Gestational Exposure to Bisphenol A and Bisphenol S Leads to Fetal Skeletal Muscle Hypertrophy Independent of Sex

Abstract

Gestational exposure to bisphenol A (BPA) can lead to offspring insulin resistance. However, despite the role that the skeletal muscle plays in glucose homeostasis, it remains unknown whether gestational exposure to BPA, or its analog bisphenol S (BPS), impairs skeletal muscle development. We hypothesized that gestational exposure to BPA or BPS will impair fetal muscle development and lead to muscle-specific insulin resistance. To test this, pregnant sheep (n = 7–8/group) were exposed to BPA or BPS from gestational day (GD) 30 to 100. At GD120, fetal skeletal muscle was harvested to evaluate fiber size, fiber type, and gene and protein expression related to myogenesis, fiber size, fiber type, and inflammation. Fetal primary myoblasts were isolated to evaluate proliferation and differentiation. In fetal skeletal muscle, myofibers were larger in BPA and BPS groups in both females and males. BPA females had higher MYH1 (reflective of type-IIX fast glycolytic fibers), whereas BPS females had higher MYH2 and MYH7, and higher myogenic regulatory factors (Myf5, MyoG, MyoD, and MRF4) mRNA expression. No differences were observed in males. Myoblast proliferation was not altered in gestationally BPA- or BPS-exposed myoblasts, but upon differentiation, area and diameter of myotubes were larger independent of sex. Females had larger myofibers and myotubes than males in all treatment groups. In conclusion, gestational exposure to BPA or BPS does not result in insulin resistance in fetal myoblasts but leads to fetal fiber hypertrophy in skeletal muscle independent of sex and alters fiber type distribution in a sex-specific manner.

bisphenols, fetal, skeletal muscle, myogenesis

Bisphenols are endocrine-disrupting chemicals (EDCs) that can interfere with endocrine signaling pathways, causing metabolic disruptions (Braun, 2017). Bisphenol A (BPA) is an ubiquitous chemical used to manufacture industrial and consumer products (Huang et al., 2018). Despite being banned in certain consumer products, BPA continues to be detected in running water, soil, air, and in human body fluids (Murata and Kang, 2018). Other bisphenols, such as bisphenol S (BPS), are also used in industrial applications such as in anticorrosive agents, canned foodstuffs, and thermal paper (Chen et al., 2016). Bisphenol A and BPS can also be detected in umbilical cord and breast milk (Chen et al., 2016), highlighting the risk of early life exposure to these compounds and the potential long-term effects to the progeny.

Previous studies have demonstrated that gestational exposure to BPA can lead to metabolic alterations in the offspring, including oxidative stress, insulin resistance, and glucose intolerance (Galyon et al., 2017; Nadal et al., 2017; Veiga-Lopez et al., 2015), as well as liver and adipose tissue-specific insulin resistance (Le Magueresse-Battistoni et al., 2018). In rodents, postnatal BPA exposure can affect skeletal muscle by interfering with the insulin receptor downstream pathway (Moon et al., 2015), upregulating IRS-1 and impairing MAPK signaling (Batista et al., 2012), or decreasing glucose transporter 4 translocation (Mullainadhan et al., 2017). In vitro, BPA exposure can suppress myogenic differentiation through the inhibition of Akt signaling (Go et al., 2018). Despite growing evidence that BPA can impair insulin homeostasis in vivo (Alonso-Magdalena et al., 2010; Veiga-Lopez et al., 2016), and suppress myogenic differentiation in vitro (Go et al., 2018), the effect of gestational BPA exposure on skeletal muscle tissue development remains largely overlooked. Even less understood is the potential effect of other bisphenol chemicals, such as BPS.

In addition to its role in energy metabolism, skeletal muscle plays a role in posture support, locomotion, and breathing (Iizuka et al., 2014). Many of these functions are not only associated with the size and type of skeletal muscle myofibers, insulin sensitivity, and metabolic rates (Haizlip et al., 2015; Schiaffino and Reggiani, 2011), but also with mitochondrial densities. For instance, among fiber types, type-I fibers (slow oxidative) have the highest, whereas type-IIX fibers (fast glycolytic) have the lowest insulin sensitivity (Schiaffino and Reggiani, 2011). Myogenesis, the development and formation of skeletal muscle fibers, begins during fetal life (Du et al., 2010), with the number of muscle fibers being established before birth in sheep and cattle (Du et al., 2010; Zhu et al., 2004) and also in some muscles in humans (Li et al., 2015). Therefore, we have investigated if gestational exposure to BPA and/or BPS alters size and type of muscle fibers and myogenic gene expression in the skeletal muscle of fetal sheep, a large precocial animal model.

Myogenic differentiation is regulated by 4 myogenic regulatory factors (MRFs) and its dysregulation can result in muscle hypertrophy or atrophy (Schiaffino et al., 2013), which can be programmed during fetal life (Du et al., 2010) by several factors, including chemical exposures (D’Amico et al., 2014). In vitro, BPA exposure can suppress myogenic differentiation (Go et al., 2018), one of the last 3 phases of myogenesis (Zammit, 2017); but whether BPA leads to impaired myogenic differentiation in vivo has not been evaluated. Therefore, we also investigated if gestational exposure to BPA and/or BPS will reduce proliferation and differentiation of fetal myoblasts. In addition, we have evaluated if prenatal BPA-induced insulin resistance (Veiga-Lopez et al., 2016) is established during fetal life by evaluating responsiveness of fetal myoblasts after gestational BPA exposure.

MATERIALS AND METHODS

Animal experimentation

This study was approved by the Michigan State University (MSU) Institutional Animal Care and Use Committee and is consistent with the National Research Council’s Guide for the Care and Use of Laboratory Animals and the ARRIVE guidelines for reporting animal research (Kilkenny et al., 2012), and conducted at the MSU Sheep Research Facility (East Lansing, Michigan; 42.7°N, 84.4°W). A time-mated pregnancy strategy was used to breed 23 healthy female Polypay × Dorsett sheep. To minimize paternal variability, a single fertile ram was used for mating. After mating, females were group-housed, randomly assigned to a treatment group, and blocked by body condition score and weight as previously described (Gingrich et al., 2018; Pu et al., 2017). Animal husbandry and diet details have been previously described (Pu et al., 2017).

Gestational BPA and BPS exposure consisted of daily subcutaneous injections of BPA (0.5 mg/kg/day; purity: ≥99%; Cat #: 239658, Lot# MKBQ5209V; Sigma-Aldrich, St. Louis, Missouri) or BPS (0.5 mg/kg/day; purity: 99.7%; Cat #: 146915000, Lot# A0337011; Acros Organics, Geel, Belgium) in corn oil from days 30 to 100 of gestation (term: ∼ 147 days). Control (C) mothers received only corn oil (vehicle). Internal doses of BPA and BPS achieved in umbilical arterial samples using the 0.5 mg/kg/day dose are targeted to achieve umbilical cord levels of ∼ 2.6 and 7.7 ng/ml, respectively (Gingrich et al., 2018; Veiga-Lopez et al., 2013), approaching the median concentration of BPA measured in maternal circulation of US women (Padmanabhan et al., 2008). Numbers of breeders used were 8, 8, and 7, in C, BPA, and BPS groups, respectively. Breeding resulted in 6 C, 6 BPA, and 4 BPS female fetuses, and 4 C, 5 BPA, and 4 BPS male fetuses. All pregnancies (n = 23) but 6, were singletons. Twinning was included in the statistical analysis as a covariate. After a washout period of 20 days, pregnancy was terminated at day 120 of gestation using a barbiturate overdose. A midline incision was performed, the uterus exposed, and the fetus quickly removed. The fetal skeletal muscle from the hind limb vastus lateralis was immediately harvested. Tissues were fixed in a 10% neutral buffered formalin solution for histological processing, flash frozen for gene expression studies, or freshly collected for myoblast isolation.

Histology

Formalin fixed fetal skeletal muscle tissue (n = 4–6 fetuses/group/sex) was paraffin embedded and hematoxylin and eosin stained as previously described (Gingrich et al., 2018). A total of 10 non-overlapping images were captured using a bright field microscope (Leica DMLB microscope with a Leica DFC480 camera [Hesse, Wetzlar, Germany] from each fetus). Image J software was used to evaluate myofiber size (>500 myofibers were quantified per fetus).

Isolation of primary skeletal muscle cells

Primary skeletal muscle cultured cell lines (n = 4–6 fetuses/group/sex) were derived from fresh fetal ovine skeletal muscle tissue and used for proliferation, differentiation, and insulin responsiveness experiments. For these experiments, n = 3–4 fetuses/group/sex were randomly selected out of the available 4–6 fetuses/group/sex). Tissue was first rinsed in warm Dulbecco’s Phosphate-Buffered Saline containing 1% antibiotic-antimycotic. After removing visible blood clots and connective tissue, skeletal muscle was minced with fine scissors, damped with fetal bovine serum (FBS; Fisher Scientific, Pittsburgh, Pennsylvania), and layered onto the bottom of 100 mm culture plates. To enhance tissue adhesion, culture plates were inverted and incubated for 4–6 h at 100% humidity. Thereafter, plates were turned upwards and growth medium (Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 [DMEM/F12], Invitrogen) supplemented with 1% penicillin/streptomycin, 1% l-glutamine, 10 mM HEPES, and 10% FBS. Cells were then harvested and frozen in liquid nitrogen until further use.

Purification of myoblasts

Myoblasts were purified using a 5-step serial plating method (Park et al., 2005) modified as follows. Skeletal muscle cells were seeded into 100 mm cell culture plates and cultured in growth medium for 1–2 days until 90% confluency (Supplementary Figure 1). Single cell suspensions were obtained after cell digestion using 0.25% trypsin, centrifuged at 1250 rpm for 5 min at RT, and resuspended in growth medium. Cells were then seeded into 60 mm cell culture plates pre-coated with 0.2% gelatin and incubated for 10 min at 39°C. Thereafter, unattached cells were plated into new gelatin-coated plates and incubated for 10 min at 39°C. After 5 serial platings, unattached cells were cultured for 1–3 days until confluency. To assess the purity of myoblasts, cells were fixed with 10% formalin and immunostained with the myoblast marker desmin (see Immunocytochemistry section) (Supplementary Figure 1).

Proliferation of myoblasts

Fetal myoblast (n = 3 fetuses/ group/sex; randomly selected out of the available 4–6 fetuses/group/sex) proliferation was determined using a 5-ethynyl-2’-deoxyuridine (EdU) cell proliferation assay kit (EdU-488; 17-10525, Millipore, Sigma-Aldrich, St. Louis, Missouri). Myoblasts were cultured in growth medium in a 96-well plate at a concentration of 10 000 cells per well. After overnight culture, the medium was replaced with 10 µM EdU solution in growth medium for 3 h at 39°C and 5% CO₂. Cells were then fixed and stained with DAPI (1:1000). Six non-overlapping images were captured from each well using a fluorescence microscope (Eclipse Ti, Nikon, Japan). Proliferation rate was defined as the percentage of EdU-positive nuclei to total DAPI-stained nuclei.

Standardization of myoblast differentiation

To optimize the ovine fetal myoblast differentiation system, the effect of horse serum supplementation at 1, 2, 5, or 7% (Cat#: 16050-122, Invitrogen, Paisley, UK) was tested. The highest myotube differentiation was achieved with 7% horse serum (Supplementary Figure 2). In brief, 12-well plates were pre-coated with Matrigel (1 mg/ml) for 1 h before use and kept at 39°C and 5% CO₂. About 50 000 cells were seeded in growth medium until ∼80% confluency. Thereafter, cells (n = 4 fetuses/group/sex; randomly selected out of the available 4–6 fetuses/group/sex) were induced to differentiate in Matrigel-coated plates with differentiation medium-containing DMEM (low glucose), 1% penicillin/streptomycin, 10 mM HEPES, and 7% horse serum. Culture medium was replaced every 2 days. After 6 days of differentiation, cells were fixed with methanol and acetone (1:1, v/v) for 20 min at −20°C, and immunostained with sarcomeric myosin to evaluate myofiber formation (see Immunocytochemistry section). About 15–20 images were randomly captured from each well using a fluorescent microscope (TE2000-U, Nikon, Tokyo, Japan). Myotube area and thickness and the number of nuclei per myotube area were quantified in sarcomeric myosin-positive myotubes. At least 60 differentiated myotubes for each cultured cell line were evaluated.

Immunocytochemistry

After fixation, cells were washed with PBS 3 times and incubated in blocking buffer (5% bovine serum albumin (BSA) diluted in PBS) for 30 min at room temperature. Cells were then incubated with the primary antibody (Supplementary Table 1) overnight at 4°C. After rinsing with PBS, cells were incubated with the secondary antibody (Supplementary Table 1) for 1 h at room temperature. After 3 additional PBS washes, cells were stained with DAPI (1:1000) for 5 min.

Gene expression

Quantitative real-time PCR (qRT-PCR) was used to evaluate mRNA expression as previously published (Pu et al., 2017). Total RNA (500 ng) was extracted and reverse transcribed into cDNA using a High Capacity cDNA Reverse Transcription Kit (Promega, Madison, Wisconsin) and quantified on an ABI-QuantStudioTM 7 Flex Real-Time PCR System. Primer sequences are provided in Supplementary Table 2. Due to the high variability of the skeletal muscle tissue (Niu et al., 2016), the relative expression of genes was calculated using 2^-ΔΔCT method and normalized against the geometric mean of ribosomal protein S15 (RPS15) and ribosomal protein large P0 (RPLP0) for cell gene expression, although the geometric mean of beta-2-microglobulin (B2M) and apoptosis inhibitor 5 (API5) or B2M and 60S ribosomal protein L27 (RPL27) was used for tissue gene expression. Each qRT-PCR reaction for each sample and primer set were run in triplicate, and the tripartite average of the threshold cycle (Ct) values was used for statistical analyses. Data are presented as the relative fold change to the expression in the female control groups.

Myoblast response to insulin stimulation

Myoblasts were cultured in 6-well plates at a concentration of 50 000 cells/well in growth medium until cells reached ∼95% confluency. To test the activation of the insulin receptor downstream pathway, cells were incubated with insulin (100 nM) for 1 h and protein expression of AKT and p-AKT measured using western blotting.

Western blotting

Protein was extracted from skeletal muscle tissue or cells as previously described (Pu et al., 2019). In brief, after protein concentration determination, samples were run on an SDS-PAGE gel (for skeletal muscle tissue: 12% in the bottom half and 6% in the top half gel; for myoblasts: 10% gel). Protein was transferred from the gel to nitrocellulose membranes and blocked with 5% non-fat milk or 5% BSA (for p-AKT), incubated with primary antibodies (Supplementary Table 1) overnight at 4°C, then with horseradish peroxidase-conjugated secondary antibodies (Supplementary Table 1) for 1 h at 37°C in the dark. Membranes were subjected to enhanced chemiluminescence and developed using a Thermo MyECL imager. Quantification of band intensities performed using ImageJ software and normalized to the target band of the reference protein cyclophilin B.

Statistical analyses

All data are presented as mean ± SEM. Appropriate transformations were applied, as needed, to account for normality of data. To evaluate significant differences among treatment groups and sex, a general mixed model with Tukey post hoc tests was used. Twinning was accounted for in the statistical model as a covariate. t-Test was used to compare gene expression differences between differentiation day (day 0 vs day 6) within sex and treatment group. To assess differences in fetal myofiber and myotube size upon gestational bisphenol exposure an empirical cumulative distribution function was calculated for each measurement and the difference among groups tested using a Kolmogorov-Smirnov test. Statistical software used was PASW Statistics for Windows release 18.0.1. Differences among cell perimeter distributions were evaluated using a permutation test with 10 000 iterations using R Statistical Computing release 3.3.0. Differences were considered significant at p < .05.

RESULTS

Gestational BPA and BPS Effects on Skeletal Muscle Tissue

To evaluate if gestational bisphenol exposure altered skeletal muscle fiber size, cross-sectional fiber area was analyzed. Permutation analyses demonstrated that fiber size was larger in BPA female and male fetuses (p < .05) and tended to be larger in BPS fetuses (p = .08) when compared with their control counterparts (Figure 1). Mean myofiber cross-sectional area within sex was larger in BPA and BPS fetuses compared with the control group. In all groups, mean fiber size was larger in females compared with males (p < .05).

Figure 1.

Effect of gestational BPA and BPS exposure on myofiber size. Representative images of hematoxylin and eosin stained skeletal muscle cross-sections in control (top), BPA- (middle), and BPS- (bottom) exposed females (left) and males (right) at gestational day 120. Myofiber area (µm²) frequency (Freq.) distribution plots in females and males of control (black), BPA-exposed (red), and BPS-exposed (blue) fetuses. Histogram represent mean myofiber area in females and males of control (open bars), BPA- (gray bars), and BPS-exposed (closed bars) fetuses. n = 4–6/group/sex. Different letters denote differences among treatments within sex (a ≠ b denote p < .05). Horizontal bars with asterisk denote differences between sexes within treatment (p < .05). Abbreviations: BPA, Bisphenol A; BPS, bisphenol S.

We evaluated if gestational bisphenol exposure altered expression of MRFs. Bisphenol A-treated females had higher MYH1 compared with control females, whereas BPS-treated females had higher MYH2 and MYH7 mRNA expression (Figure 2A). Comparisons between sex and within group demonstrate that control and BPS-exposed females had higher MYH2 mRNA expression (Figure 2A). We then evaluated if gestational bisphenol exposure altered the expression of the different (fast and slow) muscle myosin heavy chain (MHC) proteins. Fast skeletal muscle MHC protein was similar among treatments and within sex but was higher in females versus males (Figs. 2B–C). Sex- and treatment-specific changes were observed in slow MHC (Figs. 2B–C). Bisphenol A-treated males had higher slow skeletal MHC protein expression (reflective of oxidative fibers) (Figure 2C) compared with BPS-treated males. Bisphenol S-treated females had higher slow MHC expression compared with their male counterparts (Figure 2C). Percent slow versus fast MHC was similar among treatment groups within sex (Figure 2D). However, males had higher slow MHC percent compared with females in the BPA (p < .005) and control (p = .07), but not BPS, groups (Figure 2D). When comparing percent slow versus fast MHC within sex and group, males had more slow versus fast MHC in control and BPA (p < .01), but not BPS, groups (Figure 2D).

Figure 2.

Effect of gestational BPA and BPS exposure on skeletal muscle fiber type. A, mRNA (mean ± SEM) expression of genes related to myofiber type (myosin heavy chains [MYH]: MYH1, MYH2, MYH4, and MYH7) in fetal skeletal muscle tissue in females (F) and males (M) of control (open bars), BPA- (gray bars), and BPS-exposed (closed bars) fetuses. B, Representative western blots for fast and slow MHC protein in the same groups as for (A). Cyclophilin B (Cyc B) is used as a reference protein. C and D, Protein expression (mean ± SEM) and percentage of MHC type for slow and fast MHC. n = 4–6/group/sex. Different letters denote differences among treatments within sex (a ≠ b denote p < .05; a′ ≠ b′ denote p = .07). Asterisks denote differences at p < .05. Abbreviations: BPA, Bisphenol A; BPS, bisphenol S; MHC, myosin heavy chain.

In skeletal muscle, BPS-treated females had higher MRFs (Myf5, MyoD, MyoG, and MRF4), inflammation-related genes (iNOS and NF-κB), and myoblast fusion promoter (MyD88) mRNA expression compared with control and BPA-treated females (Figure 3). mRNA expression of JUNB (fiber hypertrophy promoter) was higher in BPS treated females compared with control females (Figure 3). Bisphenol A or BPS exposure did not alter myogenic regulatory genes in males. When comparing skeletal muscle tissue between sex and within group, control males had higher expression of myogenic genes (Myf5 and MRF4), inflammatory genes (iNOS, NF-κB, and HMGB1), and myoblasts fusion promoter (MyD88) (Figure 3). In the BPA exposed group, Myf5 and HMGB1 expression were higher in males versus females (Figure 3). However, in the BPS-exposed group, MyoG and NF-κB were higher in females (Figure 3).

Figure 3.

Effect of gestational BPA and BPS exposure on mRNA expression (mean ± SEM) in fetal skeletal muscle tissue in females (F) and males (M) of control (open bars), BPA- (gray bars), and BPS-exposed (closed bars) fetuses. A, Genes related to myogenic differentiation (Myf5, MyoD, Myog, and MRF4). B, Genes related to inflammation (iNOS and NF-κB). C, Genes related to myotube shape and size (JUNB, MyD88, and HMGB1). n = 4–6/group/sex. Different letters denote differences among treatments within sex (a ≠ b denote p < .05). Asterisks denote differences at p < .05. Abbreviations: BPA, Bisphenol A; BPS, bisphenol S.

Gestational BPA and BPS on Fetal Myoblasts’ Proliferation and Differentiation

All primary fetal myoblast lines had similar proliferation ability (Supplementary Figure 3). To evaluate if prenatal BPA and BPS exposure altered myogenic differentiation, primary fetal myoblasts were induced to differentiate into myotubes. Area and thickness of myotubes were larger in BPA- and BPS-exposed groups compared with the control group in both females and males, except for myotube area in BPA females (Figs. 4A–C). The number of nuclei per area, reflective of myotube differentiation, was also higher in the BPA-exposed group, independent of sex, and tended to be higher in BPS-exposed males (p = .09) (Figure 4A). Bisphenol A-treated females had higher JUNB, FOXO3, and TBX15 mRNA expression in differentiated myotubes compared with BPA-treated males (Figure 4D). Females had larger myotubes compared with males for all treatment groups (Figure 4).

Figure 4.

Effect of gestational BPA and BPS exposure on fetal myoblasts differentiation. A, Mean myotube area (left), diameter (middle), and number of nuclei per area (right) in females and males of control (open bars), BPA- (gray bars), and BPS-exposed (closed bars) fetuses. B, Representative images of sarcomeric myosin stained differentiated myotubes (day 6) on females (top) and males (bottom) of control (left), BPA- (middle), and BPS- (right) gestationally exposed fetuses. C, Differentiated myotube area (µm²) distribution plots in females and males of control (black), BPA-exposed (red), and BPS-exposed (blue) fetuses. D, Effect of gestational BPA and BPS exposure on myotube shape and size genes (JUNB, FOXO3, and TBX15) mRNA expression (mean ± SEM) in differentiated myotubes in vitro on females (F) and males (M) of control (open bars), BPA- (gray bars), and BPS-exposed (closed bars) fetuses. n = 4/group/sex. Different letters denote differences among treatments within sex (a ≠ b denote p < .05). Asterisks denote differences at p < .05. Abbreviations: BPA, Bisphenol A; BPS, bisphenol S.

Gestational BPA and BPS on Myoblasts’ and Differentiated Myotubes’ Gene Expression and Insulin Responsiveness

We evaluated if the larger myotube size observed in bisphenol-exposed myoblasts was driven by changes in myogenic regulatory genes during differentiation (Figure 5). Only Myf5 was higher in BPA versus control females before differentiation (D0). Between sex and within group, BPS-exposed females had lower Myf5 in myoblasts before (D0) and after (D6) differentiation, whereas BPA-exposed females had higher MyoG expression in myoblasts only before (D0) differentiation. We evaluated if gestational bisphenol exposure led to cell-specific inflammation (Figure 5). Before differentiation (D0), males had higher iNOS and NF-κB compared with females. After differentiation (D6), males had higher iNOS in the control group, whereas NF-κB was higher in the female BPA-treated group compared with males. The insulin-stimulated increase in p-AKT:AKT was not different among groups (Figure 6).

Figure 5.

Effect of gestational BPA and BPS exposure on myogenic genes (Myf5, MyoD, MyoG, and, MRF4) and inflammatory genes (iNOS and NF-κB) mRNA expression (mean ± SEM) in before differentiation myoblasts (day 0; D0) and after differentiated myotubes (day 6; D6) in females and males of control (open bars), BPA- (gray bars), and BPS-exposed (closed bars) fetuses. n = 4/group/sex. Different letters denote differences among treatments within sex (a ≠ b denote p < .05). Asterisks denote differences at p < .05. Aside from NF-κB, all genes changed significantly (p < .05) within treatment group and sex between D0 and D6 (not noted in figure). Abbreviations: BPA, Bisphenol A; BPS, bisphenol S.

Figure 6.

Effect of gestational BPA and BPS exposure on insulin sensitivity. Relative protein expression (mean ± SEM) in myoblasts upon insulin stimulation (100 nM for 1 h) in vitro on females and males of control (open bars), BPA- (gray bars), and BPS-exposed (closed bars) fetuses. Top panel, Representative western blots for p-AKT and AKT protein. Cyclophilin B (Cyc B) used as a reference protein. Bottom panel, p-AKT to AKT ratio. Percents reflect increase in p-AKT: AKT ratio upon insulin stimulation within group. n = 3/group/sex. Abbreviations: BPA, Bisphenol A; BPS, bisphenol S.

DISCUSSION

Gestational exposure to EDCs can lead to long-term metabolic diseases in adulthood, including insulin resistance (Nadal et al., 2017; Veiga-Lopez et al., 2016). Despite skeletal muscle being a major insulin target, the programming effects of bisphenols on this tissue have been largely overlooked. We have demonstrated that gestational BPA and BPS exposure can alter fetal skeletal muscle development resulting in muscle fiber hypertrophy in both female and male fetuses. Additionally, differentiation of primary myoblasts derived from animals exposed in utero to BPA and BPS also resulted in larger myotubes. However, these developmental changes were not accompanied with a reduction in myoblast insulin responsiveness. These effects were observed in fetuses exposed to an environmentally relevant dose and persisted for 20 days after the exposure was discontinued. Changes in myogenic gene expression were bisphenol- and sex-specific, highlighting the intrinsic differences between both bisphenolic compounds.

Gestational BPA and BPS Alter Skeletal Muscle Fiber Size and Type

The larger mean fiber size in both BPA and BPS gestationally exposed fetuses demonstrates that bisphenols, similar to other prenatal insults, such as maternal nutrition and micronutrients can also alter fetal fiber growth and myofiber formation (Berard and Bee, 2010; Dwyer et al., 1994; Oster et al., 2017). We did not test if the fiber size effect observed after bisphenol exposure stems from ESR1 binding, however, postnatal estrogen supplementation has also been shown to increase fiber diameter in rats (Suzuki and Yamamuro, 1985). This coupled with the lower skeletal muscle regenerative ability in ESR1-null mice (LaBarge et al., 2014) and the decline in muscle strength in menopausal women (Sipila et al., 2001) provides with strong evidence regarding the role of ESR1 on skeletal muscle maintenance. Bisphenol A also has thyroid receptor binding activity, but thyroid hormone effects have been mostly related to changes in fiber type rather than size (Caiozzo and Haddad, 1996).

In females, we observed that gestational BPA increased MYH1 mRNA, reflective of higher expression of fast glycolytic fibers (MHC-IIx). Fast glycolytic fibers generate ATP primarily through glycolysis, have high myosin ATP activity, are less insulin sensitive (He et al., 2001), but are more susceptible to fatigue. However, despite an upregulation in MYH1 upon BPA exposure, this was not reflected in the upregulation of fast MHC (MHC-II) protein expression likely due to a lack of upregulation in both MYH1 and MYH4. Although fast glycolytic fibers have larger diameters (van Wessel et al., 2010), which we observed in all bisphenol-exposed fetuses, only BPA-exposed females had higher MHY1 expression. Female-specific changes in fiber type genes were also evident upon BPS exposure. These changes also include higher MYH7 and MYH2 mRNA expression, reflective of slow oxidative fibers, and have stronger resistance to fatigue compared with fast glycolytic fibers (I > IIA > IIX [Westerblad et al., 2010]). However, these findings were not reflected in protein abundance of slow MHC, similar to previous studies where changes in MYH mRNA expression do not always reflect a matching change MHC proteins (ie MYH7 to MHC-I, or MYH2 to MHC-II [Yates et al., 2016]).

We observed a sex-specific effect to BPA and BPS on fiber type gene and protein expression similar to previous work (Arambula et al., 2018; Pu et al., 2017; Thongkorn et al., 2019). Given the steroid binding affinity of bisphenols, differences in steroid receptor expression may underlie this effect (Pu et al., 2017). We also observed sex-specific differences in fiber distribution within treatment groups, with male fetal skeletal muscle having higher slow MHC expression in control (p = .07) and BPA exposed (p < .05) groups compared with females. Fiber type distribution is muscle- and species-specific (Eason et al., 2000; English et al., 1999; Staron et al., 2000). We also observed that males had slower versus fast MHC within the control group. These findings are similar to that reported in human vastus lateralis distribution in young adults (females: I > IIA > IIX; males: IIA > IIX > I [Staron et al., 2000]). However, as observed in studies of nutrient restriction during pregnancy, fetal shifts in fiber types may not persist into adulthood (Daniel et al., 2007; Zhu et al., 2004).

Because gestational exposure to BPA or BPS leads to differences in fiber type distribution, and these fibers exert different responses towards insulin, we tested if gestational bisphenol exposure induces different responses of primary myoblasts to insulin. Higher MYH1 expression in BPA-exposed females is consistent with previous work where pre-, peri-, or postnatal BPA exposure resulted in insulin resistance (Nadal et al., 2017). However, we did not observe differences in insulin responsiveness after prenatal bisphenol exposure. The insulin challenge was tested on myocytes rather than differentiated myotubes because recent studies have demonstrated that insulin resistance impacts early steps of myogenic specification (Iovino et al., 2016). However, it is possible that the insulin resistant phenotype presents in differentiated myotubes or that glucose transporter 4 translocation downstream of p-AKT is altered. It is also possible that the insulin resistant phenotype gets established later in life as demonstrated in other prenatal exposure models (Cardoso et al., 2016).

Gestational BPA and BPS on Myogenic Proliferation, Differentiation, and Inflammation

During fetal life, proliferation of myogenic precursors is a critical event during skeletal muscle development (Du et al., 2010). Previous in vitro studies have shown that BPA can increase proliferation in vascular smooth muscle cells (Gao et al., 2019), but does not alter cell proliferation in human mesenchymal stem cell-derived adipocytes and osteoblasts (Dong et al., 2018). We did not observe any effect from BPA or BPS on myocyte proliferation in either sex or treatment. These findings are similar to those obtained in another mesenchymal stem cell-derived lineage (preadipocyte) (Pu et al., 2017) and suggest that the larger fiber size observed in skeletal muscle tissue after BPA or BPS exposure is likely related to myogenic hypertrophy, differentiation, or an imbalance in protein synthesis or proteolysis towards a hypertrophic phenotype. The larger area and diameter in myotubes derived from gestationally BPA- and BPS-exposed groups is consistent with the larger myofiber size in fetal skeletal muscle tissues. Altogether, these results provide with strong evidence that gestational BPA and BPS exposure may contribute to fetal skeletal muscle hypertrophy. Programming of fiber hypertrophy has been associated with AKT, mTORC1, and IGF1 signaling pathways (Brown, 2014). Whether these or alternative pathways are involved in the observed phenotype remains to be investigated. The fact that we also observed increased number of nuclei per myotube, reflective of increased fusion events, in BPA-exposed fetuses suggests that BPA may also contribute to increased myoblast differentiation.

Myogenic differentiation after 6 days in culture resulted in myotube formation and gene expression consistent with myogenic progression (Myf5 downregulation and MyoD, MyoG, and MRF4 upregulation [Zammit, 2017]) in all control primary cultures. Only BPS-exposure in female fetuses lead to an upregulation of all MRFs (Myf5, MyoD, MyoG, and MRF4), which may reflect an increased tone for both early and late myoblast differentiation. This is in alignment with the observation that BPS-exposed females also had upregulation in JUNB, responsible for fiber size maintenance and induction of rapid hypertrophy (Raffaello et al., 2010), and MyD88, responsible for myoblast fusion and skeletal muscle formation (Hindi et al., 2017). Although the signaling cues to which MRFs respond to remains poorly understood during fetal life (Hernandez-Hernandez et al., 2017), Wnt, sonic hedgehog, notch receptor, and bone morphogenetic proteins are among those that are likely upstream targets. Some of these pathways are direct targets for bisphenols or through steroid receptor binding (Baba et al., 2009; Hui et al., 2018; Medwid et al., 2018) and should be explored in future studies.

Because BPA can induce inflammation via pro-inflammatory cytokine release or the ERK-NF-κB pathway (Huang et al., 2019) and result in chronic inflammation (Bansal et al., 2017; Reddivari et al., 2017), we tested if gestational bisphenol exposure led to inflammation in skeletal muscle development and if genes involved in the inflammatory response would be altered during myogenic differentiation. Gestational BPS upregulated inflammatory genes (NF-κB and iNOS) in female skeletal muscle; a finding consistent with previous studies eluding to the immunomodulatory capacity of BPS in humans, female mice, and fish (Chen et al., 2018; Qiu et al., 2015, 2018; Tucker et al., 2018). However, inflammatory gene upregulation was not observed in primary myoblast cultures, suggesting that other cellular components of the skeletal muscle may be the primary drivers of the BPS-induced inflammatory changes observed.

CONCLUSION

This study demonstrates that gestational exposure to BPA and BPS leads to fetal skeletal muscle fiber hypertrophy independent of sex and alters fiber type distribution in a sex-specific manner. However, the functional significance and permanent nature of the hypertrophy observed in fetal tissue remain to be investigated. We have also demonstrated that gestational BPA or BPS did not alter fetal myoblasts’ responsiveness to insulin suggesting that skeletal muscle tissue-specific insulin resistance may not occur until postnatal life. Overall, this work highlights the need to understand the underlying mechanisms by which bisphenols program skeletal muscle development during fetal life, and if these changes persist into adulthood.

SUPPLEMENTARY DATA

Supplementary data are available at Toxicological Sciences online

DECLARATION OF CONFLICTING INTERESTS

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

ACKNOWLEDGMENTS

We thank Dr Ehrhardt and the Michigan State University Sheep Teaching and Research Farm for animal procurement and husbandry.

FUNDING

This work was supported by the National Institute of Environmental Health Sciences of the National Institute of Health (1K22ES026208 and R01ES027863 to A.V-L.), Michigan State University (MSU) General Funds, AgBioResearch and the United States Department of Agriculture (USDA) National Institute of Food and Agriculture. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Jiongjie Jing, a joint visiting Ph.D. student from Shanxi Agricultural University, was funded through a China Scholarship Council Fellowship. Jeremy Gingrich was supported by the MSU Office of the Vice President for Research and Graduate Studies, AgBioResearch, the College of Graduate Studies, and the College of Human Medicine.