Introduction
The increasing exposure of farmed fish to acute and chronic stressors, intensified by climate change and production intensification, highlights the urgent need for reliable biomarkers that enable early, noninvasive assessment of stress maladaptation and health. In European seabass (Dicentrarchus labrax), common aquaculture practices such as transport and prolonged temperature challenges can disrupt immune function and homeostasis. The fish skin interface represents a strategic target for biomarker discovery, acting both as the first biological barrier against environmental stressors and as an accessible matrix for minimally invasive sampling. In this context, we are identifying biomarker panels based on the skin interface to capture integrated stress responses under both acute and chronic challenge conditions. Multiomics data were jointly analysed through integrative modeling approaches, enabling the identification of composite biomarker signatures that reflect coordinated regulation across molecular levels relevant to stress resilience. These signatures are being validated using an independent trial, assessing the robustness and transferability of the identified biomarkers in a separate biological context. Together, this work advances the development of biologically grounded, minimally invasive biomarker tools with direct applicability for stress monitoring and welfareoriented decisionmaking in seabass aquaculture.
Methodology
European seabass juveniles (~65g) were subjected to two independent stress models, in INMAR (C��diz, Spain), representative of common aquaculture scenarios: an acute transport stress and a chronic thermal challenge. For the acute trial, fish were exposed to transport conditions (4h) designed to mimic standard commercial practices, with skin tissue and skin mucus samples collected from the control group and the challenged group (n = 3 tanks/group). For the chronic trial, fish were maintained under control (19 ��C) or elevated temperature conditions (27 ��C) for four weeks (n = 3 tanks/group), after which skin and skin mucus samples were collected. Skin transcriptomic profiles were generated by RNA sequencing, while miRNA profiles were obtained from skin mucus. Differential expression analyses were conducted (edgeR; FDR < 0.05) to identify stress���responsive molecular features associated with each challenge. To capture coordinated stress responses across molecular layers, transcriptomic and miRNAs datasets were integrated using Multi���Omics Factor Analysis (MOFA+), enabling the extraction of latent factors and composite biomarker signatures shared across acute and chronic stress conditions. An independent validation is currently being performed using skin samples from a separate European seabass trial conducted at RIASEARCH Lda (Murtosa, Portugal). In this trial, fish (~60g) were exposed to thermal stress (26-28 ��C) for five weeks (n = 4 tanks). Targeted qPCR analyses of selected genes and miRNA biomarkers are being carried out in skin samples to assess the robustness, reproducibility, and transferability of the identified biomarker panel in an independent biological context.
Results and Discussion
Both acute transport and chronic thermal challenges elicited marked molecular responses in the skin interface of European seabass, reflected in distinct yet partially overlapping transcriptomic and miRNAomic signatures. Acute stress primarily induced rapid modulation of genes, i.e., 1064 differentially expressed genes (DEGs), and 29 miRNAs, whereas chronic thermal exposure induced the dysregulation of 1427 genes and 10 miRNAs. Despite the different temporal dynamics of the two stress models, shared genes were consistently enriched in pathways associated with immune signaling such as cytokine signaling, stress regulation pathways such as MAPK and FoxO, along with processes linked to cell migration, apoptosis, wound healing and barrier maintenance, reinforcing the biological relevance of the skin interface as both a frontline defense tissue and a sensitive readout of stress. Moreover, several downregulated processes related to DNA damage response, DNA repair, and cell cycle were observed, suggesting the induction of senescencerelated molecular programs, which if prolonged can be maladaptive and have potential implications for longterm tissue function and resilience. Multiomics MOFA+ modeling disentangled shared and challenge���specific latent factors, demonstrating that integrated molecular patterns provided stronger discrimination between stressed and non���stressed fish than single���omic markers alone and enabling the identification of composite biomarker signatures (top weighted features) that were conserved across stress models. The analysis of an independent cohort will allow us to assess the model's performance and biomarker reproducibility. The validation by qPCR will further demonstrate the feasibility of translating multi���omics discoveries into simplified, assay���ready formats suitable for more routine monitoring. Overall, these findings reinforce the skin interface as a biologically meaningful and minimally invasive source of biomarkers and support the development of integrated biomarker panels for early stress detection and welfare���oriented decision���making in European seabass aquaculture.
Acknowledgment
Work Co funded by UKRI and by the European Union's Horizon Europe research and innovation programme (GA No 101084651 project IGNITION). T.B. received funds from FCT - Funda����o para a Ci��ncia e a Tecnologia through grant 2023.04651.BDANA.