Pathophysiologically relevant bisphenol S exposure accelerates aging by disrupting brown adipose tissue–regulated energy metabolism

Significance

This study investigates the effects of bisphenol S (BPS), a common substitute for bisphenol A (BPA), at pathophysiologically relevant concentrations on the aging process, with a particular focus on its impact on brown adipose tissue (BAT) and its role in energy metabolism and lifespan regulation. Using Caenorhabditis elegans and mice as model organisms, we demonstrate that BPS exposure accelerates aging and disrupts energy homeostasis. Our findings highlight the risks associated with pathophysiologically relevant concentrations of BPS and identify BAT as a key mediator of these effects, offering mechanistic insights into how environmental toxins can influence energy metabolism and aging.

Abstract

Bisphenol A (BPA) substitutes are widely used as food contact materials and consumer products, while the effects of pathophysiologically relevant concentrations of BPA substitutes on aging remain unclear. In this study, we used Caenorhabditis elegans (C. elegans) to investigate the effects of five BPA substitutes [bisphenol S (BPS), bisphenol B, bisphenol F (BPF), tetramethyl BPF, and 4,4′-(Perfluoropropane-2,2-diyl)diphenol] at pathophysiologically relevant exposure levels during aging and examined the underlying mechanisms using a mouse model. Our results indicated that, among the five BPA substitutes, exposure to pathophysiologically relevant concentrations of BPS (300, 450, and 600 nM) accelerated aging in C. elegans. In mice, exposure to a pathophysiologically relevant concentration of BPS (125 μg/kg/day, from 4 to 20 mo of age) similarly reduces the life and health span and accelerates aging phenotypes in multiple tissues. Further investigations demonstrated that long-term BPS exposure resulted in a significantly higher accumulation of BPS in brown adipose tissue (BAT) than in other organs. RNA sequencing analysis of BAT revealed that BPS accelerates BAT aging through multiple pathways. Importantly, transplantation of BAT from BPS-exposed mice into BPS-naive mice accelerated aging in recipients. Conversely, transplantation of BAT from unexposed mice into BPS-exposed mice significantly improved their metabolic status and delayed aging. These findings elucidate the impact of pathophysiologically relevant concentrations of BPS on the aging process and suggest that these effects are likely mediated through the disruption of BAT function.

Bisphenol compounds (BPs) are primarily used as foundational materials for the production of polycarbonate plastics and epoxy resins. They are widely present in everyday items such as plastic containers, food packaging materials, and thermal paper receipts (1, 2). Bisphenol A (BPA) is one of the most commonly used BPs. Due to concerns about its reproductive and developmental toxicity (3), many countries have imposed strict regulations on BPA use. However, various BPA alternatives such as bisphenol B (BPB), bisphenol F (BPF), bisphenol S (BPS), tetramethyl BPF (TMBPF), and 4,4′-(Perfluoropropane-2,2-diyl)diphenol (BPAF) have emerged and are widely used in production (4). Epidemiological surveys conducted in multiple countries have confirmed the widespread exposure to these BPA alternatives. Taking BPS as an example, the detection rate of urine BPS was reported to be ~90% in adolescents and adults in the United States (5, 6) and 84% in adults in China (7, 8). As the use of BPA substitutes has become more prevalent, the safety of these substitutes remains a concern. Growing evidence indicates that some substitutes may also have endocrine-disrupting effects and in some cases, their biological impacts might even exceed those of BPA (4).

Aging is not only a natural process characterized by a gradual decline in physiological functions but is also closely related to environmental stimuli. Notably, a recent study investigated the impact of high-level exposure to BPA substitutes (~1,000-fold higher than typical human daily exposure) on life span (9). The researchers found that these substitutes adversely impacted the health span parameters of Caenorhabditis elegans (C. elegans), suggesting the potential link between high-level BPA substitutes exposure and reduced health span. However, the impact of exposure to pathophysiologically relevant concentrations of BPA substitutes on the aging process remains unclear. In this study, we explored the impact of exposure to five BPA substitutes at pathophysiologically relevant concentrations on life span using C. elegans, and further analyzed the possible mechanisms using a mouse model. This research aims to provide evidence for assessing the potential impact of BPA substitutes on aging and offers a reference for developing more scientifically grounded public health policies.

Results

BPS Exposure at Pathophysiologically Relevant Concentrations Shortens the Life Span Across Species.

Building on a previous epidemiological survey (10), we first examined the effects of pathophysiologically relevant concentrations (150, 300, 450, and 600 nM) of BPA and its substitutes (BPA, BPB, BPF, BPS, TMBPF, and BPAF) on the life span of C. elegans (Fig. 1A). We found that for five of the six BPA and its substitutes (BPA, BPB, BPF, TMBPF, and BPAF), exposure to varying pathophysiologically relevant concentrations had minimal impact on the life span of C. elegans (SI Appendix, Fig. S1). However, for BPS, we surprisingly observed that exposure to 300, 450, and 600 nM significantly shortened life span by approximately 8.6%, 10.4%, and 11.5%, respectively, compared to that in the Veh group (Fig. 1B).

Fig. 1.

BPS exposure at pathophysiologically relevant concentrations shortens life span. (A) Schematic diagram of the experiment illustrating the effects of BPA substitutes on the life span of C. elegans. (B) Life span analysis of C. elegans exposed to different pathophysiologically relevant concentrations of BPS (0, 150, 300, 450, and 600 nM). (C and D) SA-β-gal staining and quantification of passage 10 (P10) IMR-90 cells following exposure to different pathophysiologically relevant concentrations of BPS (0, 450, and 600 nM). (E) Assessment of IMR-90 cell viability following exposure to varying pathophysiologically relevant concentrations of BPS (0, 600 nM, and 1 mM) for 6, 12, 24, and 48 h, using the CCK-8 assay. (F and G) Life span analysis of female and male mice exposed to different pathophysiologically relevant concentrations of BPS (0, 25, 50, and 125 μg/kg/d). All values are means ± SEM. Data were considered statistically significant at P < 0.05 calculated by using one-way ANOVA (D and E) or log-rank test (B, F, and G). The corresponding sample sizes are indicated within the plots. * P < 0.05; ** P < 0.01; *** P < 0.001.

To rule out the possibility that the life span reduction was due to the acute toxicity of BPS, we exposed passage 10 (P10) IMR-90 cells to BPS (450 nM and 600 nM) for 7 d and performed SA-β-Gal staining. Results showed that compared to cells in the Veh group, the percentage of SA-β-Gal positive cells increased markedly in BPS groups (Fig. 1 C and D), with no significant change in cell morphology. Additionally, CellCounting Kit-8 (CCK-8) cell viability assay showed that 600 nM BPS exposure did not significantly impair IMR-90 cell viability compared to the Veh group, while the 1 mM BPS (positive control) induced a marked decrease in cell viability (Fig. 1E, all P < 0.05). These results suggest that the exposure doses of BPS in our study had no acute cytotoxicity.

We further established a mouse BPS exposure model using C57BL/6 WT female and male mice (25, 50, and 125 μg/kg/d, from 4-mo-old to the end of life). Results showed that regardless of sex, mice exposed to BPS (125 μg/kg/d) had shorter life spans than those in the Veh group (Fig. 1 F and G). The mean life span was reduced by 12.4% in females and 13.9% in males (Fig. 1 F and G). Collectively, these results indicate that pathophysiologically relevant concentration (125 μg/kg/d) of BPS exposure is a driving factor for aging across species.

Pathophysiologically Relevant Concentrations of BPS Exposure Compromise Health Span and Induce Multiorgan Aging.

We next assessed the impact of BPS exposure (125 μg/kg/d) on the health span of mice. We initiated daily BPS exposure from 4 mo of age, and analyzed the changes in aging-related pathological and molecular markers when these mice reached 20 mo of age (Fig. 2A). Compared to the Veh group, BPS exposure accelerated age-related weight gain by approximately 14.8% (veh: 31.52 ± 0.65 g vs. BPS: 36.20 ± 0.83 g, Fig. 2B). Spinal structural analysis revealed that the incidence of kyphosis in the BPS-exposed group was nearly double that in the Veh group (Fig. 2C). Additionally, consistent with the more frequent hair loss or discoloration in the BPS exposure group (Fig. 2C), the frailty index (FI), a recognized tool for rapid assessment of aging, was also significantly higher in mice from the BPS exposure group, regardless of sex (Fig. 2D). Echocardiography revealed that BPS-exposed mice exhibited reduced ejection fractions, indicating impaired cardiac pumping function (Fig. 2 E and F). Since metabolic dysregulation is a hallmark of aging, we assessed the metabolic status of mice from each group. Results showed that in contrast to mice in the Veh group, BPS-exposed mice exhibited lower respiratory exchange ratio (RER) levels (Fig. 2G) and reduced activity during the dark phase (Fig. 2H), indicating an impaired basal metabolic state after BPS exposure. We also conducted a comprehensive set of behavioral tests, including grip strength (Fig. 2I), memory (Fig. 2J), physical performance (Fig. 2K), general well-being (Fig. 2L), glucose homeostasis (Fig. 2M), anxiety (Fig. 2N), and depression (Fig. 2N). Across all behavioral metrics, BPS-exposed mice performed worse than mice from the Veh group, and these changes were consistent across the sexes (Fig. 2 I–N).

Fig. 2.

BPS exposure at pathophysiologically relevant concentrations impairs health span in mice. (A) Schematic diagram of an experiment exploring the impact of BPS exposure on the health span of mice. (B–N) Changes in body weight (B), alopecia/depigmentation phenotype and curvature of the spine (C), FI scores (D), echocardiography and left ventricle ejection fraction [(E) and (F)], RER and motion [(G) and (H)], muscle strength (I) (grip-strength tests), memory (J) (Y-maze test), physical performance (K) (open field test), general well-being (L) (nest building test), glucose homeostasis (M) (glucose tolerance test and insulin tolerance test), and anxiety and depression (N) (tail suspension test, light/dark box test, elevated zero maze test, and sucrose preference test) in 20-mo-old WT C57BL/6 mice orally fed daily with BPS (125 μg/kg/d) beginning at 4-mo-old. All values are means ± SEM or n (%). Data were considered statistically significant at P < 0.05 calculated by using Student’s t test (D, F, I–L, and N), or chi-squared test (C), or two-way ANOVA (B, G, H, and M). In the bar graph, red dots represent data from female mice, while black dots represent data from male mice. The corresponding sample sizes are indicated within the plots. *P < 0.05; **P < 0.01; ***P < 0.001.

We assessed senescence markers, p16 and p21, in tissues from mice (20-mo-old) of each group. Results showed that compared to the Veh group, expressions of p16 and p21 increased significantly in the heart, liver, kidneys, lungs, and skeletal muscles of the mice from the BPS group (Fig. 3 A–C and SI Appendix, Figs. S2 and S3 A and B). We also systematically evaluated pathological changes in organs/tissues by histological staining. Results showed that compared to the Veh group, mice in the BPS group showed decreased nuclei and increased fibrosis in the myocardium (Fig. 3D and SI Appendix, Fig. S3C), increased abnormal hepatic lobules (Fig. 3E) and liver fibrosis (SI Appendix, Fig. S3D), accelerated glomerular atrophy (Fig. 3F) and fibrosis (SI Appendix, Fig. S3E), increased alveolar septal thickness (Fig. 3G) and fibrosis (SI Appendix, Fig. S3F), and decreased cross-sectional area of myofibers (Fig. 3H) and increased myofibrosis (SI Appendix, Fig. S3G); all these are key characteristics of organ aging. Overall, these series of analyses confirm that BPS exposure exerts widespread detrimental effects on the health span of mice.

Fig. 3.

BPS exposure at pathophysiologically relevant concentrations accelerates multiorgan aging in mice. (A and B) Relative mRNA levels of p16 (A) and p21 (B) in various mouse tissues, as measured by qPCR. (C) Relative protein levels of p16 and p21 in various mouse tissues, as measured by Western blotting. (D–H) Changes in HE staining in heart, liver, kidney, lung, and muscle tissues in 20-mo-old WT C57BL/6 mice orally fed daily with BPS (125 μg/kg/d) beginning at 4-mo-old. The arrows indicate abnormal hepatic lobule (E), glomerular atrophy (F), increased alveolar septal thickness (G), and decreased cross-sectional area of myofiber (H). All values are means ± SEM. Data were considered statistically significant at P < 0.05 calculated by using Student’s t test. In the bar graph, red dots represent data from female mice, while black dots represent data from male mice. The corresponding sample sizes are indicated within the plots. **P < 0.01; ***P < 0.001.

BPS Predominantly Accumulates in Brown Adipose Tissue (BAT) and Accelerates Its Aging.

To elucidate the mechanism of action of BPS, we then investigated the distribution of BPS in different tissues. Results showed that BPS was detectable in BAT, kidneys, liver, and plasma, with concentrations in BAT being approximately three to five times higher than in the kidneys, liver, or plasma, and ten times higher than in white adipose tissue (WAT) (SI Appendix, Fig. S4A). This is consistent with previous reports indicating that BPs preferentially accumulate in BAT (11, 12).

It is well known that BAT plays a central role in the regulation of body energy metabolism and its dysregulation is closely related to the development of aging-related metabolic diseases (13, 14). Recent studies revealed that BAT transplantation mitigates age-associated disorders such as muscle atrophy and obesity (15, 16), while also prolonging ovarian life span in aging mice (17–19). Given the central role of BAT in regulating aging across multiple organs and our finding that BPS predominantly accumulates in BAT, we will further investigate whether BPS modulates aging through BAT using a BAT transplantation strategy in mice. Results of the following analysis showed that the adipocytes in BAT from BPS-exposed mice (20-mo-old) were enlarged and exhibited a disorganized morphology compared to that in the Veh group (SI Appendix, Fig. S4B). Notably, there were no significant morphological changes observed in WAT (SI Appendix, Fig. S4C). Additionally, expressions of p16 and p21 in the BAT from BPS-exposed mice (20-mo-old) increased markedly compared to that in mice from the Veh group (SI Appendix, Figs. S4 D–H and S5). These results indicated that BPS exposure induces structural abnormalities in BAT and accelerates its aging.

Molecular Pathways Involved in BPS-Induced BAT Aging.

We further investigated the molecular pathways affected in BAT by pathophysiologically relevant concentration of BPS exposure, using RNA-seq analysis (SI Appendix, Fig. S6A). Compared with the Veh group, we observed widespread changes in gene transcription levels in the BATs from BPS-exposed mice. Using a screening threshold of |Log₂FC| > 1 and P < 0.05, 277 and 677 genes were found to be upregulated and downregulated, respectively (SI Appendix, Fig. S6B). Principal component analysis was performed, and PC1 revealed distinct expression patterns between Veh and BPS-exposed BAT (SI Appendix, Fig. S6C). The upregulated differentially expressed genes (DEGs) in BPS-exposed BAT were enriched in several hallmarks of aging, including 1) inflammatory pathways such as “acute inflammatory response,” “regulation of nitric-oxide synthase activity,” “cytokine binding,” “monooxygenase activity,” and “negative regulation of immune system processes.” 2) macroautophagy pathway involving “phagosome”; 3) proteostasis pathways such as “negative regulation of proteolysis” and “glycoprotein catabolic process”; and 4) cellular senescence pathway (SI Appendix, Fig. S6 D and F). Conversely, downregulated DEGs were also enriched in hallmark pathways of aging, including 1) mitochondrial pathways such as “ATP metabolic process” and “positive regulation of heat generation”; 2) intercellular communication pathways involving “cell–cell contact zone,” “transmembrane transporter complex,” and “regulation of transmembrane transporter activity”; 3) cellular senescence pathway involving “positive regulation of growth”; 4) deregulated nutrient-sensing pathways involving “cellular response to nutrient levels” and “response to glucocorticoid”; and 5) proteostasis-related pathways such as “protein digestion and absorption,” “protein serine/threonine kinase activity,” “protein polymerization” and “regulation of protein secretion” (SI Appendix, Fig. S6 E and G).

Gene Set Enrichment Analysis (GSEA) revealed that BPS exposure induced widespread age-related changes in gene transcription levels in mouse BAT, encompassing several key processes: mitochondrial dysfunction (SI Appendix, Fig. S7), chronic inflammation (SI Appendix, Fig. S8 A–C), deregulated nutrient-sensing (SI Appendix, Fig. S8 D–J), cellular senescence (SI Appendix, Fig. S8 K and L), impaired macroautophagy (SI Appendix, Fig. S8 M–P), loss of proteostasis (SI Appendix, Fig. S8 Q–T), and altered intercellular communication (SI Appendix, Fig. S8 U–X). Notably, we observed that BPS exposure significantly disrupted mitochondrial function and the regulation of glucose metabolism in BAT. This finding is consistent with our earlier observations of metabolic disturbances, including alterations in basal metabolism (Fig. 2 G and H) and glucose homeostasis (Fig. 2M), in BPS-exposed mice.

Collectively, the above analysis revealed that BPS exposure accelerated the aging and functional abnormalities of BAT, involving multiple aging hallmarks like mitochondrial dysfunction, chronic inflammation, deregulated nutrient sensing, cellular senescence, impaired macroautophagy, loss of proteostasis, and altered intercellular communication (20).

BAT Transplantation Mitigates the Detrimental Effects of BPS on Both Life Span and Health Span.

Given the key role of BAT in the regulation of energy homeostasis and the aging process, along with our observation that BPS can predominantly accumulate in BAT to accelerate BAT aging, we then conducted BAT transplantation in mice models to investigate whether BPS-induced multiorgan aging is mediated by BAT dysfunction. Briefly, 4-mo-old mice were subjected to 8 mo of Veh or BPS exposure, followed by BAT transplantation. After transplantation, these mice continued their original exposure for another 8 mo. Immunohistochemical study on BAT from donor mice before transplantation showed that after 8 mo of BPS (125 μg/kg/d) exposure, UCP-1 protein levels in the BAT were significantly reduced compared to that in the Veh group (SI Appendix, Fig. S9).

The experimental mice were divided into four groups: Veh-exposed mice receiving BAT from Veh-exposed donors (Veh-Veh); Veh-exposed mice receiving BAT from BPS-exposed donors (Veh-BPS); BPS-exposed mice receiving BAT from BPS-exposed donors (BPS-BPS); and BPS-exposed mice receiving BAT from Veh-exposed donors (BPS-Veh) (Fig. 4A). Four weeks after BAT transplantation, we were able to detect functionally active BAT using PET/CT (Fig. 4B).

Fig. 4.

BAT transplantation reverses the detrimental effects of BPS (125 μg/kg/d) on both life span and health span. (A) Schematic diagram of the effect of BAT transplantation on life span and health span following BPS exposure. (B) Representative ¹⁸F-FDG-PET/CT images showing BAT transplantation in mice. (C) Life span analysis of BAT-transplanted mice exposed to different pathophysiologically relevant concentrations of BPS (Veh or 125 μg/kg/d). (D–K) Changes in muscle strength (D) (grip-strength tests), memory (E) (Y-maze test), anxiety and depression (F) (tail suspension test, light/dark box test, elevated zero maze test, and sucrose preference test), physical performance (G) (open field test), general well-being (H) (nest building test), glucose homeostasis (I), and RER and motion [(J) and (K)] in 20-mo-old WT C57BL/6 mice receiving BAT transplantation (12-mo-old) and different pathophysiologically relevant concentrations of BPS (0 or 125 μg/kg/d) beginning at 4-mo-old. All values are means ± SEM. Data were considered statistically significant at P < 0.05 calculated by using the log-rank test (C) or one-way ANOVA (D–H) or two-way ANOVA (I–K). In the bar graph, red dots represent data from female mice, while black dots represent data from male mice. The corresponding sample sizes are indicated within the plots. *P < 0.05; **P < 0.01; ***P < 0.001.

In the life span analysis, in contrast to mice in the Veh-Veh group, Veh-BPS mice experienced a significant reduction in life span of approximately 14.9% (Fig. 4C), whereas BPS-Veh mice exhibited an approximately 14.2% increase in life span compared to mice in the BPS-BPS group (Fig. 4C). In the health span analysis, extensive experiments conducted in both female and male mice corroborated the observed life span phenomena. Specifically, mice in the Veh-BPS group showed promoted aging characteristics compared to the Veh-Veh group (Fig. 4 D–H), while mice in the BPS-Veh group displayed delayed aging characteristics compared to mice in the BPS-BPS group (Fig. 4 D–H). Notably, results of glucose homeostasis assessments (Fig. 4I) and energy metabolism measurements (Fig. 4 J and K) revealed that the metabolic state of the mice is altered following BAT transplantation, reflecting changes in the aging phenotypes. Additionally, pathological analyses showed consistent trends of changes in aging phenotypes across groups (SI Appendix, Fig. S10). Overall, these results indicate that a pathophysiologically relevant concentration of BPS exposure accelerates systemic organ aging and ultimately affects the life span by disrupting BAT-regulated energy metabolism, and BAT transplantation is an effective way to mitigate BPS impacts.

Discussion

BPS is a chemical compound used as an alternative to BPA in various products, including plastics and resins food packaging, and consumer goods. Although BPS is marketed as a safer substitute for BPA, concerns have arisen owing to their similar biological effects and potential health risks (21). Telomere attrition is an important marker of aging. A prospective birth cohort study in China indicated that for mothers aged ≥28 y, each doubling of serum BPS was associated with a 3.19% reduction in the relative telomere length of umbilical cord blood leukocytes (22). Furthermore, BPS exposure is associated with age-related diseases. For instance, a cross-sectional study conducted in South Korea, which included 3,787 adults aged ≥19 y from the Korean National Environmental Health Survey 2015 to 2017, found a significant association between BPS exposure and obesity (23). A recent epidemiological report highlights that urinary BPS is positively associated with the risk of cardiovascular diseases (24). Additionally, a cross-sectional study indicated that high levels of BPS exposure are associated with an increased incidence of nonalcoholic fatty liver disease (25). These studies indicate that the risk of age-related diseases, particularly metabolic disorders, due to BPS exposure, warrants global attention. To the best of our knowledge, the effects of pathophysiologically relevant concentrations of BPS on normal life span remain unknown. In our study, we found that BPS exposure, regardless of sex, negatively affected the life span, health span, and pathological morphology of multiple organs. Furthermore, these adverse effects could be mitigated by BAT transplantation, suggesting that targeting BAT may be an effective intervention strategy for BPS-induced aging.

To our surprise, at the same exposure level as BPS, we did not observe a significant impact of BPA on the lifespan of C. elegans. We speculate that this discrepancy may be attributed to several factors. First, studies have confirmed that BPS is as hormonally active as BPA, but BPS has been shown to elicit additional effects in vitro and in vivo, including altered organ weights, reproductive outcomes, and enzyme expression (26). A prior study has shown that the minimum effective concentration for BPS was one-tenth that of BPA in fetal mouse testes supplemented with luteinizing hormone (27). Another in vitro study has demonstrated that BPS is a more potent adipogen than BPA (28). Consistent with these in vitro findings, studies using animal models have suggested that BPS in food induces hormonal and obesogenic effects that are even more pronounced than those of BPA (29). These results indicate that the toxicological effects of BPS may surpass those of BPA in certain aspects, particularly in lipid metabolism. Additionally, pharmacokinetic studies suggest a lower biotransformation efficiency of BPS (30), indicating that BPS is more challenging for the human liver to detoxify. This implies that the toxicological effects of BPS may exceed those of BPA, even at the same or lower exposure levels. Notably, the dose–response relationships of BPA and BPS in vivo are not simply linear. Epidemiological studies have indicated a “J”-shaped dose–response relationship of BPS exposure, whereas BPA exhibits an inverted “U”-shape (31). These findings suggest that the selected exposure range in our study may fall within the chronic toxicity range of BPS, while remaining within the nontoxic range for BPA. This highlights the need for future research to expand the exposure gradient of BPA to better elucidate the effects of ultralow exposure levels on lifespan.

BAT is a specialized form of fat known for its critical role in thermogenesis and energy expenditure. BAT contains a high density of mitochondria, which are responsible for its thermogenic function (32). As aging progresses, mitochondrial function within BAT may be significantly compromised, manifesting as mitochondrial damage, reduced functionality, and disrupted energy metabolism (33). Meanwhile, mitochondrial dysfunction in BAT can directly affect BAT-regulated energy metabolism, thus accelerating the aging process (33). Recent advancements have demonstrated the efficacy of BAT transplantation as a treatment for age-associated disorders such as muscle atrophy and obesity (15, 16). Emerging studies have indicated that BAT transplantation can enhance follicle and oocyte quality as well as prolong the ovarian life span in aging mice (17–19). These findings suggest that BAT plays a crucial role in the regulation of the aging process. However, a comprehensive understanding of BAT’s role in aging and its response to BPS remains lacking.

In this study, through BAT transplantation, we demonstrated that BPS accelerates aging by disrupting BAT function. Our analysis indicated that exposure to BPS causes severe disruption of the mitochondrial function in BAT, as evidenced by the overall downregulation of several key mitochondrial pathways, including “Mitochondrial Biogenesis” “Respiratory Electron Transport, ATP Synthesis by Chemiosmotic Coupling, and Heat Production by Uncoupling Proteins,” “Electron Transport Chain Oxphos System in Mitochondria,” and “Mitochondrial Protein Import.” BPS-induced BAT dysfunction may accelerate aging through the following mechanisms: 1) mitochondrial damage impairs BAT thermogenic capability, hindering its ability to effectively regulate energy metabolism-as reflected by our results, which then affects overall health and life span (34); 2) BAT dysfunction also disrupts glucose and lipid metabolism, leading to abnormal nutrient sensing, another key hallmark of aging. This further predisposes the body to aging-related metabolic diseases, such as obesity, diabetes, and cardiovascular diseases (35, 36), ultimately reducing health span.

In contrast to prior studies, our research systematically examines the impact of pathophysiologically relevant concentrations of BPS on health and life span. Additionally, we employed a mouse model to validate the effects of BPS exposure at a pathophysiologically relevant concentration on aging. These findings offer a more nuanced understanding of the implications of daily BPS exposure for human health.

To summarize, we present a comprehensive exploration of pathophysiological BPS exposure in the aging process (SI Appendix, Fig. S11). Our findings suggest that BPS accelerates aging by disrupting BAT function, reflected in mitochondrial dysfunction, chronic inflammation, deregulated nutrient sensing, cellular senescence, impaired macroautophagy, loss of proteostasis, and altered intercellular communication. These disturbances collectively impair BAT-regulated energy metabolism, which in turn reduces overall health and life span. Further investigation into the precise mechanisms by which BPS exerts its systemic effects will be crucial in developing strategies to mitigate its potential impacts.

Materials and Methods

Animal.

All animal experiments were conducted in compliance with ethical standards approved by the Ethics Committee of the Center for Animal Experiments at West China Hospital, Sichuan University (Approval No. 20200221047), and adhered to the 3Rs principle. C57BL/6 mice (3-mo-old) were purchased from SIPEIFU (Beijing) Biotechnology Co., Ltd. [Experimental Animal Production License Number: SCXK (Jing) 2019-0010]. All animals were maintained in Specific Pathogen Free (SPF)-grade housing conditions. Mice were housed in a temperature-controlled environment (22 ± 2 °C) with a 12-h light–dark cycle and ad libitum access to food and water. To minimize potential contamination by exogenous BPA, BPA substitutes, and environmental estrogens, the mice were housed in polypropylene cages, and water was supplied in glass bottles to eliminate plastic-derived contaminants. Bedding materials comprised 60Co gamma-irradiated aspen shavings (product code: SPF-M01-003, SIPEIFU), confirmed to be free of phytoestrogens.

Cell Culture.

IMR-90 human embryonic lung fibroblasts were obtained from Cellcook Biotech Co., Ltd. (Guangzhou, China). Cell line authentication was performed using short tandem repeat profiling to confirm the identity of IMR-90 cells. Mycoplasma contamination was tested using the Myco-Lumi™ Kit (C0298S, Beyotime), and all cultures were confirmed to be free of Mycoplasma contamination before use. IMR90 cells were cultured in an IMR-90-specific culture medium (CM-0538, Procell), with the medium being replaced every other day. The cells were maintained at 37 °C and 5% CO₂ in a humidified incubator.

In Vivo and In Vitro BPS Exposure Models.

4,4′-Sulfonyldiphenol (BPS, HY-W011927, 99% purity), BPB (HY-W013935, 99% purity), and BPAF (HY-W654077, 99% purity) were obtained from MedChemExpress (USA). BPF (B47006, 98% purity) was purchased from Millipore Sigma (USA), while TMBPF (M304433, 98% purity) was purchased from TRC (Canada).

C. elegans model.

In this study, we investigated the impact of various pathophysiologically relevant concentrations of BPA and its substitutes on the life span of C. elegans. Data available on BPA have indicated that total serum levels in humans can reach up to ∼430 nM, while data on other bisphenols are limited, BPS has been reported to reach a maximal serum level of approximately 680 nM (10). Based on this information, we established a gradient of BPA and its substitutes concentrations for C. elegans cultivation: 150, 300, 450, and 600 nM. C. elegans treated with OP50 (Veh) was used as the control group.

IMR-90.

To further investigate the toxicological characteristics of BPS, we examined its cytotoxicity in IMR-90 cells. We established a gradient of BPS concentrations for IMR-90 cells, consisting of four groups: Veh (dimethyl sulfoxide), 450 nM, 600 nM (the maximal exposure level in our study), and 1 mM (positive control). The concentration of 1 mM for the positive control was based on previous toxicological studies of BPS (9).

Mice model.

Epidemiological and basic research studies have shown that BPs can affect development of humans and mammals (4, 37–39), indicating that exposure during fetal or juvenile stages may impact physical maturation and potentially shorten lifespan. These effects are distinct from the pathological processes of aging. To avoid confusion, we opted to start exposure at 4 mo of age in mice, equivalent to approximately 20 y of age in humans (40). We established an appropriate pathophysiologically relevant concentration of BPS by referencing epidemiological data on humans (6, 12). These data indicate urinary BPS concentration quartiles of 0.2, 0.4, and 1.0 ng/mL. Building on our previous research on the dose–response relationship between BPS intake and urinary BPS levels, we established oral BPS exposure doses at 25, 50, and 125 μg/kg/d (12), which respectively represent the 25th, 50th, and 75th percentiles of human BPS exposure levels. Also, these BPS doses are far below the “no observed adverse effect level” of 10 mg/kg/d and can thus be considered low (41, 42). The exposure procedure was conducted as previously described (43). Briefly, BPS was dissolved in peanut oil and was carefully incorporated into the food to ensure consistent delivery (43–45). Mice were allowed to consume food and water freely, and the exposure diets were formulated based on the average food consumption of each mouse according to the following formula: Diet Dose (μg/kg) = (Single Daily Dose [μg/kg/d] × Body Weight [kg])/Food Intake (kg/d). For example, the average food intake was approximately 2 g/d for mice weighing 20 g, then BPS was incorporated into the food at 250, 500, and 1,250 μg/kg to yield the low (25 μg/kg/d), median (50 μg/kg/d), and high (125 μg/kg/d) exposure levels, respectively. Mice in the control group were treated with an equivalent Veh (peanut oil) in the same way. During the exposure process, food intake and body weight were monitored weekly, and the diet dose was adjusted accordingly to maintain a consistent exposure level in each group.

Life-Span Analysis.

C. elegans.

The life span of C. elegans (N2 Bristol) was evaluated on Nematode Growth Medium plates supplemented with BPA, BPB, BPF, BPS, TMBPF, or BPAF, according to established protocols (46).

Mice.

Life span analysis was conducted in mice administered oral BPS supplementation, either alone or alongside other interventions.

Health-Span Analysis.

We conducted a comprehensive health-span analysis utilizing a range of methods, including the FI (47), grip strength measurement, Y-maze test, open field test, nest building test, tail suspension test, light/dark box test, elevated zero maze test, sucrose preference test, glucose tolerance test, insulin tolerance test, microCT examination, ¹⁸F-FDG-PET/CT scanning, energy expenditure, and echocardiographic evaluation. Detailed information is provided in the Supporting Information (Health-span analysis).

BAT Transplantation in Mice.

The transplantation procedure was performed on mice based on previous reports (48, 49). Briefly, mice were anesthetized using pentobarbital sodium. BAT was harvested from the interscapular region of anesthetized donor mice under sterile conditions. The harvested tissue was then transplanted into a subcutaneous pocket between the scapulae of recipient mice under anesthesia, and the incision was closed using absorbable sutures. To ensure postoperative care, all mice were administered prophylactic antibiotics (enrofloxacin, 10 mg/kg, for 3 d postsurgery). Four weeks after the procedure, graft viability and integration were evaluated via ¹⁸F-FDG-PET/CT imaging.

CCK-8 Assay.

The CCK-8 assay kit (C0041, Beyotime) was utilized to assess cell viability. A total of 100 μL of cell suspension was plated into a 96-well plate at a density of 5 × 10³ cells per well. At 6, 12, 24, and 48 h postseeding, 10 μL of CCK-8 solution was added to each well and gently mixed to ensure that no bubbles were formed. Following a 2-h incubation at 37 °C in a 5% CO₂ environment, absorbance at 450 nm was measured using a microplate reader.

BPS Quantification.

The concentrations of BPS in the heart, liver, kidney, lung, muscle, BAT, WAT, and plasma were measured using high-performance liquid chromatography–mass spectrometry at WEIPU Company (Suzhou, China), following the manufacturer’s protocol for the instruments.

Histopathological Study.

We conducted a comprehensive histopathological study, incorporating SA-β-gal assay, qPCR, western blotting, immunohistochemistry, hematoxylin-eosin (HE) staining, and Masson staining, with detailed information available in the SI Appendix, Histopathological Study.

Transcriptome Sequencing.

Transcriptome sequencing was conducted on total RNA extracted from BATs of mice (vehicle control group: n = 8; BPS-exposed group: n = 7), using the DNBSEQ platform at the Beijing Genomic Institution (China). BAT from female mice was collected exclusively during the metestrus/diestrus (quiescent phase). Total RNA was enriched for mRNA using oligo(dT)-attached magnetic beads, then fragmented and reverse transcribed into cDNA. The synthesized double-stranded DNA was then end-repaired and 5′-phosphorylated, with a protruding “A” at the 3′ end forming a blunt end, followed by ligation of a bubble-shaped adapter with a protruding “T” at the 3′ end. PCR amplification of the ligation products was followed by denaturation and the formation of single-stranded circular DNA libraries using a bridged primer. The average alignment ratio of the sample comparison genome was 98.29%. The average alignment ratio of the gene set was 53.13%, and a total of 18,062 genes were detected. The project details (including reads filtering, number of sequence alignments, mapping rate, etc.) are provided in SI Appendix, Tables S1–S3. Raw sequencing data were submitted to the NCBI Gene Expression Omnibus (GEO) under the accession number GSE277525 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE277525) (50). Differential gene expression analysis identified DEGs with |log₂FC| > 1 and adjusted P < 0.05. A heatmap displaying Z-scores of individual genes was generated using the heatmap.2 function from the gplots package (version 3.0.1) in R. DEGs were classified as upregulated or downregulated based on log₂FC for enrichment analysis. GO and KEGG pathway analyses were performed with the clusterProfiler R package. Additionally, GSEA was applied to evaluate differences in biological pathways, with results visualized using the enrichplot package when P < 0.05.

Statistical Analysis.

Analyses were performed using GraphPad Prism v8.0. Data are presented as mean ± SEM, with statistical significance set at P < 0.05. Statistical analyses were performed using Student’s t test for comparisons between two groups, one-way ANOVA for comparisons across multiple groups, two-way ANOVA for evaluating the interaction of two or more factors, or the log-rank test for survival analysis. Sample size determination did not rely on a predetermined statistical method; however, they align with those documented in prior studies addressing similar topics (51, 52). Except for RNA-seq data, all experiments were independently replicated at least three times. The sample sizes, statistical methods, and corresponding P values are detailed in the figure legends.

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

Raw sequencing data were submitted to the NCBI Gene Expression Omnibus (GEO) under the accession number GSE277525 (50).