Programming effects of intrauterine hyperthermia on adrenal gland development

Maternal heat stress during late pregnancy can lead to intrauterine hyperthermia and affect fetal hypothalamic-pituitary-adrenal axis development and function. Herein, we investigated the effects of chronic environmental heat stress exposure of Holstein cows in the last 2 mo of gestation on their offspring's adrenal gland histomorphology and transcriptome. Cows in their last 54 ± 5 d of gestation were either heat-stressed (i.e., housed under the shade of a free stall barn) or provided heat-stress abatement via active cooling (i.e., via water soakers and fans) during a subtropical summer (Temperature-Humidity Index >68). Respiration rate (RR) and skin temperature (ST) were elevated in heat-stressed dams relative to the cows with access to heat abatement (23 bpm and 2 ◦C higher for RR and ST, respectively). Heifers born to heat-stressed cows experienced heat stress in utero (HS), while heifers born to actively cooled cows did not (CL). The adrenal gland was harvested from 6 heifers per group that were euthanized at birth (d 0; n = 12) or one week after weaning (d 63; n = 12). Circulating cortisol was measured from blood samples collected weekly throughout the pre-weaning period. At d 63, heifers that experienced HS while developing in utero had heavier adrenal glands, with a greater total tissue surface area and thickness of the zona glomerulosa (ZG), fasciculata (ZF), and re-ticularis (ZR), compared with CL heifers. In addition, the adrenal gland of in utero HS heifers had less cells in the ZG, more and larger cells in the ZF and larger cells in the ZR, relative to CL heifers. Although no changes in circulating cortisol were observed through the pre-weaning period, the transcriptomic profile of the adrenal tissue was altered by fetal exposure to hyperthermia. Both at birth and on d 63, approximately 30 pathways were differentially expressed in the adrenal glands of in utero HS heifers relative to CL. These pathways were associated with immune function, inflammation, prolactin signaling, cell function, and calcium transport. Upstream regulators significantly activated or inhibited in the adrenal glands of heifers exposed to intrauterine hyperthermia were identified. Maternal exposure to heat stress during late gestation caused an enlargement of their offspring's adrenal glands by inducing ZG and ZF cell hypertrophy, and caused gene expression changes. These phenotypic, histological, and molecular changes in the adrenal gland might lead to alterations in stress, immune, and metabolic responses later in life.


INTRODUCTION
The last trimester of gestation in cattle is characterized by the greatest rate of fetal growth and fetal organ development (Bauman and Currie, 2001).Stressors disrupting maternal homeostasis during this sensitive window of fetal development can result in intrauterine stress that permanently alters postnatal phenotype, a phenomenon known as fetal programming (Godfrey and Barker, 2001).Environmental heat stress severely alters maternal-fetal homeostasis in dairy cows.Heat stress results from a cow's imbalance between heat production and heat dissipation to the surrounding environment (NRC, 2001).Exposure to chronic maternal heat stress during late pregnancy triggers physiological and epigenetic changes in the developing fetus, leading to significant reductions in milk production later in life (Dahl et al., 2019;Laporta et al., 2020).
Maternal exposure to heat stress conditions during late gestation derails the developmental progression of the fetal mammary gland, impacting the heifer's future milk production (Dado-Senn et al., 2022).Our group established that in utero exposure to environmental heat stress also results in enlarged adrenal glands in dairy heifers (Dado-Senn et al., 2021).The adrenal gland is a key organ that links the primary systems responsible for maintaining homeostasis during stress, the hypothalamic-pituitary-adrenocortical (HPA) axis and the sympatho-adrenomedullary axis (Berger et al., 2019).In other species, direct chronic stress exposure increases adrenal weight (Marti et al., 1994;Kuipers et al., 2003;Ulrich-Lai et al., 2006).Indeed, direct exposure of pubertal rats to heat stress conditions exerts mild hydropic degeneration of adrenal glands (Mete et al., 2012).Conversely, maternal cold stress during late gestation in rats reduces fetal adrenal size and enlarges the maternal adrenal gland (Dahlöf et al., 1978).Yet, it is unclear whether the overgrowth caused by fetal exposure to hyperthermia in dairy heifers is caused by hyperplasia or hypertrophy or if it affects the specialized zones of the adrenal cortex, which might impact the synthesis and secretion of specific hormones.
The secretory products of the maternal and fetal adrenal glands are involved in a myriad of physiological and metabolic functions.In cows, the release of cortisol by the fetus increases before birth (Fowden and Silver, 1995;Comline et al., 1974), which is important for fetal development, organ maturation, and preparation for life outside the womb (Liggins, 1994).After birth, the hypothalamic-pituitary-adrenal axis helps orchestrate several homeostatic mechanisms, such as ionic balance, through the actions of aldosterone produced in the zona glomerulosa (ZG) and integrating stress response via glucocorticoids and catecholamines produced in the zona fasciculata (ZF), and medulla, respectively.Research suggests that the adrenal glands may also play a role in immune function through the production of inflammatory factors, which could be an evolutionary consequence of the sensitivity of these endocrine glands to blood-borne signals, including pathogen-associated molecular patterns (Deak, 2008).Whereas short-term stress activates the HPA axis causing spikes in cortisol synthesis and release from the adrenal gland, chronic stress reduces cortisol synthesis and systemic tissue responsiveness (Schöbitz et al., 1994;Wright et al., 2010).
Alterations to the adrenal gland by direct exposure to heat stress or in utero exposure to heat stress have been reported in monogastric animals (Gley et al., 2019;Zheng et al., 2021;Maskal et al., 2021).However, there is a lack of information on the effects of maternal exposure to heat stress during late gestation on the offspring's postnatal development and function of the adrenal gland in dairy cattle.Our objective was to investigate the effects of fetal hyperthermia during late pregnancy on postnatal adrenal gland growth and function in dairy heifers.We hypothesized that maternal exposure to heat stress during late gestation would trigger morphological, cellular, and molecular changes to the offspring's adrenal glands, impacting its development and function specifically at birth and at weaning.

Experimental Design and Maternal Treatments
All experimental procedures on live animals were approved by the University of Florida Institutional Animal Care and Use Committee (#201910599) and performed in accordance with their guidelines and regulations and complying with ARRIVE guidelines.Details of the experimental design and animal care were previously reported by Dado-Senn et al. (Dado-Senn et al., 2021).Eighty-two multiparous pregnant Holstein cows were exposed to environmental heat stress conditions during a subtropical Florida summer.The hourly Temperature-Humidity Index (THI) average ranged from 75 to 85, remaining above 68 for 95% of the pre-calving experimental period.One group of cows had access to the shade of a free stall barn with natural ventilation (n = 41), while another group was provided active cooling via the addition of fans and water soakers (n = 41) for the last 54 ± 5 d of gestation.Weekly recordings of dam respiration rate (RR) and skin temperature (ST) were taken during the experimental period (pre-calving) at 12:00 p.m.Both thermal indices were elevated in heatstressed dams, relative to the cows with access to heat abatement (RR: 77.4 vs. 53.5 ± 0.63 bpm, ST: 36.0 vs. 34.1 ± 0.05 •C, respectively).All cows were fed the same total mixed rations (TMR) ad libitum during the prepartum period.The prepartum diet was based on corn silage, sorghum silage, and concentrate, and it was formulated to meet or exceed nutritional requirements for this stage according to NRC (2001).The treatments of heifers born to these cows were reflected by their intrauterine environment: heifers born to heat-stressed dams were considered in utero heat stressed (HS), and heifers born to actively cooled dams were considered in utero cooled (CL).At birth (d 0), heifers were fed 3.78 L of high-quality colostrum (Brix % > 22) within 2 h of birth.Prophylactic interventions and vaccinations were performed according to the farm standard operating protocols (see Dado-Senn et al., 2021).Six heifers from each in utero treatment were euthanized at birth (d 0, before colostrum feeding).The remaining heifers were raised as a cohort, receiving identical postnatal management and nutrition, to isolate the in utero effects.then fully weaned off milk).At 60 d of age, calves were reassigned to group-housed pens where they received ad libitum starter grain concentrate and water.In utero HS heifers consumed the same liquid intake but less starter grain concentrate relative to in utero CL heifers, particularly from 42 to 63 d of age (0.81 vs. 0.68 ± 0.04 kg/d DM).Calves were raised from October to December in a temperate environment (average hourly THI = 68, dropping to <66 during night hours).Heifers exposed to heat stress in utero had higher rectal temperature and sweating rate than those exposed to cold stress, but no difference in respiration frequency was observed (Davidson et al., 2022).A second subset of 6 heifers per in utero treatment was euthanized one week after weaning (d 63 of age).

Adrenal Gland Collection: birth and weaning
In both euthanasias (d 0 and d 63), live body weight was recorded, and major organs of interest were harvested and weighed (Dado-Senn et al., 2021).The adrenal glands were located above the kidneys, carefully dissected, washed in sterile PBS, and weighed.The left adrenal was immediately bisected after harvest and snap-frozen in liquid nitrogen for RNA sequencing.The right adrenal was fixed in 10% neutral-bufferedformalin overnight.

Blood collection and cortisol analysis
Blood samples were collected at birth (d 0, before colostrum feeding) and weekly thereafter (d 7, 14, 28, 35, 42, 49, and 56, approximately 3 h post morning feeding) until d 63 (d 63, pre-euthanasia) using serumseparating vacutainer tubes (Cat#3666430, Franklin Lakes, NJ) by venipuncture of the jugular vein, allowed to clot for 20 min and stored at −20°C.Cortisol concentration was determined as described and validated by Reinemann et al. (2003).

Histology and cell proliferation analyses
The bisected right adrenal gland was stained using hematoxylin and eosin to visualize the tissue microstructure.Whole tissue imaging was performed using the Keyence BZ-X810 microscope (Keyence, Itasca, IL) stitch feature.Images (4X) were stitched using the BZ-X800 analyzer software to evaluate the total tissue surface area.The thickness of each zone was quantified from the whole tissue image (ImageJ software, version 1.47, National Institutes of Health).Additionally, images (20X) were taken in each cortex zone (i.e., zona glomerulosa, ZG; zona fasciculata, ZF; zona reticularis, ZR).Four images from 4 random locations within each zone were taken to determine the cell number (count) by zone.Fields averaged 84,920 μm 2 for ZF, 68,724 μm 2 for ZG, and 69,967 μm 2 for ZR.Cell count was performed using the Hybrid Cell Count application (Keyence Corp., Osaka, Japan).Cell proliferation per adrenal cortex zone was assessed via immunohistochemistry (Ki67 mouse anti-human Ki67, DAKO #M7240, clone MIB-1) and quantified according to protocols described by Field et al. (2021).

RNA extraction, library generation, sequencing, and expression analyses
Total RNA was extracted from approximately 5 g of adrenal gland tissue using the TRIzol ® (Invitrogen Corp., Carlsbad, CA.) following the manufacturer's instructions.RNA concentration was assessed using a NanoDrop (#ND-2000, Thermo Scientific, Wilmington, DE) and RNA integrity was measured using an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA).Samples with 28S/18S >1 and an RNA integrity number ≥7 were used for library construction.Novogene Bioinformatics Technology Co., Ltd.(Beijing, China) performed library construction and sequencing.Libraries were constructed using the NEBNext Ultra II RNA Library Prep Kit for Illumina (#E7775; New England BioLabs, Ipswich, MA) following the manufacturer's protocol.The libraries were sequenced using the Illumina NovaSeq 6000 platform (Illumina Inc., CA), generating 150 base-pair paired-end reads.The RNA-Seq data can be accessed by NCBI GEO with accession number GSE235160.The quality of the RNA-sequencing reads was tested using the software FastQC (version 0.11.7,Babraham Bioinformatics, UK).Read trimming was performed using Trim Galore software (version 0.4.4,Babraham Bioinformatics, UK) with the following options:-paired,-length 50,-clip_R1 10,-clip_R2 10,-three_prime_clip_R1 1, and-three_prime_clip_R2 1.After edition, sequencing reads were mapped to the bovine reference genome ARS-UCD1.2using the software Hisat2 (v2.1.0;Kim et al., 2015a).The number of reads that mapped to each annotated gene in the ARS-UCD1.2was obtained using the python script htseq-count (v0.6.1p1) using the option intersection-nonempty (Anders et al., 2015).Differentially expressed genes were detected using the R package edgeR (Robinson et al., 2010).This package combines the use of trimmed mean of M-values as the normalization method of the count data, the estimation of tagwise negative binomial dispersion values using an empirical Bayesian approach and generalized linear models and likelihood ratio tests for detecting differentially expressed genes between treatments.

Ingenuity pathway analysis
The ensembl ID of differentially expressed genes (False Discovery Rate, FDR ≤0.20) was uploaded into the IPA software (QIAGEN Inc., http: / / www .qiagenbioinformatics.com) to determine the association of DEGs with Canonical Pathways and Upstream Regulators.We performed an unrestricted analysis based on z-score and p-value, calculated with a righttailed Fischer's exact test that reflects the likelihood that the association between a set of genes in our data set and an upstream regulator or a pathway is significant [P-value ≤0.05 (-log10 ≥ 1.3)].The z-score infers potential activation states of predicted transcriptional regulators.We only considered functions with a |z-score| > 2.

Statistical Analysis
Histological data, protein, and DNA content were analyzed using the MIXED procedure of SAS (v.9.4 SAS Institute Inc., Cary, NC), with the fixed effect of treatment (CL vs. HS).Days 0 and 63 were analyzed separately.Blood cortisol was analyzed using the MIXED procedure of SAS, with the fixed effects of treatment, time, and their interaction.Data are presented as least squares means ± standard error of the mean (LS means ± SEM).Cell count data were analyzed relative to the image area and are reported as number of cells/mm 2 .Cell proliferation data were analyzed using the GLIM-MIX procedure of SAS as a generalized linear mixed model with a binomial distribution and a logit link function.Results for cell proliferation are presented as the proportion of positive cells to total cells (%) per total section.Statistical significance was declared at P ≤ 0.05 and tendencies at 0.05 < P ≤ 0.10.

Adrenal gland macro-and microstructure
At birth, the adrenal gland gross weight (3.92 vs. 4.27 ± 0.50 g, P = 0.63), and as a % of birth live body weight (0.10 vs. 0.12 ± 0.01 g/kg of BW for CL and HS, respectively, P = 0.42) was similar between in utero treatments (Dado-Senn et al., 2021).At d 63 of age, the adrenal gland gross weight tended to be greater (6.50 vs. 5.11 ± 0.54 g, respectively, P = 0.09), and was greater when expressed as a % of body weight for in utero HS relative to CL heifers (0.08 vs. 0.06 ± 0.01 g/ kg BW; P = 0.04; Dado-Senn et al., 2021).

Cell proliferation
No positive ki-67 cells were observed in the adrenal medulla or zona reticularis, and the total number of positive Ki67 nuclei was not different between in uetro treatments (P > 0.76) in the ZG or the ZF (Supplementary Material Fig. S1, Laporta 2023).The number of positive ki-67 cells per tissue area was overall relatively low for both treatment groups (2.5 and 2.1 ± 1.03 positive cells for every 100 cells counted for in utero CL and in utero HS, respectively), which hindered the assessment of the proportion of Ki67-positive/total nuclei by zone.

Circulating cortisol
Cortisol concentrations decreased with time (P < 0.01), but did not differ between in utero HS and CL treatments throughout the pre-weaning period (5.40 and 6.11 ± 1.07 ng/mL, respectively; P = 0.51, data not shown).

Adrenal gland RNA-sequencing
On average, 23 million paired-end reads were obtained per sample through Illumina sequencing.Approximately 92% of the reads were successfully mapped to the bovine genome.Only uniquely mapped reads were considered in the subsequent analyses.The accession number GSE235160 can be used to have access to the sequencing data through the NCBI GEO database.

Enriched Pathways
Heifers exposed to intrauterine hyperthermia had 30 significantly enriched pathways in the adrenal gland at birth.Most genes within these pathways were upregulated in the adrenal gland of in utero HS, relative to CL heifers (Figure 4).Broad biological categories enriched were adrenal regulation of blood flow, immune function, inflammatory response, cell cycle, and cell differentiation.Within the adrenal regulation of blood flow pathway, adrenomedullin signaling and α-adrenergic signaling pathways were significantly enriched in the adrenal glands of in utero HS heifers.Fourteen enriched pathways were involved in immune function and inflammatory responses, including NFκB, IL-6, IL-13, IL-17 signaling, STAT3, Th1 and Th2 activation pathways.The ID1 signaling, endothelin-1 signaling, and GAP junction signaling were among the 7 enriched pathways involved in cell function and proliferation processes.In addition, the prolactin signaling pathway was enriched in the adrenal gland at birth, with all 8 genes being upregulated in the adrenal gland of in utero HS heifers.
After weaning, 30 pathways were significantly enriched in the adrenal glands of in utero heat-stressed heifers.Conversely, most genes within these pathways were downregulated in the adrenal gland of in utero HS, relative to CL heifers (Figure 4B).Approximately 70% of enriched pathways at this time point were associated with immune function and inflammatory responses.These pathways include B-cell maturation, dendritic cell maturation, Th1 and Th2 activation, granulocyte adhesion and diapedesis, and communication between innate and adaptive immune cells.Another 2 pathways of interest significantly enriched in the adrenal gland of in utero HS relative to CL heifers, with all downregulated genes, were the calcium transport and the unfolded protein response pathways.Meanwhile, the oxidative phosphorylation was the only pathway with all upregulated genes in the in utero HS adrenal gland at this time point.

DISCUSSION
Various prenatal (intrauterine) stressors have been shown to alter the function of the fetal HPA, leading to permanent alterations in postnatal endocrine responses to stress in several mammalian species (reviewed by Kapoor et al., 2006).In humans, programming of the HPA axis during fetal development is linked to the development of adult diseases (Barker, 2002;Ward et al., 2004;Phillips et al., 2005).In cattle, exposure to prenatal heat stress, particularly during the last trimester of gestation, results in long-lasting effects that manifest years later in reduced lifespan and higher culling rates  (Laporta et al., 2020).Indeed, exposure of the developing fetus to chronic hyperthermia has profound effects on postnatal organ growth and function (Dado-Senn et al., 2021).Herein, we sought to investigate the effects of prenatal hyperthermia on postnatal adrenal gland growth, morphology, and function in dairy cattle.
Our group has reported carry-over effects of in utero heat stress on the macro and microstructure of cells, tissues, and organs, including the mammary gland (Dado-Senn et al., 2022), skin (Davidson et al., 2022), ovaries (Roth, 2017), and the gastrointestinal tract (Ahmed et al., 2021).One key finding of the present study is the significant enlargement of the post-weaning adrenal glands in heifers subjected to prenatal heat stress and not immediately after birth.Adrenal hypertrophy is known to be one of the primary responses to stress, which is caused by the effects of a stressor on the peripheral limb of the HPA axis (Gamallo et al., 1986;Zelena et al., 2003;Ulrich-Lai et al., 2006).The response to stress is a complex network of neuronal and cellular interactions leading toward the release of catecholamines by the adrenal medulla or the production of glucocorticoids in the ZF and mineralocorticoids in the ZG of the adrenal cortex (Bechmann et al., 2021).Systemic cortisol was assessed weekly throughout the pre-weaning period as a functional output.Intrauterine HS did not alter circulating cortisol, and the physiological range was similar to that observed for dairy heifers in other studies (Huber et al., 2013;Röder et al., 2022).Previous studies have reported divergent effects of late-gestation heat stress on circulating cortisol concentrations.Some reports indicate that exposure to summer heat stress increases blood cortisol in dams compared with cows calving in cooler months (Lacetera et al., 2005), while others report no changes in circulating cortisol in the dam (Tao et al., 2012).Yet, changes in cortisol levels of the calf exposed to intrauterine hyperthermia immediately after birth have been reported (Tao et al., 2012;Laporta et al., 2017), suggesting prenatal HS may impair the HPA fetal development.Still, it is unknown if the changes persist.
The weaning period in dairy heifers involves a shift from liquid intake to solid feed, and from individual housing to group housing, adding a social component to the transition (Kim et al., 2011;Steele et al., 2016).The remarkable differences observed in adrenal gland size and cell hypertrophy triggered by prenatal HS could have been exacerbated by an altered response to the weaning period, resulting in overstimulation of the adrenal cortex.In utero HS heifers are predisposed to have elevated core body temperatures and reduced DMI (Dado-Senn, et al., 2021;Davidson et al., 2022).It is tempting to speculate that in utero HS heifers may experience some level of physiological stress to the weaning transition, resulting in an increase in the thickness of the cortical zones and cell size but not in the number of cells or changes in systemic cortisol.Yet, this speculation warrants further investigation.
The hypertrophy and hyperplasia observed in the ZF and the decreased number of cells in ZG, are in agreement with previous reports of chronic stress-inducing alterations in a subregion-specific manner (Ulrich et al., 2006).Repeated exposure to exogenous ACTH during chronic exposure to stressors increases the volume of cells and induces hyperplasia in ZF (Nussdorfer and Mazzocchi, 1983;Aguilera et al., 1996, Lehoux et B).The upstream regulators were identified using the Ingenuity Pathway Analysis (IPA, QIAGEN Inc., https: / / digitalinsights .qiagen.com)regulation z-score algorithm.Only functions with a z-score > |2 were considered.The P-value was calculated with a right-tailed Fisher's exact test and reflects the likelihood that the association between a set of genes in our data set and a pathway is significant [P-value ≤0.05 (-log10 ≥ 1.3)].The y-axis displays the upstream regulators' annotation, and the x-axis displays the z-score (positive z-scores = activation, negative z-scores = inhibition of the upstream regulators).1998), while impairing ZG function (Aguilera et al., 1996, Lehoux et al., 1998, Suwa et al., 2000).
Exposure to acute heat shock prompts a cellular response via intra-and extracellular signals that trigger gene expression changes, including increased expression of heat shock proteins, increased glucose and amino acid oxidation and decreased fatty acid metabolism, activation of the endocrine system stress response, and immune system activation (Collier et al., 2008).The enriched pathways and upstream regulators identified in the adrenal glands of in utero HS heifers suggest that this response is also observed under chronic stress, and it seems to be conserved even after the heat stress insult has ceased.Acclimatization refers to a physiological state resulting from homeorhetic responses to chronic exposure to environmental stress, and it is under endocrine regulation (Bauman and Currie, 1980;Horowitz, 2002).The endocrine system mediates acclimation to environmental heat stress with an elevation of glucocorticoids during initial HS exposure, and it could increase the concentration of cortisol after prolonged periods of heat stress (Collier et al., 2017).The enrichment of multiple pathways related to immune function and inflammatory response in the adrenal gland corroborate a systemic multi-organ response of the offspring to intrauterine hyperthermia.It also corroborates previous studies from our group that reported the effects of direct heat stress or in utero exposure to heat stress on immune function and inflammation (Tao et al., 2012;Laporta et al., 2017;Laporta et al., 2020).Maternal exposure to heat stress during gestation is known to impact offspring immune establishment and competence by reducing the apparent efficiency of absorption of immunoglobulins from colostrum (Hinson and Raven, 2006;Harvey and Sutcliffe, 2010;Tao et al., 2021;Dado-Senn et al., 2022), decreasing peripheral blood mononuclear cell proliferation, compromising Tlymphocyte response (Tao et al., 2012), and decreasing the size of primary lymphoid organs, such as the thymus (Ahmed et al., 2021;Dado-Senn et al., 2021).Our findings suggest that prenatal exposure to heat stress has programming effects on the local immune response of the offspring and on inflammatory mediators in the adrenal gland, highlighting the role of the adrenal gland in immunity.
The adrenal glands are key modulators of immune function through the traditional HPA cascade, with glucocorticoids and catecholamines regulating cytokine expression and immune cell activation (Sanders and Straub, 2002;Necela, 2004).However, there is also an inflammatory circuit in the adrenal glands, which confers to the adrenal glands an intrinsic immune sensitivity (Engström et al., 2008).For instance, interlukin-1 appears to be a critical, early mediator of immune cell activation in the adrenal gland, followed by an induction of prostaglandin E 2 (PGE 2 ) synthetic enzymes, which can influence the adrenal's response to immune stress (Engström et al., 2008).Interleukin-1, which was a predicted upstream regulator activated by in utero HS at birth, has been shown to induce arachidonic acid metabolites and PGEs through activation of the transcription factor nuclear factor-kappa B (NFκB; Liu et al., 2017), which was also an enriched pathway and an upstream regulator complex identified in the adrenal glands of in utero HS heifers at birth.Downstream molecules of the NFκB pathway, such as interleukins IL1β, IL6, and TNF, were also predicted to be activated in prenatally HS heifers' adrenal glands at birth.The NFκB pathway is a central regulator of proinflammatory gene expression, and the enrichment of NFκB signaling pathway could explain the activation of several immune and inflammatory-related pathways in the adrenal glands of in utero HS heifers.The canonical NFκB members also control T-cell differentiation and effector functions, including Th1, Th2, and Th17, which secrete different cytokines such as IL-17 secreted by Th-17 (Liu et al., 2017).Both the Th1 and Th2 activation pathways and the IL-17 signaling pathways were enriched in the adrenal glands of in utero HS heifers, supporting evidence of the intrinsic inflammatory circuit in the adrenal glands.This intrinsic inflammatory state of the adrenal gland could be triggered in response to systemic inflammation, which may occur at birth in the newborn following exposure to several pathogen-associated molecular patterns and elevated cortisol in utero.Thus, in utero, exposure to heat stress could program a more exacerbated local immune and inflammatory response in the offspring.Post-weaning, most enriched pathways were related to immune function and inflammatory response.Most DEGs within these pathways were downregulated in the utero heatstressed adrenal glands, indicating an orchestrated inhibition of these processes.
The prolactin (PRL) signaling pathway was enriched in the adrenal gland of neonatal heifers exposed to in utero hyperthermia.Mature dairy cows experiencing heat stress have increased circulating prolactin (Collier et al., 1982;Amaral et al., 2010).Prolactin regulates many physiological processes via direct and indirect effects on cellular proliferation, differentiation, and cell survival (Yu-Lee, 2002).In dairy cows, prolactin has been shown to stimulate immune cell proliferation, potentially improving cell-mediated immune function (Amaral et al., 2010).In vitro, PRL decreases mammary epithelial cell apoptosis by improving the action of insulin growth factor-1 (IGF-1; Zarzyńska et al., 2007;Sobolewska et al., 2009).Upon stress induction, PRL synergizes with ACTH in the pituitary gland (Ogle  and Kitay, 1979;Lalli and Figueiredo, 2022), stimulating adrenal hypertrophy and corticosterone synthesis (Lalli and Figueiredo, 2022).Herein, the PRL signaling pathway was enriched in the adrenal glands of heifers harvested immediately after birth (i.e., after chronic in utero hyperthermia).Ruggiero et al. ( 2021) established a conserved role for prolactin signaling in regulating adrenal function with a clear sexual dimorphic expression of PRL receptors in the adrenal glands of mice, with female adrenal potentially being more sensitive to PRL effects.In the present study, only heifers were used, but adrenal PRL receptor expression and systemic PRL were not evaluated.A study investigating the effects of hyperprolactinemia induced by domperidone reported an increased relative adrenal weight and increased cell area of the ZF only after chronic exposure to high levels of circulating PRL (Silva et al., 2004).It is possible that the observed alterations in the PRL signaling pathway of in utero HS heifers could have led to the observed immune and inflammatory responses.Prolactin has been shown to modulate the expression of genes in the JAK/Stat (Janus kinase/signal transducers and activators of transcription) pathways, modulating signal transduction by leukocytes and cytokines (Dogusan et al., 2001).Several immune pathways were altered in the adrenal at birth, possibly triggered by the immunostimulatory properties of PRL (Yu-Lee, 1997; Dogusan et al., 2001;Matera et al., 2001).Our research has provided valuable insights into the systemic and endocrine postnatal impacts of fetal adrenal gland dysfunction triggered by maternal stressors potentially leading to immune dysfunction and inflammation.
The top differentially expressed geens in the adrenal glands of in utero HS heifers were RPL15, LGI3, TREML2, H2AC6, and PTMA.The LGI3 gene encodes for the Leucine-rich glioma inactivated 3 proteins of the LGI family of proteins.In adipose tissue, LGI3 may act as a pro-inflammatory adipokine associated with increased TNF-α expression through NFκB, potentially playing a role in adipose tissue inflammation in obesity (Kim et al., 2015b).The expression of the protein LGI3 is regulated at the transcription level in the brain; thus, if the same is true for different tissues, the downregulation of LGI3 of in utero HS heifers could indicate a lesser abundance of this protein, although the biological function in adrenal glands remains unknown.The expression of GBPB2 is associated with immune infiltration and immunomodulators in glioma studies (Li et al., 2022) and has been suggested to regulate cell growth and migration of glioma cells (Ren et al., 2022).The TREML2 gene encodes for the TREM-like transcript 2 protein, which can enhance T-cell activation, and it is upregulated in response to inflammation (King et al., 2006;Thomas et al., 2016;Xu et al., 2019).
Although the role of LGI3, GBP2, and TREML2 in adrenal glands has not been determined, all these genes are associated with immune processes.The orchestrated downregulation at weaning is in accordance with the inhibition of immune-related pathways identified at weaning in the adrenal glands of in utero HS heifers.Ribosomal protein L15 is used as a reference gene in PCR analysis of bovine tissues (Wu andSaggau, 1997). However, Dalanezi et al. (2019) reported downregulation of RPL15 transcripts in oocytes matured in thermoneutral conditions versus oocytes matured in heat stress conditions.Ribosomal protein L15 encodes for a ribosome protein involved in ribosome biogenesis and protein synthesis, a rate-limiting step for cell growth and proliferation.A downregulation of RPL15 could indicate an attenuation of such processes.Herein, the expression of ki-67, a nuclear protein used as a proliferation marker, and no difference between in utero treatments was observed.In addition, the lack of proliferative responses and the enrichment of several immune-related pathways suggests that the upregulation of PTMA in adrenal glands of in utero HS could be related to immunomodulatory effects, since prothymosin-α (ProTα, encoded by PTMA), is reported to exert immunomodulatory effects when found extracellularly (Samara et al., 2016).We also report an upregulation of GABA B R2 and a concomitant enrichment of the Calcium Transport I pathway, although predicted to be inhibited due to the downregulation of genes involved.γ-Aminobutyric acid (GABA) receptors have been detected in peripheral tissues, modulating intracellular calcium concentration (Zhou et al., 2010;Mizuta et al., 2011).Mizuta et al. (2011) reported that GABA B receptors heighten transient increases in ionic calcium concentrations, increasing contraction in human airway smooth muscle.Conversely, agonists of GABA B R are used as a muscle relaxant due to their presumable action of decreasing ionic Calcium conductance (Hill and Bowery, 1981;Wu and Saggau, 1997;Curtis et al., 1997).Therefore, our findings suggest a possible epigenetic modulation of intra-adrenal GABA signaling, potentially affecting vascular activity.

CONCLUSION
Exposure of pregnant cows to high temperatures in their last trimester of gestation leads to intrauterine hyperthermia and causes macrostructural and cellular changes in the adrenal gland of the offspring.The enlarged adrenal glands result from hypertrophy of adrenal cortex cells, particularly cells of the ZG and ZF, in response to chronic fetal stress.These macro-and microstructural changes were associated with molecular alterations, specifically the gene expression signature Briefly, heifers were housed in individual sand-bedded pens under the shade of an open-sided barn and fed 0.87 kg DM milk replacer per d (MR, UF Special 28/20 Bova Medicated offered as 0.45 kg MR, mixing rate 120 g/L; 12.0% solids) over 2 feeding periods.Calf Starter (Purina Animal Nutrition LLC, Shoreview, MN) was offered ad libitum starting at 4 d of age.Weaning began at 49 d and ended at 56 d of age (0.23 kg MR mixed in 1.89 L warm water for 2 feedings per d for one week, Guadagnin et al.: IN UTERO HEAT STRESS IMPACTS THE ADRENAL GLAND Guadagnin et al.: IN UTERO HEAT STRESS IMPACTS THE ADRENAL GLANDMaterial Table

Figure 1 .
Figure 1.Histological evaluation of adrenal glands from Holstein heifers euthanized at d 63 of age (one-week post-weaning).Heifers were born to either heat-stressed or actively cooled cows during the last 54 ± 5 d of gestation (HS, n = 6, light red; CL, n = 6, light blue).(A) Hematoxylin and eosin staining of longitudinal 5μm sections of the entire adrenal gland; images were captured by stitching 4X images using a BZ-800 microscope and software.Abbreviations: ZG = zona glomerulosa, ZF = zona fasciculata, ZR = zona reticularis.(B) Violin plots highlighting the thickness of each zone of the adrenal gland.Asterisk (*) indicates P-value ≤ 0.05.
Figure 3. Volcano plot depicting differential gene expression in adrenal gland tissue of dairy heifers born to heat-stressed or actively cooled cows during the last 54 ± 5 d of gestation (HS heifers, n = 6; CL n = 6).Adrenal glands were harvested at birth (d 0) and one week after weaning (d 63).Only the volcano plot of d 63 time point is shown.The y-axis displays the -log10 P-value for each gene, and the x-axis displays the log 2 fold change for that gene relative to the CL group.Purple dots (positive log 2 fold change) indicate upregulation, and green dots (positive log 2 fold change) indicate downregulation in the HS group (FDR ≤20%); black dots indicate FDR ≥20%.
Figure 4. Significantly enriched pathways in the adrenal gland tissue of Holstein dairy heifers born to heat-stressed or actively cooled cows during the last 54 ± 5 d of gestation (HS, n = 6; CL, n = 6) at birth (d 0, A) and one-week post-weaning (d 63, B).Pathways were identified based on a differential gene expression data set using the Ingenuity Pathway Analysis (IPA, QIAGEN Inc., https: / / digitalinsights .qiagen.com)P-value and z-score algorithm through the Canonical Pathways tool.The P-value was calculated with a right-tailed Fisher's exact test and reflects the likelihood that the association between a set of genes in our data set and a pathway is significant [P-value ≤0.05 (-log10 ≥ 1.3)].The y-axis displays the name and total number of genes in each pathway, and the x-axis displays the number of differentially expressed genes (DEGs) within each pathway and the significance of enrichment (-log 10 p-value).Green bars display the number of downregulated genes, and purple bars display the number of upregulated genes in each pathway in HS relative to CL heifers.

Figure 5 .
Figure 5. Upstream regulators identified in adrenal gland tissue of dairy heifers born to heat-stressed or actively cooled cows during the last 54 ± 5 d of gestation (HS, n = 6; CL, n = 6) at birth (d 0, A) and one-week post-weaning (d 63, B).The upstream regulators were identified using the Ingenuity Pathway Analysis (IPA, QIAGEN Inc., https: / / digitalinsights .qiagen.com)regulation z-score algorithm.Only functions with a z-score > |2 were considered.The P-value was calculated with a right-tailed Fisher's exact test and reflects the likelihood that the association between a set of genes in our data set and a pathway is significant [P-value ≤0.05 (-log10 ≥ 1.3)].The y-axis displays the upstream regulators' annotation, and the x-axis displays the z-score (positive z-scores = activation, negative z-scores = inhibition of the upstream regulators).
Guadagnin et al.:  IN UTERO HEAT STRESS IMPACTS THE ADRENAL GLAND related to cell function, immune function, and inflammation responses.This study underscores the importance of providing heat abatement to pregnant dairy cows in summer, as this can lead to fetal hyperthermia and adrenal gland dysfunction in their offspring.
Guadagnin et al.: IN UTERO HEAT STRESS IMPACTS THE ADRENAL GLAND