RETRACTED ARTICLE: Effects of emulsifier, betaine and l-carnitine on growth performance, immune response, gut morphology and nutrient digestibility in broiler chickens exposed to cyclic heat stress

1. This experiment investigated the efficiency of varying doses of an emulsifier blend (EB; 0 and 1 g/kg of diet), betaine (BT; 0 and 1 g/kg of diet) and L-carnitine (CT; 0 and 0.5 g/kg of diet) in broilers subjected to circular heat stress (HS) conditions.2. A total of 1080 one-day-old male broiler chickens (Ross 308) were randomly assigned to nine treatment groups (six pens/treatment with 20 birds/pen) in a completely randomised design. The thermoneutral control broiler chickens were housed at a comfortable temperature and fed a standard diet (no additives). The other 8 groups were exposed to cyclic HS conditions (34°C) for 8 h (10:00-18:00).3. There were EB × BT × CT interactions for body weight (BW) at 24 d (P=0.038) and average daily gain (ADG) during the 10-24 d period (P=0.049), with the greatest values with concurrent supplementation of all three ingredients.4. Inclusion of EB resulted in greater (P<0.05) BW, ADG, European performance index, uniformity rate, primary antibody titres against sheep red blood cells (SRBC), duodenal villus height (VH) and villus surface area, nitrogen-corrected apparent metabolisable energy (AMEn) and apparent ileal digestibility (AID) of dry matter, crude protein and fat, but lower (P<0.05) feed conversion ratio, mortality rate and heterophile to lymphocyte ratio.5. Dietary BT supplementation improved (P<0.05) overall performance indicators, primary antibody titres against SRBC and Newcastle disease virus, serum total antioxidant capacity, duodenal VH, Jejunal VH/crypt depth, AID of dry matter and crude protein. The effect of dietary supplementation with CT was limited to an increase (P<0.05) in ADG (d 10-24) and a decrease (P<0.05) in serum malondialdehyde concentration (42 d) and jejunal crypt depth (42 d).6. In conclusion, dietary supplementation of either EB or BT alone or in combination can ameliorate some of the detrimental effects of HS on growth performance, immunity and intestinal health in broilers, while a minor positive effect on performance and antioxidant status was observed with CT supplementation.


Introduction
Heat stress (HS) has a detrimental impact on the feed intake of chickens as well as their growth and physiological characteristics (Cheng et al. 2018;Amiri et al. 2019;Abo Ghanima et al. 2020). In birds that are exposed to high temperatures for an extended period of time have reduced blood supply to the digestive tract which may cause damage to the intestinal epithelium (Song et al. 2014(Song et al. , 2016, resulting in poor digestion and absorption. As a result, such stress may be mitigated by the use of feed additives that increase digestion and absorption of nutrients whilst lowering heat increment. So far, a variety of strategies, including dietary approaches, have been used to mitigate the effects of HS in poultry husbandry. The most important nutritional methods include feeding animals less protein or amino acids (Amiri et al. 2019;Ghasemi et al. 2021), supplementing fat in their diet (Brannan et al. 2021) or incorporating various additives into their diets, such as vitamin C (Zangeneh et al. 2018), vitamin E (Rehman et al. 2017a), potassium chloride and sodium bicarbonate (Roussan et al. 2008). However, there is still a need for more research and development in this field.
Lipotropic agents, such as choline, betaine and carnitine, have been shown to improve lipid turnover and reduce hepatic fat build-up by increasing their release from the liver (Ball 1964). Betaine (BT; dimethylglycine), one of the most significant lipotropic agents, has long been recognised to be involved in fat metabolism. In addition, BT has been shown to protect cells from osmotic stress and allows regular metabolic operations under conditions that would normally cause the cell to deactivate (Metzler-Zebeli et al. 2009;Akhavan-Salamat and Ghasemi 2016). A number of biological effects of BT in broilers subjected to HS have been demonstrated recently, including stress reduction (Saeed et al. 2017), antioxidant activity (Yang et al. 2021), immune stimulation (Alagawany et al. 2022) and improved intestinal barrier integrity (Alhotan et al. 2021). L-Carnitine (CT), known as B-hydroxy-Y-N-trimethyl aminobutyrate, is a necessary micronutrient that is produced from essential amino acids like methionine and lysine (Kolodziejczyk et al. 2011). Because of its role in numerous critical metabolic processes, CT is gaining popularity as a feed supplement for enhancing animal productivity (Rehman et al. 2017b). L-Carnitine is essential to the process of producing energy because it speeds up the transport of fatty acids into the mitochondria, which is a crucial step in lipid metabolism (Derin et al. 2004). Although CT has been shown to improve antioxidant status (Sepand et al. 2016) and boost immunity (Rehman et al. 2017a) in chickens, most studies on have been inconclusive (Rehman et al. 2017b). On the basis of the beneficial effects of BT and CT, it has been postulated that impairment of performance, anti-inflammatory and antioxidant activities in broiler chickens induced by HS may be ameliorated by BT and CT supplementation.
Lysophospholipids are monoacyl forms of phospholipids that are generated when an ester bond at either the sn-1 or sn-2 position is hydrolysed by the activity of phospholipase A1 or A2. They are more hydrophilic and have enhanced oilin-water emulsifying characteristics as a result of the elimination of one fatty acid (Nieuwenhuyzen and Tomás 2008). In addition to improving gut permeability to macromolecules like proteins and dextran, lysophospholipids regulate several enzymes and control protein channel formation in the membrane by boosting ion exchanges (Maingret et al. 2000). These mechanisms help in the transmission of nutrients, which can range from minute particles, such as calcium ions, to large components, such as polysaccharides, that need to be broken down in order to be absorbed (Boontiam et al. 2017(Boontiam et al. , 2019. This results in higher nutrient bioavailability and better broiler performance, particularly under challenging conditions, such as HS. Previous studies have shown that adding lysolecithin or lysophospholipids to the feed of broiler chickens may enhance their immunity (Allahyari-Bake and Jahanian 2017; El-katcha et al. 2021), intestinal morphology (Papadopoulos et al. 2018;Solbi et al. 2021) and antioxidant status (Zangeneh et al. 2018). Since the digestive system and oxidative defences are compromised when HS is present (Sgavioli et al. 2019;Khan et al. 2021), it has been hypothesised that broiler feed could be supplemented with lysophospholipids to decrease the harmful effects of HS on the birds' health and growth. However, there is a paucity of data regarding the impact of an emulsifier blend (EB) that includes various forms of lysophospholipids on nutrient digestibility, immune response and intestinal mucosa morphology in broiler chickens, particularly when they are subjected to high environmental temperatures. In addition, based on the previously mentioned positive effects of BT and CT, it was hypothesised that the administration of either BT or CT alone or in combination with EB may aid in minimising the adverse effects of HS on growth performance and health status.
Given all the above-mentioned aspects, the objective of this study was to determine whether the addition of an EB, BT and CT, either singly or in combination, to a maizesoybean meal diet improved growth performance, immunity, antioxidant status, gut morphology and nutrient digestibility of broiler chickens exposed to HS.

Materials and methods
The Animal Care and Use Committee of Ilam University (Ilam, Iran) approved all of the animal husbandry and experimental protocols used in the study

Experimental design and diets
For this study, a total of 1080 male Ross 308 broiler chicks were obtained from a commercial hatchery. Upon arrival, broiler chickens were weighed individually (44.1 ± 0.56 g) and then they were distributed among 54 separate pens (9 treatments of 6 replicate groups of 20 birds each). The experiment was conducted in a completely randomised design with a 2 × 2 × 2 plus 1 factorial arrangement of the treatments with two levels of EB supplementation (0 and 1 g/ kg of diet), two levels of BT supplementation (0 and 1 g/kg of diet) and two levels of CT supplementation (0 and 0.5 g/kg of diet). A ninth treatment was included as a thermoneutral control (TNC), in which the birds were fed a basal diet (without any additives) and raised in an environment with a comfortable temperature range. There were three phases of feeding in this study: starter (0-10 d), grower (11-24 d) and finisher (25-42 d). All of the starter, grower and finisher diets were formulated to meet or exceed the nutrient specifications for Ross 308 broilers. Table 1 details the experimental diets in terms of their ingredients as well as their chemical composition.
The experiment was conducted in two environmentcontrolled rooms with independently controlled temperatures. The rooms were each equipped with thermostatically controlled portable electric heaters (with a power of 2000 W), each supplied with a fan for circulation of hot air. Throughout the course of the experiment, the relative humidity (RH) within the rooms, which was maintained by a humidifier with a capacity of 4.5 l, was kept at a level that varied between 55% and 60%. An automatic thermohygrometer was used to make the necessary adjustments to both the temperature and the RH. In the TNC group, the temperature was maintained at 33-34°C for the first 3 d, then gradually decreased by 3°C per week to a final temperature of 22°C, while in the HS group, it remained unchanged for 8 h (10:00-18:00) and then decreased to the same level as the TNC group for the remaining 16 h (Amiri et al. 2019). Temperature and RH ranges in both rooms over various trial periods are described as supplementary material. The rooms were mechanically ventilated using a continuousflow, pressure-controlled blower that moved air via the air intake ports. During the growth phase, the chickens had unrestricted access to water and food. The lighting schedule was 24 h/d for the first 3 d, then decreased to 23 h/d afterwards.

Growth performance
After recording BW and feed consumption on d 0, 10, 24 and 42, the average daily gain (ADG) and the average daily feed intake (ADFI) were determined for each experimental group. In order to determine the mortality rate, data on the number of deaths that occurred each day was gathered. The feed conversion ratio (FCR) for each feeding phase was calculated using the total ADG and ADFI for each pen, taking into account any dead or culled birds. The uniformity rate at 42 d was estimated using the following equation: Uniformity rate = 100 − [(standard deviation/average BW) × 100]. For each experimental group, the European performance index (EPI) was calculated as follows: EPI = liveability (%) × live weight (kg) × 100/age (d) × FCR

Blood measurements
On d 42, two broiler chickens per pen (about the average BW of the replicates) were chosen and blood samples were collected from the wing vein. One part of the blood was inserted into a venoject tube containing 0.5 cm 2 of anticoagulant to test the blood leukocyte profile and antioxidant enzymes, while the other portion of the blood was placed in anticoagulant-free tubes to separate serum. The slides for the purpose of classifying leukocytes were instantly created from an aliquoted sample of whole blood taken from each broiler chicken. After being air-dried, the slides were then fixed with methanol. After staining the slides with May-Grünwald-Giemsa (Lucas and Jamroz 1961) they were viewed under an optical microscope to differentiate the leucocyte cell counts and 100 leucocytes were counted in each sample. The ratio of heterophils to lymphocytes (H/L) was then calculated. The enzyme activities of glutathione peroxidase (GSH-Px) , superoxide dismutase (SOD) and catalase (CAT), as well as the contents of total antioxidant capacity (TAC) and malondialdehyde (MDA) in the serum, were determined spectrophotometrically using commercial kits in accordance with the instructions provided by the manufacturer. The following are the specifications for the commercial kits: For the TAC test (Randox Laboratories Ltd, Crumlin, UK) and the GSH-Px, SOD, CAT and MDA assays (Cayman Chemical Co., Ann Arbor, MI, USA). The corticosterone concentration in the serum was also determined using an enzymelinked immunosorbent assay (ELISA) kit for chicken corticosterone (ZellBio, GmbH, Ulm, Germany).
In order to determine the immune response against Newcastle disease virus (NDV), blood samples from two broiler chickens were collected from the brachial vein on d 28 and 35 (7 and 14 d after the last vaccination). The serum samples were obtained using centrifugation (2000 × g, 10 min) at 4°C. According to Iritani et al. (1991), serum anti-NDV antibody titres were evaluated using a haemagglutination inhibition test and the results were represented as the logarithm base. In order to determine the primary and secondary immunity to sheep red blood cells (SRBC), two birds per replicate were immunised intramuscularly with 0.25 ml of 10% SRBC in a PBS solution on d 14 and 35. At 7 d after each injection (d 21 and 42), 1.5 ml blood samples were taken from the brachial vein of each bird. The serum was separated using centrifugation for 10 min at 1500 × g. A haemagglutination method using 96-well, U-bottom microtitre plates was employed to determine antibody titres against SRBC according to Wegmann and Smithies (1966). Total antibody titres against SRBC were expressed as the reciprocal of the greatest serum dilution that resulted in full agglutination.

Gut morphology
After blood samples were taken at the end of the feeding trial (d 42), selected broilers (12 chickens/treatment) were killed by cutting their jugular vein. In order to conduct the morphological investigation, 2 cm slices were taken from the middle sections of each duodenum and jejunum segment, rinsed with distilled water to flush out any residual contents and then fixed in 10% neutral-buffered formalin. The intestinal tissue segments were cut into 5 mm cross-sections using a microtome and mounted on a glass slide before being analysed under a light microscope (Olympus CX31, Shinjuku). Each sample of intestinal tissue was cut into three cross-sections, and each cross-section was measured 10 times. An image-analysis software (QWinPlus v. 3.1.0, Leica Cambridge Ltd., Cambridge, UK) was used to determine the morphological measurements of villus height (VH; the distance between the base and top of the villi), villus width (VW; at the mid-point of the villus) and crypt depth (CD; the distance between the crypt-villus junction and the base of the crypt). Data taken from both the VH and the CD were used to calculate the VH/CD ratio. In addition, the value of the villus surface area (VSA) was determined by using the following equation: VSA = 2π × (VW/2) × VH.

Nutrient digestibility
To investigate in vivo nutritional digestibility, a total of 108 birds (two per replication pen) were selected at random on d 42.
Throughout the course of this investigation, that lasted for 4 d, the acid-insoluble ash (AIA) marker approach was used to determine the nutrient digestibility. On d 42, Celite (Celite*545, Merck KGaA, Darmstadt, Germany) was added to the diet at a dose of 10 g/kg, which served as an extra source of AIA. After birds were killed by cervical dislocation (on d 46), the ileal digesta (the contents of the gut between Meckel's diverticulum and roughly 10 mm above the ileal-caecal junction) was collected in plastic zip bags for further analysis. The ileal digesta from two birds from each duplicate pen were combined and a representative sample was promptly frozen at -20°C for subsequent analysis of nutrient digestibility and apparent metabolisable energy adjusted for nitrogen balance (AMEn). Following drying in an oven at 65°C for 24 h, the samples taken from the diets and the ileal digesta were then ground into a fine powder for chemical analysis. The dry matter (method 930.15), crude protein (N × 6.25; method 984.13), crude fat (method 920.39) and ash (method 942.05) in the feed and ileal samples were analysed using AOAC (2006) procedures. Gross energy in the sample was also measured using an automated adiabatic oxygen bomb calorimeter (Parr Instrument Company, Moline, IL). According to McCarthy et al. (1974), the content of AIA present in the feed and ileal samples was measured. The following equation was used to determine the apparent ileal digestibility (AID) of the various nutrients included in diets: where AIA diet and Nutr diet represented the contents of AIA and nutrient in the diet (%), while AIA id and Nutr id reflected the contents of the same AIA and nutrient in the ileal digesta (%).
The AMEn value was determined by using the following equation by Majdolhosseini et al. (2019): where GE diet was gross energy value in the diet (kcal/kg) and GE id the gross energy value in excreta (kcal/kg), IF the indigestibility factor (AIA diet /AIA id ), N diet the nitrogen concentration in the diet (%), N id the nitrogen concentration in excreta (%) and 8.22 the energy equivalent (Cal/g) of uric acid.

Statistical analysis
Using statistical software (SAS Institute, version 9.0; SAS Institute Inc., Cary, NC, USA), the data omitting the TNC treatment were analysed statistically using a three-way ANOVA for a 2 × 2 × 2 factorial design. The model includes the main effects of EB, BT, CT and their interactions. The applied mathematical model was as follows: where Y ijkl= observation, µ = overall average, A i = effect of EB dose; B j = effect of BT dose; C k = effect of CT dose, (AB) ij = interaction effect of ith EB dose × jth BT dose; (AC) ik = interaction effect of ith EB dose × kth CT dose; (BC) jk = interaction effect of jth BT dose × kth CT dose; (ABC) ijk = interaction effect of ith EB dose × jth BT dose × kth CT dose; e ijkl = error associated with each observation.
The data from all the treatments (completely randomised design) were analysed in a one-way ANOVA (comparisons between TNC and HS treatment) using the GLM Procedures of SAS (version 9.0; SAS Institute Inc., Cary, NC, USA). The experimental unit varied according to the variables that were being measured. The data for growth performance were investigated on a pen basis, whilst blood, intestinal morphology and nutritional digestibility were on an individual bird basis. The Shapiro-Wilk test and the Levene test were used to examine the normality and homogeneity of the variances, respectively. The mean separation was done with Tukey's post hoc analysis. The results are provided as least-square means (LSMeans), together with the standard errors related to each mean (SEM). Statistical significance was defined as P < 0.05 in all analyses.

Growth performance
The effects of dietary supplementation with EB, BT and CT on growth performance in broiler chickens reared under HS conditions are detailed in Tables 2, 3 and -4. There were no significant differences between the TNC and HS groups in terms of BW at 10 d or ADG during the first 10-d period. However, all HS groups had lower (P < 0.001) BW at 24 and 42 d of age, as well as lower (P < 0.001) ADG during the grower and entire experimental periods, than broiler chickens in the TNC group. During the finisher period, ADG in the TNC group was equivalent to that in the HS-8 group, but greater (P = 0.044) than that in other HS groups. All experimental groups had lower ADFI (P = 0.021) than the TNC group during the whole trial period. Broiler chickens in the HS-1 and HS-2 groups exhibited significantly greater FCR (P = 0.002) those in the TNC group during the starter and whole experimental periods. The results also showed that the uniformity rate in the TNC treatment was higher (P = 0.003) than that in the HS-1, HS-2, HS-3 and HS-5 treatments, but equivalent to other HS treatments. The EPI of the TNC group was similar to that of the HS-8 group, but significantly greater (P < 0.001) than that of the other HS groups. A lower (P < 0.001) mortality rate was also observed in the TNC group when compared to the HS-1, HS-2 and HS-3 groups, although it was equal to that observed in the other HS groups.
According to the three-way factorial analysis, interactions between EB, BT and CT were detected for BW at 24 d (P = 0.038) and ADG over the 10-24 d period (P = 0.049) in heat-stressed broilers. The main effects of dietary supplementation with EB and BT were observed (P < 0.05) in terms of BW on d 10, 24 and 42; ADG and FCR on d 0-10, 10-24 vi J. YOUSEFI ET AL.

R E T R A C T E D
and 0-42; and uniformity rate and EPI on d 42, showing that these metrics improved when either EB or BT were added to the diet. Although supplemental CT had no effect on any of the performance indicators in heat-stressed broilers for the whole trial period, it did improve BW at 24 d (P < 0.001) and ADG from d 10 to 24 (P = 0.006).

Stress indicators and humoral immune response
Data for blood stress indicators and serum antibody response are shown in Table 5. All experimental groups, except the HS-8 group, exhibited lower lymphocyte count (P < 0.001) and higher H/L ratio (P = 0.004) than the PC group. Heterophil count and the serum corticosterone concentration in all HS treatments, except for the HS-7 and HS-8 treatments, were higher (P = 0.009 and P = 0.044, respectively) than for broilers in the TNC group. In addition, the HS-1 and HS-2 groups had significantly lower secondary anti-SRBC titres (P = 0.021) than the TNC group. However, there were no significant differences among the experimental groups in terms of primary anti-SRBC titres or primary and secondary antibody responses to NDV. The main effects of CT supplementation, two-way interaction effects of EB × BT, EB × CT and the three-way interaction of EB × BT × CT were not significant for all stress and immune measures. By contrast, the secondary anti-SRBC response was influenced by BT × CT interaction (P = 0.038), so that the simultaneous addition of BT and CT supplements to heat-stressed broiler diets improved this response. As shown in Table 5, supplemental dietary EB caused a significant increase in lymphocyte count (P = 0.003) and primary anti-SRBC titres (P = 0.017), but a decrease in heterophil count (P = 0.020), H/L ratio (P = 0.011) and serum corticosterone concentration (P = 0.089). Additionally, dietary supplementation with BT enhanced lymphocyte count (P = 0.035), primary and secondary anti-SRBC titres (P = 0.010 and P = 0.021, respectively) and secondary antibody responses to NDV (P = 0.013), but lowered heterophil count (P = 0.078) and H/L ratio (P = 0.053). Table 6 shows the data for the antioxidant status in heatstressed broilers on d 42. There were no significant differences among the experimental groups in terms of serum GPx and CAT activities. In contrast, the TNC group had the lowest TAC level (P = 0.006) and SOD activity (P = 0.012) among all experimental groups. The HS-1, HS-2 and HS-5 groups had a significantly higher MDA concentration (P = 0.016) than did the TNC group.

Antioxidant status
Although ANOVA revealed no interactions in all antioxidant indices, dietary EB was associated with a trend towards higher CAT (P = 0.091) and GPx (P = 0.082) activity. Dietary supplementation with BT increased TAC level (P = 0.038) and SOD activity (P = 0.097), but decreased MDA concentration (P = 0.003) in heat-stressed broilers on d 42. Table 2. Effects of dietary supplementation with emulsifier blend (EB), betaine (BT) and L-carnitine (CT) on body weight and average daily gain of broiler chickens subjected to heat stress (HS).

Body weight (g)
Average daily gain (g/bird/d) Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, ***P < 0.001 and # tendency.
Additionally, the inclusion of CT resulted in a greater TAC level (P = 0.067) and lower MDA concentration (P = 0.020).

Gut morphology
With regards to morphological traits (Tables 7 and 8), the duodenal VH and duodenal VSA of the birds in the HS groups, with the exception of the HS-6, HS-7 and HS-8 birds, were significantly lower than those of the birds in the TNC group (P < 0.05). All HS groups, except the HS-8 group, exhibited a lower VH/CD ratio (P < 0.05) than did the TNC group. In addition, the TNC group tended to have a lower jejunal CD (P = 0.082) than all other groups. Based on the factorial analysis, the interaction effects of EB × BT, EB × CT, BT × CT and the three-way interaction of EB × BT × CT were not significant for all morphological characteristics in the duodenum and jejunum. In contrast, the addition of EB to the diet increased duodenal VH (P = 0.011), VW (P = 0.071) and VSA (P = 0.001). However, morphological parameters of the jejunum were not influenced by EB supplementation. BT increased the VH (P = 0.032), VH/CD (P = 0.082) and VSA (P = 0.097) in the duodenum, as well as the VH (P = 0.025) and VH/CD (P = 0.045) in the jejunum. The significant effect of CT was limited to the reduction of CD in the jejunum (P = 0.048) and other intestinal morphological parameters were not affected by its administration.

Nutrient digestibility
Data on AID of nutrients, as well as AMEn values, are shown in Table 9. No differences were seen for ADI ash or energy among the experimental groups (P > 0.05). In contrast, the ADI of dry matter in the TNC group was similar to the HS-6, HS-7 and HS-8 groups, but better (P = 0.044) than the other HS groups. All HS groups, except the HS-4, HS-7 and HS-8 groups, exhibited a lower (P = 0.023) AID of crude protein than that in birds in the TNC group. The TNC group exhibited a tendency towards a lower AID for crude fat (P = 0.067) than in all other groups. Additionally, the AMEn value in the TNC treatment was greater (P = 0.013) than that in the HS-1 and HS-2 treatments, but similar to that in the other HS treatments.
According to the results of the factorial analysis, neither AMEn nor digestibility was affected by CT treatment or two-way interaction effects of EB × BT, EB × CT, BT × CT, or three-way interaction of EB × BT × CT (P > 0.05). In terms of evaluating the main effects, inclusion of EB increased the AID for dry matter (P = 0.026), crude protein (P = 0.042), crude fat (P = 0.023) and energy (P = 0.055), as well as AMEn (P < 0.001), but did not affect the AID of ash. In the BTsupplemented group, the AID of dry matter and crude protein were also increased (P = 0.042 and P = 0.010, respectively), compared with the non-supplemented group. Table 3. Effects of dietary supplementation with emulsifier blend (EB), betaine (BT) and L-carnitine (CT) on average daily feed intake and feed conversion ratio of broiler chickens subjected to heat stress (HS Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, P < 0.001 and # tendency.

Discussion
The results showed a decrease in ADFI, ADG, uniformity rate and EPI in broiler chickens exposed to HS, which agreed with other publications on the negative effects of HS on growth performance (Amiri et al. 2019;Abo Ghanima et al. 2020). The poor appetite and lower feed intake of heat-stressed birds might have been linked to their inferior performance, which is a defence mechanism against HS to reduce heat production, thereby maintaining metabolic functions and homeothermy (Yahav 2009;de Souza et al. 2016). According to a previous study (Cheng et al. 2018), HS may have negatively impacted the metabolism of nutrients, like glucose, lipids, oligopeptides and amino acids, by altering mRNA expression of their transporters in relevant organs. This may explain the reduced growth performance of heat-stressed broilers. In addition, HS has been shown to enhance oxidative stress and induce lipid peroxidation and oxidative damage to cellular macromolecules (Akbarian et al. 2016;Khan et al. 2021), which may contribute to elevated mortality rates. The interaction of dietary EB, BT and CT on BW and ADG was significant during the grower phase under high-temperature conditions, suggesting that these three dietary supplements may function synergistically to accelerate growth rate. Supplementation with EB or BT was shown to be effective in alleviating the effects of HSinduced loss in growth performance; however, total recovery was not observed as the TNC group had a greater growth rate than any of the HS-groups studied. Interestingly, the EPI data showed that, when three supplements were fed together in the feed (HS-8 group), the EPI in this group equalled the TNC group, which demonstrated a protective impact on broiler chicks under HS conditions. It has been shown that increasing the degree of emulsification might boost fat absorption (Singh et al. 2009), which may have contributed to the improvement in growth performance observed in the EB-supplemented group. With oil content in the diets ranging from 1.5% to 3.5% during different growth phases, it appeared that the beneficial effects of EB may be due to improved digestion of dietary fat as well as increased energy extraction from fat sources, which resulted in better growth performance under thermal stress conditions. According to the current published literature, no study has monitored the effect of EB supplement on growth performance in broilers challenged with HS from the initial growth stage, but Zangeneh et al. (2018) reported no improving effect of dietary supplemental lysophospholipids on body weight gain and FCR of heat-stressed broilers from 21 to 38 d of age.
In accordance with the results of the present investigation, dietary supplementation with BT resulted in a greater body weight gain and a reduced FCR in broiler chicks subjected to HS (Akhavan-Salamat and Ghasemi 2016; Ghasemi and Nari 2020). This suggested that BT might be useful as an anti-HS supplement in fast-growing commercial broilers. According Table 4. Effects of dietary supplementation with emulsifier blend (EB), betaine (BT) and L-carnitine (CT) on European performance index (EPI) mortality rate of broiler chickens subjected to heat stress (HS) from 0 to 42 d of age. Means within each column followed by different lowercase letters differ significantly (at P < 0.05). n: 6 replicate pens/treatment group (20 broilers/pen). 1 EPI = liveability (%) × live weight (kg) × 100/age (d) × FCR 2 TNC, thermoneutral control group with no supplement in the diet. 3 Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, ***P < 0.001 and # tendency. Alhotan et al. (2021), supplementary BT can help to alleviate intestinal damage in heat-stressed broilers by decreasing inflammatory responses and improving mucosal barrier function, ultimately leading to improvements in performance. When exposed to circumstances that would typically render the cell inactive, betaine is thought of as an effective organic osmoprotectant that keeps cells from being damaged by osmotic stress (Metzler-Zebeli et al. 2009). Accordingly, it is possible that BT supplementation may have helped to boost broiler performance by increasing the birds' capacity to withstand high levels of hypertonicity during periods of osmotic disturbance caused by HS and retain more water inside their bodies. According to the current findings, the positive effects of CT on reducing the negative effects of HS were much less pronounced when compared to the effects of EB and BT and only an increase in growth rate was observed during the grower phase, which was not reflected throughout the entire period. The effects of CT on performance in broiler chickens have been extensively studied in the literature, with a variety of findings. For example, Babazadeh Aghdam et al. (2015) found that heat-stressed broilers fed a diet containing 300 mg/kg CT showed the greatest body weight and lowest FCR throughout the grower phase when compared to unsupplemented chickens. According to Rehman et al. (2017a), the use of CT (500 mg/kg) in the diet of Hubbard and Cobb broiler chickens exposed to HS enhanced feed intake and FCR. In contrast, a study by Murali et al. (2015) found that the performance of Van Cobb chickens was unaffected by the addition of CT to the basal diet (0 and 900 mg/kg). Such inconsistent results might have been due to a variety of factors, including the type (liquid or powder) and inclusion rate of CT, the composition of the basal diet, the sparing impact of CT on its precursor, the age and sex of the birds or other variances in management techniques.
Consistent with the current results, previous studies have shown that broilers subjected to the HS condition exhibited an increase in the blood H/L ratio, indicating an overactivation of the hypothalamic-pituitary-adrenal (HPA) axis, which induces chronic stress (Calefi et al. 2017;Amiri et al. 2019). In response to an activation of the HPA axis, corticotropin-releasing hormone (CRH) is released from the hypothalamus, which is followed by the release of adrenocorticotropic hormone (ACTH) from the anterior pituitary, which stimulates the adrenal cortex to secrete the stress hormone corticosterone (Ramiah et al. 2019;Miri et al. 2022). A hypersensitive HPA axis is certainly harmful, since it reduces growth rate and impairs immunological function (Akhavan-Salamat and Ghasemi 2016). Birds grown in HS on the EB-supplemented diet had significantly lower blood corticosterone levels and tended to have a lower H/L ratio and a higher anti-SRBC antibody response compared to the non-supplemented group, which suggested that these birds were better able to cope with the rising Table 5. Effects of dietary supplementation with emulsifier blend (EB), betaine (BT) and L-carnitine (CT) on stress indicators and antibody responses against sheep red blood cells (SRBC) and Newcastle disease virus (NDV) of broiler chickens subjected to heat stress (HS Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, ***P < 0.001 and # tendency.
x J. YOUSEFI ET AL.

R E T R A C T E D
temperatures in their growing environment. There is limited information available on the effect of lysophospholipids on immunity in heat-stressed broilers; however, Allahyari-Bake and Jahanian (2017) reported that lysophosphatidylcholine supplementation increased antibody titres to infectious bursal disease in broiler chickens reared under normal environmental conditions. As indicated in Table 5, the addition of BT had no effect on the serum corticosterone levels, but decreased the H/L ratio and boosted the primary and secondary antibody responses against SRBC and NDV. In agreement with these results, Ghasemi and Nari (2020) found that dietary BT supplementation enhanced broiler chicken's antibody responses against NDV and infectious bronchitis virus when they were exposed to HS. According to previous studies, the immune-modulating effect of BT has been attributed to the following mechanisms: (1) increasing nitric oxide production by heterophils and macrophages (Klasing et al. 2002) and (2) inhibiting the formation of prostaglandins by liver macrophages, which regulate the production of cytokines (Zhang et al. 1996). This study showed a significant interaction between BT and CT on primary anti-SRBC titres, showing that both supplements stimulated the immune system in HS-challenged broilers in a synergistic manner. It has been shown that CT may help to prevent apoptosis in immune cells, which is one of the most commonly described mechanisms for its favourable effects on immunity (Rehman et al. 2017b). The immunemodulating properties of CT may also be explained by its antioxidative characteristics (Liu et al. 2004).
According to previous reports, HS causes oxidative stress and the generation of free radicals, resulting in lipid peroxidation and oxidative damage to cellular components (Vesco et al. 2017;Khan et al. 2021). In this study, the increase in MDA level under HS conditions, which indicated oxidative stress, was accompanied by an increase in TAC level and SOD activity, which suggested a defence mechanism against oxidative stress, which was consistent with the findings of a previous study (Akhavan-Salamat and Ghasemi 2016). However, both BT and CT administration significantly reduced the oxidative damage caused by HS via increasing TAC and lowering MDA levels, which indicated that these supplements might enhance the antioxidant status of broilers under high temperatures.
Betaine is a methylating agent that has the additional benefit of sparing methionine via the betaine-homocysteine methyl transferase route (Deminice et al. 2015). It was therefore possible that BT, by restoring S-adenosyl methionine, protects cells from oxidative damage by increasing the availability of the substrate required for glutathione formation, thereby protecting them from reactive oxygen species (ROS) and reactive metabolites (Alirezaei et al. 2012). A prior investigation on broilers found that BT might operate as an antioxidant agent in the presence of HS-induced oxidative stress (Chen et al. 2020;Yang et al. 2021), which was consistent with the current results.
L-Carnitine has also been shown to have antioxidant properties in both rats (Sepand et al. 2016) and broiler chickens (Rehman et al. 2017a). By allowing lipid entry Table 6. Effects of dietary supplementation with emulsifier blend (EB), betaine (BT) and L-carnitine (CT) on antioxidant status of broiler chickens subjected to heat stress (HS). Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, ***P < 0.001 and # tendency.
into the mitochondrial matrix, it appears that CT can reduce lipid peroxidation in the body (Derin et al. 2004). Rehman et al. (2017b) demonstrated that the presence of CT could neutralise a variety of radicals, including superoxide, hydroxyl and hydrogen peroxide. According to the results of the current study, the antioxidant status of blood in heat-stressed broilers was only slightly improved by lysophospholipid with tendencies towards increased activity of GPX and CAT enzymes. Only a few studies have been published that explored the effects of lysophospholipids on oxidative stability in poultry, particularly when exposed to high temperatures. A recent study on meat-type ducks found that supplementing diets with lysolecithin boosted blood antioxidant activity when the birds were kept in a thermoneutral environment (El-katcha et al. 2021). Phospholipids have the potential to prevent lipid oxidation by binding prooxidative metals via the negative charges present on the phosphate head group (Cui and Decker 2016). As a result, lysophospholipids, which are phospholipid derivatives, have the ability to boost antioxidant activity.
The digestive system is regarded to be one of the primary target organs and it is highly vulnerable to stressors. Due to HS, pathogen binding receptors on epithelial surfaces may be mimicked, causing damage to gut epithelial tissue integrity (Song et al. 2014). In the present study, apparent damage to villi, shortened VH, deeper CD and a lower VH/CD ratio were observed in heat-stressed broilers, which was consistent with the findings of the previous studies (Amiri et al. 2019;Ghasemi et al. 2021) in cyclic HS conditions. The EB supplementation minimised the unfavourable effects of HS on duodenal morphology, while only a slight beneficial impact was seen for jejunal morphology. One explanation for this was that the duodenum is the site where emulsified lipid droplets enter the intestine and initiate fat breakdown (Majdolhosseini et al. 2019). The increase in the VH and VW of the duodenum is consistent with the findings of El-katcha et al. (2021), who discovered that feeding growing ducks with dietary lysophospholipid supplementation enhanced duodenal VH and VW. In a recent study on meat-type turkeys, supplementation with 2 g/kg of soybean lecithin also enhanced the VH, VH/CD ratio and VSA in the duodenum; nevertheless, with regard to the jejunal morphology, only CD was lowered by lecithin (Nemati et al. 2021). In contrast, supplementing broiler chicken diets with lysolecithins had no influence on the morphology of the duodenum, but did enhance the jejunal VH and VH/CD ratios (Boontiam et al. 2017). Previous studies employed different types of emulsifiers with varied inclusion levels and structures, as well as different sources and rates of dietary fat, which may have contributed to the discrepancies in findings. Despite the lack of evidence about the exact mechanism by which emulsifier supplements Table 7. Effects of dietary supplementation with emulsifier blend (EB), betaine (BT) and L-carnitine (CT) on duodenum morphology of broiler chickens subjected to heat stress (HS Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, ***P < 0.001 and # tendency.

R E T R A C T E D
improve gut morphology, it was possible that, by enhancing micelle formation in the small intestine and minimising intestinal fermentation, emulsifier supplements might have reduced damage to the villi surface (Majdolhosseini et al. 2019). Additionally, it has been reported that lysophospholipids are able to affect the lipid bilayer of cell membranes and reduce the production of inflammatory mediators, thereby enhancing gut integrity and maintaining gut morphology (Chen et al. 2019). This research indicated that there were beneficial effects of BT on intestinal morphology in both the duodenum and the jejunum of broilers subjected to HS, but the positive effect of CT was only associated with a decrease in CD in the jejunum. This suggested that BT supplementation in the diet may be effective in alleviating the suppression of intestine development caused by HS. It is possible that dietary BT could alleviate HS-induced intestinal oxidative damage through a variety of mechanisms: 1) the methyl group donor nature of BT, which could promote intestinal epithelial cell proliferation (Metzler-Zebeli et al. 2009); 2) the osmotic property of BT, which could result in an improved intestinal environment (Saeed et al. 2017); and 3) the antioxidant activities of BT, which could protect against HS-induced intestinal oxidative damage (Alagawany et al. 2022).
In the present study, HS resulted in decreased nutrient digestibility, in terms of AID of dry matter, crude protein and crude fat, as well as the AMEn value. Due to a reduction in blood supply to the digestive tract, HS-exposed chickens are known to be less able to absorb nutrients efficiently (Song et al. 2014;de Souza et al. 2016). However, HS birds have an increasing quantity of maintenance energy, which they employ to expel excess heat in order to maintain a steady body temperature (Akbarian et al. 2016;Sgavioli et al. 2019). As indicated in Table 9, heat-stressed broilers fed with EB-supplemented diets showed significant improvements in AID for crude fat and crude protein as well as the AMEn value. Although only limited data have been reported on the effects of emulsifier supplementation on nutrient digestion in broiler studies conducted under HS conditions, an in vivo study found that lysolecithin increased the digestibility of dry matter, nitrogen and energy in broilers reared under thermoneutral environmental conditions (Jansen et al. 2015). It is possible that the modulation effects in fat emulsification are responsible for the enhanced nutritional digestibility seen with inclusion of bioemulsifiers based on lysophospholipids (Haetinger et al. 2021). According to the findings of the present study, it can be said that lysophospholipid supplementation might increase nutrient availability under HS conditions by improving epithelial morphology in the intestine, as well as enhancing antioxidant status. In this investigation, the inclusion of CT at a concentration of 0.5 g/kg in the diets of heat-stressed broiler chickens had no effect on nutrient digestibility, while the inclusion of dietary BT enhanced the AID for dry matter and crude protein. Previous studies conducted under HS Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, ***P < 0.001 and # tendency.

R E T R A C T E D
conditions have indicated a correlation between BT administration and enhanced CP digestibility in broilers (Liu et al. 2019) and Japanese quails (Ratriyanto and Prastowo 2019). As previously reported, intestinal cells benefit from the osmotic properties of BT, which aid in boosting their survival and activity, ultimately resulting in improved nutrient digestion (Park and Kim 2019).
In conclusion, the three-way interaction test showed that growth rates of broiler chickens can be increased during the grower period if EB, BT and CT are administered concurrently under HS conditions, although this effect did not persist throughout the entire experimental period. Accordingly, a single administration of either EB or BT to heat-stressed broilers may improve the overall performance, immune response and gut morphology. Additionally, the results suggested that EB might be more effective than BT at increasing nutrient digestibility and AME n . In contrast, BT and CT could provide greater antioxidant protection than BT supplementation by activating antioxidant capacities and decreasing MDA levels in broilers exposed to HS conditions.

Disclosure statement
No potential conflict of interest was reported by the author(s). Table 9. Effects of dietary supplementation with emulsifier blend (EB), betaine (BT) and L-carnitine (CT) on apparent ileal digestibility (AID) of nutrients and nitrogen-corrected apparent metabolisable energy (AMEn) of broiler chickens subjected to heat stress (HS Analysed as a completely randomised design by GLM procedure of SAS. Compared with the TNC group within a column. *P < 0.05, **P < 0.01, P < 0.001 and # tendency. xiv J. YOUSEFI ET AL.