Sheep Tissue Study Links Bioactive Lipids to Cellular Functions

Researchers at the University of Vienna and the University of Veterinary Medicine Vienna have identified functional relations between bioactive lipids and specific biological processes by integrating tissue proteomics with lipid analysis in sheep. Published in Frontiers in Pharmacology, the study analyzed thirteen organs and tissues from healthy, skeletally mature female sheep to project functional annotations from proteins to correlating lipids. The team hypothesized that by combining mass spectrometry-based untargeted proteomics with bioactive lipid analysis, they could improve the functional annotation of these potent effector molecules, which often lack complete biological descriptions due to analytical challenges.

The study utilized a well-established large animal model, specifically the Merino sheep, chosen for its body size and organ dimensions that closely resemble those of humans. Samples were collected from three healthy, skeletally mature female sheep that had been euthanized for reasons unrelated to the research. The tissues analyzed included the brain, heart, kidney cortex, renal medulla, liver, lung, muscle, nerve, skin, spleen, tendon, and cartilage, along with plasma. This systematic approach allowed the researchers to investigate the association of bioactive lipid occurrence with biological functions across varying metabolic activities.

Using high-resolution mass spectrometry, the team identified 4,717 proteins and 168 distinct bioactive lipids in the samples. These lipids included fatty acids and their conjugates, oxylipins such as octadecanoids and eicosanoids, lyso-phosphatidylcholines, and lyso-phosphatidylethanolamines. Additionally, the analysis covered endocannabinoids, related fatty amides, and bile acids. The correct identification of 114 lipids was confirmed using commercially available analytical standards, ensuring the reliability of the dataset for subsequent functional correlation analyses.

A significant finding of the research was the verification of a previously uncharacterized oxylipin named 4-hydroxy-eicosatetraenoic acid, or 4-HETE. The identity of this molecule was verified through in-house chemical synthesis and gas chromatography-mass spectrometry analysis. 4-HETE was detected in heart, kidney cortex, liver, lung, muscle, and spleen tissues. This oxylipin was not previously listed in the LIPID MAPS database, a repository of well-characterized lipids, until the researchers shared their analytical standard with the database maintainers.

The researchers employed weighted co-expression network analysis to merge the proteomics and bioactive lipid expression matrices. This statistical method allowed them to form modules that correlated specific sets of proteins with bioactive lipids. By mapping UniProt IDs to gene symbols and performing Gene Ontology term enrichment analysis, the team could relate specific lipid patterns to cellular functions characteristic of given tissue types. This approach enabled the projection of functional knowledge established for proteins onto co-occurring bioactive lipids.

The analysis revealed that 18-HETE and 19-HETE were related to mitochondrial fatty acid beta-oxidation, a process essential for energy production in cells. These correlations were particularly evident in the kidney cortex and liver, tissues known for their high metabolic activity.

Other novel functional relations were identified for lipids involved in neural and muscular processes. The study found that 11,12-dihydroxy-eicosatetraenoic acid, 15-deoxy-prostaglandin J2, and 5(6)-epoxy-eicosatrienoic acid were related to chemical synaptic transmission in the brain. Furthermore, arachidonoyl ethanolamide phosphate was correlated with proteins essential for muscle contraction. Numerous lyso-phosphatidylcholines and lyso-phosphatidylethanolamines were found to be related to exosome formation, suggesting a role in extracellular vesicle biology.

Tissue levels of polyunsaturated fatty acids, pro-resolving mediators, and the newly characterized 4-HETE were associated with the cellular protein synthesis and folding machinery. This finding suggests that polyunsaturated fatty acids and their derivatives may support protein synthesis fidelity in addition to their known role in efferocytosis, the removal of dead cells. Both processes are essential contributors to the effective resolution of inflammation, highlighting a potential dual role for these lipids in maintaining tissue health.

The study also highlighted tissue-specific expression patterns of bioactive lipids. For instance, the endocannabinoid dihomo-gamma-linolenoyl ethanolamide was predominantly found in the brain, while the bile acid glycodeoxycholic acid was predominant in the liver. The prostaglandin PGE2, a regulator of renal water transport, was found in the renal medulla. These patterns corroborated known bioactive lipid functions and provided a framework for understanding how lipid abundance correlates with specific organ activities.

Principal component analysis of the proteome and bioactive lipid profiles showed clear clusters according to organ and tissue types. While muscle, cartilage, and tendon clustered away from other tissue types in both analyses, brain and nerve tissues were clearly separated on the proteome level but clustered together on the bioactive lipid level.

The researchers acknowledged that the small number of donors limits the statistical power of the study. Furthermore, the inherent correlation-based approach lacks causal evidence for the described functional implications of bioactive lipids. Dedicated study designs are required for functional validation of these associations. Despite these limitations, the study provides a comprehensive map of potential functional annotations for bioactive lipids, which may help to better understand their physiological roles and potential disease implications.

This integration of multi-omics data represents a powerful approach to understanding physiological regulatory mechanisms. By combining proteomics with bioactive lipid analysis, the team was able to validate the approach through the independent confirmation of previous knowledge and suggest numerous novel correlations. The results indicate that bioactive lipids may serve as biomarkers for specific biological processes, offering new insights into the complex interplay between metabolism and cellular function in healthy tissues.

The analytical workflow relied on distinct mass spectrometry platforms for the two omics layers. Proteomic data were acquired using a timsTOF Pro instrument operating in Parallel Accumulation-Serial Fragmentation mode, with a scan range of 100 to 1700 m/z. Bioactive lipid profiling utilized a Vanquish UHPLC system coupled to a Q Exactive HF hybrid quadrupole orbitrap mass spectrometer. For lipid separation, a reversed-phase Kinetex XB-C18 column was employed. The untargeted lipid analysis covered specific classes in negative ionization mode, including oxylipins, fatty acids, glycerophosphocholines, and bile acids, while fatty amides, monoradylglycerols, and steroids were acquired in positive ionization mode. This dual-platform approach allowed for the simultaneous characterization of the proteome and the complex lipidome within the same tissue samples.

Sample preparation involved rigorous protocols to ensure data integrity. Tissue pieces were rinsed with ice-cold phosphate-buffered saline to remove residual blood cells without flushing the capillaries. Samples for proteomics were snap-frozen in liquid nitrogen, whereas lipid analysis samples were transferred into ice-cold ethanol. For proteomic digestion, the ProtiFi S-Trap protocol was used after protein concentration determination via a bicinchoninic acid assay. Lipid extraction involved homogenization and solid-phase extraction, with peak areas normalized to internal heavy isotope-labeled oxylipin standards and sample weight. Only molecules independently identified in all replicates of at least one tissue type were included in the statistical analysis, ensuring that the resulting correlations were based on reproducible detection.

The co-expression network analysis identified 27 distinct modules, of which 18 successfully correlated proteins with bioactive lipids. Module M8 served as a validation point for the methodology, correlating CYP monooxygenase products such as 11,12-dihydroxy-eicosatrienoic acid and 14,15-dihydroxy-eicosatetraenoic acid with a cluster of monooxygenases including CYP1A2 and CYP4F2. This module was highly specific to liver tissue, mirroring the known liver-specific expression of these enzymes. In contrast, Module M2 correlated low levels of polyunsaturated fatty acids and their CYP derivatives, including the novel 4-HETE, with low levels of proteins involved in cytoplasmic translation and endoplasmic reticulum function. This module was characteristic of liver, lung, and kidney, but showed a negative correlation in muscle, tendon, and cartilage, highlighting tissue-specific metabolic differences.

The study also examined the distribution of the anesthetic embutramide, which was used to euthanize the sheep. Although not an endogenous metabolite, its inclusion provided insights into tissue pharmacokinetics. Embutramide was hardly detectable in plasma but showed highly efficient accumulation in the heart, lungs, and brain, with poor uptake in muscle, tendon, and cartilage. This pattern illustrated the redistribution dynamics of the substance and the varying access to blood supply and drug affinity properties across different tissue types.

Principal component analysis revealed that while proteomic clusters were generally better separated than lipid clusters, certain tissues showed divergent profiles. Brain and nerve tissues were clearly separated at the proteome level but clustered together in the bioactive lipid analysis. Conversely, the separation of brain and liver clusters was more distinct via bioactive lipids than via proteins. The clustering of heart and skeletal muscle also differed; heart tissue grouped with high-lipid-level tissues, while skeletal muscle grouped with low-lipid-level tissues, consistent with the higher lipid turnover rates and lipoprotein lipase levels described in cardiac tissue.

The identification of 4-HETE was further supported by its fragmentation pattern, which included a specific fragment at m/z 101.024, a known marker for oxylipins with a functional group at position C4. This pattern is also seen in DHA-derived products like 4-hydroxy-docosahexaenoic acid. The synthesis of the 4-HETE standard was verified via gas chromatography-mass spectrometry, and the standard was subsequently shared with the LIPID MAPS database, where it is now listed under the identifier LMFA03060123. The formation of 4-HETE is likely via allylic hydroxylation by CYP enzymes, as arachidonic acid lacks a double bond between C4 and C5 for potential autoxidation.

Data from this study have been deposited in public repositories to facilitate further research. Bioactive lipid data are available in the Metabolomics Workbench under Study IDs ST003163 and ST003164, with the project DOI 10.21228/M8BF0G. Proteomics data are available in the ProteomeXchange Consortium via the PRIDE repository with dataset identifiers PXD064992 and PXD065008.

Subject of Research: Functional annotation of bioactive lipids via integrated proteomics and lipidomics in ovine tissues

Article Title: Integrating tissue proteomics and bioactive lipid analysis of ovine tissue samples identifies functional relations of bioactive lipids

Article References: Hagn, G., Jenner, F., Bileck, A., Gültekin, S., Hamberg, M., Mohr, T., Gerner, C., & Gerner, I. (2026). Integrating tissue proteomics and bioactive lipid analysis of ovine tissue samples identifies functional relations of bioactive lipids. Frontiers in Pharmacology, 17, Article 1959280. https://doi.org/10.3389/fphar.2026.1959280

Image Credits: AI Generated

DOI: 10.3389/fphar.2026.1959280

Keywords: bioactive lipids, proteomics, ovine tissue, oxylipins, 4-HETE, Gene Ontology, mass spectrometry, inflammation resolution, polyunsaturated fatty acids, lipidomics, Integrating, tissue