https://orcid.org/0000-0002-1730-8883
Abstract
There is growing evidence that organophosphate esters (OPEs) can act as endocrine-disrupting chemicals. However, only a few studies have assessed the effects of OPE exposure on one of the most important endocrine glands in the body, the adrenal gland. Our aim was to test the effects of a mixture of OPEs detected in Canadian house dust on adrenal function in Sprague Dawley rats. Adult male and female rats (n = 15 per treatment group) were administered either a vehicle or an OPE mixture (0.048, 1.6, or 48 mg/kg bw/d) for 70 to 72 d via their diet. With OPE exposure, adrenal glands from male adult rats were reduced in weight, whereas those of female rats showed an increase in weight. This led us to investigate whether OPEs induce sex-specific effects on adrenal gland function and the mechanisms involved. Serum levels of two adrenal hormones, aldosterone and corticosterone, were decreased only in male serum samples. Serum levels of renin and adrenocorticotropic hormone, which regulate aldosterone and corticosterone synthesis, respectively, were assessed. Exposure to the OPE mixture decreased renin levels only in males. Serum biochemistry analysis revealed that triglycerides and LDL cholesterol levels were increased in males. Transcriptomic analysis revealed that the top affected pathways in male adrenal glands from all three treatment groups were related to potassium channels, which play a role in regulating aldosterone and corticosterone levels. The most affected pathways in female adrenal glands were related to cholesterol biosynthesis and immune functions. These results show that an environmentally relevant mixture of OPEs affects adrenal function and that these effects are sex specific.
Organophosphate esters (OPEs) are extensively used as flame retardants and plasticizers in various products including furniture, baby toys, electronics, textiles, and building materials (van der Veen and de Boer 2012; Wei et al. 2015). Since OPEs are not covalently bonded to the polymers, they gradually volatilize into the air and subsequently settle into dust. This process creates multiple exposure pathways for individuals, including ingestion, inhalation, dermal contact, and dietary intake of contaminated dust (Agency for Toxic Substances and Disease Registry (ATSDR) 2012; van der Veen and de Boer 2012; Hou et al. 2021). Among the indoor exposure pathways, ingestion of dust is the major source of human exposure to OPEs (Kim et al. 2019). Consequently, OPE metabolites are detectable in the urine samples of nearly 100% of adults and children in the United States, Europe, China, and Australia (Cequier et al. 2015; Hoffman et al. 2017; He et al. 2018a; Ospina et al. 2018; Sun et al. 2018). Moreover, OPEs have been identified in other human matrices, such as blood (reviewed by Chokwe et al. 2020; Hou et al. 2022), breast milk (Ma et al. 2019), hair, and nails (Liu et al. 2016; He et al. 2018b).
Exposure to OPEs is associated with various adverse health effects, including neurotoxicity (Doherty et al. 2019; Patisaul et al. 2021), developmental and reproductive toxicity (reviewed by Meeker and Stapleton 2010; Carignan et al. 2017; Hales and Robaire 2020; Siddique et al. 2022), carcinogenicity (Blum et al. 2019), and endocrine disruption (Rosenmai et al. 2021; Yao et al. 2021; Hu et al. 2023). The concurrent use of multiple OPEs raises concerns about the potential health risks of OPE mixtures. Studies in both humans and rodent models indicate that exposure to mixtures of OPEs can affect multiple organs, including bone (Pillai et al. 2014; Macari et al. 2020; Yan and Hales 2021), liver (Witchey et al. 2020; Aluru et al. 2021), brain (Baldwin et al. 2017; Gillera et al. 2020; Rock et al. 2020; Wiersielis et al. 2020; Marinello et al. 2022; Newell et al. 2023; Witchey et al. 2023), ovary (Wang et al. 2024), and adipose tissue (Pillai et al. 2014; Tung et al. 2017a, 2017b). Exposure to one of the most extensively studied mixtures, Firemaster 550, has demonstrated endocrine-disrupting potential, resulting in weight gain and early onset of puberty in rats at environmentally relevant levels (Patisaul et al. 2013). Despite these findings, research on the impact of OPEs on the adrenal gland, one of the most critical endocrine glands in the body, remains limited.
The adrenal gland produces hormones that regulate several vital functions, including metabolism, blood pressure, and the stress response (Rosol et al. 2001). Loss of adrenal function, as seen in adrenal insufficiency, can be life-threatening (Burton et al. 2015). The adrenal glands are among the endocrine tissues most frequently affected by chemicals (Ribelin 1984). For instance, the off-target effects of etomidate have led to fatalities due to the unrecognized inhibition of steroidogenic enzymes involved in cortisol and aldosterone synthesis (Wagner et al. 1984). Despite the critical roles of the adrenal glands, they are often overlooked in toxicological studies, which have primarily focused on reproductive and developmental endpoints (Cockburn and Leist 1999). To date, only a few studies have examined the effects of OPEs on the adrenal glands, and these studies have tested a very limited number of OPEs. In vivo studies have focused on the effects of three OPEs: Tris(methylphenyl) phosphate (TMPP), isopropylated triphenyl phosphate (IPPP), and tris(1,3-dichloro-2-propyl) phosphate (TDCIPP). Only studies using TMPP assessed both male and female subjects, finding that adrenal gland weight and the adrenal cortex were affected in both sexes (Latendresse et al. 1993, 1994, 1995; National Toxicology Program 1994). In contrast, studies using IPPP and TDCIPP tested only male rats (Wade et al. 2019; Akimoto et al. 2022). None of these studies explored the mechanisms of action of OPEs.
In vitro studies using human H295R adrenal cortical cells and PC12 rat adrenal cells have investigated a broader range of OPEs. These studies found that OPEs could induce cytotoxicity (Chang et al. 2020) and alter the production of steroid hormones, specifically testosterone and 17β-estradiol, by affecting enzymes in the steroidogenic pathway (Liu et al. 2012). Using human H295R adrenal cells, we previously demonstrated that exposure to commonly used OPEs affects both the phenotype and function of these cells (Li et al. 2023). Given the contrasting effects observed with individual OPEs on cortisol and aldosterone production levels, we assessed the impact of an environmentally relevant mixture of OPEs to better mimic real-world exposure scenarios. This mixture induced accumulation of lipid droplets as a primary target and affected the steroid hormone production levels in the cells (Li et al. 2024). This body of evidence underscores the necessity of including the adrenal glands in toxicological evaluations.
The goal of this study was to assess the impact of a Canadian household dust-derived mixture of OPEs on adrenal gland function in Sprague Dawley rats. This research represents the first in vivo assessment of real-world OPE exposure effects on the whole adrenal gland of both sexes. We assessed adrenal gland weight, histology, and function following exposure to the OPE mixture. Furthermore, we examined the molecular mechanisms underlying these effects by elucidating affected pathways in both male and female rats using RNA sequencing.
Materials and methods
OPE house dust mixture
The OPE house dust mixture, which contains 13 OPEs detected in over 85% of house dust samples collected from 144 urban Canadian homes between 2007 and 2010, was formulated by Dr Michael G. Wade (Health Canada) (Fig. 1; Fan et al. 2014; Kubwabo et al. 2021). The relative proportion of each OPE in this mixture is based on their 95th percentile values from these samples.
Composition of the Canadian household dust-based OPE mixture. Bodyweight and food weight were recorded weekly for each rat. Using these values, the average food consumption was calculated, along with the resulting proportion of the individual OPEs and the overall mixture received by the animals (mg/kg bw/d) for each dose.
Diets were prepared fresh every month and fed to the animals no later than 1 mo after preparation. To maximize the homogeneous distribution of the OPE mixture with the diet, the mixture was added to the mixing bowl containing powdered AIN-93G diet in a stepwise manner (Reeves 1997). The combined diet, at the desired mixture concentration, was mixed with 8% (by weight) deionized water, pelleted, and dried at room temperature for 20 h. Each diet was labeled with a date of preparation and a color code. The identity of the diet dose was revealed only after all in-life handling and related data were collected and analyzed. The diets were formulated to deliver mixture doses of 0.048, 1.6, or 48 mg/kg bw/d. This estimation is based on a daily food consumption rate for rats of 80 g/kg bw/d and accounts for assumptions regarding dust ingestion rates (100 mg/d) for children (16.5 kg bw), as well as the scaling of dose from humans to rodents based on body surface area (1:6.3 human to rat) rather than body weight. The low dose was selected to reflect a human-relevant exposure level, based on the estimated daily intake of ΣOPE from indoor dust in the United States in 2018 (Kim et al. 2019). The highest dose was chosen to represent a level at which some effects would be expected, based on observed changes in the adrenals of male rats treated with the individual OPE, IPPP, at 70 mg/kg bw/d for 90 d (Wade et al. 2019). This dose also incorporates a “safety factor” to address the significant variability in OPE dust concentrations. For example, the median TDCIPP concentration in dust was 1,506 ng/g, with a range spanning from 181 to 2 140 000 ng/g (Castorina et al. 2017).
Animal treatments
Seventy-two-day-old male Sprague Dawley rats and 70-d-old female Sprague Dawley rats were obtained from Charles River (St-Constant, QC, Canada) and maintained on a 12-h light:12-h dark cycle at the McIntyre Animal Resources Centre (McGill University, Montreal, QC, Canada). Food and water were provided ad libitum. All animal studies followed the principles and procedures outlined in the Guide to the Care and Use of Experimental Animals prepared by the Canadian Council on Animal Care (McGill Animal Research Centre protocol 2018-7997). Rats were allowed to acclimatize for 1 wk before being randomly assigned (n = 15 per group) to either the control group or one of the three OPE mixture treatment groups (0.048, 1.6, or 48 mg/kg bw/d). All researchers were blinded to the treatment groups for the entirety of the study. Animals and their food were weighed once a week to determine food consumption. Male rats were exposed for 70 d, which is sufficient to complete a full cycle of spermatogenesis and epididymal sperm maturation. Female rats were initially exposed for 30 d, the time required for ovarian follicles to progress from the primordial to the preovulatory stage. Following this, females were mated with males from the same dosage group and continuously exposed to either control or OPE-containing diets during gestation and lactation, until the pups were weaned on postnatal day 21.
Tissue collection
Male rats (n = 10 per group) were euthanized under isoflurane anesthesia by exsanguination via cardiac puncture. Adrenal glands were collected, weighed, and snap-frozen in liquid nitrogen for further analysis. The remaining rats (n = 5 per group) were euthanized by fixation with Modified Davidson’s solution via retrograde perfusion through the abdominal aorta. Procedures were conducted following standard guidelines (Latendresse et al. 2002; Parasuraman et al. 2010). Fixed adrenal glands were then post-fixed for 24 h, dehydrated, and stored in a 4 °C cold room until further analysis. Female rats (n = 15 per group) were euthanized under isoflurane anesthesia by exsanguination via cardiac puncture. Adrenal glands were collected, weighed, and snap-frozen in liquid nitrogen (n = 10). The remaining adrenal glands (n = 5 per group) were fixed and stored in a 4 °C cold room until further analysis.
Whole blood was transferred to Vacutainer SST tubes (BD Biosciences Canada, Mississauga, ON), allowed to clot for 30 min at room temperature, and then centrifuged at 2,000 × g for 10 min at 4 °C for serum collection. For plasma collection, whole blood was transferred to Vacutainer EDTA tubes (BD Biosciences Canada, Mississauga, ON) and then centrifuged at 2,000 × g for 10 min at 4 °C. Sera and plasma were aliquoted and stored at −80 °C until further analysis.
Histology
Fixed adrenal glands were paraffin-embedded, serially sectioned at 5 µm thickness, and stained with hematoxylin and eosin. The stained slides were scanned using the AxioScan.Z1 microscope slide scanner (Zeiss Canada, Toronto, ON) and digitized. Digitized adrenal sections were examined blindly, considering key parameters such as adrenal hypertrophy, hemosiderin, lipid droplets, hyperplasia, apoptosis, and the width of blood sinusoids. Based on the severity of the examined parameters, a score from 0 to 5 was assigned, with 0 representing the absence of pathology. The scoring criteria are provided in Table S1. The thickness of total cortex and that of zona glomerulosa, zona fasciculata, and zona reticularis was measured using QuPath (v0.4.1; Bankhead et al. 2017). Specifically, the length parallel to the adrenal radius of these areas was measured in five different sections per adrenal and the average value was used in final calculation of zone thickness.
Serum biochemistry
Serum samples were sent to the Center for Phenogenomics (TCP, Toronto, ON, Canada) for a comprehensive cardiovascular panel analysis, which included total cholesterol, HDL cholesterol, LDL cholesterol, triglycerides, and glucose. The standard biochemistry panel was conducted at the Diagnostic and Research Support Service (DRSS) Laboratory of the Comparative Medicine Animal Research Centre, McGill University (Montreal, QC, Canada).
Hormone measurements
To assess the effects of the OPE mixture on the steroid hormone-producing function of the adrenal gland, enzyme-linked immunosorbent assay (ELISA) kits were used to measure the levels of aldosterone, corticosterone, adrenocorticotropic hormone (ACTH), and renin (Aldosterone: Abcam Catalog No. ab136933, RRID: AB_2895004; Corticosterone: Arbor assays Catalog No. K014-H1/H5, RRID: AB_2877626; ACTH: Abcam Catalog No. ab263880, RRID: AB_2910221; Renin: Invitrogen Catalog No. ERA52RB, RRID: AB_3661832) in accordance with the manufacturers’ instructions. The detection limit for the assay kits was: Aldosterone 4.7 pg/ml, corticosterone 0.0175 ng/ml, ACTH 6.0 pg/ml, and renin 30.0 pg/ml. The SpectraMax Plus 384 microplate reader (Molecular Devices, San Jose, CA, United States) was used to read the ELISA plates at a wavelength of 450 nm. Aldosterone and corticosterone levels have been adjusted for cross-reactivity. The intra-assay and inter-assay coefficients of variation for female samples were 4.5% and 14.2% for the aldosterone assay; 4.6% and 14.8% for the corticosterone assay, 4.9% and 15.0% for the ACTH assay, and 4.5% and 14.4% for the renin assay. For males, they were 3.4% and 13.8% for the aldosterone assay; 3.6% and 13.4% for the corticosterone assay; 4.9% and 14.8% for the ACTH assay, and 3.5% and 13.5% for the Renin assay.
RNA-seq and pathway analysis
Adrenal glands were homogenized, and total RNA was isolated using the RNeasy Plus Mini Kit (Qiagen, Mississauga, ON, Canada). The integrity of the extracted RNA was assessed using the RNA 6000 Nano Kit (Agilent Technologies, Mississauga, ON, Canada) on the 2100 Bioanalyzer System (Agilent Technologies, Mississauga, ON, Canada). Paired-end sequencing (150 bp) was conducted at Novogene Corporation (Sacramento, CA, United States) using their NovaSeq 6000 Sequencing System (Illumina, San Diego, CA, United States). RNA-seq data have been uploaded to GEO (number pending). Differentially expressed genes (DEGs) were identified using the DESeq2 R package (version 1.20.0) (P < 0.05). Transcripts differentially expressed by more than 1.5-fold were then further analyzed using Ingenuity Pathway Analysis (IPA 2024; Qiagen, Valencia, CA, United States) software.
Statistical analyses
Data were analyzed using GraphPad Prism (version 9.4.1, GraphPad Software Inc., La Jolla, CA, United States). One-way ANOVA followed by Dunnett’s test was used to compare adrenal gland weight, plasma biochemistry data, and hormone levels. Adrenal gland weight and hormone measurements were obtained from 10 animals per experimental group, with four groups for males and four groups for females. These included a control group and three groups exposed to different OPE doses for each sex. Plasma biochemistry and RNA sequencing analysis were done for 5 animals from each treatment group. For all experiments, the minimum level of significance was taken as P < 0.05.
Results
Effects of the OPE mixture on animal and adrenal weights
Weekly weights and food consumption were monitored throughout the treatment. No changes in body weight or food consumption were observed with OPE exposure in either the male or female treatment groups when compared with the control group (Fig. S1). As expected from the literature (Piao et al. 2013), the male adrenal glands weighed less than those of the female adrenal glands at the control level (Fig. 2). Starting with the lowest OPE treatment group, the male adrenal gland showed a reduction in weight (Fig. 2A). In contrast, the female adrenal glands showed a trend of increase in weight, with high OPE treatment group showing statistical significance (Fig. 2B). The distinct sex differences observed in adrenal gland weight in response to OPE exposure led us to further investigate whether the function of the adrenal gland was affected. Histological examination did not reveal any apparent changes in the adrenal gland (Fig. S2). Apart from the decrease observed in the female fasciculata zone, exposure to the mixture did not induce alterations in the width of other zones (Fig. S3).
Effects of the OPE mixture on the weight of (A) male and (B) female adrenal glands. *P < 0.05 and **P < 0.01 compared with control; values represent means ± SEM; n = 8 to 10.
Serum biomarkers are affected by the OPE mixture
Serum samples were collected at the time of euthanasia and analyzed for biomarkers associated with pathogenesis (Figs S4 and S5). OPE treatment did not result in any significant change in most biomarkers, indicating markers of liver or kidney function, muscle damage, and serum electrolyte levels. However, markers for lipid homeostasis were affected by the OPE mixture. Exposure to the high level of OPE mixture nearly doubled the levels of triglycerides and LDL cholesterol in male serum samples (Fig. 3B and D). The same exposure level also induced a trend toward increased HDL cholesterol concentration in males (P = 0.06) (Fig. 3E). For females, the only observed change was at the low exposure level, where a trend toward increased total cholesterol (P = 0.07) was noted (Fig. 3C).
Effects of the OPE mixture on the serum level of glucose (A), triglycerides (B), total cholesterol (C), LDL cholesterol (D), and HDL cholesterol (E) concentrations in male (green) and female (yellow) treatment groups. *P<0.05 and **P<0.01 compared with control; values represent means ± SEM; n = 5.
Effects of the OPE mixture on serum hormone levels
The average baseline level of serum aldosterone in males was 550.5 pg/ml (Fig. 4A), whereas that in females was much lower at 208.2 pg/ml (Fig. 4B). In the male adrenal gland, exposure to the OPE mixture resulted in a downregulation of aldosterone levels in all three treatment groups, with the resulting levels similar to the baseline level observed in females, around 200 pg/ml (Fig. 4A). No significant effect in aldosterone levels was observed in the female treatment groups (Fig. 4B). The average baseline level of renin is similar in males (510.8 pg/ml) and females (505.6 pg/ml) (Fig. 4C and D). In males, a concentration-dependent decrease in the level of renin was observed, with the high OPE group significantly reduced the level to 356.8 pg/ml (Fig. 4C). In females, however, exposure to OPEs resulted in a trend of increase in the level of renin (P = 0.08) (Fig. 4D).
Effects of the OPE mixture on serum hormone levels. Serum levels of aldosterone (A and B), renin (C and D), corticosterone (E and F) and ACTH (G and H) in male (green) and female (yellow) Sprague Dawley rats were measured. *P < 0.05 and ***P < 0.001 compared with control; values represent means±SEM; n = 5 to 10.
The baseline level of serum corticosterone in males (377.6 ng/ml) was approximately three times higher than in females (140.3 ng/ml) (Fig. 4E and F). In the high OPE treatment group, the level of corticosterone in males was reduced to half of the control level (Fig. 4E). No effect was observed in the females (Fig. 4F). ACTH regulates the level of corticosterone produced by the adrenal gland. Although OPE exposure did not significantly affect ACTH levels in either male or female rats, a similar trend was observed in ACTH levels compared with corticosterone levels (Fig. 4G and H).
Effects of the OPE mixture on signaling pathways in the adrenal gland
RNA sequencing was done to determine the transcriptomic changes induced by the OPE mixture in the adrenal gland. Principal component analysis (PCA) plot revealed the extensive degree of separation in the transcriptomic profile of male and female adrenal glands (Fig. 5A). OPE treatment resulted in significant alterations in the overall transcriptomic profiles of both male and female adrenal glands, as evidenced by the clear separation between control and the three OPE-treated groups. The heatmap analysis further revealed distinct expression patterns; at the control level, male and female adrenal glands exhibited remarkably different expression profiles (Fig. 5B). Compared with controls, OPE treatment-induced changes in gene expression, with different sets of targeted transcripts affected in male and female adrenal glands (Tables S2 and S3). This prompted further investigation into the specific transcripts and pathways impacted by the treatment.
Principal component analysis (PCA) and heatmap of the overall expression profile of the adrenal gland exposed to the OPE mixture. (A) PCA plot showing the degree of separation in the transcriptomic profiles between the male and female OPE treatment groups. (B) Heatmap showing the transcriptomic profiles of the adrenal glands. Red indicates upregulation, green indicates downregulation. n = 5.
For the male adrenal gland, a total of 674 DEGs were identified, with 381 transcripts downregulated and 293 being upregulated by at least 1.5-fold in the low OPE treatment group (Fig. 6A). Fewer transcripts were altered in total in the middle OPE treatment group, with a similar number of transcripts downregulated (230 transcripts) or upregulated (225 transcripts). In the high OPE treatment group, 480 transcripts were affected, with somewhat more transcripts downregulated (255 transcripts) than upregulated (225 transcripts). More transcripts were specifically targeted by each treatment group than those commonly affected across groups as shown in the Venn diagram (Fig. 6B). The top affected pathways in each of the uniquely affected gene sets for three treatment groups were identified using IPA. At the low OPE exposure level, pathways related to inflammation and adrenal gland development were affected (Fig. S6A). With increasing OPE concentrations, pathways related to cancer, WNT signaling, and neurotransmitter receptors were targeted (Fig. S6B and C). Among the transcripts commonly affected, 44 out of 52 could be identified with specific functions (Fig. 6C and Table S4). Nearly half of the commonly affected transcripts are associated with the immune pathway, the apoptosis pathway, or have functions related to the adrenal gland. The remaining transcripts are related to lipid metabolism, the cell cycle, cell–cell interactions, or serve as transporters or receptors.
Differentially expressed genes (DEGs) in male treatment groups. (A) Numbers of significantly downregulated or upregulated transcripts (fold changes >1.5 and P-values <0.05). (B) Venn diagram showing the numbers of unique DEGs specific to each treatment, as well as the DEGs shared among the treatments. (C) DEGs commonly affected in male treatment groups and their related functions. Forty-four out of 52 were identified with known functions. n = 5.
Analysis of the DEGs in male adrenal glands using IPA revealed the top affected canonical pathways in all three treatment groups (Fig. 7A–C). There were insufficient data in the Ingenuity Knowledge Base to predict whether several pathways, such as leukotriene biosynthesis, prostanoid biosynthesis, and cysteine biosynthesis/homocysteine degradation, are activated or inhibited. Overall, the majority of the affected pathways were predicted to be inhibited. The only commonly impacted pathway across all three treatment groups was potassium channels. Based on known activity patterns in the Ingenuity Knowledge Base, this pathway was predicted to be activated at all three OPE exposure levels. Potassium channels play a role in the biosynthesis of aldosterone and corticosterone. When activated, as shown in Fig. 7D and Table S6, and using expression changes in low OPE treatment group as an example, more potassium will be exported out of the cell, leading to a decrease in the downstream production of corticosterone (in fasciculata cells) or aldosterone (in glomerulosa cells).
Top 10 canonical pathways in (A) low, (B) middle, and (C) high OPE group identified by the Ingenuity Pathway Analyses (IPA). Orange indicates pathway activation; blue indicates pathway inhibition; white indicates that insufficient DEGs (<4) were associated with the pathway; grey indicates unknown direction of change of pathway activity. (D) Pathway diagram showing the commonly affected pathway(s). The expression changes represent transcripts affected in the low OPE treatment group. Red indicates upregulation of the expression level of the transcript, blue indicates downregulation. n = 5. Created with BioRender.com.
The number of DEGs in the female adrenal gland indicated that more transcripts were affected in the low and high OPE treatment groups comparing to the control (Fig. 8A). In the low OPE group, 729 transcripts were affected, with 383 downregulated and 346 upregulated. The high OPE treatment group affected the greatest number of transcripts (803 in total); downregulated transcripts again accounted for a smaller number (358), whereas the remaining transcripts were upregulated (445). The Venn diagram showed a similar pattern to that observed in the male adrenal gland: There were more transcripts individually affected by each treatment group compared with those commonly affected across groups (Fig. 8B). The low OPE treatment group affected pathways related to adrenal cell movement, steroid hormone receptors, and inhibited ferroptosis (Fig. S7A). In the middle OPE treatment group, the WNT signaling pathway and pathways related to adrenal development and differentiation were targeted (Fig. S7B). The highest exposure condition targeted cancer-related pathways, autophagy, and immune pathways (Fig. S7C). Analysis of the commonly affected transcripts revealed that those involved in steroidogenesis were only affected in the female treatment groups (Fig. 8C and Table S5).
Differentially expressed genes (DEGs) in female treatment groups. (A) Numbers of significantly downregulated or upregulated transcripts (fold changes >1.5 and P-values <0.05). (B) Venn diagram showing the numbers of unique DEGs specific to each treatment, as well as the DEGs shared among the treatments. (C) DEGs commonly affected in female treatment groups and their related functions. Sixty-two out of 68 transcripts were identified with known functions. n = 5.
There are more overlaps in the top affected pathways across the three treatment groups in female adrenal glands. Analysis using IPA revealed an enrichment for canonical pathways related to cholesterol biosynthesis and the immune system (Fig. 9A–C). Cholesterol biosynthesis was predicted to be strongly activated in all treatment groups. The LXR/RXR pathway, an important regulator of cholesterol homeostasis, was predicted to be inhibited at low OPE group but activated at the middle and high exposure levels. The predicted activation of the SREBF pathway will, in turn, promote the transcription of transcripts involved in cholesterol biosynthesis. Two immune-related pathways, interferon alpha/beta signaling and the role of hypercytokinemia/hyperchemokinemia in the pathogenesis of influenza, were predicted to be inhibited. The commonly affected pathways and the expression pattern of affected transcripts in the low OPE treatment group are shown in Fig. 9D and Table S6.
Top 10 canonical pathways in (A) low, (B) middle, and (C) high OPE group identified by the Ingenuity Pathway Analyses (IPA). Orange indicates pathway activation; blue indicates pathway inhibition; white indicates that insufficient DEGs (<4) were associated with the pathway; grey indicates unknown direction of change of pathway activity. (D) Pathway diagram showing the commonly affected pathway(s). The expression changes represent transcripts affected in the low OPE treatment group. Red indicates upregulation of the expression level of the transcript, blue indicates downregulation. n = 5. Created with BioRender.com.
Discussion
The adrenal glands produce hormones that regulate diverse essential functions, ranging from stress responses to metabolism, blood pressure, and circadian rhythms. Our study reveals that exposure to an environmentally relevant mixture of OPEs had sex-specific effects on the weight and function of the adrenal glands in Sprague Dawley rats. To our knowledge, this is the first time a household dust-based OPE mixture that closely resembles human exposure has been characterized for its effects on the adrenal gland. Importantly, this is also one of the few times that both male and female adrenal glands have been assessed for the toxicological impact of endocrine-disrupting chemicals.
The first indication of an effect on the adrenal gland is a change in organ weight. Exposure to individual OPEs has been shown to affect the weight of the adrenal gland. In a 2-yr feeding study, male and female F344/N rats were exposed to TMPP at various doses. At 3 mo, the adrenal glands of female rats receiving 15 mg/kg of TMPP were reported to be significantly heavier (National Toxicology Program 1994). The same study also utilized the B6C3F1 mouse model, and at 15 mo, the adrenal glands of female mice receiving 27 mg/kg TMPP were significantly heavier, whereas the adrenal glands of male mice receiving 37 mg/kg of TMPP were lighter (National Toxicology Program 1994). This trend is similar to what we observed with Sprague Dawley rats exposed to the OPE mixture, where the adrenal glands of females increased in weight, whereas those of males decreased (Fig. 2). In studies assessing short-term exposure effects, female F344 rats exposed to 0.4 g/kg TMPP for 40 d showed heavier adrenal glands compared with controls (Latendresse et al. 1993). However, the effects of other OPEs have varied. Exposure to IPPP induced a dose-dependent increase in adrenal gland weights in both sexes of Wistar rats (Wade et al. 2019). Although only male Wistar rats were assessed, exposure to TDCIPP also resulted in an increase in the weight of the adrenal gland (Akimoto et al. 2022). The variability in observed effects depending on the OPEs assessed highlights the importance of evaluating mixtures that reflect human exposure patterns.
Lipid homeostasis appears to be a sensitive molecular event targeted by OPE exposures. Previously, we showed that the same OPE mixture induced lipid accumulation and altered the composition of lipid droplets in female H295R adrenal cells after a 48-h exposure period (Li et al. 2024). Among the affected lipids, triglycerides were significantly altered. Serum lipid levels provide insight into the net effect of OPEs on lipids. In our study, serum lipid levels in male rats were notably more affected than in females. Specifically, triglycerides and LDL cholesterol levels increased in the highest exposure group. In contrast, females exhibited a trend of increased total cholesterol level only (Fig. 3). Alterations in serum lipid levels have been reported in other models. For example, exposure to TPHP has been associated with elevated triglyceride and total cholesterol levels in adult zebrafish (Du et al. 2016). Similarly, exposure to IPPP resulted in increased total and HDL cholesterol levels in both sexes of Wistar rats (Wade et al. 2019). Additionally, the metabolite of TDCIPP was positively associated with increased total cholesterol and LDL cholesterol levels in Canadian women (Siddique et al. 2020). Indeed, higher levels of OPEs have been associated with increased risk of cardiovascular disease, with TPHP proposed as a major driver (Guo et al. 2022; Zhang et al. 2024). In the rat adrenal gland, no significant change in lipid droplets was observed. It is possible that accumulation of lipid droplets serves as an acute response of the cells exposed to the OPE mixture in vitro, but that at the tissue level, it was brought back to homeostatic state over longer periods of time.
The adrenal cortex, responsible for producing aldosterone and corticosterone, serves as the final effector organ in the hypothalamic–pituitary–adrenal (HPA) axis. In our study, we observed that exposure to the OPE mixture significantly reduced both aldosterone and corticosterone levels in male rats, with aldosterone level being the more significantly affected (Fig. 4). It is not surprising that aldosterone levels were more affected compared with corticosterone levels, as the glomerulosa zone, which produces aldosterone, is the outermost layer of the adrenal gland and is thus the first to be impacted by chemical exposures (Rosol et al. 2001). The adrenal cortex undergoes continual renewal throughout life, and the function of the female adrenal cortex in steroid hormone production might remain less affected due to its naturally faster turnover rate (Greep and Deane 1949). Notably, corticosterone also influences bone metabolism by inducing apoptosis in osteoblasts and osteocytes while prolonging the lifespan of osteoclasts (reviewed by Mitra 2011). The same OPE mixture was found to suppress endochondral ossification in murine limb bud cultures (Yan and Hales 2021). Therefore, alterations in corticosterone levels may contribute to the adverse effects on bone formation. Previously, we found that exposure to the same OPE mixture increased aldosterone and cortisol levels in adrenal H295R cells derived from female patients by disrupting the expression of enzymes in the steroidogenic pathway (Li et al. 2024). Limited human data are available on the effects of OPE exposures on aldosterone and cortisol levels. One study indicates that increased urinary levels of di-butyl phosphate (a metabolite of TPHP), bis(1-chloro-2-propyl) phosphate (a metabolite of TDCIPP), and bis(2-butoxyethyl) phosphate (a metabolite of TBOEP) are associated with increased serum cortisol levels (Ji et al. 2021).
To assess whether the observed changes in aldosterone and corticosterone levels reflected alterations in hormones upstream in the pathway, we evaluated the levels of renin and ACTH. Renin, released by the kidneys in response to low blood pressure, converts angiotensinogen into angiotensin I, which is then transformed into angiotensin II. Angiotensin II stimulates the adrenal glands to release aldosterone, promoting sodium reabsorption and increasing blood volume and pressure (Lavoie and Sigmund 2003). The decrease in the level of renin produced by the kidneys suggests that it may partly contribute to the decreased aldosterone levels observed in males. This also indicates that kidney might be one of the targets of OPEs. It has been reported that urinary metabolites of OPEs in humans are positively associated with chronic kidney disease (Kang et al. 2019). ACTH is a peptide hormone synthesized and secreted by the anterior pituitary gland in response to corticotropin-releasing hormone from the hypothalamus. ACTH binds to melanocortin 2 receptors on the adrenal cortex, leading to the production of corticosterone (Smith and Vale 2006). Although the levels of ACTH were not significantly affected in either male or female rats, we did notice that the trend in the level of ACTH after exposure to the OPE mixture was similar to the effect seen in corticosterone levels, suggesting that OPEs may affect the HPA axis both locally at the adrenal glands and at the pituitary level. Further investigations are needed to examine the effects of OPEs on the renin–angiotensin–aldosterone system and the HPA axis to elucidate the precise mechanism(s) by which OPEs operate.
The RNA sequencing analysis revealed the potential mechanisms of action of the OPE mixture. First, we observed that the overall transcriptomic profiles of the female and male adrenal glands differ at the control level (Fig. 5). Differences in adrenal transcriptomic profiles between sexes have been documented previously. For instance, in a gonadectomy study, orchiectomized rats with testosterone replacement exhibited stimulation of transcripts associated with lipid and cholesterol metabolism, whereas ovariectomized rats with estradiol replacement showed inhibition of transcripts involved in intracellular signaling pathways (Jopek et al. 2017). With OPE exposure, the pathways disrupted in male adrenal glands exhibit a distinctly different profile compared with those in females. To elucidate the mechanisms of action of OPEs, we first investigated the commonly affected pathways in males across treatment groups. Potassium channels emerged as the only consistently affected pathway (Fig. 7). Notably, circulating levels of K+ are major regulators of aldosterone and, to a lesser extent, corticosterone synthesis. Alterations in extracellular K+ concentrations lead to significant changes in the membrane potential of zona glomerulosa and zona fasciculata cells, which are controlled by multiple potassium channels (Guagliardo et al. 2012; Bandulik et al. 2015, 2019). In our dataset, the expression of several potassium channels was upregulated, potentially causing depolarization of the cell membrane and resulting in decreased levels of aldosterone and corticosterone produced. This finding may partly explain the reduced levels of aldosterone and corticosterone observed in male rats.
In female adrenal glands, a broader range of pathways were significantly affected. Key pathways impacted include cholesterol biosynthesis, lipid regulation, and the immune system (Fig. 9). Cholesterol biosynthesis has been identified as a common target of OPEs. Previous studies have demonstrated that exposure to either the OPE mixture (Li et al. 2024) or individual OPEs (Li et al. 2023) results in the accumulation of lipid droplets in H295R adrenal cells. Analysis of these droplets revealed that sterol lipids were among the most significantly upregulated lipid categories. Correspondingly, the transcript for the enzyme responsible for the rate-limiting step in cholesterol biosynthesis, Hmgcr, was also upregulated (Li et al. 2024). Additionally, activation of the SREBF pathway led to the upregulation of transcripts involved in cholesterol biosynthesis, contributing to elevated cholesterol levels. Lipid homeostasis has been consistently identified as one of the most perturbed pathways affected by OPEs across various in vivo and in vitro studies, including those involving ovarian cells (Wang et al. 2023), hepatocytes (Shen et al. 2019; Wang et al. 2020; Yu et al. 2024), macrophages (Giles et al. 2024), murine limb buds (Yan and Hales 2021), and zebrafish brains (Yan et al. 2022). Lipid homeostasis is also regulated by the LXR pathway. Our findings align with previous observations, which reported pathway inhibition following exposure to the same OPE mixture in murine limb buds (Yan and Hales 2021), and to TPHP and 2-ethylhexyl diphenyl phosphate (EHDPHP) in macrophages, where these exposures induced a foam cell phenotype by inhibiting cholesterol efflux (Hu et al. 2019). At higher exposure levels, the observed pathway activation might be due to feedback mechanisms, where increased levels of cholesterol stimulate the activation of the pathway to facilitate cholesterol export. Furthermore, pathways related to immune functions in the female adrenal glands were generally found to be inhibited. Evidence suggests that organophosphate pesticides have immunosuppressive effects (Mitra et al. 2019). Several OPEs used as flame retardants and plasticizers have also been positively correlated with the prevalence of immune-related diseases in Japanese households (Araki et al. 2014). The same mixture as the one used in this study also altered the expression of phagocytic receptors in macrophage cells, which are key players in the immune response (Giles et al. 2024). Additionally, exposure to components of this mixture, such as tris(2-butoxyethyl) phosphate (TBOEP), tris(1-chloro-2-propyl) phosphate (TCIPP), and tris(2-chloroethyl) phosphate (TCEP), has been reported to alter the expression of genes related to immune functions in the HepG2 human hepatocellular carcinoma cell line (Krivoshiev et al. 2018a, 2018b).
The inclusion of three doses allowed us to compare exposure effects at different levels. For both male and female rats, the low dose affected a greater number of transcripts (Figs 6A and 8A). In fact, in low dose treatment group, the affected transcripts had consistent changes in the direction of their expression level, making a stronger prediction of the affected pathways (Figs 7A and 9A). It is also noteworthy that higher concentrations of these chemicals were associated with the activation of additional pathways related to cancer development (Figs S6C and S7C), suggesting an elevated risk associated with higher exposure levels.
Our study provides novel evidence that exposure to a household dust-based OPE mixture adversely affects both lipid homeostasis and adrenal gland function. Transcriptomic analysis further revealed that these effects occur through sex-specific mechanisms: In males, the primary affected pathway involves potassium channels, whereas in females, disruptions in lipid homeostasis and immune functions are most affected. However, we acknowledge that human exposure patterns vary depending on geographical location, individual behaviors, and socio-economic factors. The exposure profile tested here may not be fully representative of the entire population. Additionally, since rodents lack the enzyme CYP17 and have different steroidogenic pathways (van Weerden et al. 1992), the translation of these findings into human effects must be done with caution.
Together, these findings highlight the importance of evaluating the toxicological impacts of environmentally relevant mixtures in both sexes. Furthermore, since every step in the steroid biosynthetic pathway, from the ACTH receptor and cholesterol transport to critical CYP and hydroxysteroid dehydrogenase enzymes involved in steroid synthesis, is known to be targeted by over 70 compounds (Harvey and Everett 2003; Harvey et al. 2007, 2009), including OPEs, the adrenal gland should be considered a crucial endpoint in toxicological assessments.
Acknowledgments
We thank Dr Michael G. Wade (Health Canada) for providing the OPE house dust mixture. We thank Dr Aimee Lee Katen, Dr Abishankari Rajkumar, Trang Luu, Dr Xiaotong Wang, and Dongwei Yu for their valuable assistance in animal monitoring, animal handling, and tissue collection. Serum biochemistry service was provided by the pathology core at the Centre for Phenogenomics.
Author contributions
Zixuan Li, Barbara F. Hales, and Bernard Robaire were responsible for the experimental design, data interpretation, and manuscript preparation. Zixuan Li was responsible for data acquisition and analyses. All authors approved the final version of the article.
Supplementary material
Supplementary material is available at Toxicological Sciences online.
Funding
Canadian Institutes of Health Research (CIHR), Institute for Population and Public Health Team Grant (FRN IP3-150711), Canadian Institutes of Health Research (CIHR) Project Grant (FRN 156239), and McGill University. Z.L. is the recipient of training awards from McGill University and the Centre for Research in Reproduction and Development (CRRD). B.F.H. and B.R. are James McGill Professors.
Conflicts of interest
None declared.
Data availability
None declared.
Comments