A total of 300 healthy broiler fertile eggs (Arbor Acres, Gallus domesticus) with similar body weight (average 59 g) were purchased from a local hatchery (Shandong Liuhe Breeding Co., Ltd, China) and were randomly divided into five groups, with six replicates in each group and 10 eggs in each replicate. All eggs were set in a single incubator, and six replicates (i.e. 6 trays) of each treatment group were randomly distributed on six layers of the incubator. Amniotic cavity injection was performed at 17 days of embryonic age with 0.5 ml chlorogenic acid (Sigma-Aldrich, Missouri, United States) of 4 mg/egg (4CGA) or 7 mg/egg (7CGA) or 10 mg/egg (10CGA) or sham-injected with saline (positive control group, PC), or non-injected group (negative control, NC). The dose of chlorogenic acid was calculated and determined according to a previous report by Chen (Chen et al., 2021a). The fertile eggs were incubated at 37°C under a relative humidity of 60%–70%.

Six healthy chicks with similar body weights from each replicate were selected after hatching and raised under intermittent heat stress treatment (35°C ± 1°C, 8 h/d, 60% relative humidity) during 28–42 days of age. Chicks were subjected to intermittent heat stress to better simulate the diurnal variation of ambient temperature during the summer season. The chicks were managed in environmentally controlled chambers and fed with commercial diets purchased from a local feed factory (Shandong Zhongcheng, China). Feed and water were provided ad libitum.

The in ovo injection of fertile eggs was performed in a sterile environment. A small electric drill was used to drill a hole with 1-mm diameter at the top of the egg air chamber, and a 1-ml syringe with a disposable sterile dental injection needle (0.5 × 38 TWLB L) was used to inject the solution into the amniotic cavity of the egg from the hole, with an injection depth of about 25 mm. The hole was sealed with paraffin after injection. The chlorogenic acid solution was prepared in normal saline and warmed to 37°C for injection.

The hatching time, the number of chicks hatched, and the body weight of hatched chicks were recorded. After hatching, the weekly production performance of chickens during 1–27 d without heat stress and 28–42 d during heat stress were recorded. At the age of 27 d (before heat stress) and 42 d (after heat stress), eight healthy broilers with similar body weight were randomly selected from each treatment for weighing and sampling. After fasting for 6 h, blood was collected from the wing vein using 5-ml heparinized syringe. Plasma samples were obtained after centrifugation at 400 g for 10 min at 4°C and stored at -20°C for further analysis. Thereafter, the chicks were sacrificed by cervical dislocation and exsanguination. The liver, heart, duodenum, jejunum, ileum, breast muscle, and thigh muscle were isolated and weighed. The length of the duodenum, jejunum, and ileum was measured, and about 2 cm of the middle part of each small intestine segment was cut and fixed in 4% paraformaldehyde solution for histomorphological observation. Sections of the duodenum, jejunum, and ileum were sampled, immediately snap-frozen in liquid nitrogen, and stored at -80°C for further analysis.

Plasma activities of alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), creatine kinase (CK), lactate dehydrogenase (LDH) and concentrations of total protein (TP) and albumin (ALB) were measured with commercial diagnostic kits (Sichuan Mike Biotechnology Engineering Co., Ltd, China). Plasma activities of SOD, catalase (CAT), glutathione peroxidase (GPX) and concentrations of MDA, total antioxidant capacity (T-AOC), and glutathione (GSH) were measured spectrophotometrically with commercial diagnostic kits (Nanjing Jiancheng Biological Engineering Research Institute, China).

The intestinal histomorphology was observed with hematoxylin and eosin staining as described by Wang et al. (Huang et al., 2015). Briefly, the intestinal tissue was fixed conventionally in 4% paraformaldehyde, dehydrated, and embedded in paraffin. De-paraffinated sections with a thickness of 4 μm were stained with Harris hematoxylin and eosin (Sigma-Aldrich, Missouri, United States). Sections were examined under a Chongqing Mack Optoelectronic Instrument XDS-1B inverted microscope (Chongqing, China).

Total RNA was extracted from the intestine using commercial kit (Xinsaimei, Suzhou, China). Then the concentration of the RNA was measured by spectrophotometry (Eppendorf, Hamburg, Germany), and the RNA purity was verified by calculating the ratio between the absorbance values at 260 and 280 nm (A260/280 ≈ 1.75–2.01). Next, reverse transcription was performed using total RNA (1 μg) for cDNA synthesis with the kit (Cowin, Beijing, China). The cDNA was amplified in a 20-μL PCR reaction system according to the manufacturer’s instruction (Cowin, Beijing, China). Real-time PCR was performed at the ABI QuantStudio 5 PCR machine (Applied biosystems Inc., Carlsbad, CA). Real-time PCR was performed at 95°C for 30 s of predenaturation, followed by 40 cycles, with each cycle consisting of denaturation at 95°C for 5 s and annealing and extension at 60°C for 30 s. All samples were run in duplicate, and the specificity of the amplification product was verified by the standard curve and dissolution curve. Primers used in this study were designed using Primer 5.0 software and synthesized by Sangon Biotech (Shanghai, China, Table 1). Primers against GAPDH were used as internal controls, and mRNA expression was quantified according to the comparative CT method (2^-△△Ct).

Data analysis was performed by one-way ANOVA using SAS statistical software (SAS version 9.1; SAS Institute Inc., 2004). Differences between means were evaluated using Duncan’s Multiple Range test. All the values were expressed as means ± SE. Orthogonal polynomial contrasts were used for the trend test to examine the linear and quadratic effects of different chlorogenic acid injection concentrations. p < 0.05 means that the differences were significant, and 0.05 < p < 0.1 means that there was a tendency towards significance.

The hatching peak time of the 4CGA group was 4 h earlier than that of NC group and 2 h earlier than that of PC group (Figure 1A). There was no significant differences in hatchability, egg weight at hatching, and body weight of hatching chicks among groups (p > 0.05, Figures 1B–D). There was no significant liner or quadratic effect (P _linear > 0.05; P _quadratic > 0.05).

There was no significant difference in the production performance between chlorogenic acid treatment and control (p > 0.05), but feed/gain at 1–27 d increased with the concentration of chlorogenic acid, with a linear effect (P _linear < 0.05). Although there was no significant difference between groups (p > 0.05), mortality was highest in the NC group (19.44%) and lowest in 4CGA group (8.33%). There was no significant quadratic effect (P _quadratic > 0.05) (Table 2).

At 27 d when there was no heat stress, the heart proportion of broilers was significantly increased in 10CGA (0.48%) group than that in PC group (0.38%) (p < 0.05), the jejunum proportion had a linear effect (P _linear < 0.05), and no changes were observed for other organs (p > 0.05). In ovo feeding of chlorogenic acid had no significant effect on all organs weighed in the heat-stressed broilers at 42 d of age (p > 0.05). There was no significant quadratic effect (P _quadratic > 0.05) (Table 3).

For 27 d-broilers without heat stress, 4CGA group (277.68 mg/cm) had a higher jejunum index than NC group (243.76 mg/cm) (p < 0.05). Compared with NC (140.15 mg/cm) or PC group (140.34 mg/cm), the ileum index was significantly increased in the 4CGA group (163.09 mg/cm) (p < 0.05), while no changes were found in the intestinal organ index for 42 d-heat stressed broilers (p > 0.05). There was no significant liner or quadratic effect (P _linear > 0.05; P _quadratic > 0.05) (Table 4).

For 42 d-heat stressed broilers, AST activity in the 4CGA group (253.88 U/L) decreased significantly compared with NC group (328.50 U/L) and PC group (309.75 U/L) (p < 0.05). The ALP content in all chlorogenic acid treated groups (average 1232.35 U/L) were significantly lower than that in NC group (2110.75 U/L) (p < 0.05). However, no changes were observed for other biochemical parameters (p > 0.05). There was no significant liner or quadratic effect (P _linear > 0.05; P _quadratic > 0.05) (Figure 2).

For 42d-heat stressed broilers, the MDA content of 4CGA group (0.58 nmol/ml) and 7CGA group (1.31 nmol/ml) decreased significantly compared with NC group (1.86 nmol/ml) and PC group (1.93 nmol/ml) (p < 0.05). Also, the MDA content of 4CGA group (0.58 nmol/ml) was significantly lower than that of the 7CGA group (1.31 nmol/ml) (p < 0.05). The MDA content of the different injection groups had a quadratic effect, with the 4CGA group having the lowest MDA content (P _quadratic < 0.05). Compared with the NC group (SOD 10.97 U/mL; GPX 1585.95 mol/L), SOD and GPX activities in all chlorogenic acid-treated groups (average SOD 14.15 U/mL; GPX 1973.57 mol/L) were significantly higher (p < 0.05). The CAT activity in 4CGA group (6.34 U/mL) was significantly higher than that in control groups (average 4.66 U/mL) and other chlorogenic acid-treated groups (average 4.05 U/mL) (p < 0.05). However, there were no significant changes for T-AOC level and GSH content (p > 0.05). There was no significant liner effect (P _linear > 0.05) (Figure 3).

For 42d-heat stressed broilers, the villus height of the duodenum in the 4CGA group (1743.57 μm) was significantly higher than that in the NC group (1447.81 μm) (p < 0.05, Figures 4A, D). In the jejunum, the ratio of villus height to crypt depth was significantly higher in the 4CGA group (9.36) than in the NC group (7.56) (p < 0.05, Figures 4B, F). In the ileum, the ratio of villus height to crypt depth in the 4CGA group (5.73) was higher than that in the PC group (4.40) (p = 0.0971, Figures 4C, F). There were no significant changes for crypt depth (p > 0.05, Figure 4E). There was no significant liner or quadratic effect (P _linear > 0.05; P _quadratic > 0.05).

In the duodenum of 42 d-heat stressed broilers (Figure 5A), the mRNA expressions of Nrf2 and H O -1 were significantly higher in the 4CGA group (1.56; 1.43) compared with NC (Yan et al., 2017Yan et al., 2017) and PC groups (1.1; 1) (p < 0.05). Also, the mRNA expression of Nrf2 in the 4CGA group (1.56) was significantly higher than that in the 7CGA (1.14) and 10CGA (0.92) (p < 0.05) groups. The mRNA expressions of glutathione synthetase (GSS) and SOD1 in the 4CGA group (1.37; 1.59) and 7CGA group (1.31; 1.54) were significantly higher than NC (Yan et al., 2017Yan et al., 2017) and PC groups (0.96; 0.9) (p < 0.05), while the mRNA expression of SOD2 tended to increase (p = 0.0933). Compared with the PC group (0.7; 0.88), the mRNA expressions of glutathione reductase (GSR) and GPX7 in all chlorogenic acid-treated groups (average 1.27; 1.59) were significantly higher (p < 0.05). More so, the mRNA expression of GPX7 in 7CGA group (1.67) and 10CGA group (1.57) was significantly higher than that in NC group (Yan et al., 2017) (p < 0.05). The mRNA expressions of Nrf2 and SOD1 in different injection groups had a significant quadratic effect (P _quadratic < 0.05), and the 4CGA group was the highest. Also, the mRNA expressions of GSR and GPX7 in different injection groups had a linear effect (P _linear < 0.05), and increased with the concentration of chlorogenic acid. The mRNA expression of CAT in the 4CGA group (2.01) was significantly higher than that in the NC group (Yan et al., 2017) (p < 0.05). However, there were no significant differences in mRNA expression of antioxidant-related genes between chlorogenic acid-treated groups and control groups (p > 0.05) in jejunum and ileum (Figures 5B,C).

Compared with NC (Yan et al., 2017) and PC group (1.01), the mRNA expression of ten eleven translocation 1 (Tet1) was significantly higher (p < 0.05) in all chlorogenic acid treated groups (average 1.68) in the duodenum of 42 d-heat stressed broilers (Figure 6A). The mRNA expression of Tet2 and DNA-methyltransferase three alpha (DNMT3A) in the 4CGA group (2.41; 1.88) was also significantly higher than NC (Yan et al., 2017Yan et al., 2017) and PC group (1.37; 1.07) (p < 0.05), while the mRNA expression of Tet2 in the 4CGA group (2.41) was significantly higher than that in 10CGA group (1.45) (p < 0.05). Compared with NC group (Yan et al., 2017), the mRNA expression of DNMT1 in the 4CGA group (1.66) tended to increase (p = 0.0557). However, there were no significant differences in the mRNA expression of Tet3 and DNMT3B between chlorogenic acid-treated groups and the control groups (p > 0.05); The mRNA expressions of Tet2 in different injection groups had a significant quadratic effect (P _quadratic < 0.05), and the 4CGA group was the highest. Also, the mRNA expressions of Tet1 in different injection groups had a linear effect (P _linear < 0.05), and increased with the concentration of chlorogenic acid. In jejunum and ileum (Figures 6B,C), there was no significant difference in the mRNA expression of DNA methylation-related enzymes between chlorogenic acid treated groups and control groups (p > 0.05); The jejunal mRNA expressions of Tet3 in the different injection groups had a quadratic effect (P _quadratic < 0.05), such that the 4CGA group had the lowest mRNA expression. In addition, the jejunal mRNA expression of DNMT1 had a linear effect (P _linear < 0.05) in different injection groups, thus, it was increased with increasing concentration of chlorogenic acid.

Although several studies have discussed the importance of adding chlorogenic acid to broiler diets, the present experiment studied the effects of chlorogenic acid administration via in ovo feeding for the first time. The results showed that in ovo feeding of chlorogenic acid did not have an adverse effect on the hatching performance of broilers, thus ascertaining the reliability of the in ovo feeding technique.

Heat stress disturbs the body’s homeostasis, resulting in adverse effects on physiology, immunity, and intestinal morphology, with a consequent decrease in production and increased mortality of poultry (Xu et al., 2018). Chlorogenic acid has been previously used as an antioxidant due to its antioxidant properties. In the past few years, researchers have studied the dietary addition of chlorogenic acid, but its effects on growth performance were inconsistent. Chen et al. (Chen et al., 2021b) showed that adding chlorogenic acid to the diet for 2 weeks had no significant effect on the average daily gain and average daily feed intake of chicks with acute heat stress. However, Zhang et al. (2020b) found that after adding chlorogenic acid to the diet, the average daily gain was increased, and the average daily feed intake and feed/gain were significantly reduced for broilers infected by Clostridium perfringens during 1–21 days of age. Zhao et al. (2019) also showed that the dietary addition of Eucommia ulmoides leaf extract (which contained mainly chlorogenic acid) increased the average daily gain of broilers, and reduced the feed/gain, but had no significant effect on the average daily feed intake. In this study, there was no significant difference in production performance. During the whole feeding period, the mortality rate of 4CGA group was the lowest at 8.33%, while that of the control group was the highest at 19.44%. Chlorogenic acid had no effect on the growth performance of broilers, which may be due to the small population of experimental animals and the differences in experimental conditions, animal species, and mortality rate. The association between these factors needs to be further confirmed in future studies. Evidently, in ovo feeding of chlorogenic acid was beneficial to the development of the heart, jejunum, and ileum of postnatal broilers without heat stress.

The small intestine is the most important site for animals to digest and absorb nutrients (Lv et al., 2021). Intestinal mucosal function and health can be reflected by intestinal morphological indices, including villus height, crypt depth, and the ratio of villus height to crypt depth (Jha et al., 2019b). A higher villus facilitates nutrient absorption by the intestine, and a shallow crypt is beneficial to the maturation of intestinal epithelial cells (van der Hulst et al., 1998). An increased villus height/crypt depth ratio reflects an improvement in the intestinal mucosal structure and the enhancement of intestinal mucosal absorption function (Williams et al., 2015). Zhang et al. (Zhang et al., 2018b) showed that the dietary addition of chlorogenic acid had a positive effect on the intestinal morphology of weaned piglets, as demonstrated by the increased villus height and the villus height/crypt depth in duodenum and jejunum, whereas, the duodenal crypt depth was decreased. Similarly, Ruan et al. (2014) showed that the addition of chlorogenic acid can improve the intestinal morphology of the jejunum and ileum in lipopolysaccharide-treated rats. In the present study, 4CGA increased the villus height of the duodenum and the villus height/crypt depth of the jejunum and ileum compared to the control group after 2 weeks of heat stress. These results showed that in ovo feeding of chlorogenic acid had a positive effect on the intestinal morphology of postnatal broilers suffering from heat stress.

Blood biochemical indicators generally include blood glucose, liver function, and renal function indicators, which altogether reflect the health status of the body (Liao et al., 2007). Total protein mainly consists of albumin and globulin, and their ratio indicates protein metabolism and immune function (Stohl et al., 2019; Yoshino et al., 2019). Blood ALT, AST, and ALP are usually used to evaluate the degree of liver oxidative damage, and their decreased activity is indicative of an improved redox status (Dalia et al., 2017; Pandit et al., 2022). LDH and CK mainly exist in cardiomyocytes and can be released when cardiomyocytes are damaged. Therefore blood LDH and CK are important indicators to measure myocardial injury (Tang et al., 2018; Yin et al., 2020). The present results showed that, after 2 weeks of heat stress, the blood activity of AST and ALP were reduced in CGA-treated groups, especially in the 4CGA group. This suggests that in ovo feeding of chlorogenic acid can alleviate the oxidative damage of postnatal heat stress in the liver.

Heat stress upset the balance between oxidation and antioxidant systems thereby inducing lipid peroxidation, and ROS overproduction. As such, heat stress is widely used to establish both in vivo and in vitro models of oxidative stress (Huang et al., 2015; Liu et al., 2016; Gupta et al., 2017; Zhang et al., 2018a; McGarry et al., 2018; Li et al., 2019; Yu et al., 2019), as presented in this study.

Antioxidant enzymes, especially SOD and CAT can inhibit the accumulation of reactive oxygen species and protect cells from oxidative stress. T-AOC reflects the overall antioxidant capacity of the organism (Chen et al., 2012). GSH is an important antioxidant that scavenges hydrogen peroxide and ROS (Wang et al., 2015), while GPX is a powerful free radical scavenger and an important enzyme in the antioxidant defense system (Zhang et al., 2018b). MDA is the oxidation product of lipid peroxide, and the MDA content can directly reflect the level of lipid peroxide and the degree of lipid peroxidation caused by free radicals (Lv et al., 2021). As a polyphenol, chlorogenic acid acts as an effective neutralizer of ROS by scavenging free radicals (Yao et al., 2019). Studies have shown that chlorogenic acid can increase the serum level of T-AOC, promote the activities of SOD and GPX, and inhibit MDA content, effectively alleviating oxidative stress caused by Clostridium perfringens infection (Zhang et al., 2020b). Consistently, Zhou et al. (2016) found that chlorogenic acid can increase the activities of GPX and SOD and the level of T-AOC in H₂O₂-induced intestinal mitochondrial injury in rats. The molecular mechanism of antioxidant activity of chlorogenic acid may depend on its specialized chemical structure. Chlorogenic acid can provide hydrogen atoms to reduce free radicals, inhibit oxidative reactions, and then rapidly stabilize resonance after oxidation (Liang and Kitts, 2015). In this study, after 2 weeks of heat stress, the plasma oxidation product MDA was significantly decreased and showed a significant quadratic effect, with the 4CGA group having the lowest MDA levels. The activities of antioxidant enzymes including SOD, GPX and CAT were significantly increased by the 4CGA group, indicating that in ovo feeding of chlorogenic acid would play an antioxidant role via increasing the activities of antioxidant enzymes.

We further explored the underlying mechanism of chlorogenic acid’s antioxidant effect. Nrf2 is an important transcription factor that regulates the gene expression of antioxidant enzymes (Kim et al., 2010). The Nrf2 pathway exerts an antioxidant effect by up-regulating a series of endogenous protective genes (Wu et al., 2017). HO-1 is one of the key cytoprotective enzymes in cell defense. It is a key protein and plays an important role in cell adaptation to oxidative stress caused by a variety of pathological events (Hwang and Jeong, 2010; Ibáñez et al., 2014). Many studies have confirmed that the up-regulation of HO-1 contributes to cellular defense against oxidative damage (Lee et al., 2012; Xu et al., 2015; Han et al., 2017). Liu et al. (2020) found that chlorogenic acid promoted the protein expressions of Nrf2 and HO-1, increased SOD activity, and reduced MDA concentration by activating the Nrf2 pathway in rats with cerebral ischemia-reperfusion injury. The present study showed that, after 2 weeks of heat stress, the mRNA expressions of Nrf2 and antioxidant enzyme H O -1, GSS, GSR, GPX7, SOD1, and CAT in the duodenum of 4CGA group were significantly increased, while the mRNA expression of SOD2 also tended to increase. In addition, the mRNA expressions of GSR and GPX7 in different injection groups showed a linear effect. The mRNA expressions of Nrf2 and SOD1 had a significant quadratic effects, and the 4CGA group was the highest. These results showed that chlorogenic acid activated the Nrf2 pathway and promoted the mRNA expressions of Nrf2 and H O -1, thus improving the expression of their downstream antioxidant enzymes. Altogether, the results indicate that in ovo feeding of chlorogenic acid improved the antioxidant capacity of postnatal broilers subjected to heat stress by regulating the Nrf2 pathway.

Epigenetics refers to heritable changes in gene expression that are not caused by changes in DNA sequences (Holliday, 1987). It mainly includes DNA methylation, RNA interference, and histone modification (Egger et al., 2004), among which DNA methylation is the most famous (Yin et al., 2013). DNA methylation is catalyzed by DNA methyltransferases (DNMTs) at cytosine in the CpG sequence. During DNA replication, a methyl is added to the cytosine in the complementary DNA strand of methylated cytosine by DNMT1, and a methyl is added to the unmodified cytosine in the CpG sequence by DNMT3A and DNMT3B (Li et al., 1992; Okano et al., 1999). In addition, methylated cytosine is oxidized by Tet1, Tet2, and Tet3 of the Tet family, especially in the early stage of mouse embryonic development, and transformed into 5-hydroxymethylcytosine (Tahiliani et al., 2009; Inoue and Zhang, 2011). Genomic analysis showed that the chicken genome was rich in methylated cytosine, and contained homologous genes of mouse DNMT1, 3A, 3B; and Tet1, 2 and 3 (Li et al., 2011; Tada et al., 2021). In this study, the mRNA expressions of Tet1, Tet2, and DNMT3A in the duodenum of the 4CGA group were significantly increased, while the mRNA expression of DNMT1 in the 4CGA group also tended to increase. The mRNA expressions of Tet2 in the different injection groups showed a quadratic effect with the 4CGA group having the highest expression. This indicates that the effects of in ovo feeding of chlorogenic acid may be related to the epigenetics of broilers.