2024 Volume 71 Issue 2 Pages 181-191
Vertebrate animals often exhibit sexual dimorphism in body shape. In mammals, decreases in sex hormones caused by testicular castration can affect body shape and occasionally lead to pathologies such as obesity. Post-castration obesity can also be problematic for the health of companion animals, including non-mammals. In order to understand the mechanism of post-castration obesity in vertebrates other than mammals, experimental models are required. We examined whether the Iberian ribbed newt, which has recently become a popular experimental model for amphibian research, could serve as a model for analyzing changes in body shape after castration. In newts, new testes can be regenerated after removal of differentiated testes. We analyzed changes in body shape by removing the testes under conditions in which they could regenerate or conditions in which they could not regenerate. Removal of the testes reduced blood testosterone levels. The body weight and abdominal girth of the newts were increased compared with normal male newts. Transcriptome analysis of the liver showed that a set of genes related to lipid metabolism was continuously up-regulated in castrated newts. Our study suggests that changes in body shape after castration are common in vertebrates. Iberian ribbed newts are thus a suitable model for comparative studies of the long-term physiologic- and endocrine-level effects of castration.
MANY VERTEBRATE ANIMALS exhibit sexual dimorphism in body shape [1-5]. In mammals, including humans, factors that determine body shape, such as size and balance of body parts, amount of skeletal muscle, and body fat percentage, differ between males and females [3, 6]. Sexual dimorphism in vertebrates is thought to be characterized by sex hormones, because body shape differences become apparent with sexual maturation. Hormone preparations used in the medical field are known to affect various sexual characteristics, including body shape [7, 8]. Decreases in levels of sex hormones caused by castration also exert various effects on the animal body. For example, weight gain is a prominent change that occurs after removal of the testes. Post-castration obesity is a problem for the health of companion animals; indeed, approximately half of castrated male dogs are reportedly obese [9-11].
One of the factors thought to cause obesity after castration is a decrease in the level of male sex hormones such as testosterone due to removal of the testes. In castrated dogs, levels of male hormones are reduced, and the animal’s temperament becomes calmer and less active. Castrated dogs reportedly become obese because their caloric intake exceeds their caloric expenditure due to reduced testosterone levels, a calmer temperament, and less activity [12]. However, the exact causes and mechanisms of post-castration obesity are not fully understood. Few relevant studies have been carried out using experimental biological methods, and most of the studies of obesity induced by castration have been done using companion cats, dogs, or human medical data. In addition, studies of late-onset changes in body shape are difficult to conduct using mice, which have a short lifespan. Therefore, new experimental models are required to investigate the mechanisms leading to weight gain and changes in body shape after castration.
Amphibians such as frogs, salamanders, and newts are quadrupeds that breathe via lungs. These animals have long been used as experimental models because of the many similarities in body structures and physiological mechanisms with those of mammals. If castration could be shown to induce weight gain or a change in body shape in amphibians, this would suggest that these animals would make excellent model organisms. Newts are smaller than mice, require less space for housing, and are more cost-effective to maintain. These advantages can be used to screen for drugs that effectively prevent weight gain. In addition, as an established experimental system, it would be easy to conduct experiments via genome editing or reverse genetics, in part because as amphibians, newts have powerful regeneration capability. For example, newts can regenerate their testes [13, 14]. Newts have paired testes in the abdominal cavity. Each testis consists of two regions: a spermatogenic region and glandular tissue that produces testosterone and other male sex hormones [13, 15]. In particular, newts that regenerate the testes can be analyzed to determine whether a return to the original body shape occurs with testicular regeneration. The Iberian ribbed newt, Pleurodeles waltl, was recently established as a useful model animal because of its suitability for molecular genetic research and reverse genetics, including genome editing [16, 17]. Using Iberian ribbed newts, it is thus possible to conduct bioinformatic analyses using a published reference gene model [17, 18]. Information obtained at the genetic level would enable comparative studies between newts and mammals, thereby furthering our understanding of the effects of castration on body shape.
In the present study, we examined in detail the changes in body shape associated with weight gain after testicular removal in newts and analyzed how these changes were affected by testicular regeneration. To investigate the mechanism underlying these changes, we performed a transcriptome analysis of the liver using RNA sequencing (RNA-seq).
Iberian ribbed newts (hereinafter referred to as “newts”) were obtained from a closed colony at the Hiroshima University Amphibian Research Center. The newts were reared at 25 ± 2°C in aquarium tanks and fed every 2 days with a compound feed (Kyorin, Hyogo, Japan). All procedures were carried out in accordance with the guidelines of the Institutional Animal Care and Use Committee of Hiroshima University and the national guidelines of the Ministry of Education, Culture, Sports, Science and Technology of Japan.
Testicular removal (castration) and histologyThe newts were anesthetized for 30–60 min in ethyl-3-aminobenzoate methanesulfonate salt dissolved at 0.1% in 0.1 × modified Holtfreter’s solution (a bicarbonate-buffered saline used for urodele amphibians) [18].
To access the pericardial cavity, an incision (15–20 mm long) was made in the lateral region of the abdomen skin using a scalpel, and the testes and surrounding tissue were exposed. The testes were removed with/without adjacent connective tissue (anlage of regenerating testis) using surgical scissors. Following the procedure, the chest skin was closed and sutured. The newts were incubated in a humid chamber for another week to allow for healing of the wounded lateral region of abdomen skin, and the animals were then returned to the tap water aquarium.
Tissue samples were fixed in Bouin’s fixative and processed using standard procedures for normal paraffin sectioning. The resulting 8- to 10-μm-thick sections were stained with hematoxylin-eosin (Muto Pure Chemicals Co., Ltd., Tokyo, Japan).
Measurement of blood testosterone levelsTo collect a sufficient amount of blood for analysis, blood was drawn from the heart. The ventricles were cut with scissors, and then blood was collected into a plastic tube using a pipet. The collected blood was centrifuged immediately to separate the serum (12,000 rpm, 10 min).
Testosterone levels were measured using a Testosterone ELISA kit (Arbor Assays LLC, #K080-H1) in accordance with the manufacturer’s instructions. Testosterone concentrations were calculated based on a calibration curve prepared from a standard sample provided with the kit.
Measurement of triglyceridesTriglycerides in the liver were extracted according to a previously described method [19]. The liver (50 mg) was extracted with 0.5 mL of chloroform:methanol (2:1) using a beads crusher (μT-12; TAITEC, Saitama, Japan), and 0.125 mL of distilled water was added and mixed. The sample was centrifuged at 18,000 × g for 5 min and the lower organic phase was collected and evaporated. The residues were resolved into isopropanol. Triglyceride levels in the liver and serum were measured using the Triglyceride E-Test Wako (Wako Pure Chemical Industries, Osaka, Japan).
RNA-seq and gene ontology (GO) analysesAn anesthetized newt was placed on ice, and its abdomen was opened. Blood was flushed out by cardiac perfusion with 50 mL of Holtfreter’s solution, and the liver was then promptly removed. Pieces of liver tissue were then collected, and the fragments were ground in liquid nitrogen. Total RNA was extracted using TRI reagent and purified using a NucleoSpin RNA kit (Takara Bio, Shiga, Japan), following the respective manufacturer’s instructions.
Sequencing library construction and sequencing were performed by Annoroad Gene Technology Corporation (Beijing, China). A total of 2 μg of RNA per sample was used as input material for the RNA sample preparation. Sequencing libraries were generated using an NEBNext Ultra RNA Library Prep kit for Illumina (#E7530L, NEB, MA, USA) following the manufacturer’s recommendations. The libraries were sequenced on an Illumina platform, and 150-bp paired-end reads were generated. Short reads were mapped to the reference transcripts of Pleurodeles waltl [PLEWA_ID, 18] and then quantified using edgeR [20, 21]. Identifying homologs to evaluate enriched GO terms was conducted using Trinotate, an annotation program for Trinity-assembled sequences [22, 23].
Adult newts differed in body shape between males and females. Female newts had plumper bodies than males (Fig. 1A and B). This characteristic was consistent with a report by Hayashi et al. [16]. In accordance with observations in mammalian species, male newts in which the bilateral testes were removed became plumper in the abdomen compared with normal males (Fig. 1C, arrowheads). To examine in detail the changes in body shape after castration, we measured total body length (TL) (Fig. 1D), abdominal girth (Fig. 1E), body weight (BW) (Fig. 1F), ratio of BW to TL (BW/TL, Fig. 1G), and compared these parameters to normal male and female newts. These measurements confirmed clear male-female differences. Castrated male newts showed increases in abdominal girth, BW, and BW/TL (Fig. 1E–G). Even though their TL was not significantly different from normal males, their average weight increased by 30% at 7 months after castration. Newts have a pair of fat bodies on each side of the abdominal cavity. The weight of the fat bodies of the castrated newts was not significantly different from that of normal male newts (Fig. 1H). On the other hand, the ratio of liver weight/total body length was significantly greater in the castrated newts (Fig. 1I).
Body shapes of adult male, female, and castrated-male newts
Dorsal view of the body shape of adult newts. (A) A 14-month-old male, (B) a 14-month-old female, and (C) a 14-month-old male at 6 months after castration. Arrowheads in (C) indicate swelled abdomen. (D–I) Box-and-whisker diagrams of the different body shapes of newts. Thick horizontal lines indicate the median. Vertical lines indicate ranges of the maximum and minimum values. (D–G) n = 12 (normal males), n = 10 (normal females), n = 16 (males, 24 weeks post-castration). (H) n = 5 (normal males), n = 5 (normal females), n = 11 (males, 28 weeks post-castration). (I) n = 13 (normal males), n = 11 (normal females), n = 7 (males, 33–42 weeks post-castration). **: p < 0.05, ***: p < 0.005.
We also examined changes in body shape over time after castration. In newts, testicular regeneration occurs after testes removal in which the accompanying anterior terminal tissue of the testes is left intact [13, 14]. In order to examine the relationship between changes in body shape and testicular regeneration, we divided the male newts into two groups: a non-regeneration group (the terminal tissues were removed with the testes) and a regeneration group (the terminal tissues were left).
Testes are thought to produce testosterone in newts as well as many other vertebrates [24]. To examine the changes in testosterone levels due to castration, blood testosterone concentrations were measured. In untreated newts, blood levels in males were much higher than those in females, but in castrated newts, the blood concentrations decreased to the level of females (graph of testosterone measurements shown in Fig. 2A).
Blood testosterone levels and changes in body shape over time after castration
(A) Box-and-whisker plot of blood testosterone levels in newts. Horizontal lines indicate ranges of the maximum and minimum values. Thick vertical lines indicate the median. *: 0.05 ≥ p > 0.01. n = 8 (normal males), n = 6 (normal females), and n = 8 (males, 13–32 weeks post-castration). (B–E) Changes over time in body shape of regenerating (Re.) and non-regenerating (Non-re.) newts after castration. Horizontal axes represent months after castration. Vertical lines indicate standard errors. Light-magenta areas indicate the range of maximum and minimum values for normal adult females; light-blue areas indicate the range of maximum and minimum values for normal adult males. Number of samples is shown in (B). *: 0.05 ≥ p > 0.01, **: 0.01 ≥ p > 0.001, ***: p < 0.001. (F–H) Hematoxylin-eosin staining of sections of abdominal skin. Samples were prepared from three individuals for each group to confirm the same characteristics. Brackets labeled with “s” indicate skin, and those labeled “m” indicate muscle. g: secretory gland. Scale bars: 200 μm.
The results of the analysis of subsequent changes in body shape measured over time are shown in Fig. 2B–E. Both the regeneration and non-regeneration groups showed a similar increase in TL throughout the experimental period (Fig. 2B). The BW and BW/TL increased similarly in both groups between 3 and 5 months post-castration (Fig. 2C and D). However, in the regeneration group, BW did not increase after 5 months, and the BW/TL decreased (Fig. 2C and D). In the non-regeneration group, weight gain continued until 7 months after castration (Fig. 2C).
In Iberian ribbed newts, a sex difference in the ratio of tail length to TL occurs with maturity [16]. Males have longer tails. As indicated in Fig. 2E, the ratio of tail length continuously increased in castrated males under conditions in which testicular regeneration occurs. In contrast, in castrated males in which testicular regeneration was theoretically inhibited, the ratio of tail length decreased compared with regenerative males. Castrated newts were dissected to examine the possible correlation between weight gain and the presence/absence of the testes. The presence of the left and right testes was confirmed in the regeneration group. In contrast, no testes were present in the non-regeneration group, as expected. No ovarian formation (or regeneration from the terminal tissue) occurred in the castrated males.
Unexpectedly, abdominal skin thickness differed between sexes. The skin of female newts is thicker than that of males (brackets “s” in Fig. 2F and G). The dermal layer of the female skin has well-developed collagen fibers. The skin of castrated males showed more massive fiber layers compared with the skin of females. Again, such a sex difference seemed to be dependent on the presence/absence of the testes (Fig. 2H).
Liver transcriptome analysis to examine the effect of castration on gene expressionOur results showed an increase in liver weight after castration. Then, we conducted RNA-seq analysis of the liver of male newts 3 months post-castration in comparison with normal, adult male and female newts (9–10 months old). We confirmed that the liver of normal females expressed the egg yolk protein vitellogenin at higher levels. In addition, expression of vitamin D-binding protein (group-specific component/Gc protein), the synthesis of which is known to be promoted by high estrogen levels in mammals, was significantly higher in females (Table 1), indicating that our transcriptome analysis could discriminate female- and male-type gene expression.
Genes highly expressed in normal female as compared with normal male newts (FDR < 0.05)
| PLEWA_ID | gene_name* | logFC | PLEWA_ID | gene_name* | logFC |
|---|---|---|---|---|---|
| M0328821_PLEWA04 | Vitellogenin-A2 | 18.98 | M0233684_PLEWA04 | Lanosterol synthase | 2.65 |
| M4021216_PLEWA04 | Vitellogenin-2 | 11.51 | M0391065_PLEWA04 | Arylsulfatase I | 2.65 |
| M0300545_PLEWA04 | Sodium/hydrogen exchanger 3 | 11.03 | M0328423_PLEWA04 | Metallophosphoesterase 1 | 2.53 |
| M2174133_PLEWA04 | LINE-1 retrotransposable element ORF2 protein | 10.22 | M0212119_PLEWA04 | Single-stranded DNA-binding protein, mitochondrial | 2.53 |
| M0476461_PLEWA04 | Serine dehydratase-like | 9.55 | M0401016_PLEWA04 | Leucine-rich repeat-containing protein 42 | 2.49 |
| M0089883_PLEWA04 | Voltage-gated potassium channel subunit beta-1 | 8.16 | M0170340_PLEWA04 | Microsomal triglyceride transfer protein large subunit | 2.46 |
| M0246694_PLEWA04 | Angiogenin | 6.48 | M0072589_PLEWA04 | Interleukin-8 | 2.46 |
| M0111829_PLEWA04 | Vitamin D-binding protein | 6.47 | M0333580_PLEWA04 | Monocarboxylate transporter 7 | 2.42 |
| M0047185_PLEWA04 | Solute carrier family 41 member | 6.26 | M0088895_PLEWA04 | Acyl-coenzyme A thioesterase THEM4 | 2.37 |
| M0429539_PLEWA04 | Cathepsin E | 6.02 | M0128372_PLEWA04 | Lipoprotein lipase | 2.34 |
| M0408304_PLEWA04 | Gamma-aminobutyric acid type B receptor subunit 2 | 5.88 | M0109839_PLEWA04 | Neuronal regeneration-related protein | 2.30 |
| M0389177_PLEWA04 | Dynein axonemal heavy chain | 5.02 | M0434277_PLEWA04 | Inositol-3-phosphate synthase 1 | 2.29 |
| M0204997_PLEWA04 | Transposon Tf2 polyprotein | 4.26 | M0144068_PLEWA04 | Retinol dehydrogenase | 2.27 |
| M0413885_PLEWA04 | Squalene monooxygenase | 4.18 | M0213657_PLEWA04 | Inhibitor of growth protein | 2.26 |
| M0175375_PLEWA04 | Transposon Ty3-G Gag-Pol polyprotein | 3.85 | M0104265_PLEWA04 | Neural Wiskott-Aldrich syndrome protein | 2.21 |
| M0366708_PLEWA04 | Actin | 3.73 | M3359278_PLEWA04 | Cytochrome c | 2.20 |
| M0478110_PLEWA04 | Suppressor of cytokine signaling 2 | 3.73 | M3981790_PLEWA04 | Non-histone chromosomal protein HMG-14 | 2.17 |
| M0053354_PLEWA04 | Amiloride-sensitive sodium channel subunit gamma | 3.71 | M0395533_PLEWA04 | Zinc finger MYM-type protein 3 | 2.17 |
| M0299031_PLEWA04 | Cytochrome P450 2 | 3.31 | M0471735_PLEWA04 | Endogenous retrovirus group PABLB member 1 Env polyprotein | 2.13 |
| M0247052_PLEWA04 | Lysine-specific demethylase 4B | 3.21 | M0490097_PLEWA04 | Matrix metalloproteinase-9 | 2.13 |
| M0125558_PLEWA04 | Pseudokinase FAM20A | 3.12 | M3404576_PLEWA04 | Protein MANBAL | 2.10 |
| M0242222_PLEWA04 | LINE-1 retrotransposable element ORF protein | 3.02 | M0059992_PLEWA04 | Coiled-coil domain-containing protein 86 | 2.09 |
| M0477574_PLEWA04 | Single-stranded DNA-binding protein 3 | 3.01 | M0114595_PLEWA04 | Lysophosphatidic acid receptor 3 | 2.09 |
| M0030413_PLEWA04 | Unknown protein | 3.00 | M0274460_PLEWA04 | Ethanolamine-phosphate phospho-lyase | 2.08 |
| M0080321_PLEWA04 | Krueppel-like factor 9 | 2.99 | M3296702_PLEWA04 | Vimentin | 2.08 |
| M0219336_PLEWA04 | Extracellular serine/threonine protein kinase FAM20C | 2.92 | M0489275_PLEWA04 | Proton-activated chloride channel | 2.07 |
| M0184177_PLEWA04 | ADP/ATP translocase 2 | 2.80 | M0217882_PLEWA04 | Methylsterol monooxygenase 1 | 2.06 |
| M0292554_PLEWA04 | Procathepsin L | 2.80 | M0015423_PLEWA04 | Coiled-coil-helix-coiled-coil-helix domain-containing protein 10, mitcondrial | 2.03 |
| M0105395_PLEWA04 | Sulfotransferase 1C | 2.76 | |||
| M0085954_PLEWA04 | Transmembrane protein 255A | 2.73 | M2115772_PLEWA04 | Stress-associated endoplasmic reticulum protein 1 | 2.00 |
* When the same gene was counted multiple times, the clone exhibiting the greatest change was used to represent the result.
For example, we found that expression of phosphoenolpyruvate carboxykinase 1 (PCK1) was approximately 3-fold higher in castrated males compared with normal males (Table 2). This result is consistent with a report indicating that PCK1 expression is greatly increased in castrated chickens [25]. Expression of thyroid hormone-inducible hepatic protein (Thsrp) was also up-regulated in castrated newts (Table 2). Other studies have shown that Thsrp expression is also up-regulated in chickens and cattle after castration [25, 26]. None of these groups of up- or down-regulated genes are known to be directly regulated by nuclear androgen receptor (Tables 2 and 3). We were unable to identify any genes that could characterize the liver of normal or castrated male newts. In order to characterize the expression profiles of the transcripts, we therefore conducted GO analysis.
Genes highly expressed in castrated males as compared with normal male newts (FDR < 0.05)
| PLEWA_ID | gene_name* | logFC |
|---|---|---|
| M0281511_PLEWA04 | Cytochrome P450 7A1 | 4.75 |
| M0414326_PLEWA04 | Mitochondrial uncoupling protein 2 | 4.13 |
| M0252984_PLEWA04 | Cilia- and flagella-associated protein 100 | 3.56 |
| M0505609_PLEWA04 | Phosphoenolpyruvate carboxykinase, cytosolic | 3.11 |
| M0034855_PLEWA04 | Perilipin-3 | 2.89 |
| M0365932_PLEWA04 | Diacylglycerol O-acyltransferase 2 | 2.75 |
| M0221059_PLEWA04 | Retinoid-binding protein 7 | 2.56 |
| M0312077_PLEWA04 | Cytochrome P450 2 | 2.52 |
| M0112723_PLEWA04 | Thyroid hormone-inducible hepatic protein | 2.42 |
| M0193976_PLEWA04 | ATP-citrate synthase | 2.31 |
| M0275530_PLEWA04 | Patatin-like phospholipase domain-containing protein 2 | 2.26 |
| M0238545_PLEWA04 | Fatty acid synthase | 2.24 |
| M0077249_PLEWA04 | Kynurenine/alpha-aminoadipate aminotransferase, mitochondrial | 2.13 |
| M0314022_PLEWA04 | Keratin, type I cytoskeletal 18 | 2.05 |
| M0031460_PLEWA04 | Secreted phosphoprotein 24 | 2.00 |
* When the same gene was counted multiple times, the clone exhibiting the greatest change was used to represent the result.
Genes highly expressed in normal males as compared with castrated male newts (FDR < 0.05)
| PLEWA_ID | gene_name* | logFC |
|---|---|---|
| M2312797_PLEWA04 | Max-interacting protein 1 | 4.70 |
| M0092822_PLEWA04 | Monocarboxylate transporter 13 | 3.71 |
| M0474527_PLEWA04 | Unknown protein | 3.47 |
| M0135726_PLEWA04 | Tubulin-folding cofactor B | 3.42 |
| M0135726_PLEWA04 | Zinc finger protein basonuclin-2 | 3.42 |
| M0193451_PLEWA04 | Death domain-containing protein 1 | 2.97 |
| M0212018_PLEWA04 | Endonuclease domain-containing 1 protein | 2.75 |
| M0168242_PLEWA04 | Cytochrome P450 3A | 2.58 |
* When the same gene was counted multiple times, the clone exhibiting the greatest change was used to represent the result.
Fig. 3A shows the top 10 most reliable (lower p-value) GO terms for which expression was up-regulated in castrated males compared with normal males. Of these 10 GO terms for genes up-regulated after castration, 7 terms were related to lipid storage or lipid metabolism (Fig. 3A), which strongly suggests that lipid metabolism is affected by testis removal. Fig. 3B shows the GO terms that characterize the liver of female newts. One could imagine that these terms are closely related to gene sets that are expressed more abundantly in the presence of ovaries or absence of testes. However, the GO terms shown in Fig. 3B did not agree well with those shown in Fig. 3A, indicating that castration did not simply alter the gene expression pattern of the male liver to that of the female liver.
GO terms characterizing gene expression in the liver after castration, and triglyceride levels in serum and liver. (A) GO terms up-regulated in castrated males compared with normal males in RNA-seq. (B) GO terms characterizing the liver of female newts in RNA-seq. Values of triglyceride in the serum (C), liver tissue (D), and whole liver (E). Vertical lines indicate the standard errors. n = 4 (normal males), and n = 3 (males, 33–36 weeks post-castration).
To examine possible physiological changes that occurred in the castrated newts, we measured triglyceride levels in the serum and liver tissues. The triglyceride levels in serum were not altered by castration (Fig. 3C), but the triglyceride levels in liver tissue tended to increase with castration (p = 0.095, Fig. 3D). Since the liver size was larger after castration, the triglyceride content of the whole liver increased further (p = 0.0042, Fig. 3E).
Although weight gain after castration is widely known in mammals and birds, the details of its mechanisms remain unclear. One possible reason for this is the lack of suitable animal experimental models. Our study revealed that changes in body shape, accompanied by an increase in BW, also occur in an amphibian, Iberian ribbed newts, after castration. This suggests that weight gain occurring in castrated animals may be a phenomenon common to numerous vertebrates. We also showed that the changes in gene expression that occur in the liver of castrated newts exhibit similarities to the changes in mammals, as discussed below. Our results suggest that newts are a suitable model animal for comparative studies of the effects of castration.
Weight gain in the castrated males seemed to reflect an increase in liver weight rather than fat body mass (Fig. 1H and I). Besides, the abdominal skin was also thicker in the castrated newts. Thicker skin would bring tension to the body, thereby altering body shape and contributing to weight gain. In the castrated males, tail length relative to TL was shorter than in normal males (Fig. 2E). In newts, the tail is thin and relatively lighter than the body. A higher proportion of the body being composed of the tail would result in an increase in the BW/TL ratio in castrated males. The BW/TL value may have increased due to the combined effects of the above factors, but further analysis is needed to determine if there are any other organs or tissues that exhibited significantly increased weight after castration. Although the collagen layer appears to be developed in the skin of castrated males, the mechanism of skin thickening remains unclear. Whether the metabolic changes in the liver directly affected skin thickening and collagen fiber orientation needs to be analyzed in the future. Interestingly, it has been reported that collagen accumulation and collagen fiber orientation in the skin of rats and mice differs or is altered depending on sex, castration, and obesity [27, 28]. Skin thickening should also be studied, taking into consideration that it is a common phenomenon in the mammals and newts. In addition to metabolic changes in the liver, hormonal changes or weight gain due to castration should be considered as causes of increasing of the skin thickness. Although it is also important to analyze changes in the expression pattern of collagen genes, there are multiple types of collagen genes that have not been well annotated in the newts. Therefore, as more detailed gene models are developed and annotated in the future, this will enable the discrimination of the genes related to testis-dependent collagen synthesis and accumulation in skin. Also, no well-developed subcutaneous adipose tissue was observed, but the amount of fat contained in the skin needs to be investigated.
Although studies in boars and chickens have reported that castration increases subcutaneous and abdominal fat mass, the mechanism underlying the increase in fat mass is unclear [25, 29]. Our GO analysis showed up-regulation of genes related to lipid synthesis and storage after testis removal, suggesting that lipid metabolism was altered in the castrated male animals. Our results showed that the triglyceride levels in serum and in liver tissue did not significantly differ before and after castration (Fig. 3C and D). The triglyceride levels in serum did not significantly differ, but the triglyceride in the liver tissue tended to increase with castration (Fig. 3C and D). Because the liver itself was larger in the castrated males, the triglyceride content of the whole liver was greatly elevated (Figs. 1I and 3E). Our data raise the possibility that castration alters lipid metabolism by activating or derepressing genes related to lipid metabolism and consequently increases the liver size. It has been reported that fat metabolism can be altered in castrated male cats, resulting in increased blood levels of insulin-like growth factor I, prolactin, and leptin [30]. These factors could be involved in newts as well. Another alternative possibility is that changes in body shape and weight gain result from reduced caloric expenditure due to decreased locomotion, as discussed below. Nonetheless, we confirmed that removal of the testes reduces blood testosterone levels in newts. Further studies are needed to determine whether testosterone directly regulates body shape and weight in these animals. Although our results demonstrated changes in the expression of genes related to hepatic lipid metabolism after castration, the role of androgens in hepatic fat metabolism also remains unclear. As it has been reported that hepatocyte-specific androgen receptor–knockout mice show dysregulation of lipid metabolism and develop severe fatty liver, androgens may act on the liver via the androgen receptor to regulate metabolism [31]. However, our analysis did not show significant changes in the expression levels of genes for which homologs are known to be directly regulated by the androgen receptor in mammalian models (Table 3). The physiological mechanism of androgen function in newts thus needs further investigation.
Preventing obesity associated with post-castration weight gain is especially important for the quality of life of companion animals. Reasons for castration do not include only the avoidance of unexpected reproduction but also the control of behaviors that are unique to male individuals, and simply supplementing with testosterone would lead to a return of unwanted behaviors. However, even if the direct cause of weight gain is a decrease in testosterone, testosterone supplementation in castrated animals is inappropriate as a treatment. Furthermore, our data show that female- and male-type gene expression involves a complex network (Fig. 3). Such effects have not been expected, supporting the hypothesis that newts, which exhibit a long lifespan, are useful model organisms for not only studies aimed at the management of companion animals after castration but also studies seeking to better understand human aging.
It is thought that down-regulation of physical activity brought on by behavioral changes induced by castration results in weight gain due to excess calorie intake from feed. However, it is unlikely that behavioral activity as well as calorie intake differed between castrated and control newts at 3 month after the operation. The newts spend most of their time lying at the bottom of the tank, and they are not very active by nature. No changes in locomotion, behavior, or feeding due to castration were observed. Nonetheless, our results demonstrated that castration alters lipid metabolism in the liver at the gene expression level. Hence, in order to prevent obesity after castration, rather than suppressing caloric intake, treatment strategies that return liver lipid metabolism to the normal male state should also be considered. In parallel, behavioral data such as analysis of camera images is necessary to accurately assess locomotion.
Some genes showed expression changes in newts similar to those in mammals and birds. For example, PCK1 is an enzyme that converts oxaloacetate to phosphoenolpyruvate in the TCA circuit in liver glycogenesis, and PCK1 is known to regulate metabolic reprogramming [32]. Increased expression of PCK1 has also been reported in castrated chickens, steer, and prostate cancer patients treated with androgen depletion [33, 34]. The expression of Thsrp is also known to increase after castration in chickens and cattle [25, 26]. Thsrp is associated with fat accumulation and glucose tolerance, and mice lacking Thsrp show decreased BW [35]. These results suggest that the mechanism by which androgens regulate PCK and Thsrp expression is conserved among vertebrates. The direct mechanism by which PCK-mediated changes in glucose metabolism or Thsrp-regulated lipid metabolism cause weight gain in these animals is unknown. Newts are smaller than mice and other domestic animals, which makes them more cost-effective for use in research. In addition, functional analyses of genes via genome editing are easier in newts. Further investigations of the function of genes encoding factors such as PCK and Thsrp in newts would enhance our understanding of obesity in other avian and mammalian model organisms as well. Comparative studies between such models could reveal the mechanism underlying castration-induced weight gain and provide insights as to how to prevent it. On the other hand, the newts show the interesting phenomenon of testicular regeneration after castration. In this study, the testes were treated so that they could not regenerate in order to clarify for certain the effects of testicular loss. Future work should analyze gene expression profiles at different regeneration stages to learn more about the relationship between testicular regeneration and gene expression.
In order to make an argument about comparison with the mammalian models with respect to aging, it is necessary to consider how to correlate the age of newts with that of humans. Although no studies have been conducted to correlate the lifetime of newts with that of other animals, the timing of sexual maturation is one possible benchmark. P. waltl newts reach sexual maturity 6 months after fertilization under our rearing conditions [15, 16]. Sperm can be collected for several years after sexual maturation. The newts in this study were castrated 7–8 months after fertilization and kept until a maximum of 14 months post-fertilization. Hence, the newts we used may correspond to adolescence in humans in terms of age. On the other hand, the newts have a long lifespan (20–30 years), which makes them an interesting animal model for aging studies [36], and therefore, using time-scaling with reference to this lifespan, we may consider newts of age 7–8 months to be in childhood. In any case, a detailed study that maps the lifespan of the newts to that of mammals (humans and mice) is needed.
The treatment of human prostate cancer sometimes involves surgically removing the testes. Removal of the testes in humans can cause changes in body composition, such as increased fat accumulation, and also changes in lipid metabolism or insulin resistance [37, 38]. Our newt castration models could be used to study countermeasures against side effects associated with orchiectomy in humans.
We would like to thank Kyorin Corporation (Hyogo, Japan) for kind support in providing feed for the newts. We also thank Ms. N. Morihara and Ms. F. Irisuna of the Natural Science Center for Basic Research and Development, Hiroshima University (NBARD-00201) for their assistance in qualifying the RNA samples. This research was supported by the MEXT/JSPS KAKENHI (grant number, 19K07268). The Pleurodeles waltl specimens were provided by Hiroshima University Amphibian Research Center through the National BioResource Project of MEXT.
None of the authors have any potential conflicts of interest associated with this research.
