Tributyltin perturbs femoral cortical architecture and polar moment of inertia in rat

Background Tributyltin, a well-known endocrine disruptor, is widely used in agriculture and industry. Previous studies have shown that tributyltin could cause deleterious effects on bone health by impairing the adipo-osteogenic balance in bone marrow. Methods To investigate further the effects of tributyltin on bone, weaned male SD rats were treated with tributyltin (0.5, 5 or 50 μg·kg− 1) or corn oil by gavage once every 3 days for 60 days in this study. Then, we analyzed the effects of tributyltin on geometry, the polar moment of inertia, mineral content, relative abundances of mRNA from representative genes related to adipogenesis and osteogenesis, serum calcium ion and inorganic phosphate levels. Results Micro-computed tomography analysis revealed that treatment with 50 μg·kg− 1 tributyltin caused an obvious decrease in femoral cortical cross sectional area, marrow area, periosteal circumference and derived polar moment of inertia in rats. However, other test results showed that exposure to tributyltin resulted in no significant changes in the expression of genes detected, femoral cancellous architecture, ash content, as well as serum calcium ion and inorganic phosphate levels. Conclusions Exposure to a low dose of tributyltin from the prepubertal to adult stage produced adverse effects on skeletal architecture and strength. Supplementary Information The online version contains supplementary material available at 10.1186/s12891-021-04298-2.

contaminated sites [10]. Moreover, high levels of BTs have been detected in human liver tissue [11] and blood samples [12]. Previous studies showed that exposure to high dose TBT (over 10 mg·kg − 1 ) during gestation period caused delayed ossification of the fetal skeleton [13,14]. Our previous study revealed that exposure to a low dose of TBT (50 μg·kg − 1 ) reduced the femoral bone mineral density (BMD) of rats with a downtrend of biomechanical strength [15].
Although measurement of BMD is an indispensable tool to identify individuals at high risk of injury, bone densitometry affords only a two dimensional areal view of the three dimensional mineralized mass of the skeleton [4]. Micro-computed tomography (μCT) is the most common technique for the nondestructive assessment and analysis of the three-dimensional bone architecture. There were two papers using μCT to assess the effect of TBT on bone [16,17]. However, they both used adult female rodents. Although osteopenia is more serious in women than men, there is a 20% osteoporotic fracture risk in white men [2]. Moreover, skeletal growth is rapid during adolescence, and exposure to EDs before and during puberty therefore results in greater deficits at a site than exposure after puberty [18]. In this study, we would furtherly assess the effect of TBT on bone based on femoral cancellous and cortical architecture using μCT, derived polar moment of inertia (Jo), ash content (mineral content), serum calcium ion (Ca 2+ ) and inorganic phosphate (Pi) levels, and expression of genes related to adipogenesis and osteogenesis.

Methods
Reagents TBT (purity ≥96%) was purchased from Sigma-Aldrich Chemicals Co. (St Louis, MO). Kits for determination of total protein, serum Ca 2+ and Pi were from NanJing JianCheng Bioengineering Institute (Nanjing, China). The QIAamp RNA Blood Mini Kit and QuantiTect SYBR® Green RT-PCR Kits were from QIAGEN (regional headquarter, Shanghai for Asia). All other chemicals were of analytical grade and were obtained from commercial sources.

Animals and treatment
21-day-old male SD rats were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. [License No. SCXK (Jing) 20,160,011]. After 3 days of acclimatization, rats were randomly assigned into four groups (10 rats per group) based on body weight to achieve similar average weights in different groups. Rats were treated with corn oil or TBT (0.5, 5 or 50 μg·kg − 1 ) by gavage once every 3 days from ages of 24 d to 84 d, while the established no observable adverse effect level (NOAEL) of TBT was 25 μg·kg − 1 ·day − 1 based on immunological effects [19]. All rats were killed 1 day after the final gavage (85 d). All treatments of rats in this study were performed humanely in accordance with the National Institutes of Health's Guide for the Care and Use of Laboratory Animals and followed the principles described in the "Use of Animals in Toxicology," which were approved by the Ethics Committees of the School of Public Health, Shandong University, in March 2019. The identification code was 20,190,204.

Preparation of serum and femurs
Urethane was dissolved in saline solution at a final concentration of 20% (w/v). The rats were anesthetized with urethane (~1.2 g·kg − 1 ) by introperitoneal injection. After reaching the surgical level of anesthesia, blood was obtained from the aorta ventralis. The serum was separated and stored at − 80°C for further analysis. Both femurs were isolated modestly, and then the muscular tissue was removed. Five left femora were used to assess gene expression, and the other left femora were stored at − 20°C for ash content analysis. All right femurs were weighed and measured as described in a previous study [15], and fixed in 4% (vol/vol) paraformaldehyde for further research.

RNA isolation and quantitative real-time reverse transcription-polymerase chain reaction (QPCR)
Bone marrow (BM) was flushed from the left femur (n = 5 per group), and strained through a 70-μm cell strainer. After centrifugation, total RNA was extracted using the QIAamp RNA Blood Mini Kit (Qiagen, Valencia, CA) according to the manufacturer's recommendation. The concentration and quality of RNA were determined with a NanoDrop 2000c spectrophotometer (Gene Company Limited, USA). After quality inspection, RNA was stored at − 80°C for QPCR.
QPCR was performed using QuantiTect SYBR® Green RT-PCR Kits (QIAGEN) with reactions scaled to 25 μl, and 25 ng mRNA was used in each reaction. PCRs (in duplicate) were performed using a 7500 Fast Real-Time PCR System (Applied Biosystems, Carlsbad, CA) with a program consisting of 95°C for 2 min, 40 cycles of 95°C for 10 s and then 60°C for 20 s and 72°C for 30 s, followed by melting curve analysis. The primer sequences for each gene are listed in Table S1. Genes of interest were normalized to the GAPDH gene in the same sample following the 2 -ΔΔ CT method [20]. Primers were designed using Primer-BLAST (NCBI) or Primer5 software (Rozen and Skaletsky 2000), and were verified by gradient amplification and melting curve analysis.

Micro-computed tomography (μCT)
The fixed femurs were scanned with a Scanco μCT100 scanner (Scanco Medical AG, Bassersdorf, Switzerland), at a 20 μm isotropic resolution using an integration time of 300 ms, energy of 70 kVp, and intensity of 200 μA. Gaussian filtering (sigma = 0.8, support = 1) was used to reduce background noise. For analyses of trabecular bone within the distal femur, the specimen was scanned in 700 slices and a region of interest (ROI) of trabecular bone was selected between 230 slices and 300 slices to the growth plate. Trabecular bone parameters of the femoral metaphysis, including bone volume fraction (BV/TV), connectivity density (Conn.D), structural model index (SMI), trabecular number (Tb.N), trabecular thickness (Tb.Th), and trabecular separation (Tb.Sp), were determined using Scanco's 3D analysis tools (direct model). For cortical bone analysis, the femur specimens were scanned similarly. One millimeter thick sections immediately distal to the mid-diaphysis were used as ROI. In addition, total cross sectional area (TCS.Ar), cortical cross sectional area (Ct.Ar), cortical thickness (Ct.th), periosteal circumference (Ps.Cf), endocortical circumference (Ec.Cf) and marrow area (Ma.Ar) were obtained from the analysis.

Calculation of Jo
Jo was calculated as the medio-lateral (IML) + anteroposterior (IAP) axes, following the method of Jepsen et al. [21].

Ash content
Five left femora per group were used to detect the ash content. The dried, and ash weights were determined as described previously [21]. Ash content was determined as the ash weight normalized for hydrated weight.

Serum parameters
Serum Ca 2+ and Pi levels were detected by colorimetry, following the manufacturer's instructions.

Statistical analysis
The statistical analyses were performed by SPSS software, version 21.0 for Windows (SPSS Inc., Chicago, IL, USA). Two-tailed Student's t-test was used to analyze the bone structural phenotype and Jo, which were compared between the two groups. Other indicators were analyzed by ANOVA. The significance level was set at 0.05. Once significance was established (P < 0.05), Dunnett's or Dunnett's T3 test was performed to make multiple comparisons among the groups based on the homogeneity of variance.

Effect of TBT on body weight and femur
The mean body-weight of rats is shown in Fig.1. Rat body weights increased with prolonged feeding time, however, exposure to TBT caused no significant effects on the body weights at all time points observed compared to the control. Consistent with the body weight, TBT-treatment caused no significant differences in femoral length and weight at the end of the experiment ( Table 1).

Effect of TBT on the mRNA expression of genes involved in adipogenesis and osteogenesis in BM
PPARγ is a key transcriptional regulator of fat formation [22], while Fabp4 and Angptl4 are their target genes [23,24]. Runx2 and ALP are early osteogenic markers [25,26] and osteocalcin (OC) is a late osteoblast differentiation marker [27].
Analysis with QPCR analysis showed no notable TBTrelated changes in the expression of PPARγ, Fabp4, ALP and OC at 85 d (Fig.2). The relative expression of Fabp4 showed 1.69-, 1.68-and 1.48-fold increases in respective TBT groups, but the change was not significant (P = 0.153). Although the expression of Angptl4 showed a significant decrease in the 0.5 μg·kg − 1 and 5 μg·kg − 1 TBT groups compared with the control (P < 0.05), the degree of the decline was minor (− 29.79% and − 24.25%, respectively).  along with a decreasing trend in Conn. D (− 17.56%, P = 0.277), Tb. N (− 14.92%, P = 0.193), and a subtle increase in Tb. Sp (+ 10.53%, P = 0.390) and SMI (+ 15.17%, P = 0.124), in TBT-treated rats compared to controls, although the differences were not significant (Fig.3). In contrast, the analysis of the mid-diaphysis showed that TBT-treated rats had a smaller Ct. Ar, Th. Ar and Ps. Cf than those of their control counterparts ( Fig.4; P < 0.05), but had no effects on other indexes.

Effect of TBT on ash content
There was no significant difference in ash content between the TBT groups and the control group (Fig.6).

Effect of TBT on serum Ca 2+ and pi levels
As shown in Fig.7, there were no significant changes in serum levels of Ca 2+ and Pi between the control and TBT treated rats.

Discussion
Consistent with our previous result on the BMD of the femur, μCT analysis revealed that the effect of TBT was more significant on the diaphysis than on the cancellous diaphysis in the femur of rats in this study. However, the study of Watt et al. observed a reduced femoral cortical cross-sectional area and thinner cortex with increased cancellous Tb. Th, Tb. N, and BV/TV in TBT-treated mice [16]. The discrepancies between the two studies might be explained by (1) the different dose. EDs could exhibit complex dose-response curves, and they might produce different effects at extremely low concentrations [28][29][30]. The highest concentration used in our study was 200-fold lower than the dose used in the study by Watt [34], in which endochondral and intramembranous ossification are produced simultaneously, while the process of endochondral ossification disappears after puberty [35]. Hence, the effect of TBT in this study includes the effect on skeletal development, while such an effect in the study of Watt et al. [16] is only on bone remodeling. As a long bone, the femoral diaphysis is a hollow cylinder whose size and shape can be influenced by the relative amounts of bone deposition and resorption on the periosteal and endosteal surfaces [36][37][38][39]. Therefore, a significantly smaller periosteal expansion and marrow infilling might lead to a smaller Ct. Ar, Th. Ar and Ps. Cf, which reflected the relative contribution of osteoblasts and osteoclasts working on them [33]. Indeed, in vitro studies showed that TBT not only inhibited osteoblasts [40,41] but also suppressed osteoclastogenesis and resorptive activity of osteoclasts [16,42]. Mesenchymal stem cells (MSCs) are multipotent cells contributing to osteoblast and adipocyte progenies in adult bone marrow [43]. As common progenitor cells of adipocytes and osteoblasts, the commitment towards these lineages is classically considered to be inversely related [44]. As a dual RXRα/β and PPARγ agonist [22], TBT could reprogram BM-MSCs towards adipogenesis at the cost of osteogenesis in vitro [40,41] and in vivo [45]. In contrast, TBT did not obviously change the expression of PPARγ and its target genes in this study. Consistent with the response of genes involved in adipogenesis, genes associated osteogenesis (ALP, OC and Runx2) were not significantly changed in this study. Since the relative expressive fold of Angptl4 expression was marginal only in the 0.5 and 5 μg·kg − 1 TBT groups, and had no effects in the 50 μg·kg − 1 TBT group. These data revealed that the impaired effects of TBT on the mesenchymal differentiation disappeared at 85d, suggesting that TBT treatment might have no persistent effects on the adipo-osteogenic balance in the BM of rats at the transcriptional level. This no residual effects might be partly due to the resistance produced by rats to TBT. One recent paper showed that feeding young mice a high-fat diet (HFD) significantly increased the CD45 − CD31 − Sca1 + CD24 + (a tri-potent population with stem cell-like characteristics) and adipogenic progenitor cells (APCs) frequencies for 1 day but not for 10 days [46]. Our recent study showed that TBT treatment resulted in a dose-dependent increase in lipid accumulation and adipocyte number in the BM of the femur [15]. In addition, our previous study demonstrated that TBT- induced functional perturbations of the gut microbiome were similar to the HFD styles [47]. Therefore, whether the effect of TBT on APC proliferation is similar to that of HFD needs to be further investigated.
The different effects of TBT between this study and in vitro studies [40,41] were at least partially were explained by the compensation effect of rats, because experimental animals can exert feedback mechanisms to counteract harmful effects. However, one paper [45] observed up-regulated MSCs expression of genes involved in lipogenesis in all three subsequent generations after ancestral perinatal TBT exposure. Preimplantation development is a major developmental period of epigenetic reprogramming of the genome in mammalians [48]. Indeed, exposure to TBT throughout pregnancy and lactation induced genome-wide alterations in methylation and altered the expression of metabolism-relevant genes in unexposed male descendants [49]. However, genomic methylation patterns are generally stable and heritable in somatic differentiated cells [48]. Therefore, the different exposure times might be a plausible explanation for the discrepancy between that of Chamorro-García et al. [45] and ours.
Jo is a quantity used to predict an object's ability to resist torsion. Because structures possessing the same Ct. Ar but different moments of inertia (e.g., a solid cylinder and a tube) will exhibit different mechanical characteristics in bending and torsion [50], so measuring Jo is necessary. Moreover, the level of Jo is consistent with bone strength [51]. Ct. Ar, Jo, and ash content together might explain 66-88% of the genetic variability in adult whole bone mechanical properties [21,33]. In this study, treatment with 50 μg·kg − 1 TBT caused a significant decrease in Ct. Ar and Jo, but not on ash content, which might be one possible explanation for the reduced but no significantly different biomechanical strength change induced by TBT in our previous study [15].
The skeleton is made of collagen fibers that form a scaffold where mineralization is initiated by the accumulation of Ca 2+ and Pi, mainly in the form of crystalline hydroxyapatite (HA) [52]. Calcium and phosphorus play critical roles in diverse biological processes including bone mineralization [53,54]. TBT treatment resulted in no significant changes in the serum Ca 2+ and Pi levels in this study, which was in accordance with the data of ash content.

Conclusions
In conclusion, treatment with 50 μg·kg − 1 TBT caused a significant decrease in femoral Ct. Ar, Th. Ar and Jo of rat. All the data suggest that exposure to a low dose of TBT from the prepubertal to adult stage produces adverse effects on skeletal architecture and strength. A statement for the methods All methods in the study were carried out in accordance with the NIH Guide for Care and Use of Laboratory Animals and followed the principles in the "Use of Animals in Toxicology" (ARENA/OLAW 2002).
A statement for guidelines Authors declare that the study was carried out in compliance with the ARRIVE guidelines.

Availability of data and materials
All data generated and analyzed during this study are included in its supplementary information files.