Effect of genotype on performance, vitamin and carotenoid deposition, oxidative stability, fatty acid profile and sensory characteristics in cockerels housed on litter and in mobile boxes on pasture

Abstract

The study evaluated the effect of housing system and genotype on performance and breast meat quality in 540 cockerels. Three genotypes of cockerels differing in growth intensity (slow-growing ISA Dual, medium-growing Hubbard JA757 and fast-growing Ross 308) and two housing systems (litter and mobile box on pasture) were compared. The significantly lowest feed conversion (p < .001) was recorded for Ross 308 cockerels housed on litter, while ISA Dual cockerels from mobile boxes had a 74% higher conversion ratio. The highest concentrations of lutein (p = .043) and α-tocopherol (p = .024) in breast meat were found in the cockerels from mobile boxes of genotypes ISA Dual and Ross 308, respectively. Meat stored for five days showed the highest oxidative stability of fat (p = .001) in slow- and medium-growing cockerels housed on pasture. The ISA Dual genotype housed both on litter and on pasture and the Hubbard JA757 genotype from mobile boxes had the highest proportion of polyunsaturated fatty acids in the meat (p < .001). The lowest n6/n3 fatty acid ratio (p = .045) and thrombogenic index (p < .001) and the highest hypocholesterolaemic/hypercholesterolaemic index (p < .001) were recorded in slow-growing chickens with access to pasture. The meat of cockerels fattened on litter was considered more fragrant (p = .013), more tender (p < .001), juicier (p < .001) and overall more acceptable (p = .001). It can be concluded that the ISA Dual genotype showed the highest willingness to graze compared to Hubbard JA757 and Ross 308 cockerels and made the best use of the benefits of pasture fattening, which was reflected in the high quality of the meat. The meat of conventionally housed cockerels was sensorially evaluated as better.

Highlights

• The ISA Dual genotype showed the highest willingness to graze, which was reflected in the higher content of health-promoting substances in meat.

• The more frequent use of slow-growing cockerels for the purpose of fattening will be hindered by lower performance of the cockerels and low level of meat tenderness.

Keywords:

• Housing system
• growth intensity
• chickens
• meat quality
• pasture herbage

Introduction

In the second half of the twentieth century, the intention was to breed chickens primarily for intensive growth, while the beginning of the twenty-first century saw a growing interest in the production of slower growing meat-type chickens or cockerels of dual-purpose chicken breeds. This change is related to higher interest of people in the welfare of animals and higher requirements for the maturity of meat on the part of consumers. In addition, the big ethical issue is the culling of day-old male layer chicks (Krautwald-Junghanns et al. 2018; Gremmen 2020; de Haas et al. 2021; Popova et al. 2022). Germany was the first country to ban the culling of male day-old chicks from laying lines, followed by France and other countries are considering the ban. One of the many solutions is the production of dual-purpose breeds, because the use of cockerels of these breeds is slightly more suited for meat production as the conventional layer genotypes due to higher growth intensity and the meat from these breeds is more comparable to broiler meat in taste and texture (Mueller et al. 2018).

Regarding the influence of genotype on some meat quality indicators, the results are contradictory. According to the studies of Sirri et al. (2011) and Valenta et al. (2022), the functional meat characteristics of fast-growing and medium-growing hybrids seem to be much more attractive for both industry and consumers. This meat is characterised by a lower cooking loss and a higher pH and tenderness. On the other hand, Fanatico et al. (2009) stated that breast meat from slow-growing birds was more tender than that from fast-growing birds. In addition, Devatkal et al. (2019) did not detect any differences in all the sensory attributes of meat and meat products from slow-growing broilers and commercial broilers.

From a nutritional point of view, meat from slow-growing chickens appears to be healthier. The meat of these chickens contains less fat (Chodová et al. 2021) but a higher proportion of n-3 polyunsaturated fatty acids (Sirri et al. 2011). This is a result of the higher efficiency of eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA) deposition and the higher willingness and ability to graze (pasture is a source of n-3 fatty acids) in slow-growing genotypes compared to fast-growing hybrids (Dal Bosco et al. 2012).

A factor that directly affects the welfare of the birds and can impact their behaviour and certain meat quality traits is the housing system (El-Deek and El-Sabrout 2019). The majority of poultry meat production is currently undertaken by fattening chickens in halls on litter. This system is especially suitable for fast-growing genotypes of chickens that achieve high productivity therein. An alternative housing system is free-range or mobile boxes on pasture. Free-range access is often considered beneficial for animal welfare because it allows the animals to perform natural behaviours and gives them additional space for activities such as dust- and sun-bathing (Gonçalves et al. 2017) or grazing, which can subsequently positively affect the quality of the meat. The rearing system may modify the health-promoting properties of meat (Michalczuk et al. 2016). Meat from pasture-raised chickens has a higher content of antioxidants, such as vitamins and carotenoids, n-3 fatty acids and minerals (Sossidou et al. 2015; Englmaierová et al. 2021). Although polyunsaturated n-3 fatty acids are more susceptible to oxidation than saturated fatty acids, the higher vitamin E content in meat of chickens fattened on pasture may limit the oxidation of these health-promoting fatty acids and preserve the meat sensory properties. Castellini et al. (2002) found that access of chickens to a grass paddock resulted in better sensory attributes than conventional production in terms of overall acceptability and juiciness. In contrast, increased activity of chickens in free-range system could result in relatively stronger muscle fibres, which would affect meat tenderness (Fanatico et al. 2005). Additionally, Grashorn and Serini (2006) stated that organic breast and thigh meat were juicier and less tender than conventional chicken meat but exhibited a superior flavour. The disadvantage of the free-range system is lower weight gain and poorer feed conversion efficiency compared with intensive housing systems (Bogosavljević-Bošković et al. 2012).

In relation to the growing trend of using slow-growing genotypes or dual chickens, due to the ethics of animal husbandry, welfare and higher maturity of meat, for the purpose of meat production, it is necessary to determine the combination effect of the genotype and housing system on quantity and quality of production. The effects of both factors have already been investigated separately, and therefore, differences in performance and meat quality due to different growth rates of individual genotypes and a positive effect of grazing on meat quality can be expected, but it is not well known how individual genotypes will use the nutrients obtained from grazing to increase performance and meat quality compared to fattening on litter. Especially in the case of ISA Dual cockerels, as representatives of the dual-purpose genotype whose fattening is an alternative to their culling immediately after hatching, findings regarding the effect of pasture fattening on meat quality have not yet been published. The results regarding genotype and housing system interactions can be very important in choosing suitable housing for individual chicken genotypes. Therefore, the aim of this study was to compare performance characteristics and breast meat quality characterised by vitamin and carotenoid content, oxidative stability, fatty acid profile and sensory characteristics in three genotypes of chickens differing in growth rate and housed in the conventional system on litter or in mobile boxes on pasture.

Materials and methods

Genotypes, experimental design and performance characteristics

The experiment was carried out with 540 one-day-old cockerels of three genotypes: slow-growing ISA Dual, medium-growing Hubbard JA757 and fast-growing Ross 308. The cockerels were divided into six groups (90 cockerels per group) with six replication pens (15 cockerels per pen) according to the housing system (litter and mobile box on pasture) and the genotype. One group for each genotype stayed in indoor pens for the entire fattening period. The other three groups were also initially housed in indoor pens and were then moved to floorless portable pens (15 cockerels per pen) on experimental grassland in Netluky Village (Czech Republic, altitude 284 m a.s.l.) 14 days before slaughter. Chicken genotypes were stocked sequentially to be placed on pasture at the same time (2 weeks before slaughter). The one-day-old cockerels were housed in indoor pens on litter from wooden shavings with a 16-h lighting programme, gas heating and ventilation with a temperature-controlled fan. The initial temperature in the room was 32 °C (1st day) and continuously decreased to 20 °C (21st day for Ross 308, 28th day for Hubbard JA757 and 56th for ISA Dual). Each pen was equipped with pan feeders and nipple drinkers. The environmental conditions were kept in accordance with the requirements for each genotype. The grazing part of the experiment was carried out at the turn of the months of May and June 2019, and the average temperature during the monitored period was 16.6 °C. The following species predominated in the pasture herbage: Festuca pratensis, Lolium perenne and Trifolium pratense. The portable pens (length 360 cm × width 300 cm × height 60 cm) were made according to the utility model number 29006 and were equipped with hat drinkers and feeders and were moved twice a day at 8:00 am and 4:00 pm to minimise grassland damage.

Throughout the experiment, the cockerels were fed two diets. ISA Dual/Hubbard JA757/Ross 308 cockerels were fed starter diets until 28/14/11 days of age, respectively. A grower–finisher diet was fed between 29/15/12 and 81/42/35 days of age, respectively. The ingredients and nutrient contents of the diets and freeze-dried pasture herbage are shown in Tables 1 and 2. Feed and water were provided ad libitum. Procedures performed with the animals were in accordance with the Ethics Committee of the Central Commission for Animal Welfare at the Ministry of Agriculture of the Czech Republic (Prague, Czech Republic) and were carried out in accordance with Directive 2010/63/EU for animal experiments. The protocol of this experiment was approved by the Ethical Committee of the Institute of Animal Science (Prague-Uhříněves, Czech Republic), case number 02/2019.

Table 1. Composition of chicken diets.

Ingredient, g/kg	Starter	Grower–finisher
Wheat	451.5	576.3
Maize	150.0	80.0
Soybean meal	310.5	268.5
Fish meal	10.0	–
Monocalcium phosphate	8.8	6.3
Limestone	14.4	11.2
Sodium chloride	2.8	2.5
Soybean oil	34.1	10.0
Animal fat	–	29.3
Sodium sulphate	1.1	1.2
Amino acid premix	8.0	7.7
Vitamin–mineral premix	8.8	7.0

Table 2. Analysed nutrient content of the diets and freeze-dried pasture.

Nutrient	Starter	Grower–finisher	Freeze-dried pasture herbage
Dry matter, g/kg	904.3	903.4	948.4
Crude protein, g/kg	216.1	197.0	159.6
Fat, g/kg	54.5	57.7	22.8
AME, by calculation MJ/kg	12.6	12.9	5.12
SFA, mg/100 g	1913	1941	375
MUFA, mg/100 g	2579	2616	281
PUFA, mg/100 g	5225	5673	1132
n6/n3	7.8	8.3	0.6
α-Tocopherol, mg/kg	43.2	41.9	63.5
γ-Tocopherol, mg/kg	31.5	29.0	13.2
Retinol, mg/kg	1.75	1.53	–
Zeaxanthin, mg/kg	0.59	0.62	83.2
Lutein, mg/kg	0.95	0.97	99.7

AME: apparent metabolisable energy; SFA: saturated fatty acids (caproic, caprylic, capric, lauric, tridecylic, myristic, pentadecylic, palmitic, margaric, stearic, arachidic, heneicosylic, behenic, tricosylic); MUFA: monounsaturated fatty acids (myristoleic, palmitoleic, oleic, vaccenic, 11-eicosenoic, erucic, nervonic); PUFA: polyunsaturated fatty acids (linoleic, γ-linolenic, α-linolenic, eicosadienoic, dihomo-linolenic, arachidonic, eicosatrienoic, eicosapentaenoic, docosapentaenoic, docosahexaenoic).

The length of fattening was 35 days for Ross 308 chickens, 42 days for Hubbard JA757 chickens and 81 days for ISA Dual chickens. The cockerels were weighed at 0, 14, 35, 42 and 81 days of age. The number of cockerels and their health status were checked twice a day during the experiment based on chicken activity, normal behaviour patterns (e.g. active feed and water intake, normal walking, wing stretching, energetic movements when distracted or calm and effortless breathing), voice, plumage quality, skin, stance and foot and limb formation. Feed intake was monitored on a per-pen basis. Feed conversion was calculated as the total feed intake divided by the total weight gain over the entire fattening period. Pasture herbage intake was measured one week before slaughter and evaluated using the modified method of Dal Bosco et al. (2014). Pasture herbage samples were collected in square areas (50 × 50 cm) and then calculated for the whole area of each portable pen. The amount of pasture herbage was measured before the placement of the portable pen and then after the implementation of grazing and relocation of the pen to another position.

Slaughtering, carcass characteristics and muscle sampling

The experiment was terminated and the cockerels were slaughtered at a certain age (35 days for Ross 308, 42 days for Hubbard JA757 and 81 days for ISA Dual) based on the assumption of reaching 2.3 kg according to the manual of the given hybrid. Six cockerels (1 cockerel from each pen, n = 6) with an average body weight (average weight of cockerels from each pen ± 50 g) were selected from each group and slaughtered at a commercial slaughterhouse. The cockerels were slaughtered, bled and plucked, the feet and head were cut off, and the viscera were removed. The carcases were placed in a refrigerator and stored for 24 h at 4 °C. Subsequently, carcase analysis was carried out. The carcase weight and the weights of individual parts were determined. The breast muscles were separated on the chest from the shoulder joint and sternum and stripped of the skin. The legs were separated from the torso at the hip joint. The percent composition of these parts was calculated as a proportion of the carcase weight. Breast muscle dimensions (length and thickness) were measured as follows: the maximal breast length was measured with a calliper, whereas the maximal thickness was evaluated by inserting a metal needle. The dressing percentage was calculated by dividing the carcase weight by the body weight of the cockerel. The breast muscles (pectoralis major) were dissected for the analysis of individual meat quality indicators and stored at −20 °C until the start of the analysis.

Chemical analyses of diet, freeze-dried pasture and breast muscles

The dry matter of the diet, freeze-dried pasture and breast muscles was determined by drying the samples in an oven at 105 °C to a constant weight. The ash content was determined after combustion of samples in a muffle furnace at 500 °C for 12 h. The ether extract was obtained by extraction with petroleum ether according to the Soxhlet method in a Soxtec 1043 apparatus (FOSS Tecator AB, Höganäs, Sweden), and the protein content was detected using the Kjeldahl method on a Kjeltec Auto 1030 Analyser (Tecator, Höganäs, Sweden).

The concentrations of lutein and zeaxanthin in the samples were measured by high-performance liquid chromatography (HPLC) according to a modified method of Froescheis et al. (2000). The HPLC instrument (VP series; Shimadzu, Kyoto, Japan) was equipped with a diode array detector. A Kinetex C18 column (100 × 4.6 mm; 2.6 µm; Phenomenex, Torrance, CA) was used. A gradient system was applied with acetonitrile:water:ethyl acetate (88:10:2) as eluent A and acetonitrile:water:ethyl acetate (88:0:15) as eluent B.

The α-tocopherol, γ-tocopherol and retinol contents were analysed in accordance with the European standards EN 12822 (2000) and EN 12823-1 (2000), respectively, by a Shimadzu HPLC system (VP series; Shimadzu, Kyoto, Japan) equipped with a diode array detector. The samples were subjected to alkaline saponification with 60% potassium hydroxide followed by the appropriate extraction with diethyl ether.

The lipid peroxidation levels in fresh breast meat and meat stored for five days at 4 °C were evaluated by chromatographic analysis (Shimadzu HPLC system (VP series; Shimadzu, Kyoto, Japan) equipped with a diode array detector) using the modified method mentioned in the study of Czauderna et al. (2011). The column utilised was a Phenomenex C18 column (Synergi 2.5 µm, Hydro-RP, 100 Å, 100 mm × 3 mm) (Phenomenex, Torrance, CA). Solvent A consisted of water–acetonitrile (95:5, v/v), and solvent B consisted of acetonitrile. The lipid oxidative stability was expressed as mg of malondialdehyde (MDA) per kg of meat.

For determination of cholesterol in the meat, lipids were saponified, and the unsaponified matter was extracted with diethyl ether in accordance with ISO 3596:2011. Silyl derivatives were prepared using TMCS and HMDS silylation reagents (Sigma-Aldrich, Prague, Czech Republic) and quantified on a gas chromatograph equipped with a SAC-5 capillary column (Supelco, Bellefonte, PA) that was operated isothermally at 285 °C. The fatty acid composition of the analysed samples was determined after chloroform–methanol extraction of the total lipids (Folch et al. 1957). Nonadecanoic acid (C 19:0) was used as an internal marker to quantify the FAs in the samples. Alkaline transmethylation of the FAs was performed (Raes et al. 2003). Gas chromatography of the methyl esters was performed using an HP 6890 chromatograph (Agilent Technologies, Inc., Santa Clara, CA) with a programmed 60 m DB-23 capillary column (150–230 °C) and a flame ionisation detector; split injections were performed using an Agilent autosampler (Santa Clara, CA). The fatty acids were identified by their retention times compared with standards. PUFA 1, PUFA 2, PUFA 3 and 37 Component FAME Mix (Supelco, Bellefonte, PA) were used as standards. The atherogenic index (AI) and the thrombogenic index (TI) were calculated in accordance with the methodology of Ulbricht and Southgate (1991). The ratio between hypocholesterolaemic and hypercholesterolaemic fatty acids (hypocholesterolaemic/hypercholesterolaemic index; h/H) was calculated according to the formula mentioned in Santos-Silva et al. (2002).

Sensory analysis

For sensory evaluation, the procedure described by Bureš et al. (2014) was applied. Chicken breasts without skin were evaluated by a panel of 10 selected assessors trained according to ISO 8586-1 (1993). The evaluation was performed in a sensory laboratory equipped with booths. The samples were cooked for 1 h at 180 °C without any spices or other ingredients. The samples were cut approximately into cubes (2 cm × 2 cm × 2 cm), placed in covered glass containers labelled with three-digit random numbers and served at 50 °C to the sensory panel. Water and bread were provided to the panel members to neutralise their sensory precepts. The odour, tenderness, juiciness, flavour and overall acceptability of the samples were scored. A nine-point scale was used for the assessment (1 – very undesirable, 9 – very desirable).

Statistical analyses

The data related to cockerels performance and meat quality were analysed using two-way analysis of variance (ANOVA) with the general linear model (GLM) procedure in SAS software (SAS 2003). The main effects were the housing system (H), the genotype (G) and the interaction between these two factors (H × G). For sensory evaluation, the MIXED procedure was performed. This model included the same main effects, the interaction between these two factors and the random effect of the assessor. $$Y_{\mathrm{\text{ijk}}}=\mu +{\alpha }_i+{\beta }_j+{\gamma }_{\mathrm{\text{ij}}}+{\mathrm{ e}}_{\mathrm{\text{ijk}}}$$ where Y_ijk is the value of trait; µ is the overall mean; α_i is the effect of housing system (i = 1, 2), β_j is the effect of genotype (j = 1, 2, 3), γ_ij is the effect of interaction between housing system and genotype and e_ijk represented the random residual error.

The pen was the experimental unit (n = 6). All differences between means were considered to be significant at p < .05 and tested by Duncan’s test. The results in the tables are presented as the mean and standard error of the mean (SEM).

Results

As expected, body weight was influenced by genotype (p < .001), housing system (p < .001) and the interaction of both factors (p < .001, Table 3). The fast-growing Ross 308 chickens fattened on litter showed the lowest feed conversion ratio value (p < .001), whereas the highest value was achieved in slow-growing ISA Dual chickens housed on pasture. Ross 308 and Hubbard JA757 cockerels had higher cereal diet intake (p < .001) than genotype ISA Dual. ISA Dual cockerels showed the highest pasture herbage intake (p = .001), followed by Hubbard JA757, and the lowest pasture herbage intake was determined for the Ross 308 genotype. A higher pasture herbage intake in ISA Dual cockerels can subsequently positively affect the quality of their meat.

Table 3. Effect of housing system and genotype on performance characteristics.

	Litter			Mobile box			SEM	p Value
Housing (H) Genotype (G)	ISA Dual	Hubbard JA757	Ross 308	ISA Dual	Hubbard JA757	Ross 308		Housing	Genotype	H × G
Body weight, day 0, g	37.0	41.0	40.2	37.3	41.4	39.4	0.16	–	<.001	–
Body weight, day 14, g	187	327	460	191	324	449	5.3	–	<.001	–
Body weight, day 35, g	689^e	1457^c	2276^a	713^e	1332^d	2024^b	29.1	<.001	<.001	<.001
Body weight, day 42, g	898^c	2277^a	–	941^c	2093^b	–	27.9	<.001	<.001	<.001
Body weight, day 81, g	2286	–	–	2181	–	–	16.4	.001	–	–
F:G, g/g	2.15^b	1.62^d	1.36^f	2.26^a	1.70^c	1.51^e	0.021	<.001	<.001	<.001
Feed intake, g/day/bird	59.5	83.9	84.9	58.2	82.8	85.0	0.92	.325	<.001	.412
Mortality, %	1.3	5.2	2.6	6.5	2.6	3.9	0.04	.316	.221	.752
Average pasture herbage intake, g DM/day/bird				4.05^a	3.63^b	3.50^c	0.031	–	.001	–

SEM: standard error of the mean; F:G: feed:gain; DM: dry matter.

Means followed by different letters (^a–f) differ significantly.

As is evident from Table 4, the Ross 308 genotype cockerels housed in mobile boxes on pasture achieved the highest breast yield value (p = .026). The lowest values were recorded in ISA Dual cockerels housed both on litter and in mobile boxes. Housing on litter increased the dressing percentage (p < .001) and breast muscle length (p = .022) compared to housing in mobile boxes. In terms of genotype, higher dressing percentages (p < .001) were found in Ross 308 and Hubbard JA757 cockerels. Cockerels of the slow-growing ISA Dual genotype showed higher leg yield (p < .001) and longer breast muscle (p = .041). The highest maximal thickness of breast muscle (p < .001) was found in Ross 308 chickens, followed by Hubbard JA757 chickens, and the lowest value was measured in ISA Dual cockerels. The fast-growing genotype Ross 308 is suitable for breast muscle production, while the slow-growing ISA Dual has a higher proportion of legs.

Table 4. Effect of housing system and genotype on carcase characteristics.

Housing (H)	Litter			Mobile box			SEM	p Value
Genotype (G)	ISA Dual	Hubbard JA757	Ross 308	ISA Dual	Hubbard JA757	Ross 308		Housing	Genotype	H × G
Dressing percentage, %	72.4	75.3	75.0	70.2	72.9	73.4	0.30	<.001	<.001	.640
Breast yield, %	13.2^d	19.9^c	25.0^b	12.4^d	20.1^c	27.7^a	0.85	.174	<.001	.026
Leg yield, %	30.7	28.1	28.6	31.5	29.7	28.1	0.27	.136	<.001	.128
Breast muscle
Thickness, mm	17.7	25.0	31.5	16.5	23.6	32.0	0.94	.387	<.001	.523
Length, mm	172	167	159	163	157	155	1.7	.022	.041	.795

SEM: standard error of the mean.

Means followed by different letters (^a–d) differ significantly.

Ross 308 chickens housed on litter showed significantly higher fat content (p = .001) and lower dry matter content (p = .006) compared to the other groups (Table 5). The breast meat of the fast-growing chicken genotype Ross 308 had a higher ash content (p < .001), while the meat of the ISA Dual and Hubbard JA757 cockerels was richer in protein (p < .001). Compared to the other groups, the highest values of the lutein (p = .043) and α-tocopherol (p = .024) content in meat were found in chickens from mobile boxes of genotypes ISA Dual and Ross 308. The possibility of grazing increased the content of zeaxanthin (p < .001) and retinol (p = .010) in the meat. The content of fat-soluble vitamins (p = .015 to <.001) was the highest in the meat of the fast-growing genotype Ross 308. The oxidative stability of fat in fresh and stored meat was positively (p < .001) influenced by access of chickens to pasture. In the case of meat stored for five days (p = .001), the meat of slow- and medium-growing cockerels housed in outdoor mobile boxes showed the highest oxidative stability, while the ISA Dual cockerels from the litter had the lowest meat oxidative stability. The benefits of pasture fattening were best used by ISA Dual cockerels compared to litter fattening.

Table 5. Effect of housing system and genotype on nutrient, carotenoid and vitamin contents and oxidative stability of fat in breast muscle.

Housing (H)	Litter			Mobile box			SEM	p Value
Genotype (G)	ISA Dual	Hubbard JA757	Ross 308	ISA Dual	Hubbard JA757	Ross 308		Housing	Genotype	H × G
Dry matter, g/kg	265^a	258^b	253^c	260^ab	264^a	260^ab	0.9	.084	.005	.006
Fat, g/kg DM	23.2^c	28.9^c	67.3^a	24.3^c	26.0^c	43.9^b	2.6	.007	<.001	.001
Protein, g/kg DM	889	884	839	888	882	856	3.3	.455	<.001	.083
Ash, g/kg DM	42.0	45.1	45.2	40.8	42.7	45.0	0.34	.032	<.001	.246
Lutein, mg/kg	0.101^c	0.094^c	0.109^c	0.147^a	0.134^b	0.137^b	0.0046	<.001	.001	.043
Zeaxanthin, mg/kg	0.061	0.063	0.067	0.100	0.091	0.096	0.0035	<.001	.001	.642
α-Tocopherol, mg/kg	4.91^e	5.26^e	5.45^d	6.21^b	6.01^c	6.47^a	0.195	.002	<.001	.024
γ-Tocopherol, mg/kg	0.813	0.864	1.065	1.108	0.976	1.233	0.0493	.746	<.001	.747
Retinol, mg/kg	0.044	0.040	0.050	0.054	0.052	0.056	0.0023	.010	.015	.327
MDA, day 0, mg/kg	0.334	0.329	0.364	0.292	0.249	0.278	0.0090	<.001	.197	.440
MDA, day 5, mg/kg	0.551^a	0.449^b	0.383^bc	0.315^c	0.340^c	0.393^bc	0.0162	<.001	.260	.001

SEM: standard error of the mean; DM: dry matter; MDA: malondialdehyde.

Means with different superscripts (^a–c) differ significantly.

An interaction effect of both factors was recorded on all monitored parameters related to fatty acid and cholesterol content (Table 6). Breast meat of Ross 308 cockerels housed on litter showed the highest proportion of saturated fatty acids (p < .001). Hubbard JA757 cockerels also housed on litter had the highest proportion of monounsaturated fatty acids in meat (p < .001). The significantly highest proportion of polyunsaturated fatty acids (p < .001) was recorded in the meat of the ISA Dual genotype housed both on litter and in a mobile box on pasture and in the meat of the Hubbard JA757 genotype housed in a mobile box. The lowest n6/n3 fatty acid ratio (p = .045) and TI (p < .001) and the highest h/H index (p < .001) were recorded in slow-growing chickens that were allowed to graze. The AI (p < .001) was also the lowest in these chickens but also in the Hubbard JA757 genotype with access to pasture vegetation. Meat of Hubbard JA757 cockerels fattened on litter had significantly (p = .002) lower cholesterol content. According to the sensory analysis (Table 7), cockerels housed on litter had more fragrant (p = .013) and juicier (p < .001) meat, and overall acceptability (p = .001) was also higher for this meat. Of all the tested genotypes, Ross 308 cockerels had the juiciest meat (p = .032). The most tender (p = .015) meat was ascertained in fast-growing Ross 308 cockerels housed on litter.

Table 6. Effect of housing system and genotype on composition and indices of fatty acids and cholesterol content in breast muscle.

Housing (H)	Litter			Mobile box			SEM	p Value
Genotype (G)	ISA Dual	Hubbard JA757	Ross 308	ISA Dual	Hubbard JA757	Ross 308		Housing	Genotype	H × G
SFA, %	36.1^c	38.8^b	40.7^a	35.8^c	36.2^c	38.0^b	6.6	<.001	<.001	<.001
MUFA, %	22.0^e	23.4^a	23.1^b	22.9^c	22.2^e	22.5^d	8.96	<.001	<.001	<.001
PUFA, %	41.9^a	37.8^c	36.2^d	41.3^a	41.6^a	39.5^b	11.3	<.001	<.001	<.001
n6/n3	6.92^c	7.16^b	7.55^a	6.12^d	6.86^c	7.23^b	0.120	.011	<.001	.045
AI	0.403^c	0.428^b	0.460^a	0.369^d	0.379^d	0.392^c	0.0077	<.001	<.001	<.001
TI	0.789^c	0.834^b	0.911^a	0.738^e	0.752^d	0.791^c	0.0145	.301	<.001	<.001
h/H	2.39^b	2.06^c	1.75^e	2.61^a	2.46^b	1.92^d	0.055	.001	<.001	<.001
Cholesterol, mg/kg	507^ab	362^c	564^a	495^b	461^b	508^ab	12.1	.506	<.001	.002

SEM: standard error of the mean; SFA: saturated fatty acids; MUFA: monounsaturated fatty acids; PUFA: polyunsaturated fatty acids; AI: atherogenic index; TI: thrombogenic index; h/H: hypocholesterolaemic/hypercholesterolaemic fatty acid ratio.

Means with different superscripts (^a–e) differ significantly.

Table 7. Effect of housing system and genotype on sensory analysis of breast muscle.

Housing (H)	Litter			Mobile box			SEM	p Value
Genotype (G)	ISA Dual	Hubbard JA757	Ross 308	ISA Dual	Hubbard JA757	Ross 308		Housing	Genotype	H × G
Odour	5.46	5.45	4.81	5.39	5.26	4.80	0.094	.013	.625	.928
Tenderness	4.46^d	5.80^c	7.66^a	4.65^d	4.91^d	7.05^b	0.096	<.001	.005	.015
Juiciness	4.40	4.86	6.18	4.32	4.50	6.18	0.086	<.001	.032	.606
Flavour	5.51	5.75	5.75	5.64	5.41	5.43	0.089	.998	.314	.481
Total acceptability	5.10	5.38	6.03	5.29	5.23	5.76	0.084	.001	.650	.513

Means with different superscripts (^a–d) differ significantly; SEM: standard error of the mean.

Discussion

The different growth intensity levels of the compared genotypes are evident from the live weights. When housed on litter, a live weight of 2.30 kg was reached by fast-growing Ross 308 chickens at the age of 35 days, medium-growing Hubbard JA757 chickens at the age of 42 days, and slow-growing ISA Dual chickens at the age of 81 days. Housing in mobile boxes on pasture herbage reduced the body weight of cockerels, which was probably caused by the lower average temperature during the experiment (16.6 °C) and higher movement activity that outdoor housing allows. Filho et al. (2003) showed that free-range broilers took an average of more than two days longer to reach a final body weight of 2.30 kg compared to intensively reared birds. Individual genotypes react differently to the outdoor housing system. Fast-growing organic hybrids decrease their growth potential by as much as 25% compared to intensively reared birds, whereas the decrease in slow-growing hybrids is only 8%, serving as a reason to favour less intensive genotypes in alternative broiler rearing systems (Castellini et al. 2002). This finding is consistent with our results, where outdoor housing reduced the body weight of ISA Dual cockerels by 4.6% for the Hubbard JA757 genotype by 8.1%, and the highest difference was recorded for the fast-growing Ross 308 genotype (11.1%). ISA Dual cockerels have the ability to better adapt to less favourable outdoor housing conditions compared to the other two genotypes, which may also be related to age. Buchanan et al. (2007) ascribed reduced body weight to lower feed intake. However, in the present study, an effect of the housing system on feed consumption was not found. Conversely, in the study of Ponte, Rosado, et al. (2008), the intake of forage vegetation promoted growth by improving cereal-based feed consumption. Englmaierová et al. (2021) also reported a higher body weight in chickens housed outdoors on pasture. Based on the feed conversion values, it is clear that the fast-growing genotype in ideal barn conditions was able to better use the nutrients of the mixed feed to increment body tissue formation compared to the other groups. This result is consistent with the study of Branciari et al. (2009), who compared three genotypes of chicken fattened under conventional and free-range conditions. Outdoor housing had a negative effect on the feed conversion of all compared genotypes, which can also be attributed to the lower average outdoor temperature during the duration of the experiment and the possibility of performing more activities than in indoor housing. Pasture herbage intake in cockerels ranges from 3.50 to 4.05 g of dry matter per day. Lorenz and Grashorn (2012) showed that approximately 10–15% of total feed intake in broilers may come from pasture. Genetic selection of animals for a better growth rate modified their behaviour and reduced kinetic activity (Branciari et al. 2009). Therefore, the pasture herbage intake depends on the chicken genotype. The slow-growing ISA Dual cockerels showed the highest willingness to graze, which is related to their higher mobility due to lower growth intensity and is reflected in the potential enrichment of their meat with health-promoting substances such as carotenoids, vitamins or n-3 fatty acids (Sossidou et al. 2015; Englmaierová et al. 2021). Dual cockerels show lower performance compared to commonly commercially fattened genotypes and this prevents their more frequent use. Therefore, grazing fattening is an option for these cockerels to compensate for lower performance with higher meat quality, or a higher content of health-promoting substances from pasture herbage in meat. The equipment of the pasture should be such that the chickens feel protected from predators and the sun and have enough fresh forage to stay there as long as possible. The highest breast yield was observed in Ross 308 cockerels fattened in mobile boxes on pasture. The value was twice that of the ISA Dual genotype. The higher breast muscle proportion in cockerels housed outdoors was probably caused by greater physical activity in cockerels housed in this way (Lei and Van 1997; Feddes et al. 2002; Englmaierová et al. 2021). Castellini et al. (2002) also found higher breast yield in chickens with access to a grass paddock. Semi-intensively reared birds had a significantly higher breast muscle proportion than intensively reared birds (Cheng et al. 2008). It is evident from the results that the genotype affects the representation of individual parts. Consumers prefer well-developed breast meat (Damme and Ristic 2003). Therefore, chicken selective breeding has focused on increasing breast muscle thickness (Flock 2004). In our study, Ross 308 and Hubbard JA757 cockerels achieved higher dressing percentage, breast yield and breast muscle thickness than the ISA Dual genotype, which showed higher leg yield and longer breast muscle. Dal Bosco et al. (2014) similarly stated that higher breast width, thickness and yield in combination with a lower tibia length and drumstick percentage were found in fast-growing chickens compared to slow-growing genotypes. Lower dressing percentage and breast yield associated with higher leg yield in slow-growing birds were also found in other studies (Fanatico et al. 2005, 2008). The percentage of thigh and drumstick muscles in broilers increases until approximately three weeks of age and then decreases, whereas in layer-type (slow-growing) chickens, it continues to increase, and at approximately seven weeks of age, it is higher than that in broilers (Murawska and Bochno 2007). Therefore, in the case of slow-growing chickens, especially the thigh muscles will be used for the meat industry, and so, the evaluation of meat quality should not be limited only to the assessment of the breast muscle, but also to the thigh and drumstick muscles, but the lower level of homogeneity of this part can make the evaluation difficult. In the present study, Ross 308 cockerels housed inside in the hall on litter had the highest fat content, which can be explained by the growth rate and suitable conditions in this housing system for this genotype and the lower level of physical activity. In the case of animals with high growth rates, fat is quickly incorporated into the cells and replaces water (Metzger et al. 2011). The content of fat and ash in the breast muscle increased and the protein and dry matter content decreased with increasing cockerel growth intensity. Chodová et al. (2021) also stated that selection on growth rate in fast-growing chickens could be associated with lower protein and dry matter content and higher ether extract content than is the case with medium- and slow-growing chickens. The higher protein content of the slow-growing genotype may be related to the age of the cockerels at slaughter. As the age of the animals increases, the protein content also increases, and conversely, the moisture content decreases (Metzger et al. 2011). The housing system influenced the fat and ash content in breast meat, and cockerels housed on litter exhibited higher amounts of these nutrients. The reduction in breast muscle fat content in cockerels housed in mobile boxes on pasture can be explained by lower environmental temperature, but also by greater physical activity. In addition, Mikulski et al. (2011) found a higher content of dry matter and protein in the breast meat of chickens with outdoor access. Pasture fattening of fast-growing chickens Ross 308 can increase breast muscle dry matter to the level of medium- and slow-growing genotypes.

Pasture herbage is a rich source of vitamins and carotenoids (Skřivan and Englmaierová 2014; Englmaierová et al. 2021), one of whose many functions is antioxidant activity. Pasture herbage intake is affected by the motor activity of chickens (Dal Bosco et al. 2016). From this point of view, slow- and medium-growing chickens are more suitable genotypes for outdoor housing. This corresponds to a higher intake of pasture vegetation in slow-growing ISA Dual cockerels (4.05 g DM/day/bird) and was reflected in the highest lutein content in the breast muscle of these chickens (0.147 mg/kg). In contrast, the highest α-tocopherol content in meat was observed in the commercial hybrid Ross 308 housed in mobile boxes, followed by the slow-growing ISA Dual genotype housed also in mobile boxes. The higher content of α-tocopherol in the breast meat of fast-growing chickens could be due to the higher intake of the cereal mixed feed compared to the other two genotypes, together with the non-negligible consumption of pasture. The higher representation of α-tocopherol in meat is caused by the more than 10-fold preference of the tocopherol-binding protein for α-tocopherol relative to γ-homologues, which are the most common vitamin E molecules in plant foods (Decker et al. 2000). Furthermore, pasture herbage mainly contains α-tocopherol and less γ-tocopherol. The ability to store α-tocopherol from pasture in meat is evident from a number of studies. For example, Skřivan et al. (2015) and Dal Bosco et al. (2016) ascertained an almost twofold higher α-tocopherol content in the meat of chickens with outdoor access. In contrast, Ponte, Alves, et al. (2008) did not find an effect of pasture intake on the levels of vitamin E compounds in meat. A lower intake of antioxidants (vitamins and carotenoids) and a higher representation of PUFAs in the breast muscle in slow-growing chickens fed on litter probably resulted in lower oxidative stability of five-day stored meat. γ-Tocopherol is somewhat less potent in donating electrons than α-tocopherol, and therefore, α-tocopherol is generally considered to be more effective as a chain-breaking antioxidant for inhibiting lipid peroxidation (Kamal-Eldin and Appelqvis 1996). The α-tocopherol content was significantly lowest in ISA Dual and Hubbard JA757 cockerels housed on litter. Slow-growing birds are very active, which affects their oxidative metabolism, resulting in high production of reactive oxygen species and extensive consumption of antioxidants by the body (Mancinelli et al. 2021). For slow-growing cockerels, litter housing can be more stressful and cockerels may use more antioxidants for their own purposes than for meat protection compared to pasture fattening. Therefore, the ISA Dual cockerels fattened on litter were not able to deposit a sufficient amount of vitamin E from the cereal diet into the meat to prevent fat oxidation in the meat. And so, this housing system will not be suitable for these cockerels both in terms of welfare and meat quality. Conversely, the oxidative stability of the meat of fast-growing chickens is high and comparable in both systems.

The higher content of polyunsaturated fatty acids in the breast muscle of ISA Dual cockerels is related to genetic foundation. Slow-growing strains of egg-type lines have a genetically determined higher efficiency of EPA and DHA deposition due to the FADE gene, which is involved in the formation of long-chain n-3 and n-6 (Katekhaye 2019). These strains show a higher oestrogen level, which also partly affects fatty acid elongation, resulting in a higher efficiency of EPA and DHA deposition (Alessandri et al. 2012). Furthermore, a higher content of polyunsaturated fatty acids in the breast meat of slow-growing chickens reflected the increased expression/activity of Δ5- and Δ6-desaturase (Boschetti et al. 2016). Another aspect is that the pasture herbage itself is a source of polyunsaturated fatty acids (Dal Bosco et al. 2016; Michalczuk et al. 2016; Englmaierová et al. 2020), and the slow- (ISA Dual) and medium- (Hubbard JA757) growing cockerels showed a greater pasture herbage intake, which is evidenced by the higher polyunsaturated fatty acid values in the breast meat compared to the fast-growing genotype Ross 308. From the point of view of human health, it is not only desirable to have a high intake of polyunsaturated fatty acids but also to maintain the ratio between n-6 and n-3 fatty acids at approximately 6:1 (Wijendran and Hayes 2004). This corresponds to the value achieved in the breast meat of ISA Dual cockerels. The influence of individual fatty acids on human health or on the risk of occurrence of pathogenic phenomena and on cholesterol metabolism is also comprehensively assessed by different indices, e.g. AI, TI and h/H. These indices also achieved the most favourable values in ISA Dual cockerels housed in mobile boxes on pasture. From the point of view of people’s health and the prevention of civilisation diseases, it is advisable to consume the meat of slow-growing dual cockerels that had access to pasture. These cockerels made the most favourable use of the fatty acids received through feed and grazing, and their meat can thus be considered a functional food. And this may be one of the reasons for the more frequent use of dual cockerels for meat production.

The significantly lowest cholesterol content in meat was recorded in Hubbard cockerels housed on litter. Cholesterol can be obtained directly from the diet or synthesised in cells from 2-carbon acetate groups of acetyl-CoA. The proportion of cholesterol derived from de novo biosynthesis depends on the amount of cholesterol in the diet. The composition of the supplied feed mixture was the same in all groups. The amount of cholesterol that an animal produces and stores probably depends on its genetic foundation. It is also evident from the study of Ponte, Alves, et al. (2008) that fast-growing and younger birds had greater levels of cholesterol. Furthermore, it was found that consumption of a leguminous pasture had a marginal effect on the cholesterol content of broiler meat. This is also consistent with our results. Nevertheless, the fast-growing cockerels that were allowed to graze showed the same level of cholesterol content in the meat as the dual cockerels from litter.

Sensory evaluation is the most important quality indicator for consumers. Tenderness, as one of the sensory indicators, has been shown to likely be the single most critical quality factor associated with end consumers’ satisfaction with poultry meat (Fletcher 2002). In the present study, Ross 308 cockerels housed on litter had the most tender meat. Higher tenderness is probably related to lower motor activity (Castellini et al. 2002) and higher fat content in fast-growing chickens than in slow-growing genotypes. Another factor that can affect the tenderness of meat is the age of the slaughtered chickens. Older birds are believed to have less tender and firmer meat (Farmer 1999). It is the slow-growing genotypes that are fattened to an older age. Juiciness was influenced by both genotype and housing system. Higher juiciness was recorded in the Ross 308 genotype and cockerel housed on litter. The level of juiciness can be attributed to intramuscular fat content, which was also higher in the fast-growing genotype and when housed on litter. Accordingly, Castellini et al. (2006) demonstrated that a lower intramuscular fat content was accompanied by lower meat juiciness in slow-growing broilers. Furthermore, fast-growing chickens have thicker pectoral muscles and therefore do not lose as much juice as the thinner pectoral muscles of slow-growing chickens. This is evidenced by the study of Fanatico et al. (2005), who found that slow-growing birds had poorer water-holding capacity than fast-growing birds. However, it is worth mentioning that there is no guarantee that genotypes with more juiciness, fat and energy have more tender breast meat (Hailemariam et al. 2022). In addition, we did not observe an effect of genotype or housing system on flavour. In contrast, Farmer (1999) showed that older birds have more intense meat flavours because flavour increases after the growth inflection occurs. The meat flavour increases with age, probably due to an increased concentration of nucleotides in muscle that breakdown into inosinic acid and hypoxanthine after death, which then enhance the meat flavour (Davidek and Khan 1967; Aberle et al. 2001). In the present study, in terms of overall acceptability, the breast muscle of cockerels fattened on litter earned a better rating. The same was found in the study of Alvarado et al. (2005), in which breast fillets from chickens housed on litter were preferred by evaluators over free-range fillets. Fanatico et al. (2007) did not find significant differences in overall acceptability. Conversely, the sensory panel test in the study of Castellini et al. (2002) gave significantly higher scores for overall acceptability to the breast muscles from organic chickens. Thus, the findings of the available studies are inconsistent on the issue of overall acceptability.

A limiting factor for the wider use of meat from slow-growing chickens will probably be the reduced tenderness and juiciness of the meat and the higher price of meat. However, a promising finding is that no significant differences between genotypes were found for overall acceptability. In the case of dual cockerels, the higher maturity of the meat could probably play its role, which compensated for the reduced tenderness and juiciness of the meat of this genotype. Moreover, the results of survey conducted by de Haas et al. (2021) show that consumers are willing to pay more for poultry products that do not require culling day-old male chicks.

Conclusions

There were differences among cockerel genotypes in terms of grazing activity and pasture herbage intake level. The higher pasture herbage intake in slow-growing compared to fast-growing cockerels is mainly reflected in the fatty acid profile and lutein content rather than in the content of fat-soluble vitamins in breast meat, where cereal diet intake also plays a role. Conventionally fattened slow-growing ISA Dual cockerels were not able to use vitamin E from the cereal diet to increase the oxidative stability of fat in meat compared to medium- and fast-growing genotypes. The ISA Dual genotype is suitable for pasture fattening, as it showed the highest willingness to graze, and the intake of forage vegetation had a positive effect on meat quality from the perspective of the content of health-promoting substances, especially lutein and PUFA, and increased the oxidative stability of meat. However, the use of the ISA Dual genotype for the purpose of meat production may be negatively affected by the lower performance and worse sensory evaluation of meat from pastured cockerels. In conclusion, if ISA Dual cockerels are allowed to be fattened on pasture, they provide a quality product - meat that can be considered a functional food, intended for specialised markets. Thus, pasture fattening of dual cockerels appears to be more beneficial than culling them immediately after hatching. Subsequent research could be focused on evaluating the quality of thigh and drumstick muscles, which are more represented in slow-growing chickens.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The original data of the paper are available upon request from the corresponding author.

Additional information

Funding

This research was funded by the Ministry of Agriculture of the Czech Republic, Grant Numbers MZE-RO0723 and QK1910387.