Lipidomic profiling reveals time-resolved oxylipin responses to severe burn injury

Abstract

Background

Severe burn injuries affecting more than 40% of the total body surface area (TBSA) are associated with a high risk of mortality. Clinically, burn injury is accompanied by increased systemic inflammation, adipose tissue lipolysis, and skeletal muscle catabolism that can contribute to muscle wasting. Despite the well-established role of inflammation in mediating these processes, the role of circulating oxylipins, released from various tissues and potentially linked to the inflammatory response, remains poorly understood.

Methods

In this study, we investigated the temporal dynamics of plasma oxylipins in male and female patients with severe burn injuries (n = 7, mean TBSA 61 ± 6%) over a 33-day period post-injury, using high-resolution lipidomic profiling. A total of 114 lipid species were quantified, revealing distinct time-dependent alterations in both pro-inflammatory and anti-inflammatory lipid mediators. In addition, selected oxylipins were evaluated in C2C12 myotubes to test whether they engage atrophy-related and inflammatory transcriptional programs.

Results

Specifically, levels of the pro-inflammatory oxylipins 15-hydroxyeicosatetraenoic acid (15-HETE) and its downstream metabolite 15-oxo-eicosatetraenoic acid (15-oxoETE) were elevated immediately after injury, whereas the anti-inflammatory oxylipin resolvin D1 (RvD1) peaked during the later phase of recovery. C2C12 myotubes incubated with 15-oxoETE increased expression of genes involved in muscle protein catabolism and inflammation, while co-incubation with RvD1 attenuated this transcriptional response.

Conclusion

These findings define dynamic changes in the plasma lipidome following severe burn injury and identify selected burn-regulated oxylipins that can modulate skeletal muscle stress- and catabolism-related transcriptional programs in vitro.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12944-026-03016-y.

Introduction

Burn injuries affect nearly 10 million individuals per year worldwide and are associated with significant mortality [1, 2]. Severe burn injury (> 40% total body surface area [TBSA]) frequently necessitates intensive care unit (ICU) management and prolonged hospitalization [3, 4].

Survival is the ultimate challenge as severe burns induce a profound hypermetabolic response that begins immediately after injury during the ebb phase and intensifies during the flow phase as oxygen and nutrient delivery to tissues increases [5, 6]. This sustained systemic stress is accompanied by marked skeletal muscle catabolism [7–9], which contributes to delayed recovery, prolonged weakness, and reduced quality of life [5, 10]. While there are data to suggest that many ICU patients experience skeletal muscle catabolism, this response is more pronounced in burn patients [10]. Thus, severe burn-associated hospitalization represents a major driver of ongoing muscle wasting, with inflammation acting as a central mechanism linking systemic injury to disruption of muscle protein turnover [11].

Omega-3 and omega-6 fatty acids are key mediators of systemic metabolism, inflammation, and tissue repair, and are widely used in clinical settings as biomarkers and nutritional supplements for various diseases [12]. Recent studies have identified oxylipins, bioactive lipid mediators derived from omega-3 and omega-6 fatty acids comprising more than 100 distinct species [13, 14]. These lipids are released from multiple tissues under stress and play key roles in pro- and anti-inflammatory pathways [13, 14]. Several studies investigating exercise, caloric restriction, and cachexia have reported that oxylipins regulate metabolic adaptations to skeletal muscle [3, 15–17] and, in some cases, contribute to muscle wasting [3, 17]. While temporal alterations in circulating lipids after severe burn injury have been described, how specific oxylipins mechanistically interface with skeletal muscle catabolic versus pro-resolving programs remains unclear.

In this study, we paired longitudinal oxylipin profiling in patients with severe burn injury with a skeletal muscle cell culture model to test whether selected burn-regulated lipid mediators can modulate wasting-related transcriptional programs. Plasma samples were collected during a 33-day period after injury, spanning the early “ebb phase” (−72 h) and the subsequent “flow phase” [18, 19]. Using targeted lipidomics, we quantified 114 oxidized lipid species and identified distinct temporal patterns consistent with early pro-inflammatory signaling and later resolution. We then leveraged C2C12 myotubes to test whether selected burn-regulated oxylipins can modulate muscle atrophy-related transcriptional programs and whether a pro-resolving mediator can attenuate these responses. Based on our clinical time course, we focused on the arachidonic acid–derived 15-HETE pathway and the electrophilic metabolite 15-oxoeicosatetraenoic acid (15-oxoETE), which increased early after injury, and on resolvin D1 (RvD1), which increased during the recovery phase. In vitro, 15-oxoETE increased catabolic and inflammatory gene expression in myotubes, whereas RvD1 attenuated these responses. Accordingly, this study is best interpreted as a two-stage translational analysis: first, defining temporal changes in circulating oxylipins after severe burn injury, and second, functionally testing whether selected burn-regulated oxylipins can modulate skeletal muscle catabolic and stress-related transcriptional programs in vitro.

Methods

Study design and patient demographics

In this retrospective study, we enrolled 7 patients with severe burn injuries (mean age: 48 ± 8 years; mean TBSA: 61 ± 6%) and 9 healthy control subjects (mean age: 32 ± 5 years) recruited from the medical center location (Table 1). Inclusion criteria for the burn group were TBSA > 40% [3, 4].

Table 1.

Patient demographics

Group	Sex	Age (yrs)	TBSA (%)
Burn	M	50	58
Burn	F	62	64
Burn	M	39	61
Burn	F	48	55
Burn	M	55	66
Burn	M	38	68
Burn	M	45	59
Control	F	30
Control	M	27
Control	F	28
Control	F	33
Control	F	40
Control	M	43
Control	M	28
Control	M	31
Control	M	33

Plasma samples from the burn injury patients were collected at the following time points post-injury: 2–4 days, 6–8 days, 10–12 days, 15–19 days, 25–26 days and 29–33 days. Samples were collected at the time of scheduled debridement surgical procedures. All background data, including sex, age, and TBSA, were obtained from hospital records.

Non-fasted blood samples from burn patients and healthy control subjects were collected in standard blood collection tubes treated with EDTA, and plasma was collected immediately following centrifugation. All plasma samples were stored at − 80 °C until further processing.

A waiver of consent was obtained from the University of Texas Houston Health Science Center (UTHSC) Institutional Review Board/Research Ethics Committee for all enrolled patients in accordance with UTHSC ethics number HSC-GEN-12-0059. The study was conducted in accordance with the principles of the Declaration of Helsinki.

Liquid chromatography–mass spectrometry (LC–MS)

Plasma lipid profiles were analyzed by liquid chromatography–mass spectrometry (LC–MS) using samples collected at the post-injury time points described above and baseline samples from control subjects. All lipid standards were purchased from Cayman Chemical Company, Avanti Polar Lipids, or Santa Cruz Biotechnology. C18 SPE cartridges were purchased from Biotage. All solvents were high-performance liquid chromatography (HPLC) or LC–MS/MS grade and were acquired from Sigma-Aldrich, Fisher Scientific, or VWR International. Tissue samples were homogenized in 0.1 × PBS in Omni homogenizing tubes with ceramic beads at 4 °C. Aliquots of 100 μl serum or 1 mg protein from homogenized tissue, measured using the Pierce BCA kit (Life Technologies), were used depending on the experiment. A mixture of deuterium-labeled internal standards (d4-9,10-diHOME, d4-9-HODE, d8-5(S)-HETE, d4-LTB4, and d4-PGE2) was added to each aliquot, followed by three volumes of cold methanol (MeOH). Samples were vortexed for 5 min and stored at − 20 °C overnight. Cold samples were centrifuged at 14,000 g for 10 min, and the supernatant was transferred to a new tube. Acidified H2O (3 ml; pH 3.5) was added to each sample before C18 SPE, which was performed as previously described [20–22]. The methyl formate fractions were collected, dried under nitrogen, and reconstituted in 50 μl MeOH:H2O (1:1, by volume). Samples were transferred to 0.5 ml tubes and centrifuged at 20,000 g at 4 °C for 10 min. A 35 μl aliquot of supernatant was transferred to LC–MS/MS vials for LC–MS/MS mediator lipidomics analysis. Separation of signaling lipids was performed on an Ekspert MicroLC 200 system (Eksigent Technologies) with a Synergi Fusion-RP capillary C18 column (150 × 0.5 mm, 4 μm; Phenomenex) heated to 40 °C. A sample volume of 10 μl was injected at a flow rate of 20 μl/min. Lipids were separated using mobile phases A (100% H2O, 0.1% acetic acid) and B (100% MeOH, 0.1% acetic acid) with a gradient starting at 60% B for 0.5 min, steadily increasing to 80% B by 5 min, reaching 95% B by 9 min, holding for 1 min, and then decreasing to 60% B by 12 min. MS analysis was performed on a SCIEX TripleTOF 5600 + system using the HR-MRM strategy consisting of a time-of-flight (TOF) MS experiment looped with multiple MS/MS experiments. MS spectra were acquired in high-resolution mode (> 30,000) using a 100-ms accumulation time per spectrum. Full scan MS/MS was acquired in high-sensitivity mode, with an accumulation time optimized per cycle. Collision energy was set using rolling collision energy with a spread of 15 V. Component identity was confirmed using PeakView software (SCIEX), and quantification was performed using MultiQuant software (SCIEX). Oxidized lipid values were semi-quantified relative to the corresponding spiked internal standard for normalized values. Oxidized lipid species were assessed based on MS/MS spectrum and retention time with available standard [23]. Additional details, including m/z values, internal standards, fragment ions, and retention times for each lipid species, are provided in Supplementary Table 1.

C2C12 cell culture and lipid incubation

C2C12 myoblasts were obtained from ATCC (#CRL-1772) and cultured in 4.5 g/L glucose DMEM supplemented with 10% fetal bovine serum (FBS). Cells were differentiated into myotubes using 4.5 g/L glucose DMEM supplemented with 2% horse serum for 3 days [15]. Differentiated myotubes were treated with 15-oxoETE (Cayman Chemicals, #34730) for 4 h at concentrations of 10, 30, 100, or 300 nM, with phosphate-buffered saline (PBS) as a vehicle control. For co-treatment experiments, 10 nM RvD1 (Cayman Chemicals, #10012554) was either pre-incubated for 1 h or co-incubated with 100 nM 15-oxoETE for 4 h [24–26]. RNA was isolated from C2C12 myotubes using TRIzol reagent [27–29].

Quantitative RT-PCR

Muscle catabolism-related Atrogin1, MuRF1, PSMD4, PSMA7, ATG5, and BNIP3 genes; inflammation-related F4/80, IL6, IL10, IL1β, IL4, TGFβ, CSF1, CSF1R, and MCP1 genes; endoplasmic reticulum (ER) stress-related HSPA5, HSPA1B, XBP1, ATF4, GADD34, and CHOP genes; oxidative stress-related SOD1, SOD2, SOD3, PTGS, CAT, PRDX1, PRDX2, GPX1, and GPX4 genes; and metabolism-related Pgc1a, CS, Nrf1, Nrf2, TFAM, Glut4, and HK2 genes were evaluated by qPCR. GAPDH and β-actin were used as housekeeping genes [27]. Primer sequences are provided in Supplementary Table 2.

Immunoblotting

Phosphorylated NF-κB (pNF-κB) and total NF-κB levels were assessed by Western blotting. C2C12 murine cells (day 4) were treated with 10 nM RvD1 (Cayman Chemical; #10012554) for 1 h and 100 nM 15-oxoETE (Cayman Chemical; #34730) for 4 h, or with vehicle control. Protein lysates were prepared from C2C12 cells [27]. pNF-κB (Cell Signaling Technology; #3033S; 1:1000) and total NF-κB (Cell Signaling Technology; #8242S; 1:1000) antibodies were used for detection.

Statistics

Lipid concentrations were normalized to Z-scores using the browser-based software MetaboAnalyst 6.0. Prior to volcano plot analysis, data were subjected to standard filtering and normalization procedures in MetaboAnalyst. Clusters were assigned by MetaboAnalyst according to dynamic patterns (up, down, or no change), as shown in the heat map. For volcano plot analyses, each post-burn time point was compared with control subjects across all quantified lipid species as a global lipidomics screen, and p-values were adjusted for multiple comparisons using the MetaboAnalyst multiple-testing procedure. Lipids highlighted in the volcano plots were therefore defined according to this global multiple-comparison-adjusted threshold.

For targeted individual oxylipin time-course analyses, selected lipid mediators were evaluated separately based on biological pathway relevance rather than as an unbiased screen of the full lipidomics panel. At each post-burn time point, burn samples were compared with controls using Welch’s t-test because of the small sample size and potential for unequal variance between clinical groups. These targeted oxylipin comparisons were not corrected across all 114 quantified lipid species. For C2C12 cell-culture experiments, comparisons between two groups were performed using Student’s t-test when variance assumptions were met; experiments involving more than two treatment groups were analyzed by one-way ANOVA followed by Bonferroni post hoc testing. Data are presented as mean ± SEM, with significance set at p < 0.05.

Results

Burn injury alters plasma lipids in humans in a time-dependent manner

Seven patients with severe burn injury (TBSA: 61 ± 6%) were followed over 33 days, with plasma samples taken at multiple time points in conjunction with any scheduled surgical procedure. Severe burn injury triggers systemic metabolic and inflammatory responses, which are reflected in dynamic changes in circulating lipids [15, 16]. To determine changes in lipid dynamics after burn injury, targeted signaling lipidomics was performed on plasma samples collected at six time points (days 2–4, 6–8, 10–12, 15–19, 25–26, 29–33 days) post-injury and compared to control subjects (Baseline) (Fig. 1A).

Fig. 1.

A Schematic illustration of study design. Lipid dynamics of B octadecanoids, C docosanoids, and D eicosanoids. n = 7 for burn injury and n = 9 for control. Data were collected at 2–4, 6–8, 10–12, 15–19, 25–26, and 29–33 days post-injury; day 0 denotes control subjects

To obtain an unbiased overview of how the 114 detected signaling lipids changed over time after burn injury, we performed time-course clustering in MetaboAnalyst. This analysis grouped lipids into 11 clusters with distinct temporal trajectories (Supplementary Fig. 1; Supplementary Table 3). Clusters 1 and 2 showed marked increases during the early phase (days 2–4 and 6–8) and again at later time points (days 29–33), whereas clusters 3 and 4 were predominantly elevated during the late phase (days 25–26 and 29–33). Clusters 5 and 6 peaked around days 10–12, while cluster 7 was transiently upregulated at days 6–8. Thus, clustering highlights that burn injury is associated with multiple temporally distinct “waves” of lipid regulation across the 33-day period. However, these data-driven clusters do not map cleanly onto traditional oxylipin class designations, consistent with the idea that biochemical class and temporal regulation are not necessarily aligned.

Next, to see the distinct species of lipids that are affected by burn injury, the 114 lipid species were categorized into octadecanoids, docosanoids, and eicosanoids, which are oxidized polyunsaturated fatty acid-derived lipid mediators collectively referred to as oxylipins. These physiologically active signaling lipids were mapped over the 33-day period (Fig. 1B-D). Octadecanoids increased modestly (~ 1.2-fold), peaking at day 20 before returning to baseline by days 29–33. Docosanoids, which are generally anti-inflammatory, increased nearly twofold and remained elevated until days 25–26, subsequently normalizing at days 29–33 (Fig. 1C). Eicosanoids, recognized for their pro-inflammatory properties, exhibited a bell-shaped trajectory, reaching a maximum concentration at days 15–19 and returning to baseline by day 33 (Fig. 1D). Together, these findings indicate that burn injury drives dynamic shifts in oxylipins, reflecting both pro- and anti-inflammatory processes that evolve over time.

Time-dependent upregulation of pro- and anti-inflammatory signaling lipids occurs after burn injury

Since oxylipins can be subdivided into pro- and anti-inflammatory mediators, analysis at the level of individual lipid species enables more precise evaluation of lipid-specific changes. To identify oxylipins exhibiting unique temporal regulation among the 114 lipid species, volcano plot analyses were conducted for each time point. Figure 2 summarizes the time-dependent changes in individual oxylipins after burn injury, highlighting distinct temporal patterns among pro- and anti-inflammatory lipid mediators. Relative fold changes of each lipid were binned into time frames of 2–4, 6–8, 10–12, 15–19, 25–26, and 29–33 days post-burn injury (Suppl Fig. 2A–F). Several lipids were altered at each time point, and we identified specific lipids associated with pro- and anti-inflammatory pathways that were upregulated at specific time points. For example, 15-HETE, an eicosanoid, was significantly upregulated 2–4 and 6–8 days post-injury (Suppl Fig. 2A,B). Conversely, RvD1, a docosanoid that has been reported to support skeletal muscle regeneration [25], was increased after burn injury at days 6–8, 10–12, and 15–19 post-injury (Suppl Fig. 2B–D). Similarly, 19,20-diHDPA, another docosanoid, was upregulated 25–26 days post-injury but downregulated 29–33 days post-injury (Suppl Fig. 2E,F). 13-HODE was significantly increased at days 10–12 and 15–19 post-injury (Suppl Fig. 2C,D), although the magnitude of this effect was modest (approximately 1.2-fold). 8-HETE was altered at days 15–19 and 29–33. In contrast, 12,13-dihydroxy-9Z-octadecenoic acid (12,13-diHOME), a metabolite released from adipose tissue that enhances skeletal muscle fatty acid uptake [15], was consistently downregulated throughout the 33-day period (Suppl Fig. 2A–F). Collectively, these findings highlight diverse, time-dependent changes in circulating signaling lipids after burn injury and identify candidate oxylipins for subsequent functional testing.

Fig. 2.

Temporal regulation of candidate oxylipins after burn injury and their effects in C2C12 myotubes. A Lipid dynamics of 15-HETE after burn injury. B Lipid dynamics of 15-oxoETE after burn injury. C Schematic of the 15-HETE/15-oxoETE-related pathway. D Lipid dynamics of RvD1 after burn injury. E Schematic of the RvD1-related pathway. *P < 0.05 vs. day 0. F–H Expression of genes related to F inflammation, G ER stress, and H oxidative stress in C2C12 cells incubated with 15-oxoETE with or without RvD1. For panels A, B, D, and E, statistical comparisons at each time point were performed as targeted burn-vs.-control analyses using Welch’s t-test and were not corrected across the full lipidomics panel. *P < 0.05 vs. PBS; ^†P < 0.05 vs. 15-oxoETE. n = 5 or 6/group. Data are expressed as mean ± SEM

RvD1 attenuates 15-oxoETE-associated pro-catabolic transcriptional responses in C2C12 myotubes

To prioritize oxylipins for functional testing, we focused on mediators that showed temporal regulation after injury and represented opposing inflammatory programs relevant to muscle catabolic signaling. We therefore selected an early-elevated, arachidonic acid–derived eicosanoid pathway (15-HETE and downstream metabolites) and a pro-resolving docosanoid that increased later during recovery (RvD1) for functional testing in C2C12 myotubes.

Time-dependent changes in key pro- and pro-resolving lipid mediators and their functional effects in skeletal muscle cells are summarized in Fig. 3. Although 8-HETE increased at 15–19 and 29–33 days post-injury, it was not pursued further because it did not align with our goal of identifying early-phase candidates related to the initial inflammatory response. 13-HODE increased at 10–12 and 15–19 days post-injury, but the magnitude of this change was modest. We therefore focused on 15-HETE, which was elevated during the early post-burn period and returned toward baseline thereafter (Fig. 2A). Based on this temporal pattern, we hypothesized that the 15-HETE pathway may be relevant to early inflammatory or catabolism-related signaling after burn. However, incubation with 15-HETE did not significantly alter the expression of genes involved in protein degradation in C2C12 cells (Supplementary Fig. 3). We therefore evaluated downstream metabolites of 15-HETE and identified 15-oxoETE as increased at early post-burn timepoints (Fig. 2B, C). Because 15-oxoETE is an electrophilic lipid mediator reported to modulate inflammatory signaling, we tested whether it could alter catabolism-related transcriptional programs in skeletal muscle cells. Incubation of C2C12 cells with increasing concentrations of 15-oxoETE (0, 10, 30, 100, and 300 nM) revealed that 100 nM 15-oxoETE significantly increased expression of genes associated with muscle catabolism (Supplementary Fig. 4A) but not other metabolic genes (Supplementary Fig. 4B), indicating a preferential transcriptional response under these conditions.

Fig. 3.

Expression of pro-catabolic/atrophy-related genes in C2C12 cells following incubation with 15-oxoETE in the presence or absence of RvD1 pretreatment. RvD1 attenuated 15-oxoETE-associated upregulation of catabolic gene expression. *P < 0.05, vs. PBS; ^†P < 0.05, vs. 15-oxoETE. n = 6 or 7/group. Data are expressed as mean ± SEM

The apparent difference between the global lipidomics volcano plots and the individual 15-oxoETE time-course panels reflects the use of distinct statistical frameworks: the volcano plots applied a multiple-comparison-adjusted screen across all quantified lipid species, whereas the 15-oxoETE panels show targeted comparisons of a preselected pathway mediator at individual post-burn time points.

In contrast, the pro-resolving lipid mediator RvD1 increased gradually after burn injury during the flow, or later, phase, reaching statistical significance on days 6–8. Notably, RvD1 peaked at days 25–26 at approximately ten-fold higher than basal levels and declined by days 29–33 (Fig. 2D, E). RvD1 possesses anti-inflammatory and pro-resolving properties [30–32], although its role in muscle anabolism is not fully characterized [33, 34]. Together, these data highlight the temporal dynamics of pro-inflammatory and pro-resolving lipid mediators following burn injury.

Since 15-oxoETE and RvD1 have opposing inflammatory properties, we investigated whether these lipids modulate inflammatory and stress-related transcriptional responses in skeletal muscle cells. C2C12 myotubes were incubated with 15-oxoETE, with or without RvD1 treatment, and expression of genes involved in inflammation (Fig. 2F), ER stress (Fig. 2G), and oxidative stress (Fig. 2H) were assessed by qPCR. Incubation with 15-oxoETE increased expression of multiple genes, including those involved in inflammation (F4/80, IL10, CSF1R, MCP1; Fig. 2F), ER stress (HSPA5, HSPA1B, XBP1, ATF4, GADD34, and CHOP; Fig. 2G), and oxidative stress (SOD1, SOD2, PTGS, CAT, PRDX2, GPX1, and GPX4; Fig. 2H). Importantly, when C2C12 cells were pre-incubated with RvD1, the 15-oxoETE-associated induction of several of these genes was attenuated (Fig. 2F–H). These findings suggest that RvD1 modulates inflammatory and stress-related transcriptional responses to 15-oxoETE in skeletal muscle cells. To further assess whether 15-oxoETE activates canonical inflammatory signaling in C2C12 myotubes, we examined NF-κB phosphorylation under the same treatment conditions. However, no significant change in NF-κB phosphorylation was observed (Supplementary Fig. 5). These findings suggest that the inflammatory and pro-catabolic gene expression responses observed with 15-oxoETE may not be explained solely by direct activation of NF-κB signaling under the conditions tested.

Finally, to assess whether these lipids engage molecular programs associated with muscle catabolism, we incubated C2C12 myotubes with 15-oxoETE and evaluated transcriptional markers of muscle protein catabolism. RvD1 pretreatment attenuated the 15-oxoETE-associated upregulation of catabolic genes, including those shown in Fig. 3. Notably, 15-oxoETE increased expression of pro-catabolic/atrophy-related genes, and these responses were reduced by RvD1 pretreatment (Supplementary Fig. 4, Fig. 3). These data suggest that RvD1 can suppress 15-oxoETE-associated pro-catabolic transcriptional responses in myotubes.

Discussion

These data identify time-dependent changes in circulating signaling lipids following severe burn injury. In the clinical lipidomics analysis, 15-oxoETE was increased during the early post-burn period, whereas the pro-resolving lipid RvD1 increased during the later recovery phase. Complementary in vitro experiments indicate that selected burn-regulated lipids can modulate muscle catabolism-related transcriptional programs: 15-oxoETE increased expression of pro-catabolic, inflammatory, and stress-related genes in C2C12 myotubes, whereas RvD1 attenuated several of these gene expression responses.

Unsupervised clustering provided a complementary, data-driven summary of the 114 detected oxylipins across six post-injury windows (Supplementary Fig. 1; Supplementary Table 3). Rather than separating cleanly by biosynthetic class, clusters reflected distinct temporal trajectories, consistent with multiple post-burn waves of lipid regulation. This pattern likely reflects contributions from multiple tissues, divergent production and clearance kinetics within lipid classes, and the modest cohort size. These results support focusing mechanistic experiments on strongly time-regulated candidates, including early-elevated 15-oxoETE and later-elevated RvD1.

The selection of 15-oxoETE was based on the integration of clinical lipidomics, pathway biology, and functional testing rather than statistical ranking alone. Early regulation of the arachidonic acid-derived 15-HETE pathway prompted initial testing of 15-HETE in C2C12 myotubes; however, 15-HETE did not induce protein degradation-related genes under the conditions tested. We therefore evaluated 15-oxoETE, a downstream metabolite in the same pathway. In contrast to 15-HETE, 15-oxoETE induced atrophy-related, inflammatory, ER-stress, and oxidative-stress transcriptional programs, several of which were attenuated by RvD1.

Interpretation of post-burn oxylipin trajectories should also consider the expected background variability of circulating oxylipins in healthy individuals. Prior work in healthy volunteers has shown relatively low intra-individual inter-day variability in fasting plasma oxylipins, but substantially higher intra-day variability across the postprandial period, underscoring the influence of sampling conditions (e.g., time of day and feeding status) on measured oxylipin concentrations [35, 36]. Although our clinical sampling was constrained by ICU logistics and coordinated with clinical care, the magnitude and consistency of the temporal “waves” observed after severe burn injury suggest regulated pathophysiologic shifts that exceed typical day-to-day fasting variability reported in healthy subjects.

Severe burns trigger inflammatory reactions that alter lipid metabolism, including marked elevations in blood triglycerides and free fatty acids [6, 37–39]. While routine clinical assessments capture gross lipid changes, in-depth lipidomic analyses of blood samples from burn patients have revealed alterations in both saturated and unsaturated fatty acids [11], as well as other lipid species [40]. Our findings expand these observations by characterizing post-burn dynamics of signaling lipids. Because burn injury elicits a strong immune response, and signaling lipids are central mediators of inflammation [13, 14], understanding changes in their levels and activity may provide insight into inflammatory and metabolic responses after burn injury. Here, we determined the chronological effects of burn injury on signaling lipids using lipidomic analysis at multiple clinical time points from patients with severe burns (average TBSA > 60%).

We measured lipid levels in plasma, but the tissue origins of the altered signaling lipids remain undefined. While multiple organs may contribute to these lipids in circulation, white adipose tissue (WAT) is a likely major source [41]. Because signaling lipids can be stored in triglycerides, lipolysis is a probable mechanism for their release, with burn injury promoting lipolysis. This may involve β₂-adrenergic receptor activation [6, 42, 43]. WAT-derived lipids released during lipolysis include signaling lipids as well as omega-3 and omega-6 metabolites. Lipolysis promotes the release of arachidonic acid into the bloodstream, thereby altering systemic metabolism [41, 44–47]. The oxidative pathway of this omega-6 PUFA produces bioactive lipids such as 15-HETE and, subsequently, 15-oxoETE [48]. Thus, lipolysis-induced release of arachidonic acid may lead to its conversion into inflammatory mediators such as 15-oxoETE. Interestingly, RvD1 can also be synthesized in adipose tissue, suggesting that its circulating levels could be influenced by lipolysis. Further studies are needed to determine whether 15-oxoETE, 15-HETE, and RvD1 are released from WAT and to define their tissue-specific roles in burn physiology.

Skeletal muscle can also secrete cytokines and chemokines that act locally as myokines [49, 50]. Thus, 15-oxoETE-associated induction of inflammatory transcripts in C2C12 myotubes may reflect both a cell-intrinsic stress response and a potential feed-forward signaling program within muscle. The induction of MCP1 and CSF1R is consistent with this possibility, even though NF-κB phosphorylation was not detectably altered at the time point assessed. Future studies should test whether 15-oxoETE increases secreted myokines and whether RvD1 limits these autocrine or paracrine responses.

Burn injury induces an initial ebb phase followed by a prolonged flow/recovery phase [51]. In this context, early elevation of the 15-HETE/15-oxoETE pathway and later increases in RvD1 are consistent with a temporal shift from inflammatory to pro-resolving lipid programs. Although 15-HETE and 15-oxoETE have been reported to exert anti-inflammatory effects in other settings [48, 52, 53], our data show that 15-oxoETE increases catabolic and stress-related gene expression in C2C12 myotubes under the present conditions. RvD1 attenuated several of these responses, including MCP1, HSPA5, and HSPA1B, supporting an association with resolution of inflammatory and catabolism-related transcriptional programs. Because NF-κB phosphorylation was not altered in our acute assay, these effects may involve transient, non-canonical, or NF-κB-independent mechanisms, as well as possible myokine-mediated signaling that was not directly tested here.

Notably, 15-oxoETE also induced transcripts involved in antioxidant defense (e.g., SODs, catalase, peroxiredoxins, and glutathione peroxidases), consistent with engagement of oxidative stress–responsive pathways in myotubes. However, gene expression changes alone cannot distinguish whether reactive oxygen species are increased, which ROS species predominates, or the subcellular compartment(s) responsible for ROS generation. Future studies using compartment-targeted ROS reporters/probes (e.g., cytosolic versus mitochondrial sensors) will be important to define the spatial origin of redox signaling in response to 15-oxoETE and to determine whether RvD1 modulates these compartment-specific ROS signals [54–56].

Muscle wasting after burn injury is a critical clinical problem during recovery [9, 57]. Critically ill patients can lose more than 15% of skeletal muscle mass within the first week of ICU admission [7], underscoring the rapidity of muscle loss in severe illness. Although muscle mass can be assessed by DXA, MRI, CT, bioelectrical impedance, total body water approaches, or ultrasound [58], these measures are not uniformly feasible or interpretable in patients with extensive burns who require repeated operations, wound care, fluid resuscitation, mechanical ventilation or sedation, infection surveillance, and other life-sustaining interventions. Accordingly, body composition and muscle function could not be evaluated in this retrospective cohort.

Importantly, muscle mass was not the primary clinical outcome of this study. The human lipidomics analysis should instead be interpreted as the first stage of a translational approach that identified burn-regulated circulating oxylipins, followed by targeted myotube experiments testing whether selected candidates engage catabolism- and stress-related transcriptional programs. Thus, the absence of direct muscle-mass measurements is a limitation of the retrospective cohort, not evidence against the relevance of skeletal muscle catabolism in severe burn injury. Our data support an association between later-phase increases in RvD1 and attenuation of 15-oxoETE-associated pro-catabolic and inflammatory gene expression in myotubes, but do not define the tissue source of RvD1, its cellular targets in vivo, or a compensatory release mechanism in patients.

From a translational perspective, selective pharmacologic antagonists of 15-oxoETE are not well established, and 15-oxoETE likely acts through pleiotropic electrophile-responsive signaling rather than a single dedicated receptor. Near-term strategies may therefore be more feasible by limiting excessive arachidonic acid release or oxidation, targeting downstream stress- and inflammation-responsive pathways, or augmenting pro-resolving lipid signaling. Based on the temporal pattern observed here, future burn-relevant animal studies could test early delivery of RvD1 or related pro-resolving mediators to determine whether enhanced resolution signaling blunts subsequent muscle catabolic programming and improves functional recovery. Translation will require additional efficacy, safety, pharmacokinetic, and formulation studies.

Muscle wasting reflects coordinated activation of catabolic pathways, including the ubiquitin–proteasome system, autophagy/mitophagy, and stress-responsive signaling. In C2C12 myotubes, 15-oxoETE increased canonical atrophy-related genes, proteasome components, autophagy-related mediators, and ER- and oxidative-stress transcripts, consistent with engagement of pathways implicated in muscle wasting. RvD1 attenuated several of these inflammatory, stress-related, and pro-catabolic transcriptional responses. Together, these findings support a model in which early-elevated electrophilic oxylipins such as 15-oxoETE may contribute to muscle catabolism-related signaling, whereas later-elevated pro-resolving mediators such as RvD1 may counter-regulate these programs during recovery.

These findings provide insights into oxylipin changes after burn injury and highlight several limitations. The tissue sources of circulating lipids were not established, and the cohort was small and not evenly distributed by sex; however, this is a common sex distribution for a trauma population, which is consistently approximately 75% male [59, 60]. In addition, mortality is high in these patients, and while we initially identified seven patients, one died during the 33-day period. Patients who suffer a burn injury over 60% TBSA constitute a small portion of the population (~ 5%) and have an extremely high mortality rate, resulting in inherent limitations in these human studies. Age differed significantly between the burn and control groups. However, both groups were middle-aged, which may reduce, but does not eliminate, the likelihood that age substantially influenced the overall lipid profile. Detailed clinical parameters including ICU length of stay, duration of hospitalization, relevant comorbidities, documented infections or sepsis and medication data were not available in a format suitable for meaningful analysis in this cohort and were not taken into consideration. Finally, our mechanistic studies in C2C12 myotubes primarily assessed transcriptional markers and did not directly quantify myotube size, muscle protein synthesis/degradation rates, or autophagy/proteasome flux; thus, functional wasting phenotypes remain to be established. In addition, clinical measures of muscle mass or strength were not available in this retrospective cohort. Follow-up rodent studies will be essential to define how oxylipins modulate muscle protein turnover and functional recovery after burn injury.

Conclusion

These data identify significant alterations in circulating oxylipins after severe burn injury and suggest that selected burn-regulated oxylipins can influence skeletal muscle stress- and catabolism-related transcriptional programs in vitro. Specifically, RvD1 attenuated 15-oxoETE-associated pro-catabolic signaling in myotubes, supporting further investigation of pro-resolving lipid pathways in burn-associated muscle catabolic responses. Further in vivo studies are needed to define the therapeutic relevance of these lipids after burn injury.

Abbreviations

TBSA

Total body surface area

15-HETE

15-Hydroxyeicosatetraenoic acid

15-oxoETE

15-Oxoeicosatetraenoic acid

RvD1

Resolvin D1

LC–MS

Liquid chromatography-mass spectrometry

MS

Mass spectrometry

FBS

Fetal bovine serum

PBS

Phosphate-buffered saline

ER stress

Endoplasmic reticulum stress

Atrogin1

F-box protein 32

MuRF1

Muscle RING finger 1

PSMD4

Proteasome 26S subunit, non-ATPase 4

PSMA7

Proteasome subunit alpha type-7

ATG5

Autophagy-related 5

BNIP3

BCL2/adenovirus E1B 19 kDa-interacting protein 3

F4/80

Macrophage marker encoded by Adgre1 (formerly Emr1)

IL6

Interleukin-6

IL10

Interleukin-10

IL1β

Interleukin-1 beta

IL4

Interleukin-4

TGFβ

Transforming growth factor-beta

CSF1

Colony-stimulating factor 1

CSF1R

Colony-stimulating factor 1 receptor

MCP-1

Monocyte chemoattractant protein-1

HSPA5

Heat shock protein family A (Hsp70) member 5

HSPA1B

Heat shock protein family A (Hsp70) member 1B

XBP1

X-box binding protein 1

ATF4

Activating transcription factor 4

GADD34

Growth arrest and DNA damage-inducible protein 34

CHOP

C/EBP homologous protein

SOD

Superoxide dismutase

PTGS

Prostaglandin-endoperoxide synthase

CAT

Catalase

PRDX1

Peroxiredoxin 1

GPX

Glutathione peroxidase

Pgc1a

Peroxisome proliferator-activated receptor gamma coactivator 1-alpha

CS

Citrate synthase

Nrf1

Nuclear respiratory factor 1

Nrf2

Nuclear factor erythroid 2-related factor 2

TFAM

Transcription factor A, mitochondrial

Glut4

Glucose transporter type 4

HK2

Hexokinase 2

ANOVA

Analysis of variance

12,13-diHOME

12,13-Dihydroxy-9Z-octadecenoic acid

WAT

White adipose tissue

Authors’ contributions

LAB and CEW collected patient data. RSD, LAB, SN, NPS, DHS, and KM performed the biochemical analyses. VB and MAK conducted the lipidomic analyses. RSD and DHS performed the statistical analyses. RSD, LAB, SN, CEW, and KIS interpreted all data. RSD, SN, and KIS were the major contributors to writing the manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by NIH grants R01DK133859-01A1 to KIS and American Heart Association Career Development Award (25CDA1444322) to SN.

Data availability

The analyzed dataset in the current study is available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study was approved by the University of Texas Houston Health Science Center Institutional Review Board/Research Ethics Committee (HSC-GEN-12-0059). A waiver of consent was obtained for all enrolled patients. The study was conducted in accordance with the principles of the Declaration of Helsinki.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.