Improved muscle recovery after omega-3 supplementation is associated with increased oxylipin availability

Abstract

Eccentric exercise-induced muscle damage (EIMD) impairs muscle function and recovery. While omega-3 polyunsaturated fatty acid (n-3 PUFA) supplementation has shown promise in mitigating EIMD, the role of their oxidation products (oxylipins) remains unclear. This study investigated the effects of 8-week n-3 PUFA supplementation on muscle recovery and its relationship with blood oxylipin profiles after EIMD in healthy men. Eighteen participants were randomly assigned into two groups: OMEGA (n = 9 28.0 ± 2.5 y; 2.5 g/d of DHA and 0.5 g/d of EPA) and PLACEBO (n = 9 21.4 ± 0.6 y; maltodextrin). Participants performed one bout of 100 unilateral isokinetic eccentric maximal voluntary contractions of the knee extensor muscles before and after an 8-week supplementation period. N-3 PUFA supplementation augmented plasma EPA- and DHA-derived oxylipins (p < 0.05) in the OMEGA group, which showed an attenuation of peak muscle maximum voluntary contraction loss (-15.4 ± 5.2%; p = 0.05) post-exercise compared to the PLACEBO group (1.3 ± 6.0%). There were no differences in muscles soreness between groups. Multivariate analyses identified body fat percentage and specific n-3 and n-6 PUFA-derived oxylipins as predictors of MVC loss after EIMD. In conclusion, the increased plasma levels of n-3 PUFA derived oxylipins following supplementation suggest a possible role for these lipid mediators in promoting muscle recovery.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-44339-1.

Introduction

It is well established that exercise is essential for maintaining and improving health. However, untrained people and athletes can experience muscle damage when exercise is prescribed at a load that exceeds the accustomed level for a specific muscle group, a phenomenon known as exercise-induced muscle damage (EIMD)¹. Experimentally, EIMD can be induced after eccentric actions and is caused by the rupture of muscle membranes, loss of muscle fiber structure, and inflammation of the extracellular matrix². EIMD is typically evidenced by delayed onset muscle soreness (DOMS), the appearance of specific proteins in the blood (e.g. creatine kinase, myoglobin), production of reactive oxygen species, and loss of muscle function for days or weeks^3,4. As skeletal muscle damage can reduce training frequency and adherence, limit long-term training adaptations, and increase the susceptibility to muscle injuries⁵, it is crucial to search for aids that facilitate a faster recovery of muscle function, enabling a quicker return to training or competition.

Due to their unique properties, it has been suggested that increased intake of omega-3 polyunsaturated fatty acids (n-3 PUFAs), eicosapentaenoic acid (C20:5n-3 EPA), and docosahexaenoic acid (C22:6n-3 DHA) can reduce EIMD caused by eccentric exercise^6,7. However, the evidence to date remains equivocal, with some studies reporting positive^8–13 and others reporting small or no effects^14–17. It is most likely that these discrepancies are due to the heterogeneous duration of the studies and the different doses and types of n-3 PUFAs supplemented¹⁸. In fact, most studies have supplemented for six weeks or less, with doses ranging from 0.2 g/d to 3 g/d, and with different concentrations of EPA and DHA. Furthermore, for optimal incorporation of n-3 PUFAs into human skeletal muscle membranes, a supplementation period of at least 6 weeks is required^19,20, with doses higher than 1 g/d of EPA and DHA²¹. Supporting this, a recent study observed that a 7-week supplementation period of 4 g/d of EPA or DHA improved muscle soreness and recovery following an EIMD protocol involving downhill running and plyometric lunges¹³. Thus, dose and supplementation time are key aspects of this strategy.

Importantly, n-3 PUFAs attenuate the production of certain pro-inflammatory proteins^22,23. This effect has been suggested to be attributed to the oxylipins derived from the metabolism of n-3 and omega-6 (n-6) PUFAs^6,24. Oxylipins are bioactive lipid mediators derived from the enzymatic oxidation of PUFAs and play key roles in the initiation, propagation, and resolution of a controlled inflammatory response triggered by injury or eccentric exercise^25,26. They are synthesized from n-3 and n-6 PUFAs released from immune and muscle cell membrane phospholipids by the enzyme phospholipase A₂, and subsequently metabolized by one of three enzymatic pathways: cyclooxygenase (COX), lipoxygenase (LOX), and cytochrome P450 (CYP)-catalyzed epoxygenase (CYP), giving rise to distinct classes of oxylipins²⁷. The oxylipins derived from n-6 PUFAs are generally associated with a pro-inflammatory effect related to the early or immediate phase of muscle damage²⁸, whereas oxylipins derived from n-3 PUFAs are generally recognized to decrease the production of proinflammatory molecules and, most importantly, promote the resolution of inflammation^26,28. The resolution of inflammation mediated by n-3 PUFA-derived oxylipins is hypothesized to be crucial in repairing damaged muscle tissue and restoring muscle function following EIMD⁶. Interestingly, despite the recent evidence reporting that n-3 PUFA supplementation augments blood^29,30 and skeletal muscle³¹ oxylipin profile, there is still a poor understanding of whether there is a functional relationship between systemic oxylipin shifts and muscle recovery. Investigating this association is important as it could help determine whether a specific oxylipin profile dictates the resolution of inflammation induced by muscle damage, leading to strategies that limit muscle function loss and promote faster recovery. Furthermore, it is currently unknown whether increasing the plasma levels of n-3 PUFA-derived oxylipins by supplementation with n-3 PUFAs is associated with reduced muscle damage and a faster recovery of muscle function after EIMD. Therefore, this study aimed to determine the effects of an eight-week supplementation period with n-3 PUFAs on indirect markers of muscle damage and their relationship with blood oxylipin levels after high-intensity eccentric exercise in healthy young men. We hypothesized that n-3 PUFAs supplementation would reduce the magnitude of muscle damage, which would be associated with n-3 oxylipins in plasma.

Results

Subject characteristics

A total of 18 males (n = 9 OMEGA; n = 9 PLACEBO) between 18 and 39 years of age participated in the present study (Table 1). There were no differences within and between the OMEGA and PLACEBO groups in body composition, physical performance [(maximal isometric knee strength and maximum oxygen consumption (VO_2max)], and total workload of the eccentric exercise performed (sum of all sets) before and after the 8-week supplementation period (Table 1). However, a difference in age between groups (p < 0.05), and a main effect of time for caloric intake was observed with both groups (p < 0.05). A significant Group × Time interaction was found for red blood cell (RBC) levels of EPA and DHA (p < 0.0001). Post-hoc tests revealed that RBC levels increased following supplementation in the OMEGA group (EPA: 0.592 ± 0.036 to 1.640 ± 0.054; DHA: 2.502 ± 0.086 to 9.719 ± 0.210; p < 0.0001), whereas no changes were observed in the PLACEBO group (EPA: 0.582 ± 0.029 to 0.576 ± 0.027; DHA: 2.568 ± 0.045 to 2.588 ± 0.041; Table 1).

Table 1.

Physical and biological characteristics.

	OMEGA		PLACEBO
	Pre	Post	Pre	Post
Age (years)	28.0 ± 2.5a	-	21.4 ± 0.6a	-
Body mass (kg)	82.7 ± 3.2	82.1 ± 3.1	81.6 ± 3.6	81.8 ± 3.6
Height (cm)	175.7 ± 1.8	175.5 ± 2.0	176.0 ± 2.0	176.1 ± 1.9
Body mass index (kg/m 2 )	26.8 ± 0.9	26.7 ± 0.9	26.4 ± 1.2	26.2 ± 1.3
Fat mass (% of body mass)	25.1 ± 1.5	24.3 ± 1.5	25.9 ± 3.1	25.5 ± 3.9
Fat-free mass (% of body mass)	74.9 ± 1.5	75.7 ± 1.5	74.1 ± 3.1	72.7 ± 3.8
Diet (Kcal)*	2,607 ± 482	1,747 ± 97	2,112 ± 287	1,621 ± 99
VO2max (ml/kg/min)	34.2 ± 1.5	36.7 ± 2.7	36.8 ± 2.5	37.4 ± 2.5
Muscle strength exercised leg (Nm)	267.9 ± 11.2	253.0 ± 14.1	281.4 ± 13.1	299.1 ± 10.6
Total exercise workload (kJ)	24,154 ± 771	23,011 ± 502	22,897 ± 1,280	21,456 ± 832
EPA (% of total FA in RBCs)	0.592 ± 0.036a	1.640 ± 0.054ab	0.582 ± 0.029	0.576 ± 0.027b
DHA (% of total FA in RBCs)	2.502 ± 0.086a	9.719 ± 0.210ab	2.568 ± 0.045	2.588 ± 0.041b

Maximum oxygen consumption (VO_2max), eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA); fatty acids (FA); red blood cells (RBCs). *Main effect of time (p < 0.05). Same letter represents a significant difference (p < 0.05 for age and p < 0.0001 for % of total FA in RBC).

Muscle recovery measurements

Baseline MVC strength was similar between groups (OMEGA = 285.7 ± 36.3 Nm vs. PLACEBO = 274.8 ± 35.1 Nm; p = 0.52). Both groups exhibited a decrease in MVC strength after the eccentric exercise protocol, both before and after the supplementation period (see Fig. 1A-B). However, the change in peak MVC loss from pre- to post-supplementation (ΔMVC loss) was attenuated in the OMEGA group by 15.4 ± 5.2% and was significantly different (p = 0.05) to ΔMVC loss in the PLACEBO group (1.3 ± 6.0%) (Fig. 1C). No differences were found in muscle soreness and delta peak soreness between pre- and post-supplementation in both groups (see Fig. 1D-F).

Fig. 1.

Changes in maximum voluntary contraction (MVC) strength and muscle soreness with n-3 PUFA supplementation. Knee extensor MVC strength as a percentage of baseline in the OMEGA (A) and PLACEBO (B) groups before (white circles) and after supplementation (black circles), and delta () peak MVC loss for the OMEGA and PLACEBO groups (C). Muscle soreness as a percentage of baseline in the OMEGA (D) and PLACEBO (E) groups before (while circles) and after (black circles) supplementation, and peak soreness for the OMEGA and PLACEBO groups (F). Data are expressed as means ± SEM (n = 9 per group). *Significantly different (P = 0.05) between OMEGA and PLACEBO group.

Effects of eccentric exercise on n-3 and n-6 PUFA-derived plasma oxylipins

The plasma levels of arachidonic acid (C20:4n-6 AA) were similar within and between groups at all timepoints (Supplementary Table 1). Prior to supplementation, both before- (Pre) and after- exercise (Pre-2hEx) levels of AA-derived oxylipins were comparable across groups (Supplementary Table 1). Following supplementation, only 15-hydroxy-eicosatetraenoic acid (15-HETE) was reduced in the OMEGA group relative to PLACEBO at the Post and Post-2hEx timepoints. In contrast, several oxylipins derived from linoleic acid (C18:2 n-6, LA) were consistently lower in the OMEGA group. At the Pre timepoint, the plasma levels of 9-hydroxy-ocatadecadienoic acid (9-HODE), 13-HODE, 13-octadecadienoic acid (13-oxoODE), and 9,10,13-trihydroxy-octadecenoic acid (9,10,13-triHOME) were lower compared to PLACEBO. At Pre-2hEx, 13-oxoODE was lower, while after supplementation and 2 h after exercise (Post-2hEx), 13-HODE, 13-oxoODE, and 9,12,13-triHOME were lower in the OMEGA group relative to PLACEBO (Supplementary Table 1).

The plasma levels of DHA and EPA were significantly altered by supplementation and exercise (Supplementary Table 1). Prior to supplementation, plasma DHA and EPA concentrations were comparable between groups; however, following supplementation, both DHA and EPA levels were significantly elevated in the OMEGA group at the Post timepoint compared to their own Pre values and to PLACEBO at Post. Alpha-linolenic acid (C18:3n-3 ALA) plasma levels remained similar across groups and timepoints. For DHA-derived LOX products, no differences were present at baseline between groups. After supplementation, several LOX products [10-hydroxy-docosahexaenoic acid (10-HDoHE), 13-HDoHE, 14-HDoHE, 16-HDoHE, and 17-HDoHE)], were significantly higher in the OMEGA group relative to PLACEBO at Post timepoint with 14-HDoHE and 17-HDoHE remaining elevated at Post-2hEx. Within-group changes included an increase in 16-HDoHE at Post versus Pre. Additionally, 17-HDoHE was higher in the OMEGA group than PLACEBO at Pre-2hEx. In the CYP pathway, 16,17-dihydroxy-docosapentaenoic acid (16,17-DiHDoPE), 19,20-DiHDoPE, and 20-HDoHE were elevated in the OMEGA group at Post compared to to the PLACEBO group and to its own Pre values. At Post-2hEx, 16,17-DiHDoPE and 19,20-DiHDoPE remained higher in the OMEGA group compared to PLACEBO. For EPA-derived oxylipins, the OMEGA group showed selective increases after supplementation. In the LOX pathway, 15-hydroxy-eicosapentaenoic acid (15-HEPE) was significantly elevated at Post compared to PLACEBO. In the CYP pathway, 14,15-dihydroxy-eicosapentaenoic acid (14,15-diHETE) was higher at both Post and Post-2hEx, while 17,18-diHETE and 18-HEPE were increased at Post in the OMEGA group relative to PLACEBO. The ALA-derived oxylipins of the LOX pathway, 9-hydroxy-octadecatrienoic acid (9-HOTrE) and 13-HOTrE, were lower at Pre compared to PLACEBO, and 13-HOTrE was lower at Post and Post-2hEx in the OMEGA relative to the PLACEBO group. In the CYP pathway, 12,13-epoxy-octadecadienoic acid (12,13-EpODE) was lower at Pre compared to Pre-2hEx in the OMEGA group and at Post compared to PLACEBO. Meanwhile, 15,16-EpODE was lower at Pre compared to Pre-2hEx in the OMEGA group and at Post compared to Post-2hEx in the PLACEBO group.

The sum of the enzymatically derived EPA and DHA plasma oxylipin levels showed a significant increase (p < 0.05) after the supplementation period and a significant decrease (p < 0.05) after the eccentric exercise protocol in the OMEGA and PLACEBO groups (Fig. 2A). There was also a significant difference (p < 0.05) in the plasma levels of EPA- and DHA-derived oxylipins between the OMEGA and the PLACEBO groups after the supplementation period (Fig. 2A). No differences were observed at any time point with the plasma levels of AA PUFA-derived oxylipins (Fig. 2B).

Fig. 2.

Sum of eicosapentaenoic (EPA), docosahexaenoic (DHA) and arachidonic acid (AA) derived oxylipins. (A) Plasma levels of EPA and DHA acid oxylipins. (B) Plasma levels of AA oxylipins. OMEGA (black circles) and PLACEBO (white circles). Data are expressed as means ± SEM (n = 7 per group). P < 0.05 *same treatment versus Pre; †versus PLACEBO at the same timepoint; #same treatment versus Post.

Plasma levels of oxylipins are linked to muscle recovery

A principal component analysis (PCA) was conducted to evaluate patterns in the dataset comprising the following variables, supplementation with n-3 PUFAs (yes/no), concentrations of n-3 and n-6 PUFAs and their derived oxylipins (detected in > 90% of cases before and 2 h after eccentric exercise), VO_2max, age, body fat percentage (%FAT), and body mass index (BMI), where the primary outcome was MVC loss, quantified as the area above the curve. A clear distinction was observed between the participants in the OMEGA group who performed the muscle-damaging exercise after n-3 PUFA supplementation (OMEGA post-supplementation – blue dots) and those who did not (PLACEBO pre- and post-supplementation and OMEGA pre-supplementation – red dots) (Fig. 3A). This finding suggests the potential for specific oxylipins and cardiorespiratory or metabolic factors to be associated with muscle recovery. To select the most influential factors contributing to this trait, we assessed the main variables related to MVC loss by performing LASSO and PLS regressions. From the 64 variables analyzed, twelve metabolites [9-HOTrE (2 h), 9,10 EpOME (before), 20cooh AA (2 h), 12,13 diHOME (2 h), 8-HETE (2 h), 20cooh AA (before), AA (2 h), 9,12,13 triHOME (2 h), 5-HETE (2 h), 13-HOTrE (2 h), 16,17-DiHDoPE (2 h), ALA (before), AA (before)] and three general characteristics (VO_2max, %Fat, and Age) ranked as top quartile influential variables by LASSO regression (Fig. 3B), with a strong predictive value (r² = 0.92) for MVC (Fig. 3C). Meanwhile, PLS regression selected exclusively metabolites as top quartile influential variables (Fig. 3D), with a fair predictive value (r² = 0.71) for MVC loss (Fig. 3E), and a partial overlap with the variables identified by LASSO regression (Fig. 3F). To address whether the variables selected by LASSO and PLS regressions had an additive predictive value, two linear regressions were performed. The first method employed the selected variables in common, and the second used the lowest number of variables that generated the best linear model. Common variables included seven metabolites [9-HOTrE (2 h), 20cooh AA (2 h), 12,13 diHOME (2 h), 8-HETE (2 h), 5-HETE (2 h), ALA (before), AA (before)] (Fig. 3F) with a poor predictive value (r² = 0.66) (Fig. 3G). In contrast, among the twenty-five variables resulting from combining LASSO and PLS regressions, fourteen of them [16,17-DiHDoPE (2 h), %Fat, 20cooh AA (before), 13-HOTrE (2 h), 9,12,13 triHOME (2 h), 5-HETE (2 h), 8-HETE (2 h), 12,13 diHOME (2 h), 9-HOTrE (2 h), 16-HDoHE (2 h), 11,12 DiHETrE (2 h), 9-HOTrE (before), 9-HODE (before), 9-oxoODE (before)] resulted in a highly predictive model (r² = 0.97) (Fig. 3H), combining common and non-common variables (Fig. 3F).

Fig. 3.

Plasma levels of oxylipins are linked to muscle recovery. (A) Principal component analysis (PCA) of all subjects based on 64 controlled variables showing the clustering of the OMEGA group (blue) versus PACEBO group (red) with MVC loss as the primary outcome. (B) Bar chart of LASSO regression coefficients for the top quartile of predictors of MVC loss, and (C) scatterplot of observed versus LASSO-predicted MVC loss (area above the curve), showing strong agreement (r²=0.92). (D) Bar chart of partial least squares (PLS) regression weights for the top quartile of MVC loss predictors, and (E) scatterplot of observed versus PLS-predicted MVC loss, showing moderate agreement (r²=0.71). (F) Venn diagram of variables selected by LASSO, PLS and multivariable linear regression. (G) Scatterplot of observed versus predicted MVC loss using only the seven shared metabolites (r²=0.66). (H) Scatterplot of observed versus predicted MVC loss using the combined set of 14 variables (the 7 common metabolites plus 7 additional LASSO-selected metabolites), selected by best-model multivariable linear regression yielding a highly accurate model (r²=0.97).

Discussion

The present study investigated the effects of n-3 PUFA supplementation on indirect markers of muscle damage and its relationship with plasma oxylipin levels following a single-legged high-intensity EIMD protocol in healthy men. N-3 PUFA supplementation attenuated the loss of MVC strength, while no differences were observed between groups in muscle soreness. Furthermore, n-3 PUFA supplementation increased the plasma EPA- and DHA-derived oxylipins, whereas the n-6 oxylipins remained similar in both groups. Finally, the multivariate analyses identified key oxylipins and supplementation status as potential predictors of MVC strength loss.

To determine whether supplementation with n-3 PUFAs improved muscle recovery, we employed a novel study design using a unilateral exercise model where one leg was subjected to muscle-damaging eccentric exercise before the supplementation period, while the contralateral leg was exercised with the same eccentric protocol after supplementation. This approach strengthened the methodology, as, besides including a PLACEBO group, the pre-supplementation contralateral leg in the OMEGA group also served as an internal control. To the best of our knowledge, this is the first study to use this design to assess muscle recovery following an eccentric exercise protocol in combination with n-3 PUFA supplementation. Furthermore, in the present study, the total workload of the EIMD protocol was similar before and after the 8-wk supplementation period (Table 1), and both the total workload and the maximum forced loss were comparable to those reported after 100 maximal eccentric isokinetic contraction protocols in previous studies^32,33. In addition, maximal strength and maximal aerobic capacity remained unchanged in both groups. Together, these results suggest that the individuals maintained their physical condition and performed the same amount of work, enabling us to reliably compare the magnitude of muscle damage and recovery after the EIMD protocol in both groups and in both legs.

Muscle force production capacity (i.e., MVC strength) has been identified as the best indirect marker of muscle damage^34,35. Interestingly, we found that a high dose of n-3 PUFAs (> 2.5 g/d) and a prolonged supplementation period (8 weeks) induced a ∼15% attenuation of peak MVC loss after EIMD (Fig. 1C), which is in line with previous evidence showing that n-3 PUFAs enhance muscle recovery from EIMD^8–13. However, we observed no effect of n-3 PUFA supplementation on muscle soreness (Fig. 1F), which aligns with previous studies reporting either no effects or mixed results on mitigating muscle soreness^14,36. The divergence in the present findings, between MVC loss and muscle soreness, may suggest that the pathways regulating these variables, classically used to assess muscle damage, are distinctly regulated and differentially sensitive to n-3 PUFAs. While muscle strength loss occurs immediately after EIMD and is largely attributed to excitation-contraction coupling failure, sarcomere disruption, and impaired force transmission, perceived soreness typically peaks 24–72 h post-exercise and appears to be mediated primarily by peripheral sensitization of muscle nociceptors rather than by structural damage^37,38. Consistent with these mechanistic distinctions, peak MVC loss in the present study occurred at 24 h post-exercise, whereas peak soreness was observed at 48 h (Fig. 1C and F).

While a potential confounding factor in the interpretation of the present results is the contralateral repeated-bout effect (CL-RBE), a phenomenon that occurs when a single bout of eccentric exercise provides a protective effect against muscle damage in the opposite limb during a subsequent bout, we believe that the attenuated strength loss observed in the OMEGA group cannot be explained by this effect. First, the CL-RBE is more transient than the ipsilateral RBE. While protection in the exercised limb can persist for several months, recent evidence shows that the contralateral protective effect typically dissipates within 4 weeks in both the upper and lower body^39,40. Our study utilized an 8-week supplementation period between the initial eccentric bout and the post-supplementation bout, a timeframe specifically chosen to exceed the known window of the CL-RBE, ensuring that any protective adaptations from the first bout had subsided before the second assessment. Hence, the evidence from the present and previous studies supports the position that n-3 PUFAs may act as a potential ergogenic aid for muscle recovery as long as a high dose (> 1 g/d) for a prolonged period (> 6 weeks) is administered.

In the present study, we used a fish oil supplement highly concentrated in DHA (∼5:1 DHA: EPA ratio). Our rationale was based on the higher capacity of DHA to serve as a precursor for D-series resolvins, which are critical for the resolution of inflammation during muscle regeneration^6,41. Furthermore, evidence suggests that DHA supplementation increases both EPA and DHA concentrations in red blood cell membranes, whereas EPA supplementation appears to increase EPA levels exclusively⁴². In consequence, a DHA-rich supplement could provide a high availability of substrates for both DHA and EPA-derived oxylipins, as shown in the present study. While this distinct fatty acid profile may have influenced our findings, previous evidence has reported similar attenuation of muscle strength loss after EIMD using either pure EPA or EPA/DHA combinations^11,13. Thus, future research should directly compare the effects of different n-3 PUFA ratios on the oxylipin profile and its functional association with muscle recovery following EIMD.

N-3 and n-6 oxylipins

Prior to supplementation, the lower baseline levels of specific ALA and LA-derived oxylipins observed in the OMEGA group (Supplementary Table 1) are intriguing giving their complex and context-dependent bioactivities, which can be challenging to interpret. To date, the ALA-derived oxylipins have been scantily characterized, yet there is evidence suggesting they have anti-inflammatory effects by reducing prostaglandin production⁴³. Conversely, the LA-derived oxylipins have been established as biomarkers of oxidative stress and agonists of pro-inflammatory pathways²⁷; however, they can also function as activators of the transcription factor proliferator-activated receptor gamma (PPARγ) ⁴⁴, suggesting a dual role for these oxylipins. Interestingly, prior to supplementation both groups responded in a similar fashion to the exercise damaging protocol, suggesting that that their opposing bioactivities resulted in a neutral effect.

Following supplementation, there was a significant incorporation of EPA and DHA in RBC membranes (Table 1), concomitant with an increase in plasma EPA and DHA-derived oxylipins in the OMEGA group, while the PLACEBO group exhibited no changes (Supplementary Table 1). There is growing evidence reporting the role of n-3 PUFA-derived oxylipins in promoting the resolution of inflammation and enhancing tissue repair and regeneration following injury, including muscle damage^6,7,26. The robust and consistent increase in the plasma levels of n-3 PUFA-derived oxylipins observed after the supplementation period in the OMEGA group and the lower values of n-3 PUFA-derived plasma oxylipins observed post-exercise suggest an increased uptake and metabolism in the exercised skeletal muscle. This potentially led to the faster maximum force recovery in the OMEGA compared to the PLACEBO group. Although classic specialized pro-resolving mediators were not detected, a frequent occurrence in lipidomics due to their transient nature, low concentrations, and rapid metabolic inactivation⁴⁵, the significant increases in 17-HDoHE, 14-HDoHE, and 18-HEPE are biologically relevant. These metabolites derived from DHA and EPA serve as critical molecules for the synthesis of D-series resolvins, maresins, and E-series resolvins. Therefore, it is highly likely that n-3 PUFA supplementation expanded the substrate pool available for the local synthesis of specialized pro-resolving mediators, leading to the faster resolution of muscle strength.

On the other hand, the oxylipins derived from the n-6 PUFA AA are primarily identified for their pro-inflammatory properties²⁷. In the present study, there were no differences between groups in the sum of AA PUFA-derived plasma oxylipins before and after the supplementation period (Fig. 2B). This is not surprising, as the plasma levels of AA remained similar before and after the supplementation period in both groups (Supplementary Table 1). In addition, the stable levels of n-3 and n-6 PUFA-derived oxylipins throughout the study in the PLACEBO group could explain the similar response to the muscle-damaging protocol before and after the supplementation period in this group.

It could be argued that the plasma oxylipin profile may not be an exact proxy for skeletal muscle oxylipin content; however, previous studies in humans have demonstrated comparable temporal changes between skeletal muscle and circulating oxylipins two hours following resistance-type exercise^46,47. Interestingly, these studies reported a greater abundance of n-6 and n-3 PUFA-derived oxylipins two hours after exercise, an observation that contrasts with the results of the present study (Fig. 3). Although difficult to reconcile, this divergence is potentially attributable to the distinct exercise protocols employed, as previous studies have performed either an aerobic-type exercise protocol⁴⁸ or a lower-volume, less-damaging resistance-type exercise^46,47. Consequently, the oxylipin profile after exercise appears to exhibit differential responses depending on the type and intensity of exercise⁴⁹. Future studies should investigate plasma and skeletal muscle oxylipin responses to various exercise protocols that differ in contraction type (e.g., concentric versus eccentric), intensity, and volume. This could provide a comprehensive understanding of how different types of exercise influence the production and balance of n-3 and n-6 PUFA-derived oxylipins and how they could promote recovery.

Predictive model

The clustering observed in the PCA plot suggests that supplementing with n-3 PUFAs could promote a favorable oxylipin profile, potentially protective against muscle damage (Fig. 3A). Furthermore, the identification of a strong predictive model where %FAT and specific oxylipins are linked with MVC loss (Fig. 3H), emphasizes the complex interplay between different lipid mediator families in determining the response to injury or muscle damage. The absence of links between oxylipins and muscle soreness may be related to the high variability of DOMS response in magnitude and time after EIMD in the present study, and also shown previously³⁸. Moreover, emerging evidence suggests that DOMS represents inflammation of the extracellular matrix rather than direct damage to contractile muscle fibers³⁷. In consequence, circulating oxylipins may be robustly linked to processes governing skeletal muscle damage, such as myocellular membrane remodeling, resolution of intramuscular inflammation, and restoration of excitation-contraction coupling, instead of localized extracellular matrix nociceptive processes associated with muscle soreness. The inclusion of %FAT as a predictor is not surprising, as higher adiposity is associated with poor muscle recovery⁵⁰ and changes in oxylipin profile⁵¹. Interestingly, several of the oxylipins incorporated into the predictive model exhibit distinct functional roles, reflecting a simultaneous orchestration of pro-inflammatory and anti-inflammatory processes, as well as initiation of resolution pathways and different metabolic responses. The contribution of the LA-derived oxylipins 9-HODE and 9-oxoODE illustrate these divergent roles, with the former linked to nociception and inflammation⁵², while the latter is a PPARγ ligand with anti-inflammatory activity⁴⁴. The ALA-derived oxylipins 9- and 13-HOTrE oxylipins have been shown to display pro-resolving activity²⁷, while the LA-derived oxylipin 12,13-diHOME has been shown to rise during exercise and promote fatty acid uptake and oxidation in skeletal muscle⁵³, suggesting a metabolic contribution of fatty acids to the recovery process, as reported previously with eccentric exercise⁵⁴. Furthermore, 5-HETE and 8-HETE are AA-derived oxylipins that exert a potent pro-inflammatory response promoting neutrophil recruitment and early tissue clearance²⁷, supporting their potential role in initiating tissue repair. The DHA-derived oxylipin 16-HDoHE oxylipin, although detected in other studies^55,56, its function remains unclear; however, it could have similar pro-resolving and anti-inflammatory actions to other DHA-derived hydroxy fatty acids²⁷, suggesting a role in facilitating muscle recovery. Finally, the contribution of the soluble epoxide hyrolase-derived diols 11,12-DiHETrE and 16,17-DiHDoPE most likely reflects an increased conversion of upstream epoxy-PUFAs, resulting in reduced availability of vasodilatory and pro-resolving epoxides⁵⁷. This metabolic shift could facilitate the necessary environment for the early inflammatory response to EIMD. Altogether, these findings highlight the significance of interactions between functionally distinct oxylipins in orchestrating immune responses, where pro-inflammatory signals initiate tissue repair in conjunction with pro-resolving mediators that facilitate resolution, thereby promoting efficient muscle recovery following EIMD. The strong predictive power of the present model suggests that the plasma oxylipin profile, both at baseline and two hours after exercise, is an important factor in the muscle recovery process.

Limitations

The present study has some limitations that should be acknowledged. The relatively low sample size may have reduced the study’s statistical power, potentially masking some of the beneficial effects of n-3 PUFAs and their derivatives on muscle recovery. Second, we did not obtain muscle biopsies to directly assess muscle damage, local oxylipin status, and incorporation of n-3 PUFAs into muscle membranes; however, previous studies have shown that the assessments used in the present study are valid surrogates for these parameters⁴⁷. Similarly, we did not measure direct markers of inflammation (i.e., interleukins) or muscle damage (creatine kinase and myoglobin). While their inclusion could have provided additional mechanistic insight, creatine kinase and myoglobin exhibit high interindividual variability and often correlate poorly with functional outcomes such as strength loss^58,59. Additionally, the discrepancy observed between decreased caloric intake and stable body weight throughout the intervention in the OMEGA group represents another limitation; however, it most likely reflects a methodological artifact rather than a true physiological energy deficit, giving the phenomenon of underreporting in self-reported dietary assessments⁶⁰. Moreover, the difference observed in age between groups could have potentially influenced muscle recovery; however, the likelihood that this difference confounded the present findings is minimal, as there were no between-group differences in baseline body composition, maximal isometric knee strength, VO₂max, or total eccentric workload performed, suggesting comparable neuromuscular and metabolic capacity at study entry. In addition, both groups exhibited similar functional recovery responses following EIMD, consistent with previous work indicating that age does not substantially alter the magnitude or time course of strength loss and recovery^61,62. Finally, measuring oxylipins at a single post-exercise time point (2 h) provides only a snapshot of the complex dynamic processes of oxylipin status; therefore, future studies should perform a time-course analysis comparing blood and skeletal muscle responses after EIMD with n-3 PUFAs to gain a deeper insight into the role of n-3 PUFA-derived oxylipins on muscle recovery.

Conclusions

In conclusion, this study demonstrates that a high-dose and prolonged n-3 PUFA supplementation period attenuates exercise-induced muscle damage in healthy males. The observed increase in plasma n-3 PUFA-derived oxylipins following supplementation suggests a potential role for these lipid mediators in promoting muscle recovery. Overall, these findings support the use of n-3 PUFAs as a potential ergogenic aid for mitigating strength loss after strenuous exercise, provided a sufficient dose and duration are employed.

Methods

Participants

Eighteen healthy men, who had not performed lower limb resistance training regularly in the past six months and who reported no recent history of n-3 PUFA consumption or lower limb injuries were recruited (Table 1). Individuals were excluded if they had a body mass index (BMI) > 30 kg·m^− 2 and if they trained regularly (> 3 times/week) and/or had a maximum oxygen uptake (VO_2max) ≥ 50 mL·kg^− 1·min⁻1. This study was conducted following the Declaration of Helsinki, and all participants signed an Informed Consent approved by the Universidad de O’Higgins Ethics Committee (#020/2021) before the study commencement (August 2021). All procedures were performed at the Institute of Exercise and Rehabilitation Sciences of the Universidad Andrés Bello. The study was retrospectively registered on clinicaltrials.gov #NCT07237114 (14/11/2025).

Study design

This study utilized a randomized, double-blind, placebo-controlled experimental design. The participants were randomly allocated to one of two groups: OMEGA or PLACEBO. To ensure double-blinding, the randomization sequence was generated by an independent investigator using a website (https://www.sealedenvelope.com/). A second investigator, who was not involved in participant contact or data collection, was responsible for placing the supplements into identical, coded bottles. All personnel responsible for providing the supplements and supervising the exercise intervention, as well as the participants, remained blinded to the group assignments throughout the study. A one-leg maximal isokinetic eccentric exercise protocol (10 sets of 10 repetitions at maximal eccentric isokinetic strength) was performed, with the exercised leg being randomized for each participant before (Pre) and at the end (Post; contralateral leg) of the 8-week supplementation period. This approach was adopted because the PLACEBO group and the pre-supplementation leg in the OMEGA group served as internal controls. Muscle function and soreness, and a blood sample were collected pre- and post-supplementation periods. Muscle function and soreness were assessed by blind assessors. A schematic overview of the study design and procedures is depicted in Fig. 4.

Fig. 4.

Study design. Before exercise (Pre); immediately after exercise (IM); eccentric contraction (ECC). Black arrows indicate the specific time points at which each assessment was performed.

The participants in the OMEGA and PLACEBO groups were given five daily (two in the morning and three at night) fish oil (2.5 g/d of DHA and 0.5 g/d of EPA; Omega Up TG DHA 600, Newscience, Santiago, Chile) or placebo filled with maltodextrin capsules for 8 weeks. The dose and duration of fish oil capsules have been reported to be incorporated into skeletal muscle phospholipids^20,63. To ensure double-blinding, capsules were placed into de-identified bottles and subsequently provided to the participants.

Dietary, body composition, and physical activity assessment

Participants were advised to avoid strenuous exercise, alcohol, and caffeine consumption for 24 h before each laboratory visit. Individuals were also instructed to maintain their habitual diet and lifestyle habits throughout the study. The participants completed a 3-day dietary record before and after the supplementation period to estimate their macronutrient and caloric intake. The record comprised two consecutive weekdays and one weekend day to capture variations in habitual eating patterns. Before data collection, participants received standardized oral and written instructions from a researcher on how to accurately record all food and beverage intake. These instructions included visual guides on serving sizes and portion estimation (e.g., using pictures, household measures, or weight). Body composition was measured by multifrequency bioelectrical impedance analysis (model 770, InBody Co., Seoul, Korea). Similarly, the individuals answered the International Physical Activity Questionnaire (IPAQ) to assess their physical activity levels⁶⁴.

Maximal aerobic capacity

Before (Pre) and after (Post) the supplementation period, an incremental cycling test was performed to determine VO_2max using an upright cycle ergometer (Ergoline Ergoselect100, Via Sprint 150p, Germany). The test started at a power output of 50 W for 4 min and increased by 25 W every one min until volitional exhaustion. Cadence was kept between 60 and 70 revolutions per minute (rpm), and the participants received verbal encouragement during the test³. Oxygen consumption was measured by a breath-by-breath gas analyzer (Jaeger, CareFusion, Germany) previously calibrated with known gases (CO₂=4% and O₂ = 16%). Maximal oxygen consumption criteria included volitional fatigue, a rate of perceived exertion > 18, and a respiratory exchange ratio > 1.1, following previous recommendations⁶⁵.

Exercise protocol

The maximal isokinetic eccentric exercise protocol consisted of 10 sets of 10 maximal knee extensor eccentric isokinetic contractions at 60°∙s^− 1 in the isokinetic dynamometer (IsoForce, TUR, Germany). Participants were seated with their hip and knee at 110° of flexion. The dynamometer moved the leg passively at 30°∙s^− 1 up to 10° of knee flexion, then moved the leg backward at 60°∙s^− 1 until 100° of knee flexion, during which the participant had to resist maximally the backward movement of the lever. Participants had 1 s rest between repetitions and one minute of rest between sets⁶⁶. All participants received verbal encouragement to exert maximal effort during each contraction. The total workload was recorded per set.

MVC strength

The isometric maximum voluntary contraction (MVC) strength of the knee extensor muscles of the dominant and non-dominant leg was measured at 90° of knee flexion using an isokinetic dynamometer (IsoForce, TUR, Germany). Participants performed a 5-minute warm-up on a cycle ergometer, followed by three warm-up submaximal contractions at 50%, 70%, and 80% of each participant´s self-perceived MVC with 30 s of rest between contractions. After the warm-up, participants performed three 3-s MVCs as fast and hard as possible with a 1-min rest between contractions³. Visual feedback was provided simultaneously.

Muscle soreness

Thigh muscle soreness was quantified using a 100-mm visual analog scale (VAS) in which 0 indicated no pain and 100 represented the worst pain imaginable⁶⁷. The participants were asked to mark the level of perceived pain in the quadriceps femoris muscle while sitting on and standing from a 42-cm chair three times. The average of three measurements was used for further analysis.

Blood sample collection and analyses

Blood samples were collected using a needle and Vacutainer kit and immediately centrifuged to obtain plasma (1000 g for 10 min at 4 °C) and RBC fractions (3000×g for 10 min at 20 °C). Both fractions were frozen at −80 °C and later analyzed for plasma oxylipins and RBC membrane composition.

Plasma oxylipins were analyzed by HPLC/MS/MS as previously described³⁰. Briefly, deuterated internal standards (Cayman Chemicals) were added to plasma samples and adjusted to pH < 3.0 before solid-phase extraction using Strata-X-SPE (Phenomenex) columns preconditioned with methanol, followed by pH 3.0 water. Samples were loaded on the columns, washed, and free oxylipins were eluted with 100% methanol. The eluate was then dried and resuspended in solvent A (water/acetonitrile/formic acid, 70/30/0.02 v/v/v) for oxylipin analysis by HPLC/MS/MS, using a Luna 5-µm C18 column (100 Å, 250 × 2.0 mm, Phenomenex) on a Shimadzu Nexera XR HPLC, coupled to an ABSciex QTRAP 6500 MS operating in negative mode with electrospray ionization. Oxylipins were quantified using the stable isotope dilution method⁶⁸, with quantification set at peak heights of 5 times baseline. Quantifiable oxylipins present in at least 65% of samples (Pre-Post) for any group (n = 7 per group) were included in the analysis. All oxylipins scanned for, detector response factors, instrument details and analysis conditions, signal to noise values and the lowest limit of quantitation can be found in⁶⁹ and supplementary file 1. Samples were analyzed in random order and by an individual (TW) blinded to the treatments.

Quantitative extraction of total lipids from RBC was measured according to Bligh and Dyer⁷⁰ with the addition of BHT. Fatty acid methyl esters (FAMEs) from RBC were prepared according to Morrison and Smith⁷¹. FAMEs were extracted with 0.5 mL of hexane and identified and quantified by gas-liquid chromatography in Agilent equipment (model 7890B, Santa Clara, CA, USA) equipped with a capillary column (Agilent HP-88, 100 m × 0.250 mm; I.D. 0.25 μm) and flame ionization detector. Identification and quantification of FAMEs were achieved by comparing the retention times and the peak area (%) values of unknown samples to those of commercial lipid standards (Nu-Chek Prep Inc., Elysian, MN, USA). C23:0 was used as an internal standard (Nu-Chek Prep Inc., Elysian, MN, USA), and data was processed using the Hewlett-Packard Chemstation software system.

Explicative variable discovery

To determine the potential occurrence of phenotypes associated to MVC and PUFA profiles, multidimensional reduction analysis was performed using principal component analysis. Additionally, LASSO (least absolute shrinkage and selection operator) and PLS (partial least square) regressions were applied in parallel (independently), and their results combined for selecting variables identified as important by both methods (additive and intersection approach). Continuous variables were log(2) transformed to data harmonization and then analyzed with LASSO and OLA regressions. This strategy prioritized the identification of robust variables from different traits, which were further assessed by linear regression analysis of a MVC prediction model based on selected variables. Analyses were performed using using XLSTAT (ver 2024.4.2), as previously described^72,73.

Statistical analysis

The sample size (n = 12 per group) was calculated using an alpha level of 0.05, a beta level of 0.20 (80% power), and an anticipated effect size of 1.06 based on expected differences in maximal voluntary contraction (MVC). An independent Student´s t-test was used to compare OMEGA and PLACEBO groups for baseline physical characteristics. A two-way repeated measures ANOVA was used to assess changes over time after eccentric exercise between the pre- and post-supplementation periods, with separate analyses for MVC recovery and muscle soreness within each group, and a combined analysis for RBC EPA and DHA levels. Further, a mixed-effects model ANOVA with the Geisser-Greenhouse correction to account for potential violations of sphericity to assess oxylipin responses across both groups was performed. When a significant main effect of group, time or interaction was found, multiple comparisons were made with Tukey’s post hoc test. Delta differences in peak strength loss and peak soreness were compared using an independent Student´s t-test. The significance level was set at P < 0.05. All statistical analyses were performed with GraphPad 9.0 (Prism, USA). Data are presented as mean ± standard error of the mean (SEM).

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

LP and SJ-V, contributed to the conception and design of the study. CM-F, CR, CC, YT, JA, SG, ON, SU, BC, TW, RV, HMA, LP, and SJ-V assisted with measurements. BJ-K, DV-I, LP, and SJ-V analyzed and interpreted the data, and wrote the first draft of the manuscript. All authors read and approved the final version of the manuscript.

Funding

This project was funded by FONDECYT # 1241959 (DV-I), FONDECYT #1211962 (LP) project, and FONDECYT #11220333 (SJ-V).

Data availability

All data are included in the article and Supplementary data.

Declarations

Competing interests

The authors declare no competing interests.