A total of 360 Ross 308 classic chicks (1 d old) with an average body weight (BW) of 42.5 g ± 3 were purchased from a commercial hatchery and utilized in the current experiment. Upon arrival to the Animal Physiology Lab (The University of Jordan), the chicks were brooded in electrical battery cages and provided with the required temperature, feed, and water. At 21 d of age, chicks were randomly assigned to six treatments; each treatment had six replicates (n=10 chicks/replicate). Chicks were assigned to one of the following treatments: (1) a diet containing 40 mg kg-1 of Zn from an organic source (Availa Zn^®, Zinpro Corporation) +80 mg kg-1 of Zn from a mineral premix (a total of 120 mg kg-1 of Zn) and rearing under thermoneutral (TN) conditions (Ctrl; n=60); (2) a diet containing 60 mg kg-1 of Zn from an organic source (+50 % of the organic Zn level from the Ctrl treatment) +80 mg kg-1 of Zn from a mineral premix (a total of 140 mg kg-1 of Zn) and rearing under TN conditions (50 TN; n=60); (3) a diet containing 80 mg kg-1 of Zn from an organic source (+100 % of the organic Zn level from the Ctrl treatment) +80 mg kg-1 of Zn from a mineral premix (a total of 160 mg kg-1 of Zn) and rearing under TN conditions (100 TN; n=60); (4) a diet containing 40 mg kg-1 of Zn from an organic source + 80 mg kg-1 of Zn from a mineral premix (a total of 120 mg kg-1 of Zn) and exposure to cyclical heat stress (HS) at the age of 35 d for 3 consecutive days (CHS; n=60); (5) a diet containing 60 mg kg-1 of Zn from an organic source (+50 % of the organic Zn level from the Ctrl treatment) +80 mg kg-1 of Zn from a mineral premix (a total of 140 mg kg-1 of Zn) and exposure to cyclical HS at the age of 35 d for 3 consecutive days (50 HS; n=60); and (6) a diet containing 80 mg kg-1 of Zn from an organic source (+100 % of the organic Zn level from the Ctrl treatment) +80 mg kg-1 of Zn from a mineral premix (a total of 160 mg kg-1 of Zn) and exposure to cyclical HS at the age of 35 d for 3 consecutive days (100 HS; n=60). The level of Zn in the Ctrl diet was similar to the Zn level used in other studies (Moghaddam and Jahanian, 2009; Jahanian and Rasouli, 2015) and the recommended Zn level for broilers (Council, 1994). All procedures were approved by the University of Jordan Animal Care and Use Committee. The ambient temperatures of the TN and HS conditions are shown in Fig. 1.

During the whole experiment, diets were prepared at Al-Estesharia for Poultry and Feed Co. Amman. The diet ingredients and composition are illustrated in Table 1. Different bag colors were used to allow for visual differentiation between the three diets in an attempt to prevent mixing errors. The broiler diets were formulated to meet or exceed the nutrient and energy predicted requirements (Council, 1994)

The data in the current experiment were collected in two phases: the feeding phase and the HS phase. The feeding phase of the experiment lasted for 14 d (22 to 35 d of age) where chicks were fed their respective diets ad libitum and feed intake and body weights were recorded on a daily and weekly basis, respectively. Feed was provided daily at 08:00 LT and orts were recorded daily prior to the AM feeding. During the feeding phase, all chicks were under TN conditions (22.6±1.8 ∘C; 52±7 % relative humidity, RH). Following the feeding phase, chicks of the HS treatments (HS, 50 HS, and 100 HS) were moved into an environmentally controlled chamber (at the age of 36 d) to conduct the HS phase. The HS phase lasted for 3 consecutive days (days 36 to 38); heat stress conditions were cyclical and consisted of 19.8±1.1 ∘C and 45.3 % relative humidity, from 15:00 to 11:00 LT, 33.7±0.6 ∘C and 40.9 % RH, from 11:00 to 15:00 LT (Fig. 1). Ambient temperature was controlled, but humidity was not governed, and both parameters were recorded every 5 min by a data logger (Lascar EL-USB-2-LCD, Erie, PA) in the chamber.

During HS phase, Tr were measured twice a day (after the third and fourth hours of HS) in all treatments using a standard digital thermometer (GLA M700 Digital Thermometer, San Luis Obispo, CA).

At sacrifice (day 39), fresh blood samples were collected from the wing vein and assayed immediately using an epoc^® blood analysis system (Siemens Healthineers; GmbH, Germany), which measured blood glucose and ionized calcium (iCa).

On day 39, chicks were euthanized by cutting the jugular vein. Organs and tissues were harvested immediately after sacrifice. Liver, heart, spleen, and intestine weights were recorded. Intestinal tissues were harvested within 5 min following euthanasia. The duodenum was collected 5 cm distal to the pyloric sphincter. The jejunum was collected at 15 cm before the Meckel's diverticulum. The ileum was collected at 15 cm after the Meckel's diverticulum. All the intestinal segments (10–20 cm, approximately) were flushed with cold phosphate buffer saline to remove intestinal content and fixed in 10 % neutral buffered formalin for later histological analysis.

For histological analysis, formalin-fixed duodenum, jejunum, and ileum samples were submitted to the Histology Lab at the School of Medicine, University of Jordan, within 1 week of euthanasia for sectioning and periodic acid–Schiff (PAS) staining. Two chicks were chosen from each replicate to obtain intestinal samples (12 chicks per treatment). Three slides per chick per tissue were generated. Using a microscope (Leica DM750 microscope) with an attached camera, five images per section of intestine were obtained at 100× magnification. All image processing and quantification was done using Leica LAS EZ 3.4 software. Results of the five images per intestinal section were condensed into a single average per chick.

Villus height was measured from the tip to the level of the villus–crypt interface utilizing the segmented line tool along the villus midline. Villus width was measured using a single line at mid-villus height. Crypt depth was measured with a single line from the villus–crypt interface to the laminae propria and muscularis mucosa junction. A mucosal surface area estimate was obtained using the mucosal-to-serosal amplification ratio M, as previously reported by Kisielinski et al. (2002), where 1M=villus width×villus length+villus width2+crypt width22-villus width22villus width2+crypt width22.

The effect of treatment (Ctrl, 50 TN, 100 TN, CHS, 50 HS, and 100 HS) on variables with single measures was analyzed using PROC MIXED of SAS (9.4 Inst. Inc., Cary, NC). For variables with multiple measurements over time, a repeated measures analysis with an autoregressive covariance structure and day as the repeated effect was used to determine the effects of treatment, day, and treatment–day interaction. For both single and repeated measure variables, preplanned contrasts were evaluated: TN vs. HS, TN vs. TN, and HS vs. HS using the CONTRAST statement of SAS. Additional contrasts were done for specific treatments or their combinations when relevant and their P values are reported if significant. For each variable, the residuals' distribution was tested for normality, and logarithmic transformation was performed when necessary. Results are reported as least squares means and considered different when P≤0.05 and tend to differ if P<0.10.

During the heat stress challenge, Tr was markedly increased in HS-exposed chicks (1.7 ∘C; P<0.01; Fig. 2) compared to TN chicks, which confirms that HS was successfully implemented in the current project. Initial BW (day 21) was similar in all treatments (P=0.21; Table 2). However, final BW (fBW) was increased in TN relative to the HS chicks (2 %; P=0.02; Table 2). Regardless of the Zn inclusion rate, there was a similar average daily feed intake (ADFI) for both TN and HS chicks (Table 2). Overall, HS chicks had a slightly decreased average daily gain (ADG; 2.4 %; P=0.04; Table 2) compared to TN chicks. On the other hand, the feed conversion ratio (FCR) was increased in HS chicks when compared to TN chicks (2 %; P=0.03; Table 2). Interestingly, CHS and 100 HS FCR were similar to the FCR in Ctrl chicks (Table 2). In addition, HS chicks tended to have increased dressing percentage (DP; ∼2 %; P=0.08; Table 2) compared with TN chicks.

Blood ionized calcium decreased in HS relative to TN chicks (4 %; P<0.01; Fig. 3). Relative to TN chicks, blood glucose levels in HS chicks measured directly after euthanasia were increased (16 %; P<0.01; Fig. 4).

The effects of the organic zinc inclusion rate or the environmental conditions on the internal-organ weight and their percentages relative to the body weight were also calculated and recorded (Table 3). Liver weight was decreased (13 %; P<0.01; Table 3) in 50 HS chicks relative to Ctrl. However, the liver weight as a percentage of BW was decreased (12 %; P<0.01; Table 3) in CHS and 100 HS chicks relative to 50 TN, 100 TN, and 50 HS chicks. There were no treatment differences in heart weight or heart weight as a percentage of BW (Table 3). Intestinal weight was decreased (7 %; P=0.02; Table 3) in HS birds relative to TN birds. Moreover, HS birds tended to have decreased (12 %; P=0.08; Table 3) intestinal weight relative to Ctrl birds. No differences were found in intestinal weight between 50 HS, 100 HS, and Ctrl birds (Table 3), and the intestinal weights relative to BW were similar in all treatments. Compared with Ctrl birds, chicks exposed to HS had a similar spleen weight; however, it was found that spleen weight and spleen weight as a percentage of BW was increased (28 % and 29 %, respectively; P<0.05; Table 3) in 50 HS chicks relative to CHS and 100HS chicks.

Compared with TN chicks, HS chicks had decreased villus height 6 % and 3 % in the duodenum and jejunum, respectively (P<0.01; Table 4), while in the ileum, villus height increased by 16 % in 50 HS relative to CHS (P<0.01; Table 4). Villus height in the duodenum was increased by 13 % and 6 % in 50 HS and 100 HS chicks, respectively, relative to the Ctrl chicks (P<0.05; Table 4) and by 16 % and 9 % in 50 HS and 100 HS chicks, respectively, relative to the CHS chicks (P<0.05; Table 3). Similarly, jejunum villus height was increased by 16 % and 28 % in 100 HS chicks relative to the Ctrl and CHS chicks, respectively (P<0.05; Table 4). Duodenum and jejunum villus width in TN and HS were similar. Relative to TN chicks, villus width was decreased by 20 % in HS (P<0.01; Table 4). Mucosal surface area (M index) was increased by 12 % and 34 % in 100 HS chicks relative to the CHS in the duodenum and jejunum, respectively (P<0.05; Table 4). In the ileum, the M index was increased by 12 % in 50 HS relative to CHS (P<0.01; Table 4).