Small Ruminant Research 203 (2021) 106496 Contents lists available at ScienceDirect Small Ruminant Research journal homepage: www.elsevier.com/locate/smallrumres Metabolic and behavior responses of lactating goats under heat stress Ahmed A.K. Salama *, Soufiane Hamzaoui, Elena Albanell, Xavier Such, Gerardo Caja Group of Research in Ruminants (G2R), Department of Animal and Food Sciences, Universitat Autònoma de Barcelona, 08193, Bellaterra, Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Heat stress (HS) negatively affects animal performance, but little is known about energetic metabolism and Heat stress behavior changes in dairy goats under HS conditions. Eight multiparous Murciano-Granadina dairy goats (43.3 ± Lactation 1.6 kg BW; 2 ± 0.04 L milk/d; 81 ± 3 days of lactation) were kept in metabolism cages and randomly assigned to Energetic metabolism two treatments varying in the temperature humidity index (THI). The design was crossover (two 28-d periods), Behavior Dairy goats and treatments were: 1) thermal neutral (TN; 15–20C, 40–45% humidity, THI = 59–65), and 2) heat stress (HS, 12 h/d at 37◦ C and 40%, and 12 h/d at 30◦ C and 40%, THI = 86 and 77, respectively). Jugular silicon catheters were fitted, and glucose tolerance test (0.25 g/kg BW), insulin tolerance test (4.6 μg/kg BW) and epinephrine challenge (2 μg/kg BW) were conducted. Before and after each metabolite administration, blood samples were collected for the analysis of insulin, glucose, and nonesterified fatty acids (NEFA). Also, behavior variables (position changes, duration of remaining standing, as well as eating and drinking bouts and duration) were observed at day 3 of each period by video cameras. Heat stress reduced (P < 0.01) feed intake (− 29%), milk yield (− 10%), milk fat (–12%), milk protein (–14%), and milk casein (–13%). Goats in both groups had similar blood NEFA after insulin administration, but NEFA values were greater (P < 0.05) in TN than HS goats after epinephrine infusion. The HS goats secreted lower (P < 0.05) amounts of insulin than TN goats in response to the glucose tolerance test. Additionally, TN and HS goats had similar number of eating bouts, but the duration of each bout was shorter in HS than in TN. Also, HS had greater number of drinking bouts with no differences in drinking bout durations between groups. In conclusion, body lipid tissue of HS goats became more resistant to lipolysis, making them unable to mobilize body fat reserves despite the negative energy balance. In addition, the reduction in feed intake by HS was because of the shorter time of eating bouts, whereas the greater water consumption was related to the increase in drinking bouts. 1. Introduction Currie, 1980). The reduction in insulin action allows adipose tissue lipolysis and the mobilization of nonesterified fatty acids (Bauman and Heat stress (HS) negatively impacts productivity and health of live­ Currie, 1980). However, in case of heat-stressed dairy cows (Rhoads stock (Baumgard and Rhoads, 2013; Salama et al., 2014; Sejian et al., et al., 2009) and ewes (Mehaba et al., 2021), blood nonesterified fatty 2018). Adaptations to cope with HS include, but not limited to, greater acids (NEFA) levels do not vary compared to thermal neutral (TN) ani­ sweating rate, elevated respiration rate, vasodilation with increased mals despite the reduced DMI. The lack of an elevated NEFA response blood flow to the skin surface, reduced metabolic rate, decreased dry happens even though acute HS causes an increase in blood levels of matter intake (DMI), and altered water metabolism (reviewed by cortisol, norepinephrine, and epinephrine, which are catabolic hor­ Baumgard and Rhoads, 2013; Salama et al., 2016; Sejian et al., 2018). mones that normally stimulate lipolysis and adipose mobilization Circulating T3 and T4 also decline by up to 25% (Silanikove, 1992), (Collier and Gebremedhin, 2015). Hormonal challenges and glucose which is consistent with the decrease in metabolic rate, DMI, and pro­ tolerance test have been applied in heat-stressed dairy cows (Wheelock duction under HS conditions. et al., 2010) and dairy sheep (Mehaba et al., 2021) to understand the Heat-stressed cows (Baumgard and Rhoads, 2013) and goats (Ham­ metabolic changes induced by HS. However, little is known on the zaoui et al., 2020, 2021) suffer negative energy balance (EB). One response of heat-stressed dairy goats to such metabolic challenges. typical response to negative EB is a reduction in circulating insulin Behavioral adaptive mechanisms to HS include alterations in coupled with a decrease in systemic insulin sensitivity (Bauman and standing and lying times as well as the frequency of drinking, * Corresponding author. E-mail address: ahmed.salama@uab.cat (A.A.K. Salama). https://doi.org/10.1016/j.smallrumres.2021.106496 Received 13 April 2021; Received in revised form 3 August 2021; Accepted 4 August 2021 Available online 9 August 2021 0921-4488/© 2021 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 Table 1 humidity index (THI) values. The environmental conditions were ther­ Ingredients, chemical composition, and nutritive value of the total mixed mal neutral (TN; 15–20C, 40–55% relative humidity, THI = 59–65), and ration used for dairy goats. heat stress (HS; 12-h day at 37◦ C, 40 ± 5% relative humidity, THI = 86; Item Total mixed ration and 12-h night 30◦ C, 40 ± 5% relative humidity, THI = 77). For HS Ingredient, % goats, the temperature was increased gradually (1 d at 25◦ C, 1 d at Alfalfa hay 70.0 30◦ C). Under both conditions, photoperiod was 12-h light (0800–2000 Barley, ground 14.4 h) and 12-h dark (2000–0800 h). Goats had a 3-wk pre-experimental Corn, ground 8.4 period under TN conditions for the adaptation to the diet and meta­ Soybean hulls 4.3 bolic cages. All goats remained at TN conditions for 7 days after the first Soybean meal 2.5 Molasses 0.30 period (day 1–28), and during the second period (day 36–63), goats Sodium bicarbonate 0.03 were switched to the opposite treatment. Salt 0.03 Goats were in 2 adjacent rooms with identical feeding conditions and Dicalcium phosphate 0.01 facilities. Throughout the experiment (December to March), the room Calcium carbonate 0.01 Vitamin-mineral complex 0.02 temperature for TN goats was maintained using electric heaters (3.5 kW; General Electric, Barcelona, Spain) equipped with a thermostat. The Component, % room of HS goats was provided with a temperature and humidity con­ Dry matter 89.3 trolling system (Carel Controls Ibérica, Barcelona, Spain). A continuous Crude protein 17.5 90 m3/h air turnover was maintained throughout the experiment. Ether extract 1.93 Data of environmental temperature and humidity were recorded Neutral detergent fiber 43.8 Acid detergent fiber 27.0 every 10 min throughout the experiment by a data logger (Opus 10, Lufft, Fellbach, Germany). The THI values were calculated according to Nutritive value1 NRC (1971). UFL,2 /kg 0.84 Goats were milked once daily (0800 h) with a portable milking NEL, Mcal/kg 1.47 machine (Westfalia-separator Ibérica, Granollers, Spain) provided with PDI,3 g/kg 89.5 PDIA,4 g/kg 40.1 recording jars (3 L ± 5%). Goats were individually fed ad libitum a total Ca, g/kg 8.24 mixed ration formulated according to INRA (2018). Feed was offered P, g/kg 2.60 once daily at 0900 h at 115% of expected intake. Ingredients, chemical 1 Calculated according to the French National Institute for Agricultural composition, and nutritive value of the ration are shown in Table 1. Research (Institut National de la Recherche Agronomique (INRA, 2018). Drinking water was freely available at room temperature. 2 Feed units for lactation (1 UFL = 1.76 Mcal of NEL). 3 Protein digestible in the intestine from dietary and microbial origin. 2.2. Performance variables 4 Protein digestible in the intestine from dietary origin. Daily rectal temperatures and respiration rates were recorded at rumination, urinating and defecating (Darcan et al., 2008; Sejian et al., 0800, 1200, and 1700 h. The rectal temperature was measured by a 2018). Despite the numerous published studies on the impact of HS on digital clinical thermometer (Model ICO Technology “mini color” Bar­ production and reproductive variables, little is known on the detailed celona, Spain). The respiration rate was measured by counting the in­ changes in animal behavior caused by HS, especially in dairy goats. halations and exhalations for 60 s with the aid of a digital chronometer We hypothesized that HS would result in significant changes in en­ (Model 900,400; Deltalab, Barcelona, Spain). ergetic metabolism and behavior in dairy goats. Evaluating these al­ Goats were weighted on two consecutive days at the start and end of terations would help in understanding better the performance under HS each experimental period using a digital scale (Tru-Test AG500 Digital conditions as well as establishing effective strategies to alleviate its Indicator, Auckland, New Zealand; accuracy ±20 g) to measure the negative effects. The objectives of the current study were to evaluate the change in BW. Additionally, BW values were used to calculate net en­ response of dairy goats to glucose, insulin, and epinephrine challenges ergy balance using the following equation: energy balance = net energy under HS conditions. Additionally, video recording was used to monitor intake – (NEM + NEL). goat behavior (eating, drinking, position changes) throughout the day Net energy for maintenance (NEM) was calculated using the under TN and HS conditions. following equation: NEM = (0.0406 × BW0.75) according to INRA (2018). Maintenance costs were increased by 30% for HS goats as rec­ 2. Materials and methods ommended by NRC (2001). Net energy for lactation (NEL) was calcu­ lated by using the following equation: NEL = milk yield × [0.389 + 2.1. Animals and management conditions 0.0052 (fat, g/kg – 35) + 0.0029 × (protein, g/kg – 31)] according to INRA (2018). The procedures used in the current study were approved by the Feed intake and water consumption were measured daily throughout Ethical Committee of Animal and Human Experimentation of the Uni­ the study. Feed samples were collected before the beginning of each versitat Autonoma de Barcelona (Reference 09/771). Also, animal care experimental period and were analyzed for DM, crude protein, acid conditions and management practices were according to the Spanish detergent fiber, and neutral detergent fiber according to AOAC Inter­ Royal Decree 53/2013 on the protection of animals used for experi­ national (2003). mental purposes, and the recommendations of the Ministry of Agricul­ Milk was weighed by a digital scale (Mobba industrial, Barcelona, ture, Food and Environment of Spain for the welfare of livestock. Spain) and registered daily. Milk samples were collected twice per week Eight multiparous Murciano-Granadina dairy goats in mid-lactation for the analysis of total solids, fat, protein (N × 6.38), and casein using a (43.3 ± 1.6 kg BW; 2 ± 0.04 L milk/day, 81 ± 3 days of lactation) were NIRSystems 5000 scanning monochromator (FOSS, Hillerød, Denmark) used from the herd of the experimental farm of the Universitat Auton­ according to Albanell et al., 2003). A milk aliquot was stored at − 25◦ C to oma of Barcelona. Goats were divided into two balanced groups (n = 4 determine milk osmolality using a Fiske 110 osmometer (Fiske Associ­ goats/group) according to milk yield and milk composition. The ations, Norwood, MA). experimental design was a crossover design with two periods of 28 days Urine spot samples were collected between 0800 and 0930 h, 1400 each, and two climatic treatments that differed in the temperature- and 1530 h, and 1800 and 1930 h, for 5 consecutive days during weeks 1 and 4. Urine samples (10 mL) were acidified with 1 mL of 10% H2SO4, 2 A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 composted for each goat and stored at − 25◦ C until analyses. Urine 19 of the 2nd period. On days 22, 24, and 26 glucose tolerance test (GTT; samples were diluted (1:10) and analyzed for uric acid and creatinine 0.25 g/kg of BW), insulin tolerance test (4.6 μg/kg of BW), and concentrations using commercial kits according to the manufacturer’s epinephrine challenge (2 μg/kg of BW) were performed, respectively. instructions (Biolabo SA, Maizy, France). Urine concentrations of uric Dextrose (D9434, Sigma-Aldrich, St. Louis, MO), insulin (bovine insulin acid and creatinine were used to calculate uric acid index (UAI) as a from pancreas, I6634, Sigma-Aldrich), and epinephrine (E4250, Sigma- proxy of microbial protein synthesis: Aldrich) were diluted in sterile 0.9% NaCl solution (Vitulia 0.9%, Lab­ oratorios ERN S.A., Barcelona, Spain). Solutions of glucose (500 μg/mL), UAI = (uric acid concentration / creatinine concentration) × BW0.75 insulin (230 μg/mL), and epinephrine (100 μg/mL) were sterilized by filtration through 0.22-μm polyether-sulfone filters (Millex-GP, Milli­ pore; Merck Life Science SLU, Madrid, Spain) using 1 filter for each 200 mL of solution and kept at 4◦ C. 2.3. Blood variables On the day of each challenge, blood samples were collected at − 30, − 20, − 10, 0, 5, 10, 20, 30, 45, 60, 90, and 120 min relative to the Blood samples (10 mL) were collected before feeding and milking at administration. After each blood sampling, the catheters were flushed days 7, 14, 21 and 28 from the jugular vein into vacutainers (Venoject, with heparinized 0.9% saline solution (500 IU/mL; Clexane 4000 UI, Leuven, Belgium). Plasma was obtained by centrifugation of blood for Sanofi Aventis, Paris, France). Samples were collected by syringe into 15 min at 1500 × g, and stored at –25◦ C for the analysis of blood urea N, glass tubes containing 250 units of sodium heparin and were immedi­ glucose, NEFA, β-hydroxybutyrate (BHB), lactate, and insulin. Glucose ately placed on ice. After centrifugation of whole blood for 15 min at was determined by Trinder method using Glucose GOD-PAP kit (Biolabo 1500 × g and 4◦ C, plasma was divided into different aliquots and stored SA, Maizy, France). The NEFA were analyzed by the colorimetric at − 25◦ C for subsequent analysis of plasma insulin, glucose, and NEFA enzymatic test ACS-ACOD method using a commercial kit (Wako concentrations as indicated above. Chemicals, Neuss, Germany). The BHB was determined by kinetic The area under the curve (AUC) of metabolite responses to GTT, enzymatic method using commercial kit (RANBUT, Randox®, Crumlin, insulin tolerance test, and epinephrine challenge were calculated by the UK). Insulin was measured by ELISA using a commercial kit (Mercodia trapezoidal method. The pre-challenge values (i.e. − 30, − 20, − 10, and Ovine Insulin ELISA, Mercodia®; Uppsala, Sweden). Lactate was 0 min) were averaged and considered as the basal value of each determined by enzymatic method (Olympus System Reagent®, Beckman metabolite. Values of peaks or nadirs of each metabolite after infusions Coulter®, Nyon, Switzerland). In addition, whole blood without anti­ were recorded. Clearance rates for GTT and insulin tolerance test were coagulants was collected and a single drop was immediately applied to calculated according to Kerestes et al. (2009). disposable cartridges (iSTAT Crea cartridges, Abbott Point of Care Inc., Princeton, NJ, USA) for the analysis of creatinine. 2.5. Behavior variables 2.4. Hormonal challenges and glucose tolerance test Eight digital color cameras (model VCAM—420CA, Circontrol1, Barcelona, Spain) with a focal lens (model LTC 0500/50, Philips, Indwelling jugular silicone rubber catheters (Nutricath Silicone, 60 Eindhoven, the Netherlands), fitted with infrared illuminator were used cm length and 14-gauge, Vygon, Valencia, Spain) were inserted on day (1 camera / goat). Videos were digitized and stored on SATA (Serial Advance Technology Attachment) hard disks of 500 GB each. The videos were watched using a digital recorder (VDVR-9NX Circontrol1, Barce­ lona, Spain) with screen and remote control, which allowed the manual control of the video. Due to the accuracy of the recorder, the images were displayed at a standard frame rate of 25 pictures per second. The following 6 behavior indices were measured for each goat: 1) Time spent while standing: the time duration spent by the goat standing for different activities (eating, drinking, or idling). 2) Frequency of position changings: number of times the goat changes its position from standing to lying down or vice versa. 3) Eating bouts: number of times the goat visits the feeder and eats. 4) Eating duration: the duration taken by the goat eating from the feeder. 5) Drinking bouts: number of times the goat visits the water trough and drinks. 6) Drinking duration: the duration taken by the goat drinking from the water trough. According to our previous observations (Hamzaoui et al., 2013) and the current study (see hereafter), goats suffered maximum HS during the first 3 days. Therefore, the 3rd day of each treatment (TN and HS) in each period (1 and 2) was chosen to obtain the behavior variables. The activity throughout the day was considered as diurnal (12 h day from 0800 to 2000 h) and nocturnal (12 h dark from 2000 to 0800). Data for behavior observations were recorded onto an Excel spreadsheet Fig. 1. Mean rectal temperature (A) and respiration rate (B) throughout the (Microsoft Office 2013) which enabled the calculation of the duration of day (0800, 1200, and 1700 h) in dairy goats under thermal neutral (TN; blue dashed curves) and heat stress (HS; red solid curves) conditions. The SEM each behavior variable. values were 0.07◦ C and 4 breaths/min for rectal temperature and respiration rate, respectively. (For interpretation of the references to colour in the Figure, the reader is referred to the web version of this article). 3 A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 Table 2 2019; Hamzaoui et al., 2020), but greater than losses observed during Performance of Murciano-Granadina dairy goats under thermal neutral (TN, n = late lactation (19%; Hamzaoui et al., 2013). The reduction in DMI 8) and heat stress (HS, n = 8) conditions. Values are least squares means and SE because of HS was maximum at d 5 (36%) and partial recovery was of the mean (SEM). observed thereafter (data not shown), but was always lower (P < 0.001) Effect (P <) than in TN goats. Reduced DMI in heat-stressed animals typically de­ Item TN HS SEM Treatment Period T× creases metabolic heat production since heat increment of feeding is an P1 important source of heat production (Sejian et al., 2018). Additionally, Body weight change, 5.30 –1.28 0.72 0.01 0.43 0.44 the gut fill from water in HS goats might also be related to the reduced kg DMI. Our HS goats increased their water consumption by 70%, and DM intake, kg/d 2.24 1.57 0.04 0.01 0.01 0.83 presumably this was because of the increment in heat dissipation by Energy balance, 0.74 –0.37 0.10 0.01 0.23 0.61 evaporation (panting and sweating). Mcal/d In agreement with the depressed DMI, HS goats experienced negative Urinary uric acid 10.34 6.40 1.55 0.10 0.63 0.53 index EB (–0.37 Mcal/d), whereas TN goats had a positive EB (0.74 Mcal/d). Water consumption, 5.93 10.07 0.29 0.01 0.91 0.38 This negative EB related to HS was also observed in heat-stressed dairy L/d cows (Baumgard and Rhoads, 2013) and goats (Hamzaoui et al., 2020, Milk yield, kg/d 1.72 1.55 0.04 0.01 0.01 0.73 2021). In accordance with the DMI and EB data, HS goats lost BW (− 45 Fat-corrected milk2, 1.94 1.65 0.03 0.01 0.01 0.75 kg/d g/d), whereas TN goats gained BW (+189 g/d on average) during the Milk composition, % experimental period. It should be kept in mind that changes in BW of our Total solids 12.8 11.6 0.10 0.01 0.37 0.69 TN and HS goats included the inevitable variations in digestive tract Fat 4.28 3.75 0.09 0.01 0.73 0.51 content. P rotein 3.77 3.25 0.08 0.01 0.23 0.26 Heat-stressed goats tended (P < 0.10) to have lower UAI compared to Casein 3.23 2.82 0.06 0.01 0.01 0.90 Milk osmolality, 319 299 4 0.01 0.04 0.17 TN goats, which might indicate lower microbial synthesis in the rumen mOsm/kg of HS goats (Table 2). We admit that using uric acid alone as a proxy of 1 microbial protein synthesis might not be fully accurate since uric acid Interaction of treatment (T) × period (P). 2 Corrected milk yield at 3.5 % fat = L of milk yield × [0.432 + 0.162 × (fat represents only 10–15% of urinary purine derivatives and some uric acid %)]. could be converted to allantoin by uricase (Chen and Ørskov, 2004). However, both urinary uric acid and allantoin respond similarly to experimental treatments and conditions in goats (Lindberg, 1989; 2.6. Statistical analyses Belenguer et al., 2002; Ma et al., 2014), sheep (Puchala and Kulasek, 1992), and cows (Dewhurst et al., 2010). In the current study, the Data were analyzed by the PROC MIXED for repeated measurements decrease in DMI might explain the reduction in microbial protein syn­ of SAS version 9.1.3 (SAS Institute Inc., Cary, NC). The statistical mixed thesis in the rumen of HS goats. Lindberg (1985) noted that urinary model contained the effects of treatment (TN and HS), period (1 and 2), purines in goats are highly correlated with DMI. Also, Castro-Costa et al. week (1–4), and interactions of treatment × period, treatment × week, (2015) reported that heat-stressed goats experience greater rumen and period × week as fixed effects, as well as the random effects of the temperatures and lower rumen pH, despite having the same DMI values. animal and the residual error. For rectal temperature and respiration These changes in rumen conditions because of HS could alter rumen rate measured at 0800, 1200, and 1700 h, a fixed factor of the hour of microbial populations. Differences in UAI values were unlikely related the day was added to the model. Data of performances (i.e., DMI, water to changes in digesta passage rate since Salama et al. (2016) found no consumption, and milk yield) and physiological indicators (i.e., rectal significant differences in solid digesta passage rate between TN and HS temperature and respiration rate) were analyzed on weekly average dairy goats. basis. The statistical model of video recording data included an addi­ Compared to TN, HS goats produced lower (P < 0.05) milk yield tional fixed effect of daytime (day vs. night). The model considered the (10%) and fat-corrected milk (13%) with depressed contents of total possible carryover effects of previous HS periods through the treatment solids (9%), fat (12%), protein (14%), and casein (13%) as shown in × period interaction. The statistical mixed model of metabolite chal­ Table 2. Milk yield losses detected in the current study is within the lenges contained the fixed effects of the treatment (TN and HS) and time range (3–13%) reported in heat-stressed dairy goats (reviewed by Sal­ relative to metabolite administration; the random effect of the animal; ama et al., 2014). In addition, losses in milk components caused by high the interaction of treatment × time relative to metabolite administra­ ambient temperatures have been also observed in dairy cows (Baumgard tion; and the residual error. and Rhoads, 2013) and ewes (Finocchiaro et al., 2005; Mehaba et al., 2021). The negative effects of HS on milk production could be partially 3. Results and discussion explained by decreased DMI in addition to the direct negative effects on the synthetic capacity of the mammary gland. Results in heat-stressed 3.1. Performance variables mammary cells showed reduced abundance of genes related to milk fat and protein synthesis (Salama et al., 2019). As already mentioned, Rectal temperatures and respiration rates increased (P < 0.01) from UAI tended to decrease by HS, which might decrease amino acids 0800 to 1700 h in both goat groups and were greater (P < 0.001) in HS available for milk protein synthesis, since microbial protein represents > than in TN goats (Fig. 1). Our results are similar to that observed in 50% of amino acids absorbed in the small intestine (Schwab and Bro­ lactating dairy goats from the same breed under similar environmental derick, 2017). conditions (Mehaba et al., 2019; Hamzaoui et al., 2021). Goats under HS Milk osmolality decreased (P < 0.01; Table 2) because of HS, which had the maximum values of rectal temperature and respiration rate could be related to increased water consumption (and consequently during the first 3 days and then decreased (P < 0.05), but remained more diluted milk) in HS compared to TN goats. Heat stress has been higher (P < 0.001) than the TN goats until the end of the experiment. found to decrease milk osmolality in Swedish goats (Hartmann et al., Similar trend over time under HS conditions was observed by Hamzaoui 2021) and Friesland sheep (Thompson et al., 1981). However, Mehaba et al. (2013) in dairy goats during late lactation. et al. (2021) detected an increase in milk osmolality in HS dairy Lacaune The DMI decreased (P < 0.001) by 29% because of HS treatment sheep, although HS ewes consumed significantly more water than TN (Table 2). This reduction in DMI is similar to DMI losses in heat-stressed animals. The discrepancy among studies could be related to breed dif­ Muraciano-Granadina goats at mid lactation (26–34%; Mehaba et al., ferences in their fluid balance during HS. 4 A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 Table 3 Blood metabolites of Murciano-Granadina dairy goats under thermal neutral (TN, n = 8) and heat stress (HS, n = 8) conditions. Values are least squares means and SE of the mean (SEM). Effect (P <) Variable TN HS SEM Treatment Period T× P1 Insulin, μg/L 1.17 1.42 0.286 0.54 0.87 0.46 Glucose, mg/dL 58.6 58.1 1.17 0.78 0.02 0.53 Lactate, mmol/L 0.62 0.66 0.034 0.47 0.72 0.22 Nonesterified fatty 0.10 0.11 0.022 0.67 0.34 0.52 acids, mmol/L β-hydroxybutyrate, 0.49 0.55 0.045 0.39 0.39 0.42 mmol/L Blood urea N, mg/dL 26.6 23.0 1.2 0.06 0.01 0.87 Creatinine, mg/dL 0.45 0.56 0.029 0.03 0.01 0.45 1 Interaction of treatment (T) × period (P). 3.2. Blood variables Despite the marked differences in DMI, there was no difference in blood levels of insulin, glucose, lactate, NEFA and BHB between TN and HS goats (Table 3). Heat-stressed dairy goats (Hamzaoui et al., 2013; 2020; 2021) and ewes (Mehaba et al., 2021) have also been reported to be able to keep similar blood glucose levels to TN animals. The reduced milk yield, and consequently decreased lactose secretion, could spare some glucose in blood since 80–85% of glucose in blood is used by the mammary gland for lactose synthesis in goats (Sano et al., 1985). Also, Hamzaoui et al. (2021) reported that HS reduces milk de novo fatty acids synthesis, and the resultant saved glucose might help HS goats to keep normal blood glucose levels. Additional reasons that may explain how HS goats were able to keep normal blood glucose values are discussed hereafter. Although HS goats experienced negative EB, blood levels of NEFA did not increase compared to TN goats, which has also been observed in dairy goats (Hamzaoui et al., 2013; Mehaba et al., 2019), dairy ewes (Mehaba et al., 2021), and dairy cows (Baumgard and Rhoads, 2013). The absence of significant difference in blood NEFA values might be related to similar blood insulin levels in TN and HS goats (Table 3). Insulin is a potent lipogenic and antilipolytic hormone (Baumgard and Rhoads, 2013) that prevents body fat mobilization. Insulin is necessary for the activation of the cellular stress response (Li et al., 2006), which Fig. 2. Mean plasma glucose (A) and insulin (B) response to glucose tolerance might explain why blood insulin levels did not decrease in HS goats even test of dairy goats under thermoneutral (TN; blue circles) or heat stress (HS; red though they ate less and were in negative EB. squares) conditions. * indicates a difference at P < 0.05 between TN and HS Baumgard and Rhoads (2013) reported that HS did not cause sig­ treatments. SE shown as vertical bars. (For interpretation of the references to nificant change in blood NEFA (similar to goats), but increased blood colour in the Figure, the reader is referred to the web version of this article). insulin values (such an increase was not detected in goats). Furthermore, Baumgard and Rhoads (2013) indicated that blood lactate levels are 3.3. Glucose tolerance test and hormonal challenges consistently elevated in HS animals. Nevertheless, blood lactate did not vary between TN and HS goats in our study. It should be kept in mind Glucose and insulin responses of TN and HS goats following glucose that the studies reviewed by Baumgard and Rhoads (2013) were based administration are shown in Fig. 2 and Table 4. Glucose spiked at 5 min on the comparison between HS and pair-fed TN animals, whereas our TN in both treatments, but the peak was greater (P < 0.05) in HS than in TN goats were fed ad libitum. goats (Fig. 2a). Blood glucose levels gradually decreased in both treat­ Blood urea tended (P < 0.10) to decrease, whereas creatinine values ments with similar clearance rates (1.17%/min on average; Table 4). increased (P < 0.05) in HS goats compared to TN animals (Table 3). The Glucose AUC was greater (P < 0.05) in HS than in TN goats at 30 and 90 decrease in blood urea concentration by HS could be explained by their min. Mehaba et al. (2021) also found that glucose AUC in response to lower DMI and, consequently, reduced protein intake. On the other glucose infusion tends to be greater in HS dairy ewes. Also, Wheelock hand, the increment in blood creatinine by HS might indicate muscle et al. (2010) reported that lactating HS cows have greater glucose degradation. Our HS goats were in negative EB (Table 2), but as indi­ response after glucose infusion than TN cows fed ad libitum. cated above, they did not mobilize body fat reserves (no change in blood Insulin peaked at 10 min after glucose administration in TN goats, NEFA as shown in Table 3). However, they mobilized body protein, whereas in HS goats a plateau in insulin level was observed from 10 to presumably to use some glucogenic amino acids for gluconeogenesis, 60 min, with lower (P < 0.05) insulin values at 10, 20 and 30 min resulting in keeping similar blood glucose levels. Heat-stressed dairy (Fig. 2b). This resulted in a tendency of lower (P < 0.10) insulin clear­ cows (Rius, 2018) and ewes (Mehaba et al., 2021) also experience ance rate and lower (P < 0.05) insulin AUC in HS compared to TN goats increased blood creatinine levels compared with TN animals. (Table 4). Thus, the pancreas of HS goats secreted less amounts of insulin in response to glucose infusion compared to TN goats, which could be a 5 A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 Table 4 Metabolite kinetics in response to glucose tolerance test, insulin tolerance test, and epinephrine challenge in dairy goats under thermoneutral (TN; n = 4) or heat stress (HS; n = 4). Values are least squares means and standard error of the mean (SEM). Item TN HS SEM P-value Glucose tolerance test Glucose Basal, mg/dL 58.6 64.9 5.4 0.30 Peak, mg/dL 194 211 6.6 0.05 AUC30 min,1 mg/L × min 4450 4883 171 0.05 AUC90 min, mg/L × min 9586 11,367 596 0.05 CR90,2 %/min 1.23 1.10 0.14 0.42 Insulin Basal, μg/L 0.77 0.66 0.18 0.56 Peak, μg/L 2.91 1.78 0.45 0.04 AUC30 min, μg /L × min 69.0 47.9 7.4 0.04 AUC90 min, μg /L × min 168 139 29 0.36 CR902, %/min 1.65 0.69 0.43 0.09 Fig. 4. Mean plasma nonesterified fatty acids (NEFA) response to epinephrine Insulin tolerance test challenge of dairy goats under thermoneutral (TN; blue circles) or heat stress Glucose (HS; red squares) conditions. * indicates a difference at P < 0.05 between TN Basal, mg/dL 59.6 59.2 2.9 0.90 and HS treatments. SE shown as vertical bars. (For interpretation of the refer­ Nadir, mg/dL 26.5 20.0 4.9 0.24 ences to colour in the Figure, the reader is referred to the web version of AUC120 min, mg/dL × min 5590 5223 233 0.17 this article). ISBGR3, % 55.5 66.6 7.9 0.21 Nonesterified fatty acids Basal, mmol/L 0.17 0.19 0.032 0.66 way to keep normal glucose levels in blood under HS despite the reduced Nadir, mmol/L 0.12 0.12 0.015 0.76 DMI (Table 2). Heat-stressed lactating goats (current study; Hamzaoui AUC2120 min, mmol/L × min 24.3 25.7 3.9 0.74 et al., 2013, 2020) and ewes (Mehaba et al., 2021) experienced decreased DMI, but were able to maintain similar blood glucose levels to Epinephrine challenge TN animals with no change in blood insulin concentration. Glucose Basal, mg/dL 59.3 58.3 3.1 0.78 Glucose response to insulin administration is shown in Fig. 3 and Peak, mg/dL 101 98.0 7.3 0.73 Table 4. Glucose basal levels averaged 59.4 ± 2.9 mg/dL and were AUC120 min, mg/dL × min 8862 9012 641 0.83 similar in the plasma of TN and HS goats. The level of plasma glucose Nonesterified fatty acids decreased after insulin (lipogenic signal) administration and reached the Basal, mmol/L 0.16 0.14 0.030 0.60 lowest (P < 0.001) level at 30 min (23.3 ± 4.9 mg/dL). Thereafter, blood Peak, mmol/L 0.37 0.30 0.072 0.38 AUC120 min, mmol/L × min 29.0 22.3 3.1 0.08 glucose concentration increased gradually to basal levels at 90 min and 1 remained unchanged until 120 min. The basal NEFA levels averaged Area under the curve corrected for the basal levels. 2 0.18 ± 0.03 mmol/L before the insulin injection and did not vary be­ Clearance rate from the peak to 90 min = tween TN and HS goats (Table 4), confirming that no body fat reserves ln⁡(glucose at 5 min) − ln⁡(glucose at 90 min) × 100. were mobilized in HS goats as mentioned above. 90 min − 5 min 3 No differences in AUC of glucose and NEFA in response to insulin Insulin stimulated blood glucose response = glucose at 0 min − glucose at 30 min infusion were observed between TN and HS goats (Table 4). It is well × 100. known that insulin causes cells in the liver, skeletal muscles, and fat glucose at 0 min tissue to absorb glucose from the blood. According to our results, it seems that this happens similarly in TN and HS goats. Blood glucose peak in response to the epinephrine administration was similar between TN and HS goats (Table 4). Additionally, glucose AUC did not vary, suggesting that the liver of TN and HS goats responded similarly to epinephrine in terms of glycogen breakdown and glucose release. The NEFA response of TN and HS goats following epinephrine administration (lipolysis signal) is shown in Fig. 4. When goats were administered with epinephrine, an increase in blood NEFA was observed in both TN and HS goats. The peak of NEFA in HS and TN goats was observed at 10 and 30 min, respectively. At 30, 45 and 60 min after epinephrine administration, TN goats had greater (P < 0.05) blood NEFA levels. This resulted in a tendency (P = 0.073) for a lower AUC for NEFA in HS goats (Table 4). A previous study (Collier and Gebremedhin, 2015) showed that although HS is usually accompanied by higher levels of lipolytic hormones as cortisol, no increase in NEFA was detected. Results of the current study clearly indicate that HS induces an insen­ sitivity of lipid body tissues to lipolytic hormones, preventing fat mobilization under high ambient temperature conditions. This finding Fig. 3. Mean plasma glucose response to insulin tolerance test of dairy goats further explains why NEFA did not vary between TN and HS goats under thermoneutral (TN; blue circles) or heat stress (HS; red squares) condi­ (Table 3). Heat-stressed dairy cows (Baumgard and Rhoads, 2013) and tions. SE shown as vertical bars. (For interpretation of the references to colour ewes (Mehaba et al., 2021) were also reported to have a reduced NEFA in the Figure, the reader is referred to the web version of this article). response to epinephrine administration. 6 A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 Table 5 Behavior indices during the day (12 h) and night (12 h) of dairy goats under thermal neutral (TN; n = 8) and heat stress (HS; n = 8) conditions. Values are least squares means and SE of the mean (SEM). Goats in TN were kept at 15-20◦ C throughout the day, whereas HS goats were at 37◦ C during the day and at 30◦ C during the night. TN HS Effect (P <) Items Day Night Day Night SEM Treatment Daytime T x D1 Position change2, n 7.6 6.6 22.6 15.1 1.8 0.01 0.02 0.06 Standing time, min 486 177 278 196 14 0.01 0.01 0.01 Lying time, min 234 543 442 524 17 0.01 0.01 0.01 Lying time average3, min 35.0 92.3 22.3 27.7 7.7 0.01 0.01 0.01 Eating bouts, n 32.1 8.8 26.0 16.4 2.3 0.82 0.01 0.01 Eating time, min 273 71 105 97 9 0.01 0.01 0.01 Eating time/meal, min 8.75 8.65 4.76 6.58 3.0 0.01 0.28 0.23 Drinking bouts, n 10.5 2.3 21.6 8.8 2.2 0.02 0.01 0.41 Drinking time, min 5.8 1.3 8.4 4.4 0.94 0.05 0.02 0.88 Drinking time/bout, min 0.54 0.43 0.53 0.40 0.09 0.18 0.83 0.94 1 Interactions of treatment (T) and daytime (D). 2 Counted as the change from standing to lying and vice versa. 3 Calculated as total lying time divided by the times of position changing. Fig. 5. Percentage of time remaining standing or lying down during the day (12 h) and night (12 h) in dairy goats under thermal neutral (TN) or heat stress (HS) conditions. 3.4. Behavioral indices eating, drinking, or idling) for a longer time than HS goats (68 vs. 39%), while at night, there was no difference between treatments although For the behavioral indices in the current study, the day was divided both spent a shorter time standing (26%) than during the day (Fig. 5). into 2 portions: day during which HS goats were at 37◦ C with light, and Changing the position (from lying down to standing and vice versa) night during which HS goats were at 30◦ C in dark. The TN goats were at during day and night was much greater (P < 0.01) in HS (18.9 times on 15–20◦ C throughout the day with the same light regimen as HS goats. average) than TN goats (7.1 times on average). Temple and Manteca Haley et al. (2000) and Fregonesi and Leaver (2001) reported that the (2020) reported that goats in hot environments move more to reorient time spent lying down and the duration of individual bouts are sensitive themselves in different directions to avoid the impact of direct solar measures of stall comfort and animal welfare. Therefore, information radiation and ground radiation. In this current study, both TN and HS about the duration and frequency of different activities could reflect the goats were in metabolism crates housed within climate chambers (no comfort of goats under HS conditions. solar radiation). However, the increment in the frequency of position Observed behavioral indices in TN and HS goats are shown in Table 5 changing indicates that HS goats were uncomfortable and had and Figs. 5 and 6. During the day TN goats remained standing (for extra-movements than TN goats. 7 A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 Fig. 6. Percentage of time spent while standing devoted to eating, drinking or idling during the day (12 h) and night (12 h) in dairy goats under thermal neutral (TN) or heat stress (HS) conditions. Total daily eating bouts (41.7) were similar between groups, but goats (22 vs. 28 min for day and night, respectively) than for TN goats greater (P < 0.001) during the day (29.1) than during the night (12.6). (35 and 92 min during day and night, respectively) because of the higher However, HS goats doubled their number of eating bouts during the frequency of position changing (Table 5). Darcan et al. (2008) and night when temperature decreased from 37 to 30◦ C. Throughout the Hartmann et al. (2021) also reported that goats housed in pens and day, TN and HS had a similar number of eating bouts, but the duration of exposed to HS decrease their activity and increase time laying. However, each meal was shorter (P < 0.01) in HS (5.7 min) than in TN (8.7 min) Overton et al. (2002); Zahner et al. (2004), and Cook et al. (2007) re­ goats (Table 5). This could explain the reduction in DMI observed when ported that the duration of lying behavior decreased with increasing THI goats were subjected to HS. However, the higher water consumption by in dairy cows. This discrepancy between the goat and cow studies could HS goats could be explained by the greater (P < 0.01) number of be because of species differences. drinking bouts than in the TN goats (30.4 vs. 12.8) while there was a similar time spent drinking for both groups (0.48 min/bout on average). 4. Conclusions Cook et al. (2007) showed that total time spent drinking by dairy cows increased from 0.3 to 0.5 h/d when THI increased from 56 to 74. Exposure of goats to heat stress during mid-lactation reduced milk The percentage of time devoted to different activities (eating, yield, and fat and protein contents. Although HS goats had lower feed drinking, or idling) while standing are shown in Fig. 6. During the day intakes than TN goats, they were able to keep normal blood glucose and in accordance with changes in DMI and water consumption, HS levels and did not mobilize body fat. Compared to TN goats, HS goats goats spent a greater percentage of time drinking and idling, but a lower had the same response to insulin, but their adipose tissue was less sen­ portion of time eating than TN goats. During the night, HS goats reduced sitive to lipolytic signals, which explains the lack of fat mobilization the portion of time devoted to idling and spent a greater portion of time under HS conditions. It seems that the pancreas of HS goats was less eating when ambient temperature decreased. sensitive, secreting lower insulin amounts in response to glucose Although our goats were in metabolism cages, time spent lying down administration. This finding might be a mechanism by which goats keep by TN goats (13 h/24 h) was similar to the 12− 13 h reported for healthy blood glucose levels stable under high ambient temperatures. Heat stress dairy cows housed in a free-stall barn (Cook et al., 2005; Drissler et al., had no effect on the number of eating bouts, but the observed reduced 2005). However, HS goats increased their total lying duration by 3 h (16 feed intake for the HS goats was because of a shorter duration of each h/24 h), but the duration of each lying down was shorter for the HS eating bout. The increase in water consumption under higher ambient 8 A.A.K. Salama et al. Small Ruminant Research 203 (2021) 106496 temperatures was because of the higher number of drinking bouts rather Hartmann, E., Högberg, M., Olsson, K., Dahlborn, K., 2021. Physiological and behavioural responses of Swedish domestic goats and their kids (Capra hircus) to 15 than the duration of drinking bouts. days of heat exposure. Acta Agric. Scand. A Anim. Sci. 70, 41–49. Institut National de la Recherche Agronomique (INRA), 2018. INRA Feeding System for Declaration of Competing Interest Ruminants. Wageningen Academic Publishers, Wageningen, The Netherlands. Kerestes, M., Faigl, V., Kulcsar, M., Balogh, O., Foldi, J., Febel, H., Chilliard, Y., Huszenicza, G., 2009. Periparturient insulin secretion and whole-body insulin The authors declare no conflict of interest. responsiveness in dairy cows showing various forms of ketone pattern with or without puerperal metritis. Domest. Anim. Endocrinol. 37, 250–261. https://doi. Acknowledgements org/10.1016/j.domaniend.2009.07.003. Li, G., Ali, I.S., Currie, R.W., 2006. Insulin induces myocardial protection and HSP70 localization to plasma membranes in rat hearts. Am. J. Physiol. Heart Circ. Physiol. Funding: This research was funded by the Spanish Ministry of 291, H1709–H1721. Economy and Competitiveness (Program I + D + I oriented to Society Lindberg, J.E., 1985. Urinary allantoin excretion and digestible organic matter intake in dairy goats. Swed. J. Agric. Res. 15, 31–37. challenges; Project AGL2013-44061-R) and was also supported by a PhD Lindberg, J.E., 1989. Nitrogen metabolism and urinary excretion of purines in goat kids. scholarship to Soufiane Hamzaoui from The Spanish Ministry of Science Br. J. Nutri. 61, 309–321. and Innovation (BES-2010-030089). The authors are also grateful to the Ma, T., Deng, K.D., Tu, Y., Jiang, C.G., Zhang, N.F., Li, Y.L., Si, B.W., Lou, C., Diao, Q.Y., 2014. Effect of dietary concentrate:forage ratios and undegraded dietary protein on team of SGCE (Servei de Granges i Camps Experimentals) of the Uni­ nitrogen balance and urinary excretion of purine derivatives in Dorper×thin-tailed versitat Autonoma de Barcelona for the care of the animals. Han crossbred lambs. Asian-Australas. J. Anim. Sci. 27, 161–168. Mehaba, N., Salama, A.A.K., Such, X., Albanell, E., Caja, G., 2019. Lactational responses of heat-stressed dairy goats to dietary L-carnitine supplementation. Animals 9, 567. References Mehaba, N., Coloma-Garcia, W., Such, X., Caja, G., Salama, A.A.K., 2021. Heat stress impacts some physiological and productive variables and alters metabolism in dairy Albanell, E., Caja, G., Such, X., Rovai, M., Salama, A.A.K., Casals, R., 2003. ewes. J. Dairy Sci. 104, 1099–1110. Determination of fat, protein, casein, total solids, and somatic cell count in goat’s National Research Council (NRC), 1971. A Guide to Environmental Research on Animals. milk by near-infrared reflectance spectroscopy. J. AOAC Int. 86, 746–752. National Academy of Sciences, Washington, DC, USA. AOAC International, 2003. Official Methods of Analysis of AOAC International, 17th ed., National Research Council (NRC), 2001. Nutrient Requirements of Dairy Cattle, 7th ed. Vol. I. AOAC International, Gaithersburg, MD, USA. National Academy of Science, Washington, DC, USA. Bauman, D.E., Currie, W.B., 1980. Partitioning of nutrients during pregnancy and Overton, M.W., Sischo, W.M., Temple, G.D., Moore, D.A., 2002. Using time-lapse video lactation: a review of mechanisms involving homeostasis and homeorhesis. J. Dairy photography to assess dairy cattle lying behavior in a free-stall barn. J. Dairy Sci. 85, Sci. 63, 1514–1529. 2407–2413. Baumgard, L.H., Rhoads, R.P., 2013. Effects of heat stress on postabsorptive metabolism Puchala, R., Kulasek, G.W., 1992. Estimation of microbial protein flow from the rumen of and energetics. Annu. Rev. Anim. Biosci. 1, 311–337. sheep using microbial nucleic acid and urinary excretion of purine derivatives. Can. Belenguer, A., Yanez, D., Balcells, J., Ozdemir Baber, N.H., Gonzalez Ronqui, M., 2002. J. Anim. Sci. 72, 821–830. Urinary excretion of purine derivatives and prediction of rumen microbial outflow in Rhoads, M.L., Rhoads, R.P., VanBaale, M.J., Collier, R.J., Sanders, S.R., Weber, W.J., goats. Livest. Prod. Sci. 77, 127–135. Crooker, B.A., Baumgard, L.H., 2009. Effects of heat stress and plane of nutrition on Castro-Costa, A., Salama, A.A.K., Moll, X., Aguiló, J., Caja, G., 2015. Using wireless lactating Holstein cows: I. Production, metabolism, and aspects of circulating rumen sensors for evaluating the effects of diet and ambient temperature in non- somatotropin. J. Dairy Sci. 92, 1986–1997. lactating dairy goats. J. Dairy Sci. 98, 4646–4658. Rius, A.G., 2018. Adaptations of protein and amino acid metabolism to heat stress in Chen, X.B., Ørskov, E.R., 2004. Research on urinary excretion of purine derivatives in dairy cows and other livestock species. J. Appl. Anim. Welf. Sci. 35, 39–48. ruminants: past, present and future. In: Makkar, H.P.S., Chen, X.B. (Eds.), Estimation Salama, A.A.K., Caja, G., Hamzaoui, S., Badaoui, B., Castro-Costa, A., Façanha, D.E., of Microbial Protein Supply in Ruminants Using Urinary Purine Derivatives. Guilhermino, M.M., Bozzi, R., 2014. Different levels of response to heat stress in Springer, Dordrecht. https://doi.org/10.1007/978-1-4020-2844-1_21. dairy goats. Small Rumin. Res. 121, 73–79. Collier, R.J., Gebremedhin, K.G., 2015. Thermal biology of domestic cows. Annu. Rev. Salama, A.A.K., Caja, G., Hamzaoui, S., Such, X., Albanell, E., Badaoui, B., Loor, J.J., Anim. Biosci. 3 (10), 1–10, 20. 2016. Thermal stress in ruminants: responses and strategies for alleviation. In: Cook, N.B., Bennett, T.B., Nordlund, K.V., 2005. Monitoring indices of cow comfort in Villalba, J.J., Manteca, X. (Eds.), Animal Welfare in Extensive Production Systems, free-stall-housed dairy herds. J. Dairy Sci. 88, 3876–3885. 1st ed. 5M Publishing, Sheeld, UK, pp. 11–36. Cook, N.B., Mentink, R.L., Bennett, T.B., Burgi, K., 2007. The effects of heat stress and Salama, A.A.K., Duque, M., Wang, L., Shahzad, K., Olivera, M., Loor, J.J., 2019. lameness on time budgets of lactating dairy cows. J. Dairy Sci. 90, 1674–1682. Enhanced supply of methionine or arginine alters mechanistic target of rapamycin Darcan, N., Cedden, F., Cankaya, S., 2008. Spraying effects on some physiological and signaling proteins, messenger RNA, and microRNA abundance in heat-stressed behavioural traits of goats in a subtropical climate. Ital. J. Anim. Sci. 7, 77–85. bovine mammary epithelial cells in vitro. J. Dairy Sci. 102, 2469–2480. Dewhurst, R.J., Davies, L.J., Kim, E.J., 2010. Effects of mixtures of red clover and maize Sano, H., Ambo, K., Tsuda, T., 1985. Blood glucose kinetics in whole body and mammary silages on the partitioning of dietary nitrogen between milk and urine by dairy cows. gland of lactating goats exposed to heat stress. J. Dairy Sci. 68, 2557–2564. Animal 4 (5), 732–738. Schwab, C.G., Broderick, G.A., 2017. Protein and amino acid nutrition in dairy cows. Drissler, M., Gaworski, M., Tucker, C.B., Weary, D.M., 2005. Freestall maintenance: J. Dairy Sci. 100, 10094–10112. effects on lying behavior on dairy cattle. J. Dairy Sci. 88, 2381–2387. Sejian, V., Bhatta, R., Gaughan, J.B., Dunshea, F.R., Lacetera, N., 2018. Adaptation of Finocchiaro, R., van Kaam, J.B.C.H.M., Portolano, B., Misztal, I., 2005. Effect of heat animals to heat stress. Animal 12, s431–s444. stress on production of Mediterranean dairy sheep. J. Dairy Sci. 88, 1855–1864. Silanikove, N., 1992. Effect of water scarcity and hot environment on appetite and Fregonesi, J.A., Leaver, J.D., 2001. Behaviour, performance and health indicators of digestion in ruminants: a review. Livest. Prod. Sci. 30, 175–194. welfare for dairy cows housed in strawyard or cubicle systems. Livest. Prod. Sci. 68, Temple, D., Manteca, X., 2020. Animal welfare in extensive production systems is still an 205–216. area of concern. Front. Sustain. Food Syst. 4, 545902. Haley, D.B., Rushen, J., de Passillé, A.M., 2000. Behavioural indicators of cow comfort: Thompson, G.E., Hartmann, P.E., Goode, J.A., Lindsay, K.S., 1981. Some effects of acute activity and resting behaviour of dairy cows in two types of housing. Can. J. Anim. fasting and climatic stresses upon milk secretion in Friesland sheep. Comparative Sci. 80, 257–263. Biochem. Physiol 70A, 13–16. Hamzaoui, S., Salama, A.A.K., Albanell, E., Such, X., Caja, G., 2013. Physiological Wheelock, J.B., Rhoads, R.P., Vanbaale, M.J., Sanders, S.R., Baumgard, L.H., 2010. responses and lactational performances of late-lactation dairy goats under heat stress Effects of heat stress on energetic metabolism in lactating Holstein cows. J. Dairy Sci. conditions. J. Dairy Sci. 96, 6355–6365. 93, 644–655. Hamzaoui, S., Caja, G., Such, X., Albanell, E., Salama, A.A.K., 2020. Milk production and Zahner, M., Schrader, L., Hauser, R., Keck, M., Langhans, W., Wechsler, B., 2004. The energetic metabolism of heat-stressed dairy goats supplemented with propylene influence of climatic conditions on physiological and behavioural parameters in glycol. Animals 10, 2449. dairy cows kept in pen stables. J. Anim. Sci. 78, 139–147. Hamzaoui, S., Caja, G., Such, X., Albanell, E., Salama, A.A.K., 2021. Effect of soybean oil Supplementation on milk production, digestibility, and metabolism in dairy goats under thermoneutral and heat stress conditions. Animals 11, 350. 9