Experimental animals were healthy Arbor Acres (AA) male chickens purchased from Anhui Hewei Agricultural Development Co., Ltd. (Anhui, China). The marigold extract was obtained from Guangzhou Leader Biotechnology Co., Ltd. (Guangzhou, China). The extract used in this study was made from lutein ester which was extracted from marigold flowers by saponification and adsorption in carrier of defatted rice bran and silica. The effective content of lutein is 20 g/kg.

A total of 320 one-day-old AA chickens was randomly divided into 5 groups with 8 replicates (cages) of 8 chickens each. The chickens of control group were fed with basal diet and other test groups were fed with basal diet supplemented with 0.075%, 0.15%, 0.30%, and 0.60% marigold extract respectively (the corresponding concentrations of lutein were 15, 30, 60, and 120 mg/kg). The experiment was divided into starter phase (1 to 21 d) and grower phase (22 to 42 d). The composition and nutrient content of the basal diet is given in Table 1.

Chickens were housed in wire cages with 24 h lighting. Temperature and humidity were adjusted according to normal rearing conditions. All chickens were vaccinated by routine procedure and had free access to water and feed. Body weights of chickens and feed consumption were recorded per replicate on the morning of 1 d, 22 d, and 42 d. Then the growth performance expressed as feed intake (FI), body weight gain (BWG) and feed/gain (F/G) was calculated in starter phase (1 to 21 d), grower phase (22 to 42 d) and the whole period (1 to 42 d). The experimental protocol was approved by the Institutional Animal Care and Use Committee of Nanjing Agricultural University.

One chicken with body weight near to average weight was selected from each cage on the morning of 42 d. All of the 40 chickens were killed by exsanguination, and the pH of thigh muscle was measured by HI 9125 portable pH meter (Hanna Instruments Inc., Cluj-Napoca, Romania) after 45 min post-mortem and then the muscle samples were packed into poly-ethylene bags for the subsequent determination of meat quality. Simultaneously, a few thigh muscle and liver tissue samples free of connective and fat tissues were cut into eppendorf tubes, stored at −80°C for the determination of antioxidant capacity.

Two chickens per replicate were selected randomly every seven days from 3 to 6 weeks and the yellow scores of shank and beak were evaluated using Kemin color fan (Kemin Industries [Asia] Pte Limited; color score from 3 to 14). On the morning of 42 d, the yellowness (b*) value of skin was measured before slaughter by the Commission Internationale de L’Eclairage (CIE) lab system using a colorimeter (CR400, Konica Minolta Sensing, Inc., Osaka, Japan; calibrated with a white plate). According to the method of Castaneda et al [4], the measurements were taken in the lateral apterylum region which is free of any major blood vessels or feathers. The meat color (L*, lightness; a*, redness; b*, yellowness) of thigh muscle was also assessed from a cut surface by CR-400 Chroma Meter at 24 h post-mortem, as described by Li et al [10]. Each muscle was measured 3 times at various points and the mean was calculated.

Firstly, thigh muscle and liver tissues were homogenized with a PRO 200 homogenizer (Pro Science Co., Ltd., Monroe, CT, USA) under low temperature conditions. Then, the homogenates were centrifuged to obtain supernatants and stored at −20°C for subsequent analysis. The total antioxidant capacity (T-AOC), superoxide dismutase (SOD), glutathione peroxidase (GSH-PX) and malondialdehyde (MDA) contents were measured using the commercial assay kits produced by Nanjing Jiancheng Bioengineering Institute (Nanjing, China).

The pH values of thigh muscles were measured at 45 min and 24 h after slaughter using HI 9125 portable pH meter (Hanna Instruments Inc., Romania) and expressed as pH_{45 min} and pH_{24 h} respectively. The pH probe was inserted into the crevice made by surgical knife at a 45-degree angle 3 times at various points on each sample and a mean value was calculated.

Drip loss of thigh muscles was measured in accordance with the method of Gao et al [11] with some modifications. About 30 g of muscle sample (2 cm×3 cm×5 cm) was cut from the same location of thigh muscle and then weighed (m1). The samples were hung at 4°C in sealed aerated polyethylene bags ensuring that meat was not in contact with the bag and the direction of muscle fibers was perpendicular to ground. The surface moisture of samples was dried with filter paper and then the weight (m2) was measured again 24 h later. The following formula was used to calculate the drip loss: Drip loss rate (%) = (m1–m2)/m1×100.

Shear force of thigh muscles was measured by C-LM3 digital meat tenderness meter (Engineering College of Northeast Agricultural University, Haerbin, China). Cooked meat was cut parallel to the direction of muscle fibers into 2 shaped strips (1 cm×1 cm×3 cm). Each strip was sheared two times perpendicular to direction of muscle fibers by tenderness meter and the force was expressed in Newtons (N). Each sample was measured 4 times at various points and the average value was calculated.

All data were analyzed by IBM SPSS Statistics 20 software and expressed as means and standard error of the mean (SEM). Comparisons of means were performed by Duncan’s comparison test of one-way analysis of variance. The level of significance was evaluated at p<0.05. The effect of supplementation levels of marigold extract was determined using orthogonal polynomial contrast test for linear and quadratic effects.

Compared with the control, there was no significant difference in FI, BWG, and F/G of all test groups in all phases (Table 2).

The yellow scores of shank and beak in all the groups increased gradually with the advancement of age from 3 week to 6 week (Figures 1 and 2). There was a high positive correlation between color scores of shank and additive amounts of marigold extract. The shank color scores of 0.60% and 0.30% supplementation groups had no significant differences at the last week. Beak color score was the largest in 0.60% marigold extract supplementation followed by 0.30%, 0.15%, 0.075%, and the control.

No significant difference in L* value of thigh muscles was observed between all four treatment groups and the control (Table 3). Compared with the control, marigold extract supplementation increased a* value of thigh muscle, especially in the 0.60% group (linear, p<0.01). Furthermore, marigold extract significantly increased b* value of thigh muscle and skin compared with the control (linear, p<0.01), except for the 0.075% group. The response of a* value of thigh muscle and b* value of thigh muscle and skin to supplemental marigold extract all were linear (p<0.01), and the a* value and b* value were maximized by adding 0.60% marigold extract.

In comparison to the control group, supplementation with marigold extract significantly linearly increased T-AOC and activities of GSH-PX of liver in all four treatment groups (Table 4) (p<0.01). Meanwhile, the SOD activity significantly increased in all other treatment groups (linear, p<0.01), except for 0.075% group. Dietary marigold extract decreased the concentrations of MDA, especially in 0.60% group (linear, p<0.01).

In comparison to the control, supplementation with marigold extract increased T-AOC in thigh muscle, especially in 0.30%, 0.60% groups (linear, p<0.01), and significantly increased activities of SOD in all four treatment groups (Table 4) (linear, p<0.01). There were no significant differences in activities of GSH-PX. The concentrations of MDA significantly decreased in all the treatment groups (linear, p<0.01).

Addition of marigold extract had no significant effect on pH_{45 min} or on pH_{24 h} in the thigh muscle (Table 5). Compared with the control group, marigold extract supplementation significantly decreased (p<0.01) linearly the drip loss of thigh muscle in all the treatment groups. On the other hand, supplementation with marigold extract significantly decreased the shear force of thigh muscle in all the test groups (linear, p<0.01) except for 0.075% group.

In terms of food, color, classified in freshness, is one of the crucial criteria determining consumers’ choice [12]. In addition, meat color, which is the external manifestation of a variety of physiological and biochemical changes in muscle, is the most intuitive index of meat quality. Moreover, shank and beak are typically used to evaluate pigmentation of chickens in production. The results of this experiment showed that the addition of marigold extract significantly increased the yellow scores of shank and beak. This implies that the experimental groups deposited more yellow xanthophyll pigments such as lutein. For shank, the addition of 0.30% showed the same result as 0.60% group in the end, which means the yellowness of shank can reach the maximum when the marigold extract content was at 0.30%. This also indicates that different tissues have different deposition efficiency of marigold extract pigments. In the present study, we also found that the addition of marigold extract significantly increased the b* value of the thigh muscle and skin. This might be due to deposition of marigold extract pigments in the muscle and skin, and many previous studies have obtained similar results [13–15]. Castaneda et al [4] also indicated that natural pigments were generally more effective than synthetic ones in improving the yellowness of animal skin. The marigold extract increased the a* value of thigh muscle in this study. Vitamin E could directly prevent lipid oxidation and then delay myoglobin oxidation [16]. The increase of lipid peroxidation products will accelerate the oxidation of bright red oxygenated myoglobin to brown metmyoglobin, and then the color of meat will become dark brown [17]. Furthermore, lutein has stronger antioxidant capacity than vitamin E [18]. So, the increased a* value might be due to the decreased muscle peroxidation. On the other hand, Rajput et al [19] concluded that the increase of a* value may be linked to higher levels of hemoglobin in the tissue.

Antioxidant capacity is an important factor affecting the health of the animal, and the antioxidant capacity of muscle after slaughter is one of the most critical factors affecting meat quality. Lipid peroxidation caused by high levels of free radicals not only can induce oxidative stress and increase the content of MDA which is the end product of lipid oxidation [20], but also causes the deterioration of meat. In addition, the oxidation of lipid is closely related to meat color and the accumulation of antioxidant substances can increase the color stability of meat [21]. So, improving the antioxidant properties of muscle is of great significance to improve meat quality and increase meat storage time. Studies have indicated that the addition of antioxidants in the diet can alleviate the adverse effects of oxidative stress on broiler production and improve the meat quality of broilers [22]. T-AOC is a reflection of the body or tissue antioxidant properties. SOD can remove superoxide anion free radicals and GSH-PX can catalyze the decomposition of hydrogen peroxide.

The groups with dietary marigold extract supplementation had higher antioxidant capacity of livers and thigh muscles in all four antioxidant indices, which were consistent with the findings of Shanmugasundaram and Selvaraj [23] and Rajput et al [19]. Lutein and zeaxanthin have a number of unsaturated bonds, so they can quench reactive oxygen species (ROS) [24], thereby inhibiting the activity of oxygen free radicals and preventing reactive oxygen free radicals from destroying normal cells. Chew and Park [25] also reported that lutein has a unique effect in preventing damage of the biofilm from free radical quenching singlet oxygen and free radical capture. Natural carotenoids can mitigate lipid oxidation of breast muscle [19]. Dietary supplementation of lutein prevented an increase of LPS-induced thiobarbituric acid reactive substances [23]. This is consistent with the decreased MDA observed in the current trial. Thus, lutein with the capacity of reducing oxidative stress in birds is an excellent antioxidant [26].

Meat quality is usually measured by pH, meat color, water holding capacity and tenderness. Water holding capacity refers to the ability of muscle tissue to maintain moisture and directly affects the eating quality of meat, such as taste, tenderness, nutrition level and so on [27]. Consumers consider tenderness as one of the most important eating qualities of meat. Muscle tenderness is affected by many factors with the content and chemical structure state of connective tissue and the myofibrils and sarcoplasmic reticulum being the material basis of tenderness. This study suggested that supplementation of marigold extract reduced drip loss as well as shear force, improved the water holding capacity and the tenderness of thigh muscle. No significant effect was observed on pH. Rajput et al [19] also found that natural carotenoid improved the meat’s water holding capacity of broilers affected by a coccidiosis challenge. The enhancement of water holding capacity might have been caused by the improvement of antioxidant in muscle. Study has also indicated that carotenoids are regarded to improve muscle water holding capacity [19]. Proteolysis and even protein oxidation directly affects the water holding capacity in muscle tissue [28]. Otherwise, the calpain activity which plays an important role in water holding capacity can be modulated by redox state [29]. The improvement of tenderness also might have been caused by the increase in antioxidant ability, or affected by water holding capacity of muscle, which can directly affect the tenderness of meat [27]. Rowe et al [30] had reported that muscle oxidation can reversibly inhibit the activities of muscle protein degradation enzymes and other protein hydrolases after postmortem, and then reduce muscle tenderness.

Meat quality is related to oxidant property and the enhancement of its antioxidant property ameliorates meat quality through different ways [31]. The rate and degree of lipid oxidation in postmortem muscle depends on the antioxidant capacity of the body, while SOD can block lipid oxidation, improve the meat quality and prolong the shelf life of meat. Besides, it has been reported that ROS could reduce collagen synthesis and increase matrix metalloproteinase-2, which has a negative effect on meat quality, and ROS could be eliminated by lutein [32]. The decline of MDA content in muscle is an indirect effect of the antioxidant capacity, which might be another reason [31].