All procedures involving animals in this study were conducted in strict accordance with institutional, national, and international regulations governing the ethical use of animals in research. The experimental protocol was reviewed and approved by the Ege University Local Animal Experiments Ethics Committee (Ege Üniversitesi Hayvan Deneyleri Yerel Etik Kurulu – HADYEK), Izmir, Turkey (Approval No. 2020-082; March 2020). Approval was granted following a comprehensive assessment of the study design, animal handling procedures, housing conditions, sampling frequency, and potential welfare impacts.

Throughout the study, all goats were monitored daily by trained farm personnel and a veterinarian to detect early signs of stress, discomfort, or health complications. Blood sampling was performed by the same experienced veterinarian using minimal manual restraint to minimize handling-related stress. No invasive procedures beyond standard jugular venipuncture were conducted, and animals experienced no prolonged or severe distress. Environmental conditions, including ventilation, shade availability, and access to clean water, were continuously maintained to meet welfare standards, especially during periods of high temperature–humidity index (THI).

This research complied with Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes and with the European Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes. Reporting followed the Animal Research: Reporting of In Vivo Experiments 2.0 guidelines to ensure transparency, reproducibility, and ethical accountability. All animals remained the property of the research farm, were returned to routine management after the study, and were not euthanized or subjected to any terminal procedures as part of the research.

The 4-month experimental study (May–August 2021) was conducted at the Faculty of Agriculture, Farm Animal Production, Research, and Application Center, Ege University, Izmir, Turkey. Izmir, on the Aegean coast, has a Mediterranean climate with hot, dry summers. Although the core summer months are June–August, animals in this region are exposed to high temperatures for 4–5 months beginning in May; therefore, May–August was selected to capture the onset, peak, and late phases of seasonal HS.

Daily ambient temperature (AT; °C) and relative humidity (RH; %) were recorded using a digital thermo-hygrometer (Hasvet Medical, Antalya, Türkiye) positioned at animal height (~1.5 m). Measurements were taken three times per day (08:00, 14:00, 20:00), and daily averages were used to calculate the THI using the following equation:

THI = T – (0.31 – 0.31 × RH) × (T – 14.4)

Where, T = AT (°C) and RH = RH (%)

Heat-load categories were defined as follows [10]:

No HS: THI < 22.3

Thirty clinically healthy Saanen goats from the Ege University research farm were enrolled. All animals received routine herd health care, including annual vaccinations and regular veterinary monitoring. Only nonpregnant does were included.

Goats were assigned to three age groups (n = 10 per group):

Young: 1–2 years, parity 1–2

At the start of the study (May), the average days in milk (DIM) for young, middle-aged, and old goats were ~70, 80, and 85 days, respectively. DIM increased by ~30 days per month, reaching ~160–175 days by August, indicating that all animals remained in mid-lactation. Monthly body weight (BW) and parity records were retrieved from farm records to assess group-wise BW changes under progressive HS (PHS).

The barn had an internal height of 2.5 m and consisted of four lactating-goat pens (96 m² each; total 384 m²). The gable roof was constructed from galvanized sheets, with internal heights ranging from 6 m (peak) to 4 m (sides). Each of 12 pens had a 10 m² outdoor exercise area (total 120 m²).

Goats were managed in a semi-open system that integrated shaded concrete pens and outdoor exercise yards. Indoor resting space provided approximately 1.5–2.0 m² per animal. Deep-litter bedding and concrete floors were cleaned daily. Ventilation was provided by open sidewalls, ridge openings, and mechanical ventilation systems. Goats had access to shaded outdoor areas during the day and returned indoors for milking and overnight housing. Automatic waterers supplied clean drinking water ad libitum.

All goats received the same total mixed ration throughout the study, consisting of:

0.8 kg of pelleted concentrate

Individual allotments were controlled using a radio-frequency identification (RFID)-based feeding system. Feeding occurred twice daily after milking. Water and mineral blocks were available ad libitum. The diet’s chemical composition, including dry matter, crude protein, neutral detergent fiber, acid detergent fiber, and metabolizable energy, was analyzed using standard Weende and Van Soest methods.

DAMY; kg/day) was automatically recorded using an electronic milking system integrated with RFID ear tags. Morning and evening yields were summed to calculate DAMY. Monthly averages per goat were used to evaluate seasonal trends and to assess relationships between DAMY, DIM, and endocrine/cellular markers.

Blood samples (2 mL) were drawn from the jugular vein twice monthly (~every 15 days) between 08:30 and 09:30 h to minimize circadian variability, particularly for CORT. Samples were collected without anticoagulant, allowed to clot, and centrifuged at 3,000 rpm (approximately 1,200 × g) for 10 min at 4°C. Serum aliquots were stored at −20°C until analysis of T3, T4, CORT, and HSP70. Storage duration did not exceed 5 months, and repeated freeze–thaw cycles were avoided.

Serum T3, T4, CORT, and HSP70 concentrations were quantified using commercial ELISA kits (Everon Life Science Co., New Delhi, India): Goat T3 ELISA Kit (Cat. No: EA0027Go), Goat T4 ELISA Kit (Cat. No: EA0011Go), Goat CORT ELISA Kit (Cat. No: EA0007Go), and Goat HSP70 ELISA Kit (Cat. No: E0057Go). All samples were run in duplicate, with mean values used for statistical analysis. Calibration curves were constructed for each assay, and internal quality-control sera were included. Sensitivities and detection ranges matched physiological caprine values. Intra- and inter-assay coefficients of variation were <10% and <15%, respectively; plates failing quality-control criteria were repeated.

All analyses were performed using SPSS Version 26.0 (IBM Corp., Armonk, NY, USA). Continuous variables (T3, T4, CORT, HSP70, and DAMY) were assessed for normality using the Kolmogorov–Smirnov and Shapiro–Wilk tests; homogeneity of variance was evaluated with Levene’s test.

Endocrine and cellular markers were analyzed using a general linear model (GLM) with age-group (young, middle-aged, old) and month (May–August) as fixed factors. The age × month interaction was tested. Goat ID was included as a random factor in the repeated-measures model. When the interaction was nonsignificant, main effects were interpreted. Tukey’s Honestly Significant Difference test was used for post hoc comparisons (p < 0.05).

DAMY was analyzed using repeated-measures analysis of variance, with month as the within-subject factor and age as the between-subject factor. Sphericity was assessed using Mauchly’s test, and the Greenhouse–Geisser correction was applied when appropriate.

Pearson correlation analysis assessed associations between DAMY and physiological markers. Correlation coefficients (r), p-values, and biological interpretations were reported. Statistical significance was defined as p < 0.05, and 0.05 ≤ p < 0.10 was considered a biological trend. Descriptive statistics are presented as mean ± standard error of the mean (SEM), with standard deviation calculated as SEM × √n.

Descriptive climatic data showed a progressive rise in heat-load from May to August. The lowest daily maximum temperature increased from 25.11°C in May to 39.48°C in August, representing a nearly 14°C rise. Mean monthly temperatures were 21.50°C, 27.18°C, 31.82°C, and 30.81°C in May, June, July, and August, respectively, while mean RH averaged 40.65%. As summer progressed, the THI increased from moderate to severe HS categories, and monthly differences in maximum temperature, AT, average humidity, and THI were statistically significant (p < 0.05), confirming sustained HS throughout the study. RH declined from May (54.9%) to July (38.2%), reflecting the typical arid summer pattern of the Mediterranean region.

THI showed a highly significant month effect (p < 0.0001), increasing from “no HS” in May (20.4) to “moderate HS” in June (24.8) and “severe HS” in July and August (28.5 and 27.9, respectively). These findings confirm that goats experienced persistent and intensifying HS throughout the experimental period (Table 1).

BW remained stable within each age-group, with minor fluctuations (<5 kg), indicating maintenance of overall body condition despite progressive HS.

DAMY declined across all age groups. Mean DAMY decreased from 2.59 ± 0.43 kg/day in May to 2.30 ± 0.39 kg/day in June, 2.11 ± 0.40 kg/day in July, and 1.88 ± 0.40 kg/day in August. Young goats produced 2.69, 2.41, 2.24, and 1.97 kg/day; middle-aged goats produced 2.69, 2.47, 2.27, and 2.06 kg/day; and old goats produced 2.40, 2.01, 1.84, and 1.60 kg/day from May to August, respectively (Table 2).

Repeated-measures analysis confirmed a significant month effect on DAMY (p < 0.05), with no strong age × month interaction, indicating similar seasonal declines across age groups. DAMY decreased significantly over the season (F_3,87 = 141.58, p < 0.0001), indicating that the reduction was due to PHS rather than random variation.

Clear endocrine adaptations to summer HS were observed. T4 concentrations declined across all age groups, with the largest reduction in middle-aged goats (from 92.96 nmol/L in May to 61.82 nmol/L in August; p < 0.01). Young and old goats showed decreasing trends, although the reductions in old goats were not significant. GLM analysis showed significant effects of month and age on T4 (p < 0.05), with a notable age × month interaction driven by the pronounced decline in middle-aged goats.

T3 showed a similar pattern. Young and middle-aged goats exhibited significant decreases in T3 across the summer months (p < 0.01), particularly in July and August, whereas the decline in old goats was mild and nonsignificant. These reductions reflect heat-induced downregulation of thyroid activity, a known mechanism for reducing endogenous heat production. Detailed values are shown in Table 3.

CORT levels tended to increase toward peak summer months, suggesting activation of the HPA axis under thermal load. However, the effects of month, age, and their interaction were not statistically significant (p > 0.05), reflecting high interindividual variability (Table 3). All CORT concentrations remained within physiological ranges. These findings support the view that CORT is a more variable HS indicator than T3, T4, or HSP70 under field conditions.

HSP70 concentrations increased progressively from May to August. Young goats showed a significant rise from 13.32 ng/mL in May to 17.85 ng/mL in August (p < 0.05). Middle-aged and old goats also showed upward trends, though these were nonsignificant. The GLM indicated a significant month effect and a tendency for an age × month interaction, driven by the stronger HSP70 response in young goats. These findings suggest greater cellular chaperone activity and stress resilience in younger animals.

The integrated results revealed a consistent physiological response to HS in Saanen goats: (1) reductions in T3 and T4, (2) modest and variable increases in CORT, (3) elevated HSP70 levels—particularly in young goats—and (4) a progressive decline in DAMY over the summer (Table 3).

Exploratory analyses suggested that goats with lower T3/T4 tended to show reduced DAMY, whereas higher HSP70 was associated with greater thermal strain. Correlation patterns reflected the complexity of physiological responses during HS (Table 4).

Correlation analysis based on age-group × month means (n = 12) showed:

T4 was strongly and positively correlated with DAMY (r = 0.78, p = 0.0027).

Higher thyroid activity was therefore associated with superior lactational performance, whereas increased HPA-axis activation was associated with lower productivity under HS.

This study assessed the endocrine and cellular responses of Saanen goats exposed to progressive natural summer heat, emphasizing age-dependent patterns in T3, T4, CORT, and HSP70 and their associations with DAMY. By monitoring the same goats over four consecutive Mediterranean summer months and differentiating three age groups, the study provides novel insight into life-stage–specific thermoregulatory strategies in high-producing dairy goats.

Marked reductions in thyroid hormones were observed in young and middle-aged goats, consistent with previous reports showing that HS suppresses thyroid activity to reduce metabolic heat production [5, 8]. In middle-aged goats, T4 declined by approximately one-third from May to August, the greatest reduction across age groups, suggesting that animals at peak lactation may rely heavily on thyroid downregulation as an energy-saving strategy under sustained HS. Older goats showed smaller, nonsignificant reductions, likely due to lower baseline metabolic and lactational demands that limit further suppression.

Similar age-dependent decreases in T3 and T4 have been reported in goats and other small ruminants, particularly in high-producing or mid-lactation animals [1, 2, 11]. These findings support the view that thyroid suppression is a conserved endocrine mechanism that reduces endogenous heat production and conserves energy and water during thermal stress.

CORT concentrations showed an upward trend toward the hottest months but were highly variable, with no significant effects of month or age-group. This pattern mirrors findings from other field studies, in which single-time-point CORT measurements often show large interindividual variation and inconsistent statistical significance during HS [12, 13]. CORT is strongly influenced by circadian rhythm, acute handling, and individual stress-coping capacity, all of which likely contributed to the observed variability.

The interaction between the HPA-axis and the hypothalamic–pituitary–thyroid axis is relevant. CORT can suppress thyrotropin-releasing hormone and thyroid–stimulating hormone, thereby reducing T3 and T4 synthesis [14–16]. Thus, even a modest increase in CORT may have contributed to declines in thyroid hormones. Future studies incorporating serial CORT sampling or metrics such as heart rate variability could provide a more accurate assessment of HPA responses to HS.

HSP70 increased across all age groups, with the most pronounced rise in young goats. This strong induction suggests a more vigorous cellular protective response in younger animals, consistent with evidence that younger organisms possess greater proteostatic capacity and enhanced stress-induced chaperone responses [6, 9]. In contrast, middle-aged and older goats showed smaller increases in HSP70, which may indicate reduced cellular plasticity or reliance on endocrine rather than cellular mechanisms for thermal adaptation.

Previous studies in goats and cattle identified HSP70 as a sensitive biomarker of thermal load, showing elevated expression during acute and chronic HS [6, 13, 17]. The present findings extend this literature by demonstrating age-dependent differences in serum HSP70 responses during natural summer HS and by integrating these results with endocrine and production traits.

Based on combined endocrine, cellular, and production data, we propose an age-dependent two-tier adaptation framework for HS in Saanen goats.

Young goats: Strong HSP70 induction and moderate thyroid suppression suggest that cellular defense mechanisms dominate their thermotolerance strategy. Their robust proteostatic capacity may help maintain physiological function despite high THI.

This crossover pattern, in which cellular defense is dominant in young goats and endocrine adjustment is dominant in middle-aged goats, highlights that HS adaptation is not uniform across age groups. These distinctions are important for developing age-specific management strategies and selecting heat-resilient genotypes [18].

As expected, DAMY declined from May to August, mirroring the increase in THI. Although correlations between DAMY and physiological markers were modest, goats with greater thyroid suppression or higher HSP70 tended to show greater production losses, indicating heightened thermal strain [19].

Key strengths of the study include its four-month longitudinal design, integration of endocrine, molecular, and production data, and focus on age-specific responses under authentic field conditions. Nonetheless, some limitations warrant consideration:

CORT was measured at a single morning time point; additional sampling or hair CORT could better represent HPA activity.

Future research should use multi-omics approaches (e.g., transcriptomics, metabolomics) to identify molecular networks underlying heat resilience and determine whether T4 suppression or HSP70 induction serves as a reliable phenotypic marker for breeding thermotolerant lines. Controlled studies manipulating THI, along with behavioral and feed-intake monitoring, would clarify causal relationships between physiological changes and production outcomes.