractome showed a strong association between enzymes were up-regulated on the mucus of fish fed the SDPP spliceosomal complex (six up-regulated DEGs; zero down- diet. These results were in agreement to the described increased regulated DEGs) and the ribonucleoprotein complex (seven skin metabolic activity revealed by the transcripteractome. The up-regulated DEGs; three down-regulated DEGs) (Figure S3). third group of proteins belonged to the structural function of Taking collectively both integrative clusters analyses of biological epidermal mucus, most of them participating in the process of processes and cellular component for DEGs, a strong mucus exudation. Thus, ten up-regulated proteins were relationship was found between protein transport, vesicle identified as different keratin types (I and II). However, six of compartment, protein localization to membrane, cell junction these keratin forms were located with a markedly lower structures, and membrane budding (Figure 3). molecular weight than expected (spots 14, 20, 22, 28, 31 and 32) and they could be described as “keratin fragments” resulting Proteinteractome: The Skin Mucus from own mucus enzymatic activity. Within the rest of structural Proteomic Functional Network proteins, two additional groups were proposed. The first one The mucus proteome analysis determined the relative abundance related to the cell exocytosis process, which includes up- for 35 DESs on the exuded skin matrix in fish fed the SDPP diet. regulated actin forms and cell motility-related proteins. The Details on protein identification are supplied in Table S4 second one group of proteins related to cytoskeleton providing the protein identity, gene symbol, fold change, the organization were also up-regulated except for a catenin form, theoretical/observed MW and pI, together with the accession a protein belonging to cadherin cell junction complex, which was number, identified peptides, score, sequence coverage, species of down-regulated. Frontiers in Immunology | www.frontiersin.org 8 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis FIGURE 3 | Integrative ameboid graphic for the biological processes and cellular components obtained from the skin transcripteractome analysis in gilthead sea bream fed with SDPP supplemented diet. Each node represents one differential expressed gene (DEG) obtained from the skin transcriptomic analysis. The DEG modulatory profile is represented with green (upregulated) and red (downregulated) into each node. The integrative cluster analysis groups the DEGs classified into the GO biological process or cellular component enrichment analysis. The set of these GO enrichment pathways are represented with different colors (bottom). The functional network statistics (Network stats) details are indicated (bottom). At the proteome functional network, 25 DESs were processes, including “epidermal cell differentiation” represented, including the 71.4% of the total 35 DESs (GO.0009913) and “epidermis development” (GO.0008544). (Figure S5). Among them, 22 DESs (21 up-regulated DESs; Importantly, the analysis showed a tight relationship of these one down-regulated DESs), whereas they interacted each other processes with the “skin development” process (GO.0043588) totalizing 65 edges in the functional network. Only three DESs (Figure S5), confirming the high specificity of the mucus (one up-regulated DESs; one down-regulated DESs) showed exuded proteins. Processes related to “de novo post- no interaction with the functional network (Figure S5). The translational protein folding” (GO.0051084), “protein import GO enrichment analysis showed that fish fed with the SDPP into mitochondrial outer membrane” (GO.0045040), diet presented a modulatory effect on several biological “nucleoside diphosphate phosphorylation” (GO.0006165), Frontiers in Immunology | www.frontiersin.org 9 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis and “purine ribonucleoside metabolism” (GO.0046128) “sarcomere” (GO.0030017), “myofibril” (GO.0030016), and “cell completed the biological enrichment panel for DESs (Figure leading edge structures” (GO.0031252) (Figure S6; Table S6). Their S5). The full list of the Gene Ontology (GO) biological process positional distribution in the proteinteractome showed that the enrichment analysis for skin mucus proteome is shown in “extracellular region” (GO.0005576) covered almost all the nodes Table S5. included in the network: 17 DESs (16 up-regulated DESs; one Importantly, at the cellular component level, most of DESs were down-regulated DESs) from the 22 possible interacting proteins classified in GO terms as “extracellular region” (GO.0005576), (Figure S6). indicating their involvement in exudative processes (Figure S6). Similar to the DEGs, the relationship between the biological Other GO identified for DESs were “membrane-bound vesicle” processes and cellular component GO terms was evaluated by an (GO.0031988) and the “melanosome compartment” (GO.0042470), integrative cluster analysis (Figure 4). The results showed an FIGURE 4 | Integrative ameboid graphic for the biological processes and cellular components obtained from the skin proteinteractome analysis in gilthead sea bream fed with SDPP supplemented diet. Each node represents one differential expressed spots (DES) obtained from the skin proteomic analysis. The DES modulatory profile is represented with green (upregulated) and red (downregulated) into each node. The integrative cluster analysis groups the DESs classified into each of the biological processes or cellular components indicated in color (on top). Keratin, Type I Cytoskeletal 10 (KRT10) and T-complex 1 (TCP1) genes are exclusively clustered into membrane-bound vesicle and extracellular region, respectively. The functional network statistics (Network stats) details are indicated (bottom). Frontiers in Immunology | www.frontiersin.org 10 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis intimate association between the biological processes (including “defense response”, “response to external stimulus” and epidermal and skin development and differentiation) and “anatomical structure development”, indicating a unspecific, but metabolism (into the context of mucus exudation). Importantly, broad spectrum of the effects of dietary SDPP on the mucosa (skin these data confirms the nature of the skin mucus sample analyzed. and mucus). For instance, a low number of nodes were identified sharing “defense response” with “RNA splicing” (1 node) and Skin Mutiomics-Based Interactome “protein transport” (3 nodes). In the case of “response to external Analysis: Merging the Tissue and Exuded stimulus”, none common node was registered with “RNA splicing”, Mucus Into a Biological Context just one node related to “ribonucleoprotein complex biogenesis” of Response and three nodes linked to “protein localization to membrane”. To better understand the skin mucosa functionality on fish fed Similarly, the “anatomical structure development” showed two SDPP, a multi-omics-based interactome analysis performed. common nodes with “RNA splicing”, three common nodes with This integrative tool was used to merge the transcriptome “protein transport” and three nodes with “protein localization to response determined at tissue level and the proteome profile at membrane”. The “cellular catabolic process” showed a major the mucosal level in order to interpret data from a holistic linkage with “biogenesis” (12 nodes) and “protein transport” (5 perspective. The multinteractomics analysis was constituted by nodes) processes, although it was also associated to a minor extend 115 nodes consisting in 93 DEGs (54 up-regulated; 39 down- to “defense response” (5 nodes), “response to external stimulus” (2 regulated) and 22 DESs (20 up-regulated; two down-regulated) nodes) and “anatomical structure development” (2 nodes) that in turn were responsible of 313 edges in the functional processes. “De novo post-translational protein folding” process network. In particular, 88 nodes formed part of the main was located in the middle of the main core in the interaction network: 67 DEGs (43 up-regulated DGEs; 24 multinteractome, merging with the process of “RNA splicing”. In down-regulated DGEs) and 21 DESs (19 up-regulated DESs; summary, the multinteractome at the level of biological processes two down-regulated DESs). This data indicated that our strategy showed a clear interaction with low redundancy in the number of elucidated the integrated context of response, considering nodes that interacted with the above-mentioned processes; thus, expressed genes and synthesized proteins, in fish fed the SDPP suggesting a complementarity biological response at transcriptional diet for 76.9% of the DEGs with gene annotation and 60% of the and proteome level between those mechanisms modulated in fish DESs. On the other hand, 26 DEGs and one DES showed no fed with SDPP (Figure 5). interaction with the main functional network obtained In the case of the cellular component enrichment analysis, (Figure S7). most of the GO terms already identified on the transcriptomic As shown in Figure S7, the GO enrichment analysis showed response were also recognized in the multinteractome. The several biological processes that were also identified in the “cell junction structure” (GO.0030054; 17 nodes) detected from transcripteractome (“RNA splicing”; “ribonucleoprotein complex the multinteractome celular component enrichment analysis was biogenesis”; “intracellular protein transport”; “protein transport”; probably due to the additive effect of the “anchoring junction” “protein localization to membrane”; “cellular catabolic process”) (GO.0070161; nine nodes) and “focal adhesion” (GO.0005925; and in the proteinteractome (“de novo post-translational eight nodes) processes (Figure S8). No variations in the protein folding”; “skin development”). All of them increased number of nodes were found for the “spliceosomal-related their total number of genes/proteins represented, except in processes” and the “coated vesicle membrane” compared to the the case of the “RNA splicing” and “ribonucleoprotein complex transcripteractome profile. The complete list of the GO terms for biogenesis” that showed no variations. Importantly, four the cellular component enrichment from the multinteractome new biological processes were identified: “defense response” analysis is indicated in the Table S8. (GO.0006952), “innate immune response” (GO.0045087), In agreement with the transcripteractome, the integrative “response to external stimulus” (GO.0009605), and “anatomical cluster analysis for the cellular component enrichment structure development” (GO.0048856). The complete list showed a high relationship with the “spliceosomal complex” of the GO terms for the biological processes enrichment (Figure 6). The “extracellular exosome” covered most of the obtained from the multinteractome analysis is indicated in the interactome; i.e., 36 nodes of the 50 possible (Figure 6). Table S7. Among them, 17 nodes corresponded to DEGs (12 up- According to the biological process enrichment, the integrative regulated DEGs; five down-regulated DEGs). According to cluster analysis showed no variations compared to the data obtained the exuded matrix nature of the skin mucus, most of the DESs in the transcripteractome for the interaction between “RNA were included in the “extracellular exosome” with 19 DESs (17 splicing” and “ribonucleoprotein complex biogenesis” processes up-regulated DESs; two down-regulated DESs). This result (Figure 5). The same output than in the transcripteractome was revealed the relevance of an intimate coordination between also registered between the “ribonucleoprotein complex” and the the transcriptomic and proteomic responses resulting in “protein transport process”, meanwhile 10 common nodes were the exudate of mucus on the surface of the skin’s epithelial determined between “protein transport” and “protein localization to tissue in teleost fish. On the other hand, the cell junction membrane”. The new biological processes identified using multi- structure was also identified in the multinteractome, thus omics showed that the distribution of “innate immune response” representing 13 nodes from a total of 17 possible (Figure 6). nodes in the multinteractome was disperse, as well as for the Collectively, Figure 7 represents the summary scheme of Frontiers in Immunology | www.frontiersin.org 11 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis FIGURE 5 | Integrative ameboid graphic for the biological processes obtained for the skin mutiomics-based interactome analysis in gilthead sea bream fed with SDPP supplemented diet. Each node represents one differential expressed gene (DEG) obtained from the skin transcriptomic analysis (circles) or one differential expressed spots (DES) obtained from the skin proteomic analysis (bold circles). The modulatory profile for DEGs or DES is represented with green (upregulated) and red (downregulated) into each node. The integrative cluster analysis groups the DEGs or DESs classified into each of the biological processes indicated in color (on top). The functional network statistics (Network stats) details is also indicated (bottom). global mucosa functionality showing a series of processes epithelium, response to external stimulus, and immune promoted and compartments modulated by SDPP, such as defense was also registered in fish fed the SDPP diet. RNA splicing (a gene expression-related process), biogenesis, protein transport and localization to membrane, and cell Skin Histology and Somatic junction structure. Regarding specific mucus properties, an Growth Performance augment also in the modulation of genes and proteins In order to evaluate whether the molecular response observed involved in the anatomical structure development of the on the multi-interactome had consequences on skin anatomy, a Frontiers in Immunology | www.frontiersin.org 12 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis histological analysis of this mucosal tissue was conducted. The diet in comparison to the control group (Figure 8). epidermis from all analyzed fishes had a normal appearance Importantly, this result is consistent with the favoring of the with no visible lymphocytes and macrophages scattered across anatomical structure development of the skin obtained from the stratified squamous epithelium with scattered mucous cells, the multi-interactome analysis. Compared to the control diet, irrespective to control or SDPP dietary regimes. The skin in the thickness of the stratum spongiosum of the dermis was both experimental groups had a normal histological higher in fish fed the SDPP diet (p < 0.05), indicating that this organization, being both layers, the epidermis and dermis, stratum is the main target of the SDPP in the skin mucosa. In clearly differentiated. Importantly, the thickness of the contrast, no differences in the thickness of the stratum epidermis was higher (p < 0.05) in the animals fed the SDPP compactum were found between both dietary groups (p < FIGURE 6 | Integrative ameboid graphic for the cellular components obtained for the skin mutiomics-based interactome analysis in gilthead sea bream fed with SDPP supplemented diet. Each node represents one differential expressed gene (DEG) obtained from the skin transcriptomic analysis (circles) or one differential expressed spots (DES) obtained from the skin proteomic analysis (bold circles). The modulatory profile for DEGs or DES is represented with green (upregulated) and red (downregulated) into each node. The integrative cluster analysis groups the DEGs classified into each of the biological processes indicated in color (on top). The functional network statistics (Network stats) details is also indicated (bottom). Frontiers in Immunology | www.frontiersin.org 13 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis A B FIGURE 7 | Summary for the data obtained in our study. (A) Integrated model of response for the effect of SDPP supplemented diet in gilthead sea bream skin mucosa based on multi-interactome analyses. (B) Representation for the most relevant results obtained from the skin histological analysis and somatic growth in gilthead sea bream fed with SDPP supplemented diet. Frontiers in Immunology | www.frontiersin.org 14 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis FIGURE 8 | Histological analysis in the skin of gilthead sea bream (Sparus aurata) fed diets containing spray-dried porcine plasma (SDPP). Histological organization of the skin in gilthead seabream (Sparus aurata) fed (A) the control diet and (B) the SDPP supplemented diet. (C) Histological measurements for epidermis thickness, dermis (stratum compactum; stratum spongiosum), and mucous cells density. Different letters within the same row indicate the presence of statistically significant differences between two experimental groups (t-test; P < 0.05). Data are expressed as mean ± standard deviation (n = 4). 0.05). In addition, the density of epidermal mucous cells, TABLE 2 | Somatic growth and feed efficiency in gilthead sea bream (Sparus related to mucus production, was higher (p < 0.05) in the aurata) fed diets containing spray-dried porcine plasma (SDPP). SDPP group than in fish fed the control diet. Control diet SDPP diet From a productive point of view, results of growth performance and feed efficiency parameters are summarized in BW (g) 82.7 ± 3.2 b 88.2 ± 1.6 a Table 2. In particular, BWf and SGR values were 6.2 and 4.1% SL (cm) 14.6 ± 0.2 14.8 ± 0.1 K 2.66 ± 0.6 2.72 ± 0.3 higher in fish fed the SDPP diet in comparison to the control SGR (% BW/day) 1.63 ± 0.03 b 1.70 ± 0.04 a group (p < 0.05), respectively. No significant differences were FCR 1.21 ± 0.05 a 1.09 ± 0.07 b found in terms of standard length (SL) and the condition factor Different letters within the same row indicate the presence of statistically significant K (p > 0.05). In addition, CR values were lower in fish fed the differences between two experimental groups (t-test; p < 0.05). Data are expressed as SDPP diet in comparison to those fed the control diet (p < 0.05). mean ± standard deviation (n = 4). BW: body weight; SL: standard length; K: Fulton’s Both SGR and FCR were of the major interest parameters on condition factor; SGR: specific growth rate; FCR: food conversion ratio. aquaculture production, indicating that the SDPP group grew better and were more efficient in feed use (p < 0.05). 49). In the present study, we evaluated the potential benefits of the dietary administration of SDPP in the skin mucosa in a teleost fish model of importance from biological and DISCUSSION productive point of views (50). The holistic ‘systems biology’ approach proposed in this current study combines the skin Organisms interact with the surrounding environment at transcriptome and the mucus proteome as a whole response multiple body sites, including the nasal and oral cavities, the coordinated from the epidermal in its strategic role as outmost- digestive and genitourinary tracts, and the skin surfaces. In layer barrier. This information, together with skin histology contrast to the skin of mammals, fish epidermis is considered a analysis, provides further insights into the teleost skin mucosal tissue (13), since it behaves as a mucosal surface that mucosa functionality. contains associated-lymphoid tissue with abundant mucus- When included in animal diets, SDPP products resulted in an producing cells, lacks keratinization, and harbor living improvement of somatic growth, feed efficiency parameters, and epithelial cells in direct contact with the water medium (12, supported the immune system. It is well known that inclusion Frontiers in Immunology | www.frontiersin.org 15 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis levels of 4–8% are recommended for optimal results in higher underlying cell processes that improved the skin barrier function. vertebrates like pigs (25), poultry (24, 26), cats (51) and rats (52). In addition, GO results revealed that several biological processes Recently, it was also assayed its beneficial effects on several were also up-regulated such as those related to (1) RNA freshwater and marine fish species of importance due to their metabolism (“RNA splicing”, GO.0008380; “mRNA economic value such as rainbow trout (30), Nile tilapia (32) and processing” GO.0006397; “ribonucleoprotein complex gilthead sea bream (31, 53). Under the present experimental biogenesis”, GO.0022613; and “ribosome biogenesis”, conditions, our results were in agreement with the available GO.0042254); (2) protein fate to membrane (“protein literature on different teleost species, confirming the beneficial localization to membrane”, GO.0072657; “membrane budding” effects of dietary SDPP inclusion on somatic growth performance (GO.0006900); and “intracellular protein transport”, and feed efficiency. In addition, we also evidenced that dietary GO.0006886); (3) “ribonucleoprotein complex” (GO.0030529), SDPP promoted skin development by increasing the thickness of and 4) “spliceosomal complex” (GO.0005681). The above- the epidermis and the stratum spongiosum of the dermis, mentioned results based on transcriptomic data demonstrated providing new evidence of the dietary supplement at mucosal that SDPP supported the structure and function of the skin by level. Similarly to the antecedent obtained from intestinal promoting intercellular junctions that provide contact and/or mucosa (53), our study showed that dietary SDPP increased adhesion between neighboring cells or between a cell and the the density of mucous cells in the epidermis. The above- extracellular matrix, conferring strength and adhesiveness to the mentioned results on growth, diet utilizationand mucosal different layers of the epidermis and dermis (58). The results may tissue development might be attributed to the nutritional be attributed to the nutritional profile of SDPP, rich in proteins profile of SDPP that includes growth factors, immunoglobulins and functional peptides (59) that may stimulate cell protein and bioactive peptides (27, 54, 55). Although the benefits of turnover and exudation machinery in mucosal tissues. dietary SDPP in livestock and model species are well known (27, One of the most distinctive features of body mucosal tissues is 51, 52), there is limited information about the mode of action the production and secretion of mucus by goblet and club cells and mechanistic links of this feed ingredient at mucosal level, (60). Mucus protects the underlying epithelium from chemical, especially with regard to the skin-associated lymphoid tissue. enzymatic and mechanical damage (61), whereas in the skin it Thus, we performed a transcriptomic analysis of the skin by also reduces swimming drag forces (62). In fish, the epidermis is means of SAQ microarrays (39) in order to provide insight into responsible for the production and maintenance of the mucous the beneficial effects of SDPP. This analysis was complemented layer via the synthesis and secretion of mucins- the high with the proteome analysis of the mucus exuded by epidermal molecular weight glycoproteins densely coated- which formed mucous cells using SDS-PAGE (44) with the purpose of a support matrix-web equivalent to the mammalian mucus (15). determining whether different transcriptomic profiling resulted Soluble proteins, other metabolites and microbiota are trapped in in different mucous skin proteins. Data from both methodologies this exuded mucous. The functionality of fish mucus has been were finally integrated into a multi-omics-based interactome deeply studied (15, 63–65). In this way, the mucus proteome has analysis in order to provide a holistic approach of the effects of classified proteins as structural, metabolic and protection-related SDPP at the skin level. This strategy allowed providing further functions (44). In the present study, mucus proteome changed in insight into the adaptation, responsiveness and trade-offs of the gilthead seabream fed the diet containing SDPP in comparison to skin at cellular level. the control group. In the two-dimensional proteome analysis, A detailed trancripteractome was built from Biological proteins with different pI or relative molecular mass are processes and Cellular components based on GO annotations identified as separate spots. Such information, together with in order to elucidate how the diet acted on the maintenance of the quantity (abundance and fold change) and identity of those healthy skin mucosal tissue. The transcriptomic analysis revealed proteins, allowing us to determine the changes at post- that at the cellular component level, DEGs obtained translational modification level. It may also determine the corresponded to the “extracellular exosome” (GO.0070062), cleaved proteins resulting from mucus proteolytic activity. In “anchoring junction” (GO.0070161) and “focal adhesion” our study, the current proteome gel-based approach limited the (GO.0005925). These processes might be correlated with the number of detected DESs to the 300 most abundant proteins. role of skin in chemical and physical protection due to mucus However, changes in the abundance of lower molecular mass exudation by mucous cells, as well as physical barrier by the presents as mucus soluble proteins can be not discarded. enhancement of tight junctions at cell-cell level and/or between Beyond the non-specific biological processes determined by the cell and the extracellular matrix. These characteristics favor the Gene Ontology for the proteinteractome, specific skin mucus the protection of the organism in front of the environment proteins have been grouped into relevant functional groups. fluctuations (12, 15). The effect of SDPP promoting the Thus, two specific groups related to mucus formation have integrity of mucosal tissues has also been reported in previous been identified: (1) “cytoskeleton related proteins” (TPM1, studies in mammals where an increase of tight junction TUBA1A, PPL and CTNMA1) previously described as molecular markers was observed after the administration of structural mucus related proteins (44, 66); and (2) proteins SDPP. However, these studies focused on the intestine and involved in exocytosis processes and cell motility like ACTA1 none of them evaluated the modulatory effects of SDPP on the and ACTB, which have been extensively reported in the fish skin (56, 57). Furthermore, no evidences are still available on the mucus proteome (44, 67, 68). Although the function of ANXA10 Frontiers in Immunology | www.frontiersin.org 16 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis (member of the calcium and phospholipid binding proteins) in protective role of mucus, with putatively higher antioxidant and the mucus deserves further investigation, this protein has been antibacterial properties. associated to exocytosis, as well as differentiation and cellular The multi-omics-based interactome was conducted based on proliferation processes (69). In addition, ACTR1B was found transcriptomics and proteomics data and their respective significantly up-regulated in the skin mucus of fish fed SDPP. interactomes in order to provide a holistic approach of the This is a conserved protein related to actin and dynactin complex effects of dietary SDPP on the skin functionality and integrity and has been found in different cellular compartments. It has (77). As far as we know, this study is the first one conducted on been reported in vesicular structures in the cytoplasm (70) as an animal mucosa, and especially the skin. In brief, this well as in neutrophil degranulation and in the antigen processing integrative tool was used to merge the effects of the diet on the and presentation of exogenous peptide antigens via Major tissue functionality at transcriptome (tissue level) and proteome Histocompatibility Complex (MHC) class II. Nevertheless, profile (mucosal matrix level). The resulting multi-interactome considering their cellular functionality, the above-mentioned was composed of 115 nodes (including 93 DEGs and 22 DESs) proteins may be related to vesicle formation, enhancing the with a strong interrelationship in 313 edges and higher liberation of products from epidermal mucous cells to skin enrichment values of protein-protein interactions (PPI). mucus. This data matches with our results obtained from the Interestingly, new identified biological processes were transcriptome analysis concerning the protein fate. Different highlighted considering the skin and mucus differential fragments of keratin fragments (KRT1, KRT10, KRT36) expression together. In particular, the “anatomical structure considered as structural components of the epidermis were development” process (GO.0048856; including 37 nodes) is differentially expressed in the mucus proteome. Cleaved related to the progression of an anatomical structure from an keratins produced by proteolysis via extracellular proteases initial condition to its mature state (GO Term definition). have been proposed to have putative antimicrobial function as Although few approaches existed in the literature on membrane pore-forming peptides in higher vertebrates (71). interactomes from mucosal tissues, this “anatomical structure This is of special relevance since one defence mechanism development” GO process was also identified in the gene commonly used by the skin is the production of antimicrobial expression profiling of olfactory ensheathing glial cells from peptides that can kill invading pathogens and activate the host the olfactory mucosa (78) that have the ability to promote immune response (15). Furthermore, SDPP also promoted other regeneration in the nervous system (79). Thus, the presence of defensive exuded products as proteins with chaperone activity this new biological process would explain the bigger epidermis like HSP70, HSP90, TCP1 and GRP78 (44, 67). Their and dermis thickness found in the skin of fish fed the SDPP diet, function is also linked to the immunoproteasome and to the as well as its enhanced physical barrier function. The “response MHC class I pathway, which in turn are related to an activation to external stimulus” process (GO.0009605; 22 nodes) was of innate immune responses. For instance, GRP78 has a another “new” cluster and demonstrated the collaborative work potent immunological activity when released from the internal of skin cells and mucus exuded matrix to work as interface with environment of the cell into the extracellular space. Specifically, the surrounding environment. It is also an evidence of their it feeds anti-inflammatory and pro-resolutory signals in immune tight regulation in front of an epidermal challenge and/or networks (72). Most of metabolic-related proteins found up- stressor. Additionally, more relevant was the identification of regulated in the skin mucus have no direct functionality in this the biological processes related to “innate immune response” protective layer; nonetheless, the resultant products of their (GO.0045087) and “defense response” (GO.0006952), which did enzymatic activity may have an important intracellular not appear when both transcripteractome and proteinteractome signaling functionality (66). Four of the over-expressed were analyzed separately. The innate immune response registered proteins were grouped within the cell redox activity, including an intimate relationship with the defence response. In fact, all their BHMT, GMPS, PHGDH and PDIA3. BHMT mediates betaine nodes were contained on its cluster, indicating that this response and homocysteine transformation, which is involved in process was mostly related to an immunological context. These methionine biosynthesis and has been described as a precursor results are of special relevance since the skin requires both intact of glutathione biosynthesis (73, 74). GMPS synthetises structural and immunological barriers to protect the organism from glutamine, which is the precursor of glutamate, one of the external aggressions including pathogens. In this sense, the efficacy molecules needed in glutathione formation. PHGDH is of SDPP in livestock nutrition has been associated to an improved involved in the serine biosynthesis, which has been described barrier function of the gut mucosa and the modulation of the to affect glycine formation in mouse, with an effect on mucosal immune response (27). However, no previous data on the mitochondrial glutathione activity (75). PDIA3 is also involved impact of SDPP on the skin was available. Thus, our study confirms in cell redox homeostasis (74). The presence of these groups of that the beneficial role of SDPP is not only restricted to the intestine proteins in skin mucus could be related to an up-regulation of but it also affects other mucosal tissues of the organism. The the antioxidant defence system and to a more metabolically immunological-promoting effects of SDPP might be attributed to active tissue as the upregulation of GMPS indicated (involved in the immunoglobulin-rich fraction of plasma (80) although other the synthesis of purine nucleotides) (76). In summary, the biological peptides may be involved. The skin multi-interactome DESs found in the proteome of skin mucus from fish fed suggested that the above-mentioned improvements focused no just the SDPP diet indicated that this feed ingredient promoted the in a specific cellular way of action, but they also affected several Frontiers in Immunology | www.frontiersin.org 17 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis pathways, including gene expression, biogenesis, vesicle formation, ETHICS STATEMENT protein transport, and protein localization to membrane, all of them corresponding to classical vertebrate response of innate and non- The animal study was reviewed and approved by IRTA’S specific defenses. committee of ethics and animal experimentation. AUTHOR CONTRIBUTIONS CONCLUSIONS The conceptualization of the experiment was developed by FER-L, The mucous tissue presents two main matrices: the tissue and the AI, JP, and EG. The methodology carried out was originally secreted mucus that fulfills a wide range of functions mainly proposed by FER-L, AI, and EG. The feeding trial was aimed at the protection of the epithelial layer. In the current performed by JF and EG, while sampling was conducted by AI, study, we studied for the first time both matrices together in the BO-G, JF, and EG. Microarrays hybridizations, raw data output skin mucosa using a holistic multi-omics approach. The skin and processing, and data analysis was made by FER-L and EV-V. transcripteractome revealed a gene expression profile that favors The procedure related to proteomic skin exuded mucus, including molecular mechanisms associated to transcriptional processes, raw data output and processing, and data analysis was done by biogenesis, vesicle formation, protein transport, and protein AI, BO-G, LF-A, IS, SS-N. FER-L, EV-V, and LP made the localization to membrane, whereas mucus proteome enhanced conceptualization, methodology and interpretation of the the protective role of mucus, with putatively higher antioxidant integrative analysis including interactomes, GO terms and antibactericidal properties. The multi-interactome analysis enrichment, and cluster analysis. The conceptualization and (integrating data from skin and mucus) evidenced the design of figures and tables were in charge of FER-L, AI, BO-G, interrelationship and synergy between the skin metabolism and EV-V, JB, and EG. Funding was obtained by EG. All the authors the exuded mucus functions, improving tissue development, the contributed in the data analysis. FER-L, AI, BO-G, EV-V, and EG increase in the thickness of epidermis and the stratum wrote the original draft. All authors contributed to the article and spongiosum of the dermis and goblet cell density, the innate approved the submitted version. defenses and environment recognition. These responses are sustained on a series of processes and compartments related to protein transport and localization to membrane, and FUNDING structural support at cell junction level. Additionally, SDPP positively impacted on animal performance, growth and This study had the financial support of (1) MINECO (project feed utilization. numbers AGL2014-51839-C5-5-R and AGL2015-70637-R); (2) Thus, the integrative perspective followed in our study shows DIETAplus project of the JACUMAR (Junta de Cultivos the stimulation of the cell protein turnover and the activation of Marinos, MAPAMA; Spanish government), which is co-funded the exudation machinery in the skin mucosa. This evidence with FEMP funds (EU); (3) ERC (European Research Council) reflects an intimate crosstalk between skin tissue and its in MedAID project (Mediterranean Aquaculture Integrated exuded mucus in response to a stimulus. In our study, the Development; Grant Agreement nb. 727315); and (4) Fondecyt utilization of SDPP as dietary supplement and its effect upon regular (project number 1211841; ANID; Goverment of Chile). the teleost skin as model of study. The multi-omics-based EV-V was granted with DYCIT-USACH Postdoctoral fellowship interactome analysis increases the power of detecting true (no. 022043IB). JF has been subsidized by the Industrial PhD causal genes and regulatory networks and pathways involved, program of the Generalitat de Catalunya and TECNOVIT- providing a comprehensive understanding of the biological FARMFAES. Collaboration between Ibero-American context of response that takes place. According to Suravajhala researchers has been done under the framework of the network et al. (77), it may even increase the chance to determine the effect LARVAplus “Strategies for the development and improvement of a variable of interest upon animal health and welfare. Overall, of fish larvae production in Ibero-America” (117RT0521) funded this strategy is applicable for evaluating the effect of any by the Ibero-American Program of Science and Technology for experimental variable on any mucosal tissue functionality, Development (CYTED, Spain). The funder bodies were not including the wide range of applications this assessment may involved in the study design, collection, analysis, interpretation provide on the studies in the mammalian mucosa. of data, the writing of this article or the decision to submit it for publication. DATA AVAILABILITY STATEMENT SUPPLEMENTARY MATERIAL The datasets presented in this study can be found in online The Supplementary Material for this article can be found online repositories. The names of the repository/repositories and accession at: https://www.frontiersin.org/articles/10.3389/fimmu.2020. number(s) can be found in the article/Supplementary Material. 613824/full#supplementary-material Frontiers in Immunology | www.frontiersin.org 18 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis REFERENCES 24. Campbell JM, Quigley JD, Russell LE, Kidd MT. Effect of spray-dried bovine serum on intake, health, and growth of broilers housed in different 1. Teles M, Reyes-Ló pez FE, Fierro-Castro C, Tort L, Soares AMVM, Oliveira M. environments. J Anim Sci (2003) 81:2776–82. doi: 10.2527/2003.81112776x Modulation of immune genes mRNA levels in mucosal tissues and DNA 25. Frugé ED, Bidner TD, Southern LL. Effect of incremental levels of red blood damage in red blood cells of Sparus aurata by gold nanoparticles. Mar Pollut cells on growth performance and carcass traits of finishing pigs. J Anim Sci Bull (2018) 133:428–35. doi: 10.1016/j.marpolbul.2018.06.007 (2009) 87:2853–9. doi: 10.2527/jas.2008-1703 2. Khansari AR, Balasch JC, Vallejos-Vidal E, Parra D, Reyes-Ló pez FE, Tort L. 26. Henn JD, Bockor L, Vieira MS, Ribeiro AML, Kessler AM, Albino L, et al. Comparative Immune- and Stress-Related Transcript Response Induced by Inclusion of porcine spray-dried plasma in broiler diets. J Appl Poult Res Air Exposure and Vibrio anguillarum Bacterin in Rainbow Trout (2013) 22:229–37. doi: 10.3382/japr.2012-00613 (Oncorhynchus mykiss) and Gilthead Seabream (Sparus aurata) Mucosal 27. Pé rez-Bosque A, Polo J, Torrallardona D. Spray dried plasma as an alternative Surfaces. Front Immunol (2018) 9:856. doi: 10.3389/fimmu.2018.00856 to antibiotics in piglet feeds, mode of action and biosafety. Porc Heal Manag 3. Carbajal A, Reyes-Ló pez FE, Tallo-Parra O, Lopez-Bejar M, Tort L. (2016) 2:16. doi: 10.1186/s40813-016-0034-1 Comparative assessment of cortisol in plasma, skin mucus and scales as a 28. Moretó M, Pé rez-Bosque A. Dietary plasma proteins, the intestinal immune measure of the hypothalamic-pituitary-interrenal axis activity in fish. system, and the barrier functions of the intestinal mucosa. J Anim Sci (2009) Aquaculture (2019) 506:410–6. doi: 10.1016/j.aquaculture.2019.04.005 87:E92–100. doi: 10.2527/jas.2008-1381 4. Elias PM. Skin barrier function. Curr Allergy Asthma Rep (2011) 8:299–305. 29. Rodriguez C, Blanch F, Romano V, Saborido N, Rodenas J, Polo J. Porcine doi: 10.1007/978-3-642-12264-4_4 immunoglobulins survival in the intestinal tract of adult dogs and cats fed dry 5. Slominski A, Wortsman J. Neuroendocrinology of the skin. Endocr Rev (2000) food kibbles containing spray-dried porcine plasma (SDPP) or porcine 21:457–87. doi: 10.1210/er.21.5.457 immunoglobulin concentrate (PIC). Anim Feed Sci Technol (2007) 139:201– 6. Le Guellec D, Morvan-Dubois G, Sire JY. Skin development in bony fish with 11. doi: 10.1016/j.anifeedsci.2007.01.012 particular emphasis on collagen deposition in the dermis of the zebrafish 30. Johnson JA, Summerfelt RC. Spray-dried blood cells as a partial replacement (Danio rerio). Int J Dev Biol (2004) 48:217–31. doi: 10.1387/ijdb.15272388 for fishmeal in diets for rainbow trout Oncorhynchus mykiss. J World Aquac 7. Roberts RJ. The Anatomy and Physiology of Teleosts. Fish Pathol (2012) 17– Soc (2000) 31:96–104. doi: 10.1111/j.1749-7345.2000.tb00703.x 61. doi: 10.1002/9781118222942.ch2 31. Gisbert E, Skalli A, Campbell J, Solovyev MM, Rodrı́guez C, Dias J, et al. 8. Schempp C, Emde M, Wölfle U. Dermatology in the Darwin anniversary. Part Spray-dried plasma promotes growth , modulates the activity of antioxidant 1: Evolution of the integument. J Ger Soc Dermatol (2009) 7:750–8. defenses, and enhances the immune status of gilthead sea bream (Sparus doi: 10.1111/j.1610-0387.2009.07193.x aurata) fingerlings. J Anim Sci (2015) 93:278–86. doi: 10.2527/jas2014-7491 9. Subramanian S, Ross NW, MacKinnon SL. Comparison of antimicrobial 32. de Araú jo EP, de Carvalho PLPF, de Freitas JMA, da Silva RL, Rocha MKHR, activity in the epidermal mucus extracts of fish. Comp Biochem Physiol - B Teixeira CP, et al. Dietary spray-dried plasma enhances the growth Biochem Mol Biol (2008) 150:85–92. doi: 10.1016/j.cbpb.2008.01.011 performance, villus:crypt ratio and cold-induced stress resistance in Nile 10. Iwasaki A. Mucosal Dendritic Cells. Annu Rev Immunol (2007) 25:381–418. tilapia (Oreochromis niloticus). Aquaculture (2017) 479:675–81. doi: 10.1146/annurev.immunol.25.022106.141634 doi: 10.1016/j.aquaculture.2017.07.003 11. Abelli L, Coscia MR, De Santis A, Zeni C, Oreste U. Evidence for hepato-biliary 33. Joyce AR, Palsson B. The model organism as a system: Integrating “omics” transport of immunoglobulin in the antarctic teleost fish Trematomus bernacchii. data sets. Nat Rev Mol Cell Biol (2006) 7:198–210. doi: 10.1038/nrm1857 Dev Comp Immunol (2005) 29:431–42. doi: 10.1016/j.dci.2004.09.004 34. Kohl M, Megger DA, Trippler M, Meckel H, Ahrens M, Bracht T, et al. A 12. Parra D, Reyes-Lopez FE, Tort L. Mucosal Immunity and B Cells in Teleosts: practical data processing workflow for multi-OMICS projects. Biochim Biophys Effect of Vaccination and Stress. Front Immunol (2015) 6:354. doi: 10.3389/ Acta - Proteins Proteomics (2014) 1844:52–62. doi: 10.1016/j.bbapap.2013.02.029 fimmu.2015.00354 35. Waters KM, Pounds JG, Thrall BD. Data merging for integrated microarray 13. Xu Z, Parra D, Gó mez D, Salinas I, Zhang Y-A, Jørgensen LVG, et al. Teleost and proteomic analysis. Briefings Funct Genomics Proteomics (2006) 5:261–72. skin, an ancient mucosal surface that elicits gut-like immune responses. Proc doi: 10.1093/bfgp/ell019 Natl Acad Sci (2013) 110:13097–102. doi: 10.1073/pnas.1304319110 36. Ishii N, Tomita M. “Multi-Omics Data-Driven Systems Biology of E. coli,”. In: 14. Rakers S, Gebert M, Uppalapati S, Meyer W, Maderson P, Sell AF, et al. “Fish SY Lee, editor. Systems biology and Biotechnology of Escherichia coli. Springer matters”: The relevance of fish skin biology to investigative dermatology. Exp Science+Business Media B.V., Netherlands (2009) p. 41–58. doi: 10.1007/978- Dermatol (2010) 19:313–24. doi: 10.1111/j.1600-0625.2009.01059.x 1-4020-9394-4 15. Esteban MÁ . An Overview of the Immunological Defenses in Fish Skin. Int 37. Park SJ, Lee SY, Cho J, Kim TY, Lee JW, Park JH, et al. Global physiological Sch Res Netw Immunol (2012) 2012:853470. doi: 10.5402/2012/853470 understanding and metabolic engineering of microorganisms based on omics studies. 16. Holmgren J, Czerkinsky C. Mucosal immunity and vaccines. Nat Med (2005) Appl Microbiol Biotechnol (2005) 68:567–79. doi: 10.1007/s00253-005-0081-z 11:S45–53. doi: 10.1038/nm1213 38. Ferná ndez-Alacid L, Sanahuja I, Ordó ñez-Grande B, Sá nchez-Nuño S, Viscor 17. Turner JR. Intestinal mucosal barrier function in health and disease. Nat Rev G, Gisbert E, et al. Skin mucus metabolites in response to physiological Immunol (2009) 9:799–809. doi: 10.1038/nri2653 challenges: A valuable non-invasive method to study teleost marine species. 18. Gó mez D, Bartholomew J, Sunyer JO. Biology and mucosal immunity to Sci Total Environ (2018) 644:1323–35. doi: 10.1016/j.scitotenv.2018.07.083 myxozoans. Dev Comp Immunol (2014) 43:243–56. doi: 10.1016/ 39. Teles M, Reyes-Ló pez FE, Balasch JC, Tvarijonaviciute A, Guimarães L, j.dci.2013.08.014 Oliveira M, et al. Toxicogenomics of Gold Nanoparticles in a Marine Fish: 19. Abd-El-Rhman AMM. Antagonism of Aeromonas hydrophila by propolis and Linkage to Classical Biomarkers. Front Mar Sci (2019) 6:147. doi: 10.3389/ its effect on the performance of Nile tilapia, Oreochromis niloticus. Fish fmars.2019.00147 Shellfish Immunol (2009) 27:454–9. doi: 10.1016/j.fsi.2009.06.015 40. Reyes-Ló pez FE, Romeo JS, Vallejos-Vidal E, Reyes-Cerpa S, Sandino AM, 20. ll in all out.1–7. Tort L, et al. Differential immune gene expression profiles in susceptible and 21. Nickerson JG, Dugan SG, Drouin G, Perry SF, Moon TW. Activity of the resistant full-sibling families of Atlantic salmon (Salmo salar) challenged with unique b -adrenergic Na+/H+ exchanger in trout erythrocytes is controlled by infectious pancreatic necrosis virus (IPNV). Dev Comp Immunol (2015) novel b 3- adrenoreceptor subtype. Am J Physiol (2003) 2852003:R526–35. 53:210–21. doi: 10.1016/j.dci.2015.06.017 doi: 10.1152/ajpregu.00146.2003 41. Thomas PD, Campbell MJ, Kejariwal A, Mi H, Karlak B, Daverman R, et al. 22. Vallejos-Vidal E, Reyes-Ló pez F, Teles M, MacKenzie S. The response of fish PANTHER: A library of protein families and subfamilies indexed by function. to immunostimulant diets. Fish Shellfish Immunol (2016) 56:34–69. Genome Res (2003) 13:2129–41. doi: 10.1101/gr.772403 doi: 10.1016/j.fsi.2016.06.028 42. Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S, et al. 23. Lallès JP, Bosi P, Janczyk P, Koopmans SJ, Torrallardona D. Impact of The GeneCards suite: From gene data mining to disease genome sequence bioactive substances on the gastrointestinal tract and performance of analyses. Curr Protoc Bioinform (2016) 54:1.30.1–1.30.33. doi: 10.1002/cpbi.5 weaned piglets: A review. Animal (2009) 3:1625–43. doi: 10.1017/ 43. Consortium TU. UniProt: A worldwide hub of protein knowledge. Nucleic S175173110900398X Acids Res (2019) 47:D506–15. doi: 10.1093/nar/gky1049 Frontiers in Immunology | www.frontiersin.org 19 February 2021 | Volume 11 | Article 613824 Reyes-López et al. Skin Multi-Omics-Based Interactome Analysis 44. Sanahuja I, Ibarz A. Skin mucus proteome of gilthead sea bream: A non- 64. Dash S, Das SK, Samal J, Thatoi HN. Epidermal mucus, a major determinant in invasive method to screen for welfare indicators. Fish Shellfish Immunol fish health: A review. Iran J Vet Res (2018) 19:72–81. doi: 10.22099/ijvr.2018.4849 (2015) 46:426–35. doi: 10.1016/j.fsi.2015.05.056 65. Reverter M, Tapissier-Bontemps N, Lecchini D, Banaigs B, Sasal P. Biological 45. Szklarczyk D, Gable AL, Lyon D, Junge A, Wyder S, Huerta-Cepas J, et al. and ecological roles of external fish mucus: A review. Fishes (2018) 3:41. STRING v11: Protein-protein association networks with increased coverage, doi: 10.3390/fishes3040041 supporting functional discovery in genome-wide experimental datasets. 66. Jurado J, Fuentes-Almagro CA, Guardiola FA, Cuesta A, Esteban MÁ , Prieto- Nucleic Acids Res (2019) 47:D607–13. doi: 10.1093/nar/gky1131 Á lamo MJ. Proteomic profile of the skin mucus of farmed gilthead seabream 46. Benjamini Y, Hochberg Y. Controlling the False Discovery Rate: A Practical (Sparus aurata). J Proteomics (2015) 120:21–34. doi: 10.1016/j.jprot.2015.02.019 and Powerful Approach to Multiple Testing. J R Stat Soc Ser B (1995) 57:289– 67. Cordero H, Brinchmann MF, Cuesta A, Meseguer J, Esteban MA. Skin mucus 300. doi: 10.1111/j.2517-6161.1995.tb02031.x proteome map of European sea bass (Dicentrarchus labrax). Proteomics 47. Rivals I, Personnaz L, Taing L, Potier MC. Enrichment or depletion of a GO (2015) 15:4007–20. doi: 10.1002/pmic.201500120 category within a class of genes: Which test? Bioinformatics (2007) 23:401–7. 68. Micallef G, Cash P, Fernandes JMO, Rajan B, Tinsley JW, Bickerdike R, et al. Dietary doi: 10.1093/bioinformatics/btl633 yeast cell wall extract alters the proteome of the skin mucous barrier in atlantic 48. Binns D, Dimmer EC, Huntley RP, Barrell DG, O’Donovan C, Apweiler R. salmon (Salmo salar): Increased abundance and expression of a calreticulin-like QuickGO: a web-based tool for Gene Ontology searching. Bioinformatics protein. PloS One (2017) 12:e0169075. doi: 10.1371/journal.pone.0169075 (2009) 25:3045–6. doi: 10.1002/pmic.200800002 69. Gerke V, Moss SE. Annexins: From structure to function. Physiol Rev (2002) 49. Agren G, Lund I, Thiblin I, Lundeberg T. Tail skin temperatures reflect coping 82:331–71. doi: 10.1152/physrev.00030.2001 styles in rats. Physiol Behav (2009) 96:374–82. doi: 10.1016/j.physbeh. 70. Clark I, Giniger E, Ruohola-Baker H, Jan LY, Jan YN. Transient posterior 2008.11.003 localization of a kinesin fusion protein reflects anteroposterior polarity of the 50. Basurco B, Lovatelli A, Garcı́a B. “Current Status of Sparidae Aquaculture,”. Drosophila oocyte. Curr Biol (1994) 4:289–300. doi: 10.1016/S0960-9822(00) In: Sparidae: Biology and Aquaculture of Gilthead Sea Bream and other Species. 00068-3 (2011); eds: Pavlidis M.A. and Mylonas C.C., Blackwell Publishing, Oxford, 71. Tam C, Mun JJ, Evans DJ, Fleiszig SMJ. Cytokeratins mediate epithelial innate UK, p. 1–50. doi: 10.1002/9781444392210.ch1 defense through their antimicrobial properties. J Clin Invest (2012) 122:3665– 51. Rodrı́guez C, Saborido N, Ró denas J, Polo J. Effects of spray-dried animal 77. doi: 10.1172/JCI64416DS1 plasma on food intake and apparent nutrient digestibility by cats when added 72. Shields AM, Panayi GS, Corrigall VM. Resolution-associated molecular to a wet pet food recipe. Anim Feed Sci Technol (2016) 216:243–50. patterns (RAMP): RAMParts defending immunological homeostasis? Clin doi: 10.1016/j.anifeedsci.2016.03.026 Exp Immunol (2011) 165:292–300. doi: 10.1111/j.1365-2249.2011.04433.x 52. Pé rez-Bosque A, Miró L, Polo J, Russell L, Campbell J, Weaver E, et al. Oral 73. Ibarz A, Martı́n-Pé rez M, Blasco J, Bellido D, De Oliveira E, Ferná ndez-Borràs plasma proteins attenuate gut inflammatory effects induced by S. aureus J. Gilthead sea bream liver proteome altered at low temperatures by oxidative enterotoxin B challenge in rats. Livest Sci (2010) 133:242–5. doi: 10.1016/ stress. Proteomics (2010) 10:963–75. doi: 10.1002/pmic.200900528 j.livsci.2010.06.076 74. Ji C, Kaplowitz N. Betaine decreases hyperhomocysteinemia, endoplasmic 53. Tapia-Paniagua ST, Balebona M del C, Firmino JP, Rodrı́guez C, Polo J, reticulum stress, and liver injury in alcohol-fed mice. Gastroenterology (2003) Moriñigo MA, et al. The effect of spray-dried porcine plasma on gilthead 124:1488–99. doi: 10.1016/S0016-5085(03)00276-2 seabream (Sparus aurata) intestinal microbiota. Aquac Nutr (2020) 00:1–11. 75. Vandekeere S, Dubois C, Kalucka J, Sullivan MR, Garcı́a-Caballero M, Goveia J, doi: 10.1111/anu.13039 et al. Serine Synthesis via PHGDH Is Essential for Heme Production in 54. Abdel-Tawwab M. Incorporating Roasted Coffee Bean into Nile Tilapia Diets Endothelial Cells. Cell Metab (2018) 28:573–87. doi: 10.1016/j.cmet.2018.06.009 Does Not Improve Growth Performance. J Appl Aquac (2015) 27:87–93. 76. Yin J, Ren W, Huang X, Deng J, Li T, Yin Y. Potential Mechanisms doi: 10.1080/10454438.2015.1007021 Connecting Purine Metabolism and Cancer Therapy. Front Immunol (2018) 55. Barnes PJ. Beta-adrenergic receptors and their regulation. Am J Respir Crit 9:1697. doi: 10.3389/fimmu.2018.01697 Care Med (1995) 152:838–60. 1995. doi: 10.1164/ajrccm.152.3.7663795 77. Suravajhala P, Kogelman LJA, Kadarmideen HN. Multi-omic data integration 56. Arthington JD, Jaynes CA, Tyler HD, Kapil S, Quigley JD. The use of and analysis using systems genomics approaches: Methods and applications bovine serum protein as an oral support therapy following coronavirus In animal production, health and welfare. Genet Sel Evol (2016) 48:1–14. challenge in calves. J Dairy Sci (2002) 85:1249–54. doi: 10.3168/jds.S0022- doi: 10.1186/s12711-016-0217-x 0302(02)74189-1 78. Guérout N, Derambure C, Drouot L, Bon-Mardion N, Duclos C, Boyer O, et al. 57. Pé rez-Bosque A, Amat C, Polo J, Campbell JM, Crenshaw J, Russell L, et al. Comparative gene expression profiling of olfactory ensheathing cells from olfactory Spray-Dried Animal Plasma Prevents the Effects of Staphylococcus aureus bulb and olfactory mucosa. Glia (2010) 58:1570–80. doi: 10.1002/glia.21030 Enterotoxin B on Intestinal Barrier Function in Weaned Rats. J Nutr (2006) 79. Hayashi KG, Hosoe M, Kizaki K, Fujii S, Kanahara H, Takahashi T, et al. 136:2838–43. doi: 10.1093/jn/136.11.2838 Differential gene expression profiling of endometrium during the mid-luteal 58. Brandner JM, Haftek M, Niessen CM. Adherens Junctions, Desmosomes and phase of the estrous cycle between a repeat breeder (RB) and non-RB cows. Tight Junctions in Epidermal Barrier Function. Open Dermatol J (2010) 4:14– Reprod Biol Endocrinol (2017) 15:20. doi: 10.1186/s12958-017-0237-6 20. doi: 10.2174/1874372201004020014 80. Pierce JL, Cromwell GL, Lindemann MD, Russell LE, Weaver EM. Effects of 59. Almeida FN, Htoo JK, Thomson J, Stein HH. Comparative amino acid spray-dried animal plasma and immunoglobulins on performance of early digestibility in US blood products fed to weanling pigs. Anim Feed Sci weaned pigs. J Anim Sci (2005) 83:2876–85. doi: 10.2527/2005.83122876x Technol (2013) 181:80–6. doi: 10.1016/j.anifeedsci.2013.03.002 60. Long Y, Li Q, Zhou B, Song G, Li T, Cui Z. De Novo Assembly of Mud Loach Conflict of Interest: JP is APC Europe, S.L. employer. FR-L was employed by (Misgurnus anguillicaudatus) Skin Transcriptome to Identify Putative Genes Consorcio Tecnológico de Sanidad Acuı ́cola, Ictio Biotechnologies S.A. Involved in Immunity and Epidermal Mucus Secretion. PloS One (2013) 8: e56998. doi: 10.1371/journal.pone.0056998 The remaining authors declare the research was conducted in the absence of any 61. Desseyn JL, Aubert JP, Porchet N, Laine A. Evolution of the large secreted gel- commercial or financial relationships that could be construed as potential conflicts forming mucins. Mol Biol Evol (2000) 17:1175–84. doi: 10.1093/ of interest. oxfordjournals.molbev.a026400 62. Wainwright DK, Lauder GV. “Mucus Matters: The Slippery and Complex Copyright © 2021 Reyes-López, Ibarz, Ordóñez-Grande, Vallejos-Vidal, Andree, Surfaces of Fish,”. In: SN Gorb, EVG Gorb, editors. Functional surfaces in Balasch, Fernández-Alacid, Sanahuja, Sánchez-Nuño, Firmino, Pavez, Polo, Tort Biology III: Diversity of the Physical Phenomena. Springer. p. 223–46. Available and Gisbert. This is an open-access article distributed under the terms of the Creative at: http://www.springer.com/series/8430. Commons Attribution License (CC BY). The use, distribution or reproduction in other 63. Ekman DR, Skelton DM, Davis JM, Villeneuve DL, Cavallin JE, Schroeder A, forums is permitted, provided the original author(s) and the copyright owner(s) are et al. Metabolite profiling of fish skin mucus: A novel approach for minimally- credited and that the original publication in this journal is cited, in accordance with invasive environmental exposure monitoring and surveillance. Environ Sci accepted academic practice. No use, distribution or reproduction is permitted which Technol (2015) 49:3091–100. doi: 10.1021/es505054f does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 20 February 2021 | Volume 11 | Article 613824