Abstract

   The content of androgen from adrenal is elevated under castration, and
   the mechanisms of compensatory secretion of adrenal androgen remain
   unknown. This study was designed to compare the transcript profiles
   between adrenals from noncastrated, orchiectomized and immunocastrated
   Yiling goats. Fifteen goats were randomly divided into three groups:
   pVAX-asd injection (control) group,
   pVAX-B2L-(G[4]S)[3]-kisspeptin-54-asd immunization (PBK-asd) group, and
   surgical castration (SC) group. Subsequently, serum was collected every
   two weeks after the initial immunization for hormone assays. At week 14
   after immunization, adrenal glands were collected for transcriptome
   sequencing and qPCR. Serum testosterone concentration was significantly
   reduced in PBK-asd and SC group, demonstrating the effectiveness of
   castration. Both surgical and immunized castration resulted in adrenal
   hyperplasia, and thickness of adrenal cortex elevated. The specific
   genes involving castration were enriched in many pathways, including
   Steroid hormone biosynthesis pathway. The adrenocorticotropic hormone
   (ACTH), which promotes the production of adrenal steroids, and
   dehydroepiandrosterone (DHEA), a steroid hormone secreted by adrenal
   glands, both increased after castration. Further construction of
   co-expression network for transcription genes and traits (including
   adrenal weight and cortex thickness, ACTH and DHEA concentration)
   showed that the trait-related genes were enriched in multiple
   steroid-related pathways. These results showed that adrenal
   compensatory hyperplasia and androgen secretion caused by castration
   may involve in ACTH-induced steroid hormone synthesis.

   Subject terms: Transcriptomics, Vaccines

Introduction

   Castration is a common management technique in livestock and poultry
   production to reduce aggressive behavior, improve carcass quality, and
   obtain higher economic benefits^[32]1. However, surgical castration is
   increasingly restricted due to animal welfare concerns. As a result,
   alternatives such as immunocastration have been developed.
   Immunocastration is a functional castration method that modifies the
   hormone balance of the hypothalamic-pituitary–gonadal axis (HPG axis)
   by reducing the secretion of gonadotropin-releasing hormone (GnRH),
   luteinizing hormone (LH), and follicle-stimulating hormone (FSH). This
   ultimately leads to a decrease in gonadal hormone levels, resulting in
   gonadal atrophy and functional inhibition^[33]2. Kisspeptin, produced
   by the KISS1 gene with its cognate receptor G protein-coupled receptor
   (GPR54), serves as an upstream hormone of the HPG axis^[34]3. Knockdown
   of KISS1 can cause gonadal atrophy in animals, making it a potential
   target for immunocastration^[35]4.

   In male animals, androgens are secreted mainly in the testis and
   adrenal gland^[36]5. After castration, androgen secretion from testis
   is inhibited, resulting in a significant decrease in serum androgen
   levels. However, studies show that the adrenal gland becomes active and
   begins to secrete increasing amounts of androgens, playing a
   compensative role^[37]6. The adrenal gland is composed of about 80%
   cortex, which is divided into three layers: Zona reticularis (ZR), Zona
   fasciculata (ZF) and Zona glomerulosa (ZG). The ZG cells mainly secrete
   mineralocorticoid such as aldosterone while the ZF cells secrete
   glucocorticoids, primarily cortisol (corticosterone in rats). The ZR
   cells mainly secrete sex hormones like dehydroepiandrosterone (DHEA)
   and estradiol, and a small amount of glucocorticoids^[38]7–[39]9. DHEA
   can be converted to testosterone in peripheral tissues, while
   testosterone can reduce the activity of the adrenal gland or inhibit
   the secretion of adrenal androgen^[40]10. When testosterone secretion
   is inhibited, adrenal gland may compensate by producing androgen
   precursor substances^[41]6. Several researches in rodents have shown
   that castration induces compensatory hyperplasia of adrenal cortex,
   enhancing the secretory activity of adrenal cortex cells, and thus can
   compensate for the lack of testosterone^[42]11,[43]12. However, it has
   not been reported whether the adrenal glands of goat undergo similar
   changes in rodents after castration.

   The hypothalamus–pituitary–adrenal (HPA) axis is a complex
   neuroendocrine system that regulates the secretion of adrenal
   corticosteroids^[44]13. In castrated adult rats, there is an increase
   in the levels of androstenedione, testosterone, and
   dihydrotestosterone, which are metabolized by DHEA from the adrenal
   glands increase^[45]11. Non-castrated males generally exhibit lower
   overall activity in the adrenal axis compared to castrated males, and
   castration can result in increased activity in the HPA
   axis^[46]14,[47]15. The adrenal glands of castrated animals have an
   enhanced capacity to release corticosterone or
   cortisol^[48]12,[49]14,[50]16,[51]17.

   Transcriptome sequencing is a powerful tool for identifying
   differentially expressed genes associated to important economic traits
   of animals. It involves analyzing all RNA molecules transcribed by the
   cell, tissue, or organism under specific physiological or environmental
   conditions^[52]18. Functional annotation and enrichment analysis of
   these genes can provide insights into the molecular basis underlying
   various biological processes^[53]19. To date, only one study has
   investigated the modulatory effects of sex hormones on the adrenal
   transcriptome profile of gonadectomized rats. This study found that
   testosterone stimulated the expression of genes associated with lipids
   and cholesterol metabolism in male rats, while estradiol inhibited gene
   expression in intracellular signaling pathways^[54]20. As we know, no
   study has been conducted on the effects of castration on the
   transcriptome profile of the goat adrenal gland. Therefore, our study
   performed transcriptome sequencing on the adrenal glands of
   orchiectomized and immunocastrated Yiling goats to investigate specific
   adrenal genes involved in castration. The study analyzed the
   relationship between these genes and serum ACTH and DHEA levels,
   provided the first comprehensive transcriptional database of the
   adrenal gland following castration, hoping to shed light on the
   molecular mechanisms underlying this process.

Results

Effect of KISS1 immunocastration

   The serum anti-kisspeptin antibody and testosterone concentration in
   the experimental goats were presented in Table [55]1. At the second
   week after initial immunization, the PBK-asd group produced higher
   anti-kisspeptin antibody than the control and surgical castration
   groups (p < 0.05). The antibody titer reached its peak at 10 weeks and
   lasted until the 14 weeks after the initial immunized (p < 0.05). From
   the beginning to the end of the experiment, the serum testosterone
   concentration in the surgical castration group was significantly lower
   than the control group (p < 0.05). Compared with the control group, the
   serum testosterone concentration of the PBK-asd group was significantly
   decreased from 4 to 14 weeks (p < 0.05).

Table 1.

   The serum anti-kisspeptin antibody and testosterone concentration.
   Week Traits Group
   C PBK-asd SC
   2 Anti-kisspeptin antibody (OD450) 0.18 ± 0.02^a 0.70 ± 0.30^a
   0.16 ± 0.03^b
   4 Anti-kisspeptin antibody (OD450) 0.16 ± 0.04^a 0.93 ± 0.26^b
   0.13 ± 0.03^a
   6 Anti-kisspeptin antibody (OD450) 0.12 ± 0.04^a 1.75 ± 0.35^b
   0.04 ± 0.02^a
   10 Anti-kisspeptin antibody (OD450) 0.16 ± 0.04^a 1.78 ± 0.21^b
   0.14 ± 0.08^a
   14 Anti-kisspeptin antibody (OD450) 0.14 ± 0.02^a 1.12 ± 0.11^b
   0.15 ± 0.03^a
   2 Serum testosterone level (ng/ml) 1.50 ± 0.39^a 1.27 ± 0.58^a
   0.03 ± 0.003^b
   4 Serum testosterone level (ng/ml) 1.86 ± 0.35^a 1.01 ± 0.47^b
   0.03 ± 0.003^c
   6 Serum testosterone level (ng/ml) 2.07 ± 0.31^a 0.38 ± 0.09^b
   0.03 ± 0.000^b
   10 Serum testosterone level (ng/ml) 2.33 ± 0.47^a 0.53 ± 0.13^b
   0.03 ± 0.002^b
   14 Serum testosterone level (ng/ml) 2.57 ± 0.39^a 0.96 ± 0.18^b
   0.026 ± 0.002^c
   [56]Open in a new tab

   The serum anti-kisspeptin antibody and testosterone concentration in
   control (n = 5), PBK-asd immunized (n = 5) and surgical castrated
   (n = 5) goats from 2 weeks after initial immunization to the end of the
   experiment at 14 weeks. All data are expressed as mean ± SE. Different
   superscript letter in the same row indicates a significant difference
   (p < 0.05).

Adrenal section and cortical thickness

   Hematoxylin-eosin (HE) stains of goat adrenal showed very clear pink
   cortex and darker medulla, and clear ZG, ZF and ZR of cortex
   (Fig. [57]1). The ZG was relatively thin, and the cells were arranged
   into columnar and lumpy. The ZF was thick, and the cells were arranged
   in a cable shape. The thickness of ZR was between ZG and ZF, and the
   boundary between ZR cells and ZF was not obvious, but the boundary
   between ZG and ZF was clear. Compared with the control group
   (Fig. [58]1a), the adrenal cortex was thicker in the immunocastration
   (Fig. [59]1b) and surgical castration (Fig. [60]1c) groups. We randomly
   chosed several places of cortex in each group to measure its thickness
   and found that adrenal cortex thickness in the surgical and immune
   castration group were significantly higher than that in the control
   group (p < 0.01, Fig. [61]1D). No difference of cortex thickness was
   found between the two castration groups. Castration increased adrenal
   weight (p < 0.05), but there was no difference of adrenal weight
   between the two castration groups [adrenal weight (g): C 0.63 ± 0.01;
   PBK-asd 0.71 ± 0.03; SC 0.75 ± 0.02; in each group n = 5, mean ± SE].

Figure 1.

   [62]Figure 1
   [63]Open in a new tab

   Adrenal section and cortical thickness. HE stains of adrenal gland. (a)
   is control group (C), (b) is immunocastration group (PBK), (c) is
   surgical castration group (SC), (d) is the adrenal cortical thickness
   in the three groups. All groups are observed and photographed under a
   4X object lens. The pink part on the periphery is adrenal cortex,
   darker parts of the interior is adrenal medulla (M). In each panel, the
   part of cortex from left to right is the zona glomerulosa (ZG), zona
   fasciculata (ZF), and zona reticularis (ZR), respectively. All data are
   expressed as mean ± SE. Significance is indicated by ** (p < 0.01) vs
   control group.

The expression profiles of differential genes in adrenal

   The adrenal tissues of Yiling goats were sequenced for transcriptome
   analysis to investigate the expression profiles of genes. The
   correlation analysis demonstrated that the samples obtained from PBK
   and SC groups formed a distinctive cluster, the control group belonged
   to another separate cluster (Fig. [64]2a). Compared to the control
   group, the immunocastration group and the surgical castration group had
   843 and 772 differentially expressed genes (DEGs), respectively, in
   which 493 DEGs overlapped (Fig. [65]2b). There were 158 up-regulated
   and 654 down-regulated DEGs in the immunocastration group while there
   were 30 up-regulated and 686 down-regulated DEGs in the surgical group
   (Fig. [66]2c). Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway
   enrichment analysis was performed based on the overall differential
   genes of the SC group or PBK group (Fig. [67]2d). The result showed
   that the number of pathway terms from PBK versus C groups was 250 and
   that from SC versus C groups was 260, in which 237 terms overlapped,
   including steroid hormone related pathways, namely Steroid hormone
   biosynthesis, Cholesterol metabolism.

Figure 2.

   [68]Figure 2
   [69]Open in a new tab

   The expression profiles of differential genes in adrenal. Expression
   profiles of differential genes in adrenal. (a) Heat map of correlation
   coefficient for samples based on the gene expression level. Control
   group (C) 28, 36, 46, 49, 50; Surgical castration group (SC): 25, 31,
   35, 45, 48; Immunocastration group (PBK): 26, 30, 38, 40, 43. (b) The
   Venn diagram for the numbers of DEGs in castration groups compared with
   control group; (c) Volcano map of DEGs in castration groups compared
   with control group. Red indicates the up gene while green indicates the
   down gene. (d) The bubble diagram with Gene number demonstrating the
   involved KEGG pathways. Generatio is the percent of DEGs to the whole
   genes in the pathway. Gene count is the number of the DEGs in the
   pathway.

Quantitative real-time PCR

   We randomly selected 5 DEGs in PBK and SC groups to verify the accuracy
   of the gene expression analysis. The results showed that the expression
   of these 10 DEGs in qRT-PCR was consistent with the results of
   transcriptome sequencing (Fig. [70]3), which proved the reliability of
   the high-throughput RNA-Seq data.

Figure 3.

   [71]Figure 3
   [72]Open in a new tab

   Quantitative real-time PCR. Validation of the RNA-seq results. PBK and
   SC show the specific gene expression level of RNA-seq and qPCR of PBK
   vs C and SC versus C groups. C: control group; SC: surgical castration
   group (SC); PBK: immunocastration group; fpkm: fragments per kilobase
   million mapped reads. Data are shown as means ± SE.

Serum hormone of HPA axis

   The ACTH and DHEA hormones in the serum of goats were shown in
   Fig. [73]4. The ACTH level in the SC group was significantly higher
   than that of the control group from the 6th week to the end of the
   experiment (p < 0.05). And from the tenth week to the end of the
   experiment, the ACTH level in the PBK group was significantly higher
   than that of the control group (p < 0.05). The values of serum DHEA in
   SC and PBK groups were higher than that of the control group from the
   fourth week to the end of the experiment, but only the values at eighth
   week were significantly higher than that of the control group
   (p < 0.05). The results showed that both surgical castration and
   immunocastration resulted in the increment of ACTH secretion from
   pituitary gland and DHEA secretion from adrenal gland.

Figure 4.

   [74]Figure 4
   [75]Open in a new tab

   Serum hormone of HPA axis. Serum ACTH and DHEA levels in goats of the
   immunocastration group (PBK), the control group (C) and the surgical
   castration group (SC). All data are expressed as mean ± SE. Different
   letters at the same week indicate significant differences between the
   groups (p < 0.05).

Co-expression network construction and module mining

   The weighted gene co-expression network analysis (WGCNA) method was
   used to cluster the expression genes after castration (Fig. [76]5a).
   The genes were clustered into six modules clustered by co-expression,
   and two modules were selected as the main modules (Fig. [77]5b,c). The
   black module was associated with the traits of adrenal, including
   adrenal weight, adrenal cortex thick, and the hormone secretion of ACTH
   and DHEA. The green module was also related to the adrenal weight,
   adrenal thick and secretion of DHEA. For black module (883 genes), the
   terms of KEGG pathway included MAPK signaling pathway, and Wnt
   signaling pathway, which is related to ACTH function. For green module
   (1677 genes), the enriched pathway included Prostate cancer, MAPK
   signaling pathway, Estrogen signaling pathway, Choline metabolism in
   cancer, Prolactin signaling pathway, Insulin signaling pathway, Renal
   cell carcinoma and Cortisol synthesis and secretion, all of which are
   related to ACTH and DHEA function.

Figure 5.

   Figure 5
   [78]Open in a new tab

   Co-expression network construction and module mining. Co-expression
   network construction and module mining. (a): Gene clustering tree based
   on hierarchical clustering. Modules corresponding to branches are
   labeled with colors indicate by the color bands underneath the tree. A
   total of six modules are identified; (b): Module-traits relationship.
   Each row corresponds to a module feature gene, and each column
   corresponds to a feature. Each cell contains the corresponding
   correlation and p-value; red indicates positive correlation and green
   indicates negative correlation; (c): The functional enrichment results
   for each module. The terms with the lower FDR are shown in the figure.

Discussion

   In this study, it was observed that serum testosterone level decreased
   sharply after surgical castration. The kisspeptin antibody and
   testosterone concentration in serum can be used as indicators to
   evaluate the effectiveness of immunocastration. The serum testosterone
   concentration in goats significantly decreased after KISS1 DNA vaccine
   injection, while the anti-kisspeptin antibody increased. These results
   proved the effectiveness of surgical castration and immunocastration.
   Previous researches on rodents have shown that castration can cause
   adrenal hyperplasia. A study conducted on Sahara gerbils during the
   breeding season demonstrated a significant increase in the height of
   adrenal cells, nucleus diameter, and overall activity of the adrenal
   axis in both surgically castrated and immunized gerbil groups compared
   to the control male gerbil group^[79]21. Similar results were observed
   in pigs, where the overall activity of the adrenal axis was higher in
   both surgically castrated and immunized male pigs compared to the
   control male pigs^[80]15. Our research on Yiling goats showed a
   significant thickening of the adrenal cortex in castrated goats,
   indicating adrenal compensatory hyperplasia in both surgically
   castrated and immunized goats.

   Here, we aimed to investigate the pathways and genes involving in the
   changes induced by castration in the adrenal gland. Transcriptome
   analysis revealed that surgical castration and immunocastration
   resulted in 843 and 772 differentially genes, respectively, with 493
   genes overlapping between the two methods. Subsequent functional
   annotation analysis identified several pathways, including Steroid
   hormone biosynthesis and Cholesterol metabolism, which were later found
   out to be related to adrenal function enhancement after castration.
   Steroid hormones are involved in many important physiological processes
   and play an important role in maintaining homeostasis^[81]22. All
   steroid hormones, including those synthesized by the adrenal cortex,
   are derived from cholesterol^[82]23. ACTH plays a crucial role in
   enhancing the steroidogenic capacity of adrenal cells by stimulating
   adrenal hypertrophy and hyperplasia^[83]24, which provides a cellular
   site for steroidogenesis^[84]25. After the release of ACTH from the
   pituitary gland, the intracellular cAMP level increases, leading to the
   activation of cAMP-dependent protein kinase (PKA) and ultimately
   resulting in the increased biosynthesis of steroid
   hormones^[85]26,[86]27. ACTH acts primarily on the zona fasciculata and
   zona reticularis of the adrenal cortex to promote the synthesis of
   glucocorticoids and androgen precursors, respectively^[87]28. DHEA is
   the most abundant circulating steroid hormone secreted by the adrenal
   cortex and its production is regulated by ACTH^[88]29. Castration can
   result in adrenal hyperplasia due to the stimulation of ACTH, which
   leads to variation in DHEA concentrition. Based on the findings of
   transcriptome analysis above, we measured the concentration of steroid
   hormones in serum, including ACTH and DHEA, and found that both
   hormones increased following castration. The decrease in testicular
   androgen after castration may trigger a feedback response in the
   pituitary gland, resulting in increased secretion of ACTH. The abundant
   ACTH stimulates the proliferation of adrenal cortical cells, which
   produce pregnenolone through cholesterol side chain cleavage, mediated
   by the enzyme encoded by CYP11A1 (P450scc)^[89]30. Pregnenolone is
   further converted to 17-hydroxypregnenolone and then to DHEA under the
   action of microsomal P450C17 and 17,20-lyase of P450C17,
   respectively^[90]31. DHEA is a weak androgen secreted by the adrenal
   cortex and can be converted to testosterone in peripheral
   tissues^[91]9. Our findings suggested that the ACTH-mediated
   steroidogenesis pathway mag be involved in enhancing adrenal secretion
   after castration. This indicates that the HPA axis activity is
   enhanced, and the adrenal gland secretes more androgen as a
   compensatory response to castration.

   Castration in male rats may lead to a decrease in inflammatory
   responses, possibly due to compensatory secretion of glucocorticoids
   and the inhibition of COX-2 expression^[92]32. COX-2 isoenzyme is
   localized in the PVN of the hypothalamus, which plays a role in
   modulating the response of CRH to HPA axis^[93]33. The HPA axis is
   mainly controlled by neurons in the PVN, which releases CRH and drives
   ACTH release from the anterior pituitary^[94]34,[95]35. Glucocorticoids
   secreted by the adrenal cortex have anti-inflammatory effects, and DHEA
   has also been shown to attenuate inflammatory responses^[96]36–[97]38.
   This study suggests that castration may lead to a decrease in
   inflammatory response in male rats. This could be attributed to an
   increase in compensatory secretion of glucocorticoids after castration,
   which can counteract the effect of some inflammatory factors and
   inhibit the expression of COX-2. Additionally, binding of ACTH to its
   receptors can promote activation of signaling cascades that initiate
   adrenal-specific steroidogenic transcriptional programs, which may
   explain the increment in ACTH-controlled steroidogenic enzymes after
   castration^[98]39. Overall, these findings suggest a potential
   association among the HPA axis, inflammation, and castration in male
   rats. Binding of ACTH to its receptors leads to the activation of PKA
   and MAPK-dependent signaling cascades that collaborate to initiate
   adrenal-specific steroidogenic transcriptional programs. This process
   ultimately leads to glucocorticoids synthesis and secretion from the
   adrenal cortex^[99]40. After castration, there is an increase in ACTH
   secretion and a high expression of ACTH-controlled steroidogenic
   enzymes. COX-2 gene expression is inhibited by ACTH-mediated steroid
   hormone biosynthesis. COX-2 may be involved in the negative regulation
   of steroidogenesis by MAPK because the expression of COX-2 was
   down-regulated in adrenal tissues after castration^[100]41. The
   activation of COX-2 signaling pathway may involve in the biosynthesis
   of steroid hormones mediated by ACTH. Thus, these two pathways may
   cooperate to participate in the compensatory process of the adrenal
   gland.

   In this paper, we found that several genes of the Wnt family existing
   in co-expression network modules. Canonical Wnt signaling plays a
   crucial role in maintaining the homeostasis of the adrenal cortex and
   is involved in the differentiation of progenitor cells in a variety of
   organ systems^[101]42. The adrenal hypertrophy caused by castration in
   male mice is a result of increased production of
   corticosteroid^[102]43. During times of higher physiological demand for
   steroid production, endocrine signals transcend paracrine inputs and
   promote differentiation and steroidogenesis^[103]44. It has been
   demonstrated that cAMP/PKA inhibits canonical Wnt signaling
   pathway^[104]45. The ACTH signaling pathway antagonizes canonical Wnt
   signaling in the adrenal cortex. Canonical Wnt signaling can also be
   suppressed by noncanonical Wnt signaling and has been associated to
   adrenal steroid hormone production^[105]46. Therefore, it can be
   hypothesized that noncanonical Wnt signaling plays a synergistic role
   with ACTH in the production of adrenal steroid hormones. Based on these
   findings and our RNA-seq results, we hold the opinion that the adrenal
   ACTH signaling pathway is enhanced after castration, and this pathway
   collaborates with COX-2 signaling pathway and Wnt signaling pathway to
   stimulate steroid hormone secretion and participate in the compensatory
   secretion of adrenal androgen.

Conclusion

   Our results showed that both immunocastration and surgical castration
   can cause adrenal hyperplasia and compensatory androgen secretion in
   goat, and the potential reason for this may be the up-regulation of
   ACTH-induced steroid hormone pathway.

Materials and methods

Ethics statement

   The study was carried out in accordance with the guidelines for the
   care and use of experimental animals for the scientific purpose set by
   the Ministry of Science and Technology (Beijing China, No.398, 2006).
   The institutional animal care and use ethics committee of Huazhong
   Agricultural University approved the protocol (HZAUGO-2019-006). All
   methods are reported in accordance with ARRIVE guidelines.

Vaccines and animals

   The immunocastration vaccine pVAX-B2L-(G[4]S)[3]-kisspeptin-54-asd
   (PBK-asd) and empty plasmid vaccine (pVAX-asd) were previously
   constructed by Wassie et al.^[106]47 The PBK-asd is a recombinant
   kisspeptin-54 and B2L vaccine, which is an effective immunocastration
   vaccine, can inhibit testicular development and animal sexual behavior.
   The 15 healthy male Yiling goat (5 month, 17 ± 1.8 kg) were provided by
   Yichang Bailihuang Husbandry Co. Ltd (Yichang, Hubei, China, five goats
   each group). All the goats were sheltered in the same house with
   similar environmental conditions. All goats were in the spermatogenesis
   stage throughout the experiment. They were provided ad libitum water
   and fed 3–4 kg/day silage and peanut vine in ratio of 6: 4 supplemented
   with concentrate (1% of their body weight) containing 64% corn, 20%
   soya bean, 8% rice bran, 5% premix, 1.5% salt, and 1.5% soda ash.

Vaccine immunization and sample collection

   The goats were randomly divided into three groups: pVAX-asd injection
   (control group), PBK-asd immunization and surgical castration. Surgical
   castration was performed two weeks before the plasmid injection. The
   goats in the immunized treatment and control group were injected
   intramuscularly with PBK-asd and pVAX-asd plasmid (1 mg/dose),
   respectively. Both PBK-asd and control group goats received a
   two-booster immunization with 3-week intervals.

   Blood samples were collected from the jugular vein every two weeks from
   day 0 of immunization to the end of the experiment (lasting 14 weeks).
   Blood samples (5 ml) were collected using a coagulant collecting vessel
   (Aosaite, Shandong, China) which were stored at 4 °C overnight,
   separated for serum by centrifuging at 3000 rpm for 10 min at 4 °C and
   stored at −20 °C. Serum was used for hormone assays by ELISA. The goats
   were anesthetized at 14 weeks after the initial immunization by
   intramuscular injection with 0.1 ml/kg xylazine hydrochloride (Shengda,
   Changchun, Jilin, China), and all efforts were made to minimize
   suffering. Both adrenal glands were collected from each goat, the left
   adrenal gland was fixed with 4% paraformaldehyde, the right adrenal
   gland was immediately frozen in liquid nitrogen and then stored at
   −80 °C. Half of the collected adrenal gland was frozen in liquid
   nitrogen to extract RNA for transcriptome sequencing and fluorescence
   quantification, and the another half was fixed for H&E analysis.

Hematoxylin–eosin stains of adrenal gland

   The adrenal was fixed with 4% paraformaldehyde solution for 2 weeks,
   washed with water, dehydrated using graded ethanol, vitrified by
   dimethylbenzene and deposited in holly oil. The tissues were sliced
   into sections of 4 µm thick and stained with HE. Each adrenal section
   was observed with Eclipse-Ci™ microscope (NIKON, Chiyoda, Tokyo,
   Japan). Image Pro Plus 6.0 was used to trace the ZG band edges and
   ZR-medulla boundaries throughout the section and the average thickness
   was measured.

Library construction and RNA sequencing

   Total RNA was reverse-transcribed using the PrimeScript RT reagent Kit
   with gDNA Eraser (Toyobo, Osaka, Japan). The cDNA was amplified and
   purified with Qubit dsDNA high-sensitivity kit, then the fragments of
   approximately 200 bp in length were selected for library construction
   with NEBNext Ultra RNA library preparation kit. The libraries were
   paired-end sequenced using the Illumina HiSeq X Ten platform from
   Megagenomics Company (Beijing, China).

Sequencing data filtering

   The Raw data obtained by sequencing were filtered, and Clean Reads were
   obtained by removing connectors and low-quality Reads. Clean reads were
   aligned with reference genome sequences using Hisat2 software for
   transcript assembly. FPKM (Fragments Per Kilobases of transcript per
   Million mapped reads) was used to homogenize gene expression.

Differential gene expression analysis

   The differentially expressed genes (DEGs) were calculated using the
   DESeq2 package^[107]48. We compared the control samples with
   immunocastration group or surgical castration group to obtain the DEGs.
   The P value for each gene was obtained based on the model of negative
   binomial distribution. The fold changes were estimated using the DESeq2
   package. The screening criteria for DEGs were p < 0.05 and |fold
   change|≥ 2. Then the specific DEGs were combined to one DEG union set
   for the subsequent analysis.

Determination of serum hormones

   Testosterone, anti-kisspeptin, ACTH and DHEA were measured by
   enzyme-linked immunoassay (ELISA) kits (Huaying Bioengineering Co.
   Ltd., Beijing, China) as described in the instructions. The detection
   range of testosterone was 0.06–2.88 ng/ml, the sensitivity was 1.0U/L
   for anti-kisspeptin. The ACTH and DHEA hormone levels were measured
   using antibody against goat ACTH, DHEA. The detection range of ACTH and
   DHEA were 3–85 ng/L and 0.32–17.62 ng/ml. All inter- and intra-assay
   coefficients of variation (CV) were less than 10%.

Weighted correlation network analysis

   Gene co-expression networks were constructed using the R package
   weighted correlation network analysis (WGCNA)^[108]49. The Fragments
   Per Kilobase of exon model per Million mapped fragments (FPKM) values
   of all grouped genes were inputted for analysis. To determine the
   functions of the genes in the trait-associated modules, genes were
   subjected to web server KOBAS
   ([109]http://kobas.cbi.pku.edu.cn/kobas3/genelist/) for pathway
   enrichment. p < 0.05 was the threshold used for the significant terms.

Quantitative real-time PCR

   The cDNA was diluted 10 times and then used as the template of qRT-PCR.
   The reaction system contained 5 μl of 2 × Hieff® qPCR SYBR Green Master
   Mix (Yeasen Biotech, Shanghai, China), 1 μl of cDNA, 0.5 μl of upstream
   primer, 0.5 μl of downstream primer and deionized water in a final
   volume of 10 μl. The PCR was conducted at 94 °C for 5 min, followed by
   35 cycles of 94 °C for 10 s, 55 °C for 20 s and 72 °C for 6 s, and
   5 min extension. The gene expression level was determined by the
   2^-ΔΔCt algorithm, and the goat β-actin gene was used as an internal
   control^[110]50. Each sample had three biological replicates, and the
   gene expression level was presented as the means ± standard errors (SE)
   (n = 3). The primer sequences of the selected genes were shown in Table
   [111]2.

Table 2.

   Primer sequences for qRT-PCR test.
   Gene Forward primer (5′-3′) Reverse primer (5′-3′) Product length (bp)
   Gene bank accession number
   ACTB^a gtcaccaactgggacgacat catcttctcacggttggcct 132
   [112]NM_001314342.1
   COX1 acgtcgatacacgggcttac ctcatatcatggcgggggac 129 [113]NC_005044
   CYTB caaagccactctcacccgat tgtggggttgttcgatcctg 115 [114]NC_005044.2
   ND1 cctacccataccctaccccc tgtgctactgctcgtaaggc 150 [115]NC_005044.2
   SSU72 ctgccagtgtatccagcaca agtagaagcagacggtgtgc 110 [116]NC_030823.1
   MT1A atggacccgaactgctcct agcagctcttcttgcaggag 100 [117]XM_005692001
   DES gcgcaggatcgaatctctca gctgtgaggtctggtttcga 138 [118]XM_013972881
   ACTN2 cttgatcactgcgcatgagc gggttgcttgagctgatcct 135 [119]XM_018042499
   CDC6 cgtctcggacagtggacaaa ggttgtcatcacccagacgt 135 [120]NC_030826.1
   CYP26B1 tgaccatgcaggagctgaag cagcagctgcatgataagcg 98 [121]NC_030818
   TRIM17 acagcctgagtgtgcagtac ggcaccacatcctcttggaa 95 [122]NC_030814.1
   [123]Open in a new tab

   ^aGene ACTB was used as an internal control.

Data analysis

   Data were analyzed using the software program IBM SPSS Statistics
   (Version20). The normality of the variance of the data were checked
   using Shapiro–Wilk test. The data of antibody and hormone concentration
   were analyzed by One-way ANOVA and Tukey’s multiple comparison post-hoc
   tests. The relative quantitative method was used for gene expression
   analysis, the control group was default to 1, and the results of the
   treatment group were measured by their ratio to the control group. All
   data were expressed as mean ± SE, and statistically significant was
   considered when p < 0.05.

Acknowledgements