Growth performance, oxidative stress and immune status of newly weaned pigs fed peroxidized lipids with or without supplemental vitamin E or polyphenols

Background This study evaluated the use of dietary vitamin E and polyphenols on growth, immune and oxidative status of weaned pigs fed peroxidized lipids. A total of 192 piglets (21 days of age and body weight of 6.62 ± 1.04 kg) were assigned within sex and weight blocks to a 2 × 3 factorial arrangement using 48 pens with 4 pigs per pen. Dietary treatments consisted of lipid peroxidation (6% edible soybean oil or 6% peroxidized soybean oil), and antioxidant supplementation (control diet containing 33 IU/kg DL-α-tocopheryl-acetate; control with 200 IU/kg additional dl-α-tocopheryl-acetate; or control with 400 mg/kg polyphenols). Pigs were fed in 2 phases for 14 and 21 days, respectively. Results Peroxidation of oil for 12 days at 80 °C with exposure to 50 L/min of air substantially increased peroxide values, anisidine value, hexanal, and 2,4-decadienal concentrations. Feeding peroxidized lipids decreased (P < 0.001) body weight (23.16 vs. 18.74 kg), daily gain (473 vs. 346 g/d), daily feed intake (658 vs. 535 g/d) and gain:feed ratio (719 vs. 647 g/kg). Lipid peroxidation decreased serum vitamin E (P < 0.001) and this decrease was larger on day 35 (1.82 vs. 0.81 mg/kg) than day 14 (1.95 vs. 1.38 mg/kg). Supplemental vitamin E, but not polyphenols, increased (P ≤ 0.002) serum vitamin E by 84% and 22% for control and peroxidized diets, respectively (interaction, P = 0.001). Serum malondialdehyde decreased (P < 0.001) with peroxidation on day 14, but not day 35 and protein carbonyl increased (P < 0.001) with peroxidation on day 35, but not day 14. Serum 8-hydroxydeoxyguanosine was not affected (P > 0.05). Total antioxidant capacity decreased with peroxidation (P < 0.001) and increased with vitamin E (P = 0.065) and polyphenols (P = 0.046) for the control oil diet only. Serum cytokine concentrations increased with feeding peroxidized lipids on day 35, but were not affected by antioxidant supplementation (P > 0.05). Conclusion Feeding peroxidized lipids negatively impacted growth performance and antioxidant capacity of nursery pigs. Supplementation of vitamin E and polyphenols improved total antioxidant capacity, especially in pigs fed control diets, but did not restore growth performance.


Background
Swine diets often include fats or oils to increase the energy density of the diet. Fats and oils are very dense energy sources with a low cost per unit of energy [1]. However, supplemental lipids are prone to peroxidation due to oxygen attack on unsaturated fatty acids when exposed to high temperatures [2,3]. Unsaturated lipids are more susceptible to peroxidation than saturated fatty acids [2]. Thus, vegetable oils are more prone to peroxidation than more saturated fats, such as choice white grease and tallow [3]. High amounts of vegetable oils are used in the preparation of foods for human consumption. This cooking process at high temperatures produces high amounts of peroxidized oils [4]. Commonly, spent cooking oils are recycled into lipid sources for use in animal feed [5].
Natural antioxidants in the body play an important role to delay or prevent oxidation and blocking and capturing formed radicals in cells [22]. Vitamin E is a fatsoluble vitamin and natural antioxidant [23,24]. Likewise, plant-based polyphenols are antioxidants with high potential against oxidation [25]. Several studies reported positive effects of vitamin E or plant polyphenols on meat oxidation in broilers [26,27], meat quality in finishing pigs [9], and antioxidant activity in piglets [28,29]. Moreover, the supplementation of vitamin E plus synthetic antioxidant blends or synthetic antioxidants to diets containing peroxidized vegetable oil for weaned piglets and finishing pigs improved growth and reduced lipid and protein oxidation [9,30].
We hypothesized that supplemental antioxidants could alleviate the negative effects of peroxidized lipids by improving the antioxidant status of pigs and reducing oxidative stress. The objective of the current study was to investigate the potential antioxidant effects of vitamin E or plant polyphenols when supplemented to diets containing peroxidized oil on growth performance, immune and oxidative status, and antioxidant function of weaned pigs.

Animals, treatments and experimental design
This study was conducted at the Swine Educational Unit (Raleigh, NC). One-hundred and ninety-two 3-week old crossbred pigs (96 gilts and 96 barrows) with an initial body weight of 6.62 ± 1.04 kg were used. At weaning, piglets were blocked by initial body weight and sex and randomly assigned within blocks to a 2 × 3 factorial arrangement using an experimental allotment program [31]. Factors consisted of: 1) dietary inclusion of 2 types of oils [6% refined, bleached, and deodorized soybean oil (control oil) or 6% of the same soybean oil that had been peroxidized (peroxidized oil)] and 2) inclusion of 3 antioxidant treatments consisting of a control diet (33 IU/kg of DL-α-tocopheryl acetate; CON), the control diet with an additional 200 IU/kg of DL-α-tocopheryl acetate (Rovimix®, DSM Nutritional Products, Parsippany, NJ; VITE) or the control + 400 mg/kg of a blend of plantbased polyphenols (Promote® AOX™ 50, Cargill, Inc., Wayzata, MN; POL), based on the manufacturers' recommendation. The vitamin E concentration in the control diet was representative of current industry practices [32] and supplementation with 200 IU/kg was based on previous research [33].
The peroxidized oil was created using 163 kg of refined, bleached, and deodorized soybean oil (Kirkland, Washington, WA), equally divided into two metal barrels. A heater was placed into each barrel and was set to maintain a constant temperature of 80°C for 12 days. In addition, a polyvinyl chloride pipe with 1 mm holes was submerged in each barrel to continuously bubble air through the oil at a constant flow rate of 50 L/min for 12 days. The dietary treatment feeds were manufactured at the North Carolina State University Feed Mill Educational Unit (Raleigh, NC). Diets were based primarily on corn-soybean meal and were formulated to meet or exceed all nutrient requirements for piglets as suggested by the NRC (2012) [1]. At the time of mixing, 0.1% of liquid antioxidant (Rendox®, Kemin Industries, Inc., Des Moines, IA) containing tertiary butyl hydroquinone (TBHQ) was added to the control oil and the peroxidized oil prior to mixing of feed to avoid further peroxidation. A basal mix containing all ingredients, except the oil sources and supplements was created and equally divided into six batches. Final treatment diets were then created by mixing control oil or peroxidized oil with the basal mix and within these, no supplement, vitamin E, or polyphenols were added at the appropriate concentrations. Diets were fed in 2 phases throughout the 5-week nursery period, with phase 1 being fed immediately after weaning for 14 days and phase 2 being fed the next 21 days (Table 1). All piglets were provided ad libitum access to feed and drinking water.
Pigs and treatments were randomly allotted within body weight and sex blocks into 2 identical temperaturecontrolled rooms (Aerotech®, Aero Speed 1.2, Pittsburgh, PA) containing 24 pens each (48 pens total with 8 pen replicates per treatment). Pigs were housed 4 pigs per pen (2 gilts and 2 barrows). The size of each pen was 0.91 m × 1.52 m and each pen contained a stainless-steel single-sided feeder with 2 feeding spaces (Staco, Inc., Schaeffer town, PA) and 2 stainless steel water nipples for weaned piglets (Hog Slat, Newton Grove, NC). Environmental temperatures, ventilation rate, and pigs were checked every morning. Temperatures were set at 32°C for the first week and then reduced by 2°C throughout each week until the temperature reached 24°C.

Sampling and measurements
Individual pig body weight (BW) was measured on days 0, 7, 14, 21, 28 and 35 to calculate average daily gain (ADG) for each pen. Average daily feed intake (ADFI) was measured from the difference between the sum of feed additions and feed remaining at the end of the week or phase and divided by 7 days or days in the phase. Feed efficiency (G:F) was calculated by dividing ADG by ADFI weekly and for each phase.
Blood samples from one randomly selected pig per pen were collected by venipuncture using 20-gauge × 3.8 cm multiple use drawing needles (Vacuette®, Greiner bio-one, Kremsmunster, Austria) on day 14 and day 35 (at approximately 09:00). Blood was collected into 10 mL-vacuum tubes for serum (BD Vacutainer®, Franklin Lakes, NJ). Blood was centrifuged at 4,267×g for 10 min at 4°C using a refrigerated centrifuge (Centra GP8R, Thermo IEC, MA) and serum was collected. Serum was aliquoted into 6 tubes of 2 mL capacity (Biotix®, Neptune, 3472.X, Mesa Rim, San Diego, CA) and stored at − 80°C for further analysis.

Oil analysis
Representative oil samples from control and peroxidized oil were collected and analyzed by New Jersey Feed Laboratory Inc. (Trenton, NJ) for moisture, insoluble impurities, unsaponifiable matter, free fatty acids, peroxide value (initial, and at 4 and 20 h using the active oxygen method (AOM)), and anisidine value using AOAC [34] and AOCS [35] procedures and by Kemin Industries, Inc. (Des Moines, IA) for oxidative stability index, hexanal, 2,4-decadienal and tertiary butyl hydroquinone.

Statistical analysis
The data for growth performance and serum measurements were analyzed using the Proc MIXED procedure of SAS (v.9.4, SAS Institute. Inc., Cary, NC). For growth performance measurements, pen was used as the experimental unit and weight block was the random effect. The statistical model included block, type of oil, supplementation and the interaction between oil type and supplementation. For serum measurements, individual pig was considered the experimental unit and the data were analyzed using the Proc MIXED Procedure. The statistical model included block, type of oil, supplementation, day of collection (as repeated measure) and the appropriate interactions. Preplanned comparisons of individual treatments were conducted using the least significant difference method following a significant F-test. Statistical significances were considered at P < 0.05 and tendencies at 0.05 ≤ P ≤ 0.10.

Chemical analysis of peroxidized soybean oil
Results of chemical analysis of control and peroxidized soybean oil showed marked differences in peroxidation measures (

Growth performance
Two pigs, one from a pen fed the peroxidized oil treatment plus vitamin E, and one from a pen fed the control oil diet plus polyphenols, were removed from the study due to failure to thrive and losses in body weight (BW).

Serum oxidative status markers
Lipid peroxidation, supplementation and day of sampling did not significantly affect serum 8-OHdG (P ≥ 0.05; Table 4). Serum protein carbonyl was increased (P < 0.001) on day 35 when peroxidized oil was fed (3.35 vs. 2.41 pmol/mg protein for peroxidized and control oil, respectively), but peroxidation of oil did not affect protein carbonyl when measured on day 14 (2.08 vs. 1.98 pmol/mg protein, for peroxidized and control oil, respectively). Feeding peroxidized oil decreased serum MDA concentration on day 14 (8.43 vs. 12.7 8 μmol/L for peroxidized and control oil, respectively), but did not impact MDA concentration when measured on day 35 (12.5 vs. 12.0 μmol/L) (interaction; P < 0.001).

Serum vitamin E concentrations and total antioxidant capacity
Lipid peroxidation decreased serum vitamin E concentrations by 41.8% (P < 0.001; Fig. 1) and this reduction in vitamin E concentration was greater on day 35 compared to day 14 (day ×peroxidation interaction, P < 0.001). Supplementation of vitamin E increased serum vitamin E concentrations (P < 0.001) compared to control and polyphenols treatments (1.98 vs. 1.25 and 1.26 mg/kg, respectively) and this increase was greater for the control diet (84.6%) compared to the diets with peroxidized oil (22.3%) (interaction, P < 0.001). The threefactor interaction between lipid peroxidation, supplementation and day of measurement was not significant (P = 0.101). Serum TAC was decreased by 10.6% (P < 0.001) by lipid peroxidation compared to control oil (Table 4). Additionally, serum TAC was greater when measured on day 35 compared to day 14 (+ 18%; P < 0.001). Supplementation of vitamin E (P = 0.046) and polyphenols (P = 0.065) to diets containing control soybean oil increased TAC, but this was not the case when supplemented to diets containing peroxidized oil (interaction, P = 0.045). In addition, TAC was increased by polyphenol supplementation compared to the control diet (P = 0.016) when measured on day 35, but not day 14 (interaction, P = 0.025).

Discussion
Feeding peroxidized oil to pigs negatively affects growth performance [12,36,37] and this decrease appears to be directly related to the extent of peroxidation. In the present study, growth performance was compromised to the largest extent from day 14 through day 35 (Phase 2), but effects were limited for the initial 14 days (Phase 1). The first phase diet was complex in composition, containing whey permeate as a source of lactose, spraydried plasma protein, and other highly digestible animalbased proteins ingredients. These ingredients are highly palatable, which is especially important for the first 2 weeks after weaning [38,39]. Perhaps these ingredients allowed pigs to maintain feed intake in the presence of negative odor and taste associated with aldehydes in peroxidized oil that are responsible for rancid odor and off flavor [40]. On the other hand, there may be a threshold beyond which the impact of peroxidized lipids manifests itself. The fact that detrimental effects of dietary peroxidized soybean oil was expressed more in the long term is supported by Anjum et al. [14] and Lu et al. [30] in broilers and pigs, respectively. DeRouchey et al. [36], Liu et al. [11] and Rosero et al. [12] reported reductions in ADFI in pigs when feeding peroxidized lipids and these effects were dose-dependent. The reduction in growth rate is due, in part, to reduced feed intake, but is also likely associated with the negative impacts in the intestine caused by dietary peroxidized oil when fed to piglets [12]. In addition, peroxidized lipids have been shown to decrease digestibility of fat and gross energy in pigs [13], which can negatively impact daily gain and feed efficiency. The degree of lipid peroxidation in the present study was consistent with previous reports and is of practical relevance to the animal production industry [41].
The addition of peroxidized oil in the diet for pigs causes oxidative stress [9,10], but the supplementation of dietary vitamin E [42] and polyphenols [43] could ameliorate the negative effects caused by oxidation [9,37]. In the present study, supplementation of vitamin E was not effective in improving pig performance, regardless of the peroxidation status of the soybean oil fed. Feeding polyphenols tended to improve gain:feed in pigs fed peroxidized lipids during the first week of the study only and tended to improve gain:feed for the overall study, regardless of the oil source. These results agree partially with those published by Boler et al. [9], who found no effects of supplementation with tertbutylhydroquinone and ethoxyquin blend antioxidant to a diet containing peroxidized corn oil in finishing pigs. Rooke et al. [44] reported that an antioxidant blend or antioxidant blend plus vitamin E improved growth performance when feeding 5% of dietary peroxidized oil and showed regeneration of vitamin E by vitamin C and glutathione peroxidase. Peroxidation products could reduce the function of fat-soluble vitamins [30].
Several authors reported that peroxidized oil reduced serum concentrations of vitamin E due to catabolism of the vitamin E during oxidative stress [8,9,17,37,45]. This fact agrees with the results found in the present study. Serum vitamin E was greatly reduced with peroxidation, especially after prolonged exposure, showing the ability of vitamin E to react against oxidation. In addition, the reduction in feed intake in pigs fed peroxidized oil (and thus reduced intake of vitamin E) contributes to decreased serum vitamin E concentrations. Soybean oil contains significant amounts of tocopherols, which are destroyed by peroxidation. Lindblom and coworkers [13] reported a 48% reduction of total tocopherols in thermally peroxidized soybean oil, which will  smaller in the latter. On the other hand, supplementation of polyphenols did not impact serum vitamin E concentration, regardless of the peroxidation status of the oil fed, suggesting that polyphenols were not able to regenerate vitamin E once it was oxidized. This could be related to low absorption rate of dietary polyphenols as reported by Gessner et al. [46], the type of polyphenols used, or functioning of polyphenols as an antioxidant that is independent from vitamin E. Total antioxidant activity broadly measures antioxidant actions in the body including hydrophilic and lipophilic antioxidants and can be highly dependent on uric acid concentrations. On the other hand, TAC does not measure antioxidant enzymes, such as superoxide dismutase, glutathione peroxidase, and catalases [47], which are important components of the antioxidant system. Vitamin E is positively correlated with TAC and partially contributed to TAC results in the present study. Indeed, TAC was reduced by peroxidation, similar to serum vitamin E concentrations. Values of TAC were higher on day 35 compared to day 14, indicating increased antioxidant capacity as pigs grew and reflecting increased feed consumption. In addition, serum TAC was increased when vitamin E and polyphenols were supplemented to pigs fed control oil demonstrating effectiveness of both in improving antioxidant status. However, serum TAC was not impacted when vitamin E or polyphenols were supplemented to pigs fed peroxidized oil, suggesting that antioxidants effectively reacted against oxidation. Interestingly, on day 35, serum concentrations of TAC were higher for the polyphenols treatment than the control or dietary vitamin E treatment, suggesting that antioxidant efficiency of polyphenols increased with time. Contrarily, Lipiński et al. [48] concluded that polyphenols are deposited in the organs which participate in their metabolism, with plasma concentrations remaining lower during long periods or high doses in pigs. Moreover, no improvements in TAC were reported by Gessner et al. [46], who used 1% of dietary polyphenols in diets of piglets. Also, during metabolism of polyphenols (flavonoids) uric acid is produced as part of TAC [47]. Thus, high levels of TAC resulting from polyphenols treatments may be due to metabolic products derived from the commercial polyphenols used in this study.
Peroxidized lipids damage proteins, leading to malfunction and alteration in their structure, increasing protein carbonyls. Protein carbonyl is a biomarker with high stability, and it is formed during the early period of oxidative stress [49]. In the present study, protein carbonyl in serum was increased by lipid peroxidation, particularly on day 35, but supplementation with antioxidants did not ameliorate this effect. Likewise, Lindblom et al. [21] and Lu et al. [30] reported increased protein carbonyl in serum of pigs fed peroxidized oil. Boler et al. [9] did not observe differences in protein carbonyls in plasma when using a synthetic antioxidant blend supplemented to diets containing peroxidized corn oil, when compared with a peroxidized corn oil diet without antioxidant in pigs. A study conducted by Lu et al. [30] showed significant decreases in plasma carbonyls on day 55 in growing pigs fed peroxidized lipids when using dietary vitamin E in combination with a synthetic antioxidant blend, but this was not the case on day 118, indicating that long-term supplementation of peroxidized oil may overwhelm the antioxidant system resulting in increased markers of oxidative stress.
Malondialdehyde is a three-carbon cytotoxic molecule, produced by peroxidation of polyunsaturated fatty acids [49]. The serum concentrations of MDA, as a marker of lipid peroxidation status, have been shown to increase due to the consumption of peroxidized oil as reported in previous studies [9,14,16,30]. In the present study, serum MDA concentrations decreased with feeding with peroxidized oil when measured on day 14 and 35, and MDA levels were higher on day 35 than day 14. In contrast, Chang et al. [17] did not observe differences in serum MDA due to lipid peroxidation in weaned piglets. Hung et al. [37] conducted a meta-analysis using 23 studies conducted in swine and reported that serum MDA concentrations generally increased with the supplementation of peroxidized lipids. However, only few of those studies reported serum MDA concentrations as an outcome variable in that meta-analysis.
8-OHdG is an oxidized nucleoside of DNA detected during DNA oxidation [49]. Comparing with other oxidative status markers tested in this study, serum 8-OHdG was not affected by lipid peroxidation or antioxidant supplementation. Likewise, Lindblom et al. [21] and Chang et al. [17] did not report differences in DNA damage using various levels of peroxidation of supplemental oil. The site of DNA damage occurs in nuclear and mitochondrial DNA in tissue and in DNA of lymphocytes [50,51]. Lymphokines which are derived from lymphocyte stimulation and synthesized by lymphocytes clones, include IL-1, IL2, IL-3, IL-4, IL-5, IL-6, TNF-α and IFN-γ [52]. In addition, peroxidation caused lipid oxidation and inflammation in intestinal cells due to activation of nuclear factor kappa B, producing an increase in cytokines levels [53]. In the present study, cytokines increased with lipid peroxidation only on day 35 and could be related to the reduction in growth performance observed during the end of the study. This suggests that peroxidation over a prolonged period may have induced inflammation in pigs, with a subsequent negative impact on growth performance. It should be noted that serum measurements are single point measurements in time, whereas growth performance is cumulative and provides an assessment of total impacts over time. Similar to our results, Rosero et al. [12] did not find differences in the pro-inflammatory cytokine TNF-α when peroxidized oils were added in the diet for piglets. The high levels of the majority of cytokines on day 14 compared to day 35 may be due to the early weaning effects on pigs [54]. Interestingly, cytokines increased with peroxidation on day 35 and could be related to the reduction in growth performance at the end of the study.

Conclusion
The addition of peroxidized soybean oil negatively affected growth performance, especially during the later stages of the study. Moreover, feeding peroxidized lipids reduced serum vitamin E concentrations and total antioxidant capacity, increased protein oxidation, and increased serum cytokine concentrations after 35 days of exposure. Supplementation of vitamin E increased serum concentrations of vitamin E and both vitamin E and polyphenols improved total antioxidant capacity, especially in pigs fed control diets. Nonetheless, no significant improvements in growth performance or oxidative status were detected at the concentrations of vitamin E and polyphenols supplemented.