Dietary omega-3 fatty acid supplementation does not impair vitamin E status or promote lipid peroxidation in growing horses

Abstract

Omega-3 (n-3; ω-3) fatty acids (FA) are often included in the diet for their potential health benefits. However, because oxidative potential is increased with the degree of unsaturation in vitro, polyunsaturated FA such as eicosapentaenoic acid (EPA; 20:5n-3) and docosahexaenoic acid (DHA; 22:6n-3) may be at increased risk of lipid peroxidation. We aimed to determine the effects of dietary n-3 FA supplementation on antioxidant status and lipid peroxidation in yearling horses. Quarter Horses (mean ± SEM; 14.6 ± 0.2 mo) were randomly assigned to receive no n-3 FA supplementation (CON; n = 6) or 60 mg n-3/kg body weight from milled flaxseed (FLAX; n = 6) or encapsulated fish oil (FISH; n = 6). All horses received a basal diet of mixed grain concentrate fed individually at 1.5% body weight (dry matter basis) and ad libitum bahiagrass pasture forage. Blood samples were obtained before and after 70 d of supplementation to evaluate vitamin E, selenium, lipids, antioxidant status, and oxidative stress. Data were analyzed using a mixed model ANOVA with repeated measures. Supplementation with n-3 FA did not reduce serum vitamin E or Se and, in fact, elevated (P ≤ 0.0003) vitamin E status in FISH horses. At day 70, serum triglycerides were lower in FISH and FLAX horses than CON horses (P ≤ 0.02) and F2-isoprostanes were lower in FISH than CON horses (P = 0.0002). Dietary n-3 FA had no effect on cholesterol, reduced and oxidized glutathione, glutathione peroxidase, and thiobarbituric acid-reactive substances. In growing horses fed to meet their vitamin E requirements, supplementation with 60 mg n-3/kg body weight did not negatively affect vitamin E status or promote lipid peroxidation. Elevated vitamin E status in horses fed FISH, coupled with lower serum F2-isoprostanes, further suggest that the longer-chain, highly unsaturated n-3 FA, EPA and DHA, may actually attenuate lipid peroxidation.

Introduction

A variety of fat sources possessing differing fatty acid (FA) profiles are routinely added to equine diets. For example, corn oil, a commonly used source of fat in equine rations, contains large amounts of omega-6 (n-6) FA. By comparison, most forages, flaxseed, and marine fish oil and algae are good sources of n-3 FA. In recent years, there has been increased interest in supplementation of n-3 FA due to their recognized health benefits. In humans, these benefits include decreased risk for cardiovascular disease (Manuelli et al., 2017), reduced inflammation and enhanced immune function (Yu et al., 2017), and support of muscle repair and remodeling, including enhanced protein synthesis (Heaton et al., 2017). In horses, dietary n-3 FA improved clinical outcomes for horses with respiratory disease (Nogradi et al., 2015), enhanced the early inflammatory response (Vineyard et al., 2010), modulated metabolic parameters in horses with equine metabolic syndrome (Elzinga et al., 2019), but showed mixed results on the ability to attenuate exercise-induced inflammation (Hess et al., 2019).

Despite the reported benefits of dietary n-3 FA, there is concern that n-3 FA supplementation may increase the cell membrane’s susceptibility to lipid peroxidation and oxidative stress. Oxidative stress within the cell refers to damage caused to cellular components such as DNA, proteins, and lipids by reactive oxygen species and free radicals generated with metabolism or inflammation. If this damage is allowed to progress, it will lead to the destruction of cells and tissues. Elevated oxidative damage has been implicated in cardiovascular disease, cancer, atherosclerosis, diabetes, neurodegenerative disorders, and aging in humans (Marrocco et al., 2017). In performance horses, exercise-induced production of reactive oxygen species may lead to an increase in oxidative damage, which could result in fatigue and injury (Lykkesfeldt and Svendsen, 2007).

Because of their greater degree of unsaturation, the n-3 FA, alpha-linolenic acid (ALA; 18:3), eicosapentaenoic acid (EPA; 20:5), and docosahexaenoic acid (DHA; 22:6) are thought to be at particular risk for oxidative damage (Al-Gubory, 2012). Therefore, it is common industry practice to supplement with higher levels of an antioxidant such as vitamin E to quench the peroxy radicals when any type of fat is added to the diet of horses (Stratton-Phelps et al., 2003). However, studies in humans and dogs have produced inconclusive results regarding the effects of n-3 FA on vitamin E status or lipid peroxidation (Wander et al., 1997; Higdon et al., 2000; Ottestad et al., 2012). Other research has suggested that n-3 FA supplementation does not adversely affect, and may actually attenuate, oxidative stress (Mori, 2004).

Therefore, the objectives of this study were to determine the effects of n-3 FA supplementation as milled flaxseed or fish oil on oxidative damage and antioxidant status in growing horses. We hypothesized that n-3 FA supplementation would not impair antioxidant status, nor enhance susceptibility to oxidative damage, and that the source of n-3 FA would not induce differential responses.

Materials and Methods

Horses

This study was reviewed and approved by the Institutional Animal Care and Use Committee at the University of Florida (project number D908). Eighteen American Quarter Horse yearlings (nine fillies, nine geldings) with a starting age of 14.6 ± 0.2 mo (mean ± SEM) and body weight (BW) of 392 ± 5 kg were utilized in this study. Fillies and geldings were group-housed separately on two adjacent 8-ha pastures at the Institute of Food and Agricultural Sciences, Equine Sciences Center in Ocala, FL.

Dietary treatments

Horses were blocked by sex and paired by age and randomly assigned to one of three treatments for 70 d: milled flaxseed (FLAX, n = 6; Pizzey’s Milling, Angusville, Manitoba, Canada), encapsulated fish oil (FISH, n = 6; JBS United, Sheridan, IN) or no n-3 FA supplementation (CON, n = 6). The basal diet consisted of a custom formula grain concentrate (OBS Feeds, Ocala, FL) fed individually at 1.5% BW (dry matter basis) split equally into two meals at 0700 and 1500 h. The FLAX and FISH supplements were fed at 0.261 and 0.906 g/kg BW to provide 60 mg total n-3/kg BW. The total daily amount of each supplement was split equally between morning and evening feedings and hand-mixed into the concentrate. Horses were allowed ad libitum access to bahiagrass (Paspalum notatum) pasture forage from May to July, which was during the active growing season. Horses had been maintained on the same grain concentrate (without supplementation) and pasture for at least 6 mo prior to the start of the experiment. Nutrient composition of basal feeds and n-3 FA supplements is presented in Table 1. Feedstuff FA analysis has been previously reported (Vineyard et al., 2010). Given known concentrate intake (1.5% BW) and NRC (2007) estimates of 2.5% BW total DM intake for yearling horses, pasture intake was estimated to be 1.0% BW. Treatment diets provided similar dry matter, digestible energy, crude protein, selenium, and vitamin E intakes and met or exceeded requirements for growing horses (NRC, 2007) (Table 2).

Table 1.

Nutrient composition of basal diet feeds and the milled flaxseed and encapsulated fish oil supplements¹

	Basal diet		Supplements
Nutrient concentration	Grain concentrate	Pasture forage	Milled flax	Encapsulated fish oil
DE, Mcal/kg	3.40	2.26	3.00	3.70
Crude fat, %	4.2	2.7	37.7	21.5
Crude protein, %	16.6	16.3	22.9	11.8
NDF, %	25.3	60.2	40.0	9.9
ADF, %	12.3	35.1	19.0	5.9
Ca, %	0.83	0.45	0.24	0.33
P, %	0.72	0.39	0.59	0.15
Zn, mg/kg	168	34	41	33
Cu, mg/kg	35	9	11	4
Se, mg/kg	0.75	0.10	0.86	0.35
Vitamin E, IU/kg	104	140	38	202
Fatty acid, g/100 g total fat
18:2 n-6 (LA)	48.1	17.1	16.5	6.8
18:3 n-3 (ALA)	3.0	51.0	61.1	2.6
20:4 n-6 (ARA)	0	0	0	1.0
20:5 n-3 (EPA)	0	0	0	14.4
22:5 n-3 (DPA)	0	0	0	2.3
22:6 n-3 (DHA)	0	0	0	11.5
Other fatty acids2	48.9	31.9	22.4	61.4

¹Values are presented on a 100% dry matter basis and represent a mean of monthly samples obtained May through August.

²C12:0, C14:0, C16:0, C16:1n-7, C18:0, and C18:1n-9.

Table 2.

Estimated daily intake of select nutrients for yearling horses receiving each dietary treatment¹

Nutrient intake	CON diet	FLAX diet	FISH diet
DM, kg	9.8	9.9	10.2
DE, Mcal	28.9	29.2	30.2
Crude fat, g	353	391	429
Crude protein, g	1,615	1,638	1,657
Se, mg	4.8	4.9	4.9
Vitamin E, IU	1,160	1,164	1,232
Total n-6 fatty acids, g	137	143	143
Total n-3 fatty acids, g	61	85	85

¹Based on initial mean body weight of 392 kg; actual dry matter intake of concentrate at 1.5% BW, milled flax at 0.261 g/kg BW, and encapsulated fish oil at 0.906 g/kg BW; and pasture intake estimated at 1% BW.

Sample collection

To allow investigators adequate time for sample collection and processing, horses were equally divided into three sampling groups with treatments balanced between groups, and blood samples were obtained over three consecutive days. Blood samples were obtained before (day 0) and after 70 d of supplementation for determination of vitamin E status, markers of lipid peroxidation, antioxidant biomarkers, and oxidative stress. On each day of sampling, approximately 40 mL of blood was collected from each horse by jugular venipuncture into 10-mL evacuated tubes (Vacutainer, Becton Dickinson Co., Franklin Lakes, NJ) containing either no anticoagulant for harvesting serum or in tubes containing sodium heparin for harvesting of plasma and red blood cell lysate. Blood tubes were immediately placed on ice, where they remained until processing within 2 h of collection by centrifugation for 12 min at 1,800 × g at room temperature. Serum, red blood cell lysate, and plasma were harvested and stored at −80 °C until analyses were performed.

Vitamin E and selenium status

Serum selenium (Se) and vitamin E analyses were performed by the Michigan State University Veterinary Diagnostic Laboratory (Lansing, MI). Serum Se was determined using the fluorometric method with an ultraviolet detector. Vitamin E was measured as alpha-tocopherol by HPLC with a fluorescence detector. As recommended by Horwitt et al. (1972), vitamin E status is reported as total serum vitamin E and as the ratio of serum vitamin E to total serum lipids (cholesterol + triglycerides). Serum cholesterol and triglycerides were determined colorimetrically using commercially available kits following manufacturer instructions (cholesterol: Cholesterol E Kit, Wako Diagnostics, Mountain View, CA; triglycerides: Cayman Chemical Company, Ann Arbor, MI). Serum cholesterol and triglycerides were each analyzed on a single 96-well plate with an intra-assay CV of 3.4% and 3.7%, respectively.

Antioxidants and oxidative stress markers

Antioxidant biomarkers reduced (GSH) and oxidized (GSSH) glutathione were measured fluorometrically in erythrocyte lysate using previously published methods (Browne and Armstrong, 1998a). Briefly, GSH and total glutathione were extracted from each sample by meta-phosphoric acid. In the presence of 7.5 mM o-phthaldialdehyde (OPT; cat #P1378, MilliporeSigma, Burlington, MA), total glutathione was quantified in a 0.1 N NaOH buffer and GSH was quantified in a buffer containing 0.1 M sodium phosphate monobasic and 5 mM EDTA at pH 8.0. Total glutathione and GSH are presented relative to hematocrit in each sample, which was determined by microcapillary centrifugation of whole blood. Oxidized GSH (GSSH) was calculated by subtracting GSH from total glutathione. Glutathione peroxidase (GPx) activity was determined in erythrocyte lysate colorimetrically by cumene hydroperoxide reduction as previously described (Paglia and Valentine, 1967). Intra-assay CV was 2.1%, 1.7%, and 1.3% for GSH, GSSH, and GPx, respectively; inter-assay CV was 4.3%, 2.7%, and 3.4% for GSH, GSSH, and GPx, respectively.

Oxidative stress was assessed by plasma thiobarbituric acid reactive substances (TBARS) as a measure of total oxidative stress, plasma total lipid hydroperoxides (LHP) as a measure of lipid peroxidation, and serum F2-isoprostanes (F2-ISO) concentrations generated from free radical peroxidation of arachidonic acid (ARA). Plasma TBARS were measured fluorometrically as previously described (Yagi, 1998). Total LHP was calculated by totaling quantities of peroxidation products of linoleic acid (13-hydroperoxy-9,11-octadecadienoic acid [HpODE] and 9-HpODE) and ARA (15-hydroxyeicosatetraenoic acid [HETE], 12-HETE, 11-HETE, 8-HETE, 5-HETE, 12-hydroperoxyeicosatetraenoic acid [HpETE] and 5-HpETE) for each sample as determined by HPLC with a fluorescent detector (Browne, 1998; Browne and Armstrong, 1998b). F2-isoprostanes were determined using a commercially available competitive enzyme immunoassay (Cayman Chemical Company). F2-isoprostanes were analyzed in triplicate, with an intra-assay CV of 5.4% and an inter-assay CV of 4.5%.

Statistical analyses

Statistical analysis was performed using SAS software (Version 9.4, SAS Institute, Inc., Cary, NC). Data were analyzed with mixed model analysis of variance with repeated measures. The model included treatment, time, sex, and the interaction of treatment and time as fixed effects and horse within treatment as a random effect. Sex was not significant for any variable examined and was subsequently removed from the model. Outliers were determined using PROC BOXPLOT in SAS and were removed if greater than two standard deviations from the mean. Data are presented as means ± SEM. Differences between treatment groups were declared significant at P ≤ 0.05.

Results

As we reported previously, the FA composition of plasma and erythrocyte membranes was significantly altered by n-3 FA supplementation in a manner that mimicked the n-3 FA provided by FLAX or FISH (Vineyard et al., 2010). This finding confirmed that dietary FA were digested and absorbed into circulation and incorporated into cell membranes.

Vitamin E and selenium status

Serum levels of vitamin E, cholesterol, triglycerides, and Se of horses were not different between treatments prior to the initiation of n-3 FA supplementation. After 70 d of supplementation, vitamin E in horses fed FISH increased (P = 0.0002) to be nearly 2-fold higher (P ≤ 0.01) than horses fed FLAX or CON (Figure 1A). Serum triglycerides increased in CON horses (P = 0.03) and decreased in FISH horses (P = 0.0002) from day 0 to 70, resulting in CON having greater serum triglycerides than horses receiving FISH or FLAX at d 70 (P ≤ 0.02; Table 3). Serum cholesterol was unaffected by time or treatment through the 70-d experiment (Table 3). As a result, when serum vitamin E was expressed as a ratio to total serum lipids (cholesterol + triglycerides), the same pattern was observed as non-normalized serum vitamin E; vitamin E relative to total lipids increased in horses fed FISH (P = 0.0003) by day 70 to be greater (P ≤ 0.008) than horses fed CON and FLAX (Figure 1B). Serum vitamin E was not different between FLAX and CON horses, nor did it increase in these treatments from day 0 to 70 (Figure 1A and B). Serum Se decreased in all horses from day 0 to 70 (P < 0.0001) but remained within physiologically normal levels (Stowe and Herdt, 1992), and was not different between treatments at either day 0 or 70 (Figure 1C).

Figure 1.

Mean ± SEM serum vitamin E (A: μg/mL, B: μg/mg cholesterol + triglycerides) and C) serum selenium (Se) in yearling Quarter Horses before (day 0) and after 70 d of no n-3 supplementation (CON, n = 6) or 60 mg/kg body weight n-3 fatty acid supplementation per day in the form of milled flax (FLAX, n = 6) or encapsulated fish oil (FISH, n = 6). Overall effects of time (P = 0.007, P = 0.012, P < 0.001), treatment (P = 0.175, P = 0.114, P = 0.442), and time × treatment (P = 0.007, P = 0.009, P = 0.262), for panels A, B, and C, respectively, as determined using a repeated-measures ANOVA. ^abBars with unlike letters differ (P ≤ 0.05). ^cdDay 0 differs from day 70 across treatments (P < 0.0001).

Table 3.

Mean and SEM serum cholesterol and triglycerides, plasma TBARS, 5-HETE, and total LHP, and erythrocyte lysate reduced (GSH) and oxidized (GSSH) glutathione and glutathione peroxidase (GPx) concentrations in yearling Quarter Horses before (day 0) and after 70 d of no supplementation (CON, n = 6) or supplementation with milled flax (FLAX, n = 6) or encapsulated fish oil (FISH, n = 6)

					P-value1
Variable	Treatment	Day 0	Day 70	SEM	Time	Treatment	Time × Treatment
Serum
Cholesterol, mg/dL	CON	94.3	97.3	3.3	0.315	0.383	0.864
	FLAX	90.2	90.2
	FISH	87.0	92.0
Triglycerides, mg/dL	CON	23.2	28.7a*	2.6	0.014	0.290	0.0005
	FLAX	22.4	17.6b
	FISH	29.6	17.0b*
Plasma
TBARS, nmol/mL	CON	1.9	1.6	0.1	0.173	0.331	0.491
	FLAX	1.6	1.5
	FISH	1.6	1.6
5-HETE, µmol/L	CON	34.7	99.3a*	11.7	0.024	0.132	0.028
	FLAX	29.0	45.7b
	FISH	40.4	33.2b
Total LHP, µmol/L	CON	791	683	25	0.007	0.083	0.876
	FLAX	692	620
	FISH	766	672
Erythrocyte lysate
GSH, mg/dL	CON	9.3	6.3	1.9	0.142	0.102	0.833
	FLAX	8.0	4.9
	FISH	13.0	11.9
GSSH, mg/dL	CON	3.2	4.2	1.0	0.268	0.307	0.854
	FLAX	1.8	4.1
	FISH	1.0	1.8
GSH/GSSH	CON	4.69	2.23	0.86	0.032	0.119	0.799
	FLAX	3.27	0.08
	FISH	1.61	0.06
GPx, IU/g Hb	CON	183	185	8	0.247	0.661	0.852
	FLAX	183	194
	FISH	190	198

¹Data were analyzed using a mixed model ANOVA with repeated measures.

^abWithin day 70, means with unlike letters differ between treatments (P < 0.05).

*Within a treatment, day 0 differs from day 70 (P < 0.05).

Antioxidants and oxidative stress markers

Plasma 5-HETE increased in CON horses from day 0 to 70 (P = 0.002) but did not change over time in FISH or FLAX horses, resulting in greater 5-HETE in CON than FISH or FLAX at day 70 (P ≤ 0.02; Table 3). Plasma 12-HETE decreased from d 0 to 70 (71.4 ± 9.5 vs. 3.8 ± 9.5 μmol/L for day 0 and 70, respectively; P = 0.0002), as did 12-HpETE (31.4 ± 4.2 vs. 15.8 ± 4.2 μmol/L for day 0 and 70, respectively; P = 0.02). There was no effect of time or treatment on plasma 8-HETE, 11-HETE, 15-HETE, 5-HpETE, 9-HpODE, or 13-HpODE (data not shown). Plasma total LHP decreased from day 0 to 70 in all horses (P = 0.007) but was not different between treatment groups (Table 3). However, serum F2-ISO increased by day 70 in CON horses (P = 0.0002), tended to increase in FLAX (P = 0.06), but did not change over time in FISH horses (Figure 2). This resulted in FISH supplemented horses having lower serum F2-ISO compared to CON (P = 0.01), with FLAX being intermediate between concentrations observed for FISH and CON at day 70 (Figure 2). Addition of n-3 FA to the diet had no effect on erythrocyte GSH, GSSH, or GPx, or plasma TBARS (Table 3).

Figure 2.

Mean ± SEM serum F2-isoprostanes in yearling Quarter Horses before (d 0) and after 70 d of no n-3 supplementation (CON, n = 6) or 60 mg/kg body weight n-3 fatty acid supplementation per day in the form of milled flax (FLAX, n = 6) or encapsulated fish oil (FISH, n = 6). Overall effects of time (P = 0.0008), treatment (P = 0.226), and time × treatment (P = 0.03) as determined using a repeated-measures ANOVA. ^abc Bars with unlike letters differ (P ≤ 0.05).

Discussion

The results of this study showed that provision of n-3 FA at a level sufficient to elicit changes in cell membrane composition (Vineyard et al., 2010) did not negatively affect vitamin E or Se status nor increase lipid peroxidation in growing horses when dietary requirements of these nutrients were met. Vitamin E and Se both act as antioxidants within the cell. Vitamin E quenches radicals in the lipid bilayer of the cellular membrane (Bisby and Ahmed, 1989), while Se acts as a constituent of the selenoenzyme, GPx, to catalyze the reaction of hydrogen peroxide to non-toxic water (Klotz et al., 2003). Given the differences in localization and function of vitamin E and Se, they work in concert when faced with an oxidative insult. Long-chain PUFA are thought to be at particular risk for oxidation due to their high number of double bonds (Holman and Elmer, 1947; Cho et al., 1987), which may increase the need for antioxidants such as vitamin E and Se. However, the work on which this theory was based utilized PUFA ester systems to determine rates of oxidation. More recent work has indicated that certain PUFA, including EPA and DHA, may actually be more stable than PUFA containing less double bonds when in the aqueous phase (Miyashita et al., 1993). This would support our finding that serum vitamin E was highest in the horses receiving the fish oil supplement, but not at the expense of an alternate antioxidant, Se. Previous work in horses showed no effect of 12 wk of supplementation with 100 mg n-3/kg BW on plasma vitamin E in mature horses (average age: 12 y; (Hall et al., 2004). Aged dogs similarly showed no effect of 8 wk of n-3 supplementation at up to 8.2 g n-3/kg diet on plasma vitamin E when adjusted for plasma lipids (Wander et al., 1997). Less work exists investigating the effects of n-3 supplementation on Se status, but in broilers, supplementation with 1% n-3 FA for 42 d did not affect serum Se concentrations (Kumar et al., 2015). The relationships between n-3 supplementation, age, and vitamin E and Se status warrant further investigation. However, dietary supplementation of n-3 FA does not appear to negatively impact vitamin E nor Se status in young, non-exercised horses.

Further support that the oxidant/antioxidant balance was not disrupted in horses in the current study was the lack of increase in oxidative stress markers, TBARS, F2-ISO, and total LHP. Work in humans and dogs has shown an increase in TBARS when supplementing with fish oil (Wander et al., 1997; Higdon et al., 2000), suggesting that dietary supplementation with longer chain n-3 FA increased oxidative damage. However, the TBARS assay measures an array of thiobarbituric acid reactive substances, including lipid hydroperoxides and aldehydes. Therefore, it may not be the best biomarker for oxidative stress to lipids due solely to an increase in dietary n-3 FA (Mori, 2004). F2-isoprostanes, on the other hand, are prostaglandin-like metabolites of free radical peroxidation of ARA and are believed to be a more reliable measure of lipid peroxidation and oxidant stress in vivo (Mori, 2004). Interestingly, the fish oil supplement was the only dietary source of ARA in the current study (Table 1) and horses receiving FISH had higher plasma and red blood cell ARA concentrations (Vineyard et al., 2010). Yet horses supplemented with FISH had lower serum F2-ISO than yearlings receiving no n-3 supplementation. This finding supports the contention by Mori (2004) that F2-ISO concentrations are not dependent upon the amount of ARA available in the cell membrane, but are more of a reflection of total lipid peroxidation. In support, plasma F2-ISO concentrations were 15% lower after fish oil supplementation than after sunflower oil (high in oleic acid, C18:l n-9) supplementation and 8% lower than after safflower oil (high in linoleic acid, C18:2 n-6) supplementation in postmenopausal women (Higdon et al., 2000). In the current study, although not statistically different, horses receiving an equivalent amount of n-3 as FLAX had serum F2-ISO concentrations intermediate between CON and FISH. The reduced F2-ISO observed in horses fed FISH, but not FLAX suggest that dietary supplementation of longer-chain highly unsaturated FA, such as EPA and DHA, may attenuate production of reactive oxygen species.

The hypothesis that longer chain PUFA may decrease reactive oxygen species production is further supported by work in mice showing that 6 wk of fish oil supplementation was partially protective against ischemia-induced brain injury (Zhang et al., 2014). In response to oxidative insult, the transcription factor nuclear factor E2-related factor 2 (Nrf2) translocates to the nucleus, where it binds antioxidant response elements to induce antioxidant gene and protein expression, including heme oxygenase 1 (HO-1). The neuroprotective effect in mice was found to be attributable to Nrf2-mediated upregulation of HO-1 expression following ischemia. Further, it was shown that 4-hydroxy-2E-hexenal (4-HHE), a peroxidation product of n-3 FA was a more potent inducer of Nrf2 activation than 4-hydroxy-2E-nonenal (4-HNE), a peroxidation product of n-6 FA (Zhang et al., 2014). While this work focused on neuronal injury, the decrease in systemic F2-ISO following fish oil supplementation of yearlings in the current study may be due to a similar mechanism.

Overall, total LHP was not different between treatment groups. However, horses in the control group showed an increase in 5-HETE over the 70-d trial, while 5-HETE remained constant in horses receiving n-3 FA supplementation. Arachidonic acid and EPA are the primary substrates for a range of pro- and anti-inflammatory eicosanoids, the key mediators of inflammation (Schmitz and Ecker, 2008). The primary ARA-derived eicosanoid is prostaglandin E₂ (PGE₂), but other downstream molecules exist, including 15-, 12-, and 5-HETE (Calder, 2009). While eicosanoids downstream of ARA possess anti-inflammatory properties (e.g., PGE₂ has been shown to inhibit production of tumor necrosis factor α [TNFα] and IL-1; Miles et al., 2002), they are primarily associated with acute pro-inflammatory responses (Tilley et al., 2001; Schmitz and Ecker, 2008). Alternatively, downstream eicosanoids of EPA and DHA include metabolites that are more commonly associated with anti-inflammatory processes (EPA: prostaglandin-3-, leukotriene-5- and thromboxane-3-series; DHA: resolvins, docosatrienes and neuroprotectins). When long-chain PUFA, such as EPA and DHA, are incorporated into the diet at greater rates, ARA is partially replaced by the longer chain PUFA, resulting in a lower production of ARA-derived eicosanoids such as PGE₂ and 5-HETE and increased production of EPA metabolites such as 5-hydroxyeicosapentoic acid (5-HEPE) (Calder, 2008). In human umbilical vascular endothelial cells, 5-HEPE increased Nrf2 nuclear translocation and expression of the gene encoding HO-1, HMOX1 (Nagahora et al., 2017), thereby potentially inducing an antioxidant response. While EPA- and DHA-derived metabolites were not measured in the current study, the elevation in 5-HETE in CON horses suggests an increase in ARA-derived peroxidation products that was not noted in horses receiving n-3 FA supplementation. It is possible FISH and FLAX horses had greater 5-HEPE production, leading to an increase in cellular membrane protection against oxidants.

Although the ratio of GSH to GSSH decreased over time in the current study, presumably due to growth of the horses, n-3 FA supplementation had no effect on GSH, GSSH, the GSH to GSSH ratio, or GPx activity. Coupled with the lack of increase in oxidative stress markers (TBARS, F2-ISO, and LHP), these results confirm that n-3 FA supplementation did not induce an antioxidant response. Previous work in similarly aged horses demonstrated comparable levels of whole blood GSH as those reported here (EL-Deeb and El-Bahr, 2010). Interestingly, older Standardbred horses (aged 4 to 12 yr) had GSH concentrations around 25 mg/dL in RBC lysate but similar GSSH levels (1.5 mg/dL; Kirschvink et al., 2002). Once in training, horses achieved much higher GSH concentrations compared to the current study but similar or lower GSSH concentrations. For example, healthy 2- to 3-yr-old Thoroughbred, Saddlebred, and Standardbred horses in training showed total whole blood glutathione concentrations of 40 to 50 mg/dL but GSSH concentrations around 1 to 2 mg/dL (de Moffarts et al., 2005; Kirschvink et al., 2006). The elevation in GSH concentrations after training is likely an adaptive response to repeated exercise bouts, leading to greater antioxidant capacity. The question then becomes whether dietary antioxidants are needed during exercise training or if the body’s adaptations are sufficient to quench exercise-induced excess radical production. In humans, fish oil supplementation during exercise training has shown mixed results (Philpott et al., 2019). Studies report ameliorated exercise-induced oxidative stress and muscle stiffness (Gray et al., 2014; Tsuchiya et al., 2019), while others report no beneficial effects of n-3 FA beyond those noted with exercise training (Cornish et al., 2018). These studies differed in the supplementation rate as well as the age of the subjects, so much more work is needed in this area. Horses supplemented with n-3 FA without the additional provision of antioxidants were reported to have lower heart rates during exercise and showed improved stride length at the trot (O’Connor et al., 2004; Woodward et al., 2007). Oxidative responses to n-3 FA supplementation in exercising horses may not mirror the responses noted in non-exercised, growing horses in the current study but should be investigated due to the reduction in peroxidation products noted in these yearlings.

In conclusion, the current study provides no evidence that growing horses will endure additional oxidative stress when supplemented with milled flax or fish oil to provide 60 mg n-3/kg BW/d, assuming dietary vitamin E requirements are met. Further, results agree with other research that suggests EPA and DHA may actually reduce oxidative stress. Because the n-3 FA supplements used in this study did not supply appreciable amounts of total fat, nor was the basal diet high in fat, additional research is needed to determine if n-3 FA supplementation alters vitamin E status or lipid peroxidation when added to a high fat diet in young horses. In addition, research is needed to determine if n-3 FA supplementation alters lipid peroxidation during times of oxidative stress, such as in response to training.

Glossary

Abbreviations

ADF

acid detergent fiber

ARA

arachidonic acid

ALA

alpha-linolenic acid

BW

body weight

DE

digestible energy

DHA

docosahexaenoic acid

DM

dry matter

DPA

docosapentaenoic acid

EPA

eicosapentaenoic acid

F2-ISO

F2-isoprostanes

FA

fatty acid

GSH

reduced glutathione

GSSH

oxidized glutathione

GPx

glutathione peroxidase

HETE

hydroxyeicosatetraenoic acid

HEPE

hydroxyeicosapentoic acid

HpETE

hydroperoxyeicosatetraenoic acid

4-HHE

4-hydroxy-2E-hexenal

4-HNE

4-hydroxy-2E-nonenal

HO-1

heme oxygenase 1

HpODE

hydroperoxy-9,11-octadecadienoic acid

LA

linoleic acid

LHP

lipid hydroperoxides

NDF

neutral detergent fiber

Nrf2

nuclear factor E2-related factor 2

PGE₂

prostaglandin E₂

TNFα

tumor necrosis factor α

Conflict of interest statement

The authors declare that they have no conflicts of interest.