- Email: [email protected]
Fibroblast growth factor 21 secretion enhances glucose uptake in mono(2-ethylhexyl)phthalate-treated adipocytes
Toxicology in Vitro 59 (2019) 246–254
Contents lists available at ScienceDirect
Toxicology in Vitro journal homepage: www.elsevier.com/locate/toxinvit
Fibroblast growth factor 21 secretion enhances glucose uptake in mono(2ethylhexyl)phthalate-treated adipocytes Jhih-Wei Hsua,b, Szu-Ching Yeha, Feng-Yuan Tsaia, Hsin-Wei Chena, Tsui-Chun Tsoua, a b
T
⁎
National Institute of Environmental Health Sciences, National Health Research Institutes, Zhunan, Miaoli 350, Taiwan Department of Life Science, National Central University, Zhongli, Taoyuan City 320, Taiwan
A R T I C LE I N FO
A B S T R A C T
Keywords: Endocrine disruptor Fat cells FGF21 Glucose metabolism MEHP Plasticizers
Previous studies revealed that cellular accumulation of mono(2-ethylhexyl)phthalate (MEHP) disturbed energy metabolism in adipocytes, where glucose uptake was significantly increased. The present study aimed to determine the mechanisms underlying the increased glucose uptake. MEHP-treated 3T3-L1 adipocytes exhibited a significantly increased glucose uptake activity. Immunoblot analysis suggested that the insulin-induced signals were not responsible for the increased glucose uptake. qPCR analysis revealed that both Glut1 and Glut4 genes were highly expressed during adipogenesis; Glut1 mRNA levels in MEHP-treated adipocytes were significantly increased. Moreover, MEHP-treated adipocytes exhibited significantly increased levels of fibroblast growth factor 21 (FGF21) in both mRNA and secreted protein. FGF21 is a peptide hormone with pleiotropic effects on regulation of insulin sensitivity and glucose/lipid homeostasis. We found that MEHP, FGF21, and lactate in culture medium together enhanced Fgf21 gene expression in MEHP-treated adipocytes. FGF21 signaling requires fibroblast growth factor receptor (FGFR) and βKlotho. Fgfr family and βKlotho genes were actively expressed during adipogenesis; mRNA levels of Fgfr3 and Fgfr4 genes in MEHP-treated adipocytes were significantly increased. Roles of FGF21/FGFR and phosphoinositide 3-kinase (PI3K)/AKT signal axes in regulation of glucose uptake were determined. We demonstrated that FGF21/FGFR signals played the major roles in up-regulation of the basal glucose uptake in MEHP-treated adipocytes. The in vitro evidence suggests that cellular FGF21 secretion enhances the basal glucose uptake in MEHP-treated adipocytes.
1. Introduction Epidemiological studies revealed the association between phthalate exposure and prevalence of metabolic diseases including obesity and its complications, e.g., insulin resistance and type 2 diabetes mellitus (James-Todd et al., 2012; Kim et al., 2013; Wang et al., 2013). In the Taiwan Maternal and Infant Cohort Study (TMICS), the early life exposure to phthalates was associated with decreased thyroid hormone levels in young children (Huang et al., 2017), in which the association is evident especially for di(2-ethylhexyl)phthalate (DEHP). Animal studies showed that maternal exposure to DEHP deregulated blood pressure, adiposity, cholesterol metabolism, and social interaction in mouse offspring (Lee et al., 2016). Dietary DEHP accelerated atherosclerosis in apolipoprotein E-deficient mice (Zhao et al., 2016), and
induced glucose metabolic disorder in rats (Martinelli et al., 2006; Xu et al., 2018). Both epidemiological and animal studies suggests the possible impacts of phthalate exposure on human health, especially metabolic disorder. Analysis of humans ingested with phthalates revealed that the majority of phthalates (> 90%) was excreted rapidly via urine in the first 24 h, where monoesters were the major metabolites (Koch et al., 2012). Adipose tissue is the major storage sites of lipophilic pollutants, e.g., phthalates. Previous studies revealed that phthalate accumulation in adipose tissues in both human (Zhang et al., 2003) and rats (Zeng et al., 2013). Therefore, despite the short biologic half-lives of phthalates, adipose tissue could be an important target of phthalates in vivo. Adipose tissue is a metabolically dynamic organ, acting as a crucial integrator of glucose homeostasis (Rosen and Spiegelman, 2006). Mice
Abbreviations: AIM, adipogenesis-inducing medium; AMM, adipogenesis-maintaining medium; [14C]2-DOG, 2-[1-14C]-deoxy-D-glucose; CHC, 2-cyano-3-(4-hydroxyphenyl)-2-propenoic acid; DEHP, di(2-ethylhexyl)phthalate; Dex, dexamethasone; DMEM-HG, Dulbecco's modified Eagle's medium with high glucose formula; Elk-1, Ets-like protein-1; FBS, fetal bovine serum; FGF21, fibroblast growth factor 21; FGFR, fibroblast growth factor receptor; IBMX, 3-isobutyl-1-methylxanthine; IRβ, insulin receptor β; IRS1, insulin receptor substrate 1; MEHP, mono(2-ethylhexyl)phthalate; mTORC1, mammalian target of rapamycin complex 1; PBS, phosphate-buffered saline; PI3K, phosphoinositide 3-kinase; PPARγ, peroxisome proliferator-activated receptor γ; RSK, ribosomal S6 kinase; SRF, serum response factor ⁎ Corresponding author at: National Institute of Environmental Health Sciences, National Health Research Institutes, 35 Keyan Road, Zhunan, Miaoli 350, Taiwan. E-mail address: [email protected] (T.-C. Tsou). https://doi.org/10.1016/j.tiv.2019.04.021 Received 9 November 2018; Received in revised form 29 March 2019; Accepted 17 April 2019 Available online 19 April 2019 0887-2333/ © 2019 Elsevier Ltd. All rights reserved.
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
were incubated in adipogenesis-inducing medium (AIM) (DMEM-HG with 0.25 μM dexamethasone, 0.5 mM 3-isobutyl-1-methylxanthine, 1.5 μM insulin, and 10% FBS) for 3 days (D3), in adipogenesis-maintaining medium (AMM) (DMEM-HG with 1.5 μM insulin and 10% FBS) for 2 days (D5), and then in DMEM-HG with 10% FBS for 6 days (D11). At D11, > 95% of the cells were fully differentiated into adipocytes (Fig. 1b). For MEHP treatments, the cells were treated with 10 to 300 μM MEHP as indicated or with 0.1% DMSO (as the solvent controls) for 6 days (from D5 to D11). Lipid droplet formation in adipocytes was determined with Oil Red O staining as described (Hsu et al., 2010) with minor modifications (Chiang et al., 2016). Oil Red O stock solution (0.5% of Oil Red O in isopropanol) was prepared and stored at 4 °C for use. Adipocytes were washed with phosphate-buffered saline (PBS) twice and fixed in 10% formalin solution for 1 h at room temperature. Oil Red O working solutions were freshly prepared by mixing 60% of Oil Red O stock solution with 40% of deionized water. The fixed cells were washed twice with PBS and stained with Oil Red O working solution for at least 1 h at room temperature. Following staining, cells were washed with deionized water for photography.
with perinatal DEHP exposure resulted in obesity in offspring (Hao et al., 2012). In vitro studies revealed that mono(2-ethylhexyl)phthalate (MEHP), the primary metabolite of DEHP, promoted adipogenesis of adipocytes via peroxisome proliferator-activated receptor γ (PPARγ) activation (Feige et al., 2007; Campioli et al., 2011). Moreover, MEHP treatments altered lipid metabolism (Ellero-Simatos et al., 2011) and energy metabolism (e.g., glucose uptake) (Chiang et al., 2016; Chiang et al., 2017), and elicited inflammatory responses (Manteiga and Lee, 2017) in adipocytes. The studies clearly suggest that MEHP not only enhances adipogenesis via PPARγ but also exhibits significant effects on various biological activities in adipocytes. Fibroblast growth factor 21 (FGF21), a peptide hormone, is most abundantly expressed in liver and is also detected in adipose tissue, skeletal muscle (Itoh, 2014), and pancreas (Johnson et al., 2009). Animal studies revealed that Fgf21-transgenic mice were resistant to dietinduced obesity, and administration of FGF21 reduced plasma levels of glucose and triglycerides in both ob/ob and db/db mice (Kharitonenkov et al., 2005). Fgf21 was transcriptionally up-regulated in adipose tissue of db/db mice treated with PPARγ agonists (Muise et al., 2008). Moreover, in vitro evidence indicated that FGF21 stimulated glucose uptake and modulated GLUT1 expression in adipocytes (Kharitonenkov et al., 2005) and βKlotho was required for FGF21 signaling through fibroblast growth factor receptors (FGFR) (Suzuki et al., 2008). These studies together suggest FGF21 as a pivotal glucose regulator in adipocytes. We previously demonstrated that 3T3-L1 adipocytes treated with MEHP ≥30 μM for 6 days exhibited a markedly increased glucose uptake activity (Chiang et al., 2016). Moreover, it was noted that significantly increased FGF21 levels were detected in both DEHP-treated mice and MEHP-treated adipocytes (Chiang et al., 2017). The in vivo and in vitro evidence suggested the potential involvement of FGF21 in systemic regulation of glucose homeostasis in response to DEHP/MEHP treatments. The present in vitro study explored how MEHP enhanced glucose uptake in adipocytes via FGF21 and its associated signaling molecules.
2.3. Glucose uptake assay Following MEHP treatments and/or inhibition of PI3K, AKT, or FGFR as indicated, glucose uptake activity in 3T3-L1 adipocytes was determined as previously described (Chiang et al., 2016). Briefly, adipocytes were starved in serum-free DMEM-HG for 4 h, washed twice with Krebs–Ringer phosphate buffer (128 mM NaCl, 4.7 mM KCl, 1.65 mM CaCl2, 2.5 mM MgSO4, and 5 mM Na2HPO4, pH 7.4, 37 °C), and then maintained in the same buffer for glucose uptake assay. The cells were left untreated (for determination of the basal glucose uptake activity) or treated with insulin (100 nM) (for determination of the insulin-induced glucose uptake activity) for 10 min and then glucose uptake was started by addition of [14C]2-DOG (0.1 μCi/well) for 10 min at 37 °C. Cells were washed three times with ice-cold PBS and lysed in 800 μl of lysis solution (0.5 M NaOH and 0.1% SDS). For determining cellular uptake of [14C]2-DOG, lysate samples were assayed with a Topcount NXT Scintillation Counter (Packard Instrument Company, Meriden, CT, USA).
2. Methods and materials 2.1. Chemicals and cells
2.4. qPCR analysis of gene expression MEHP (M542490) was purchased from Toronto Research Chemicals Inc. (North York, Canada). 2-[1-14C]-deoxy-D-glucose ([14C]2-DOG) (NEC495A050UC) was purchased from PerkinElmer Life Sciences (Boston, MA, USA). Insulin (I6634), 3-isobutyl-1-methylxanthine (IBMX) (I5879), dexamethason (Dex) (D4902), Oil Red O (O0625), 10% formalin solution (HT5012), PD161570 (an FGFR tyrosine kinase inhibitor) (PZ0109), LY294002 [a phosphoinositide 3-kinase (PI3K) inhibitor] (L9908), and AKTi (an AKT1/2 inhibitor) (A6730) were purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA). The monocarboxylic acid transport inhibitor 2-cyano-3-(4-hydroxyphenyl)2-propenoic acid (CHC) (ab146008) was from Abcam (Cambridge, MA, USA). Dulbecco's modified Eagle's medium with high glucose formula (DMEM-HG) (12800–017), bovine serum (16170–078), penicillin/ streptomycin (15140–122), and fetal bovine serum (FBS) (10091–148) were from Gibco/Invitrogen (Carlsbad, CA, USA). 3T3-L1 cells (BCRC60159) were purchased from the Bioresources Collection and Research Center, Food Industry Research and Development Institute (Hsinchu, Taiwan). The cells are routinely maintained in DMEM-HG with 10% bovine serum and 1% penicillin/streptomycin in a humidified atmosphere containing 5% CO2 at 37 °C.
3T3-L1 cells during or after adipogenesis were collected as indicated times and gene expression of aP2, C/EBPα, Fgf21, Fgfr1, Fgfr2, Fgfr3, Fgfr4, Glut1, Glut4, βKlotho, PPARγ, Srebf1, and β–actin by using qPCR analysis of mRNA levels. Following RNA isolation and cDNA synthesis, quantification of cDNA with the LightCycler-FastStart DNA Master SYBR Green I system (Roche Diagnostics GmbH) was performed by using a LightCycler 480 instrument (Roche Diagnostics GmbH). DNA sequences of mouse-specific primer sets for qPCR were summarized (Table S1 in Supplementary Material). Melting curve analysis was used to confirm the amplification specificity. All qPCR assays were performed in duplicate for at least three times (n ≥ 3) with β–actin as the internal control. Relative quantification analysis was applied to determine the ratio change of a specific gene expression with the LightCycler analysis software. 2.5. Immunoblot analysis Following treatments, the cells were lysed and cell lysates were subjected to SDS-PAGE and immunoblot analysis according to standard protocols. The blot was incubated with an antibody against pIRβ(Y1150/Y1151) (#04–299) (Millipore), IRβ (#3025) (Cell Signaling), p-IRS1(Y632) (sc-17196) (Santa Cruz), IRS1 (#2382) (Cell Signaling), p-AKT(T308) (#2965) (Cell Signaling), p-AKT(S473) (#9271) (Cell Signaling), AKT (sc-1618) (Santa Cruz), α–tubulin (as the loading controls of total protein) (GTX112141) (GeneTex), βKlotho
2.2. Induction of adipogenesis and phthalate treatments Induction of adipogenesis in 3T3-L1 cells was performed as previously described (Hsu et al., 2010; Chiang et al., 2016) with modifications as shown in Fig. 1a. Two days postconfluence (D0), the cells 247
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 1. MEHP-treated adipocytes exhibit an increased glucose uptake activity. (a) Adipogenesis protocol and cell treatments in this study. Following the standard adipogenesis protocol as described in “Materials and method”, 3T3-L1 cells were treated 0.1% DMSO (the control) or 10 to 300 μM MEHP from D5 to D11. (b) Representative bright field (D0 and D11) and Oil Red O-stained (D11) images of the control adipocytes are presented. (c) mRNA levels of aP2, C/EBPα, PPARγ, and Srebf1 in the control adipocytes during adipogenesis were analyzed with qPCR. Results are expressed as fold changes (vs. D0) and presented as means ± SD (n ≥ 4). (d) At D11, both the basal and 100 nM insulin-induced glucose uptake levels in adipocytes were determined. Results are expressed as fold changes (vs. the basal glucose uptake in the controls) and presented as means ± SD (n = 4). ***p < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
2.8. Statistical analysis
(GTX122197) (GeneTex), or PPARγ (sc-7273) (Santa Cruz) at 4 °C overnight. HRP conjugated secondary antibodies were then used to reveal the specific protein bands with Western Lightning Chemiluminescence Reagent Plus (PerkinElmer Life Sciences).
Each experiment was performed independently at least three times. All analyses were performed with GraphPad Prism version 7.0 for Windows (GraphPad Software, La Jolla, CA, USA). All data were presented as means ± SD. Comparisons between groups were performed with ANOVA followed by Bonferroni's multiple comparison test. Differences were considered statistically significant when p value < 0.05.
2.6. ELISA analysis of FGF21 in conditioned medium Following treatments, culture medium was collected and stored at −80 °C until analysis. FGF21 levels in culture medium were determined with the Quantikine ELISA Kit (MF2100) (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instruction manual.
3. Results and discussion 3.1. MEHP-treated adipocytes exhibit an increased glucose uptake activity
2.7. Measurement of lactate in conditioned media Adipogenesis of 3T3-L1 cells was determined by using oil droplet formation and expression of adipogenic marker genes. Following the standard adipogenesis protocol (Fig. 1a), oil droplet formation in cells at D11 was evident as observed with light microscopy combined with Oil Red O staining (Fig. 1b). qPCR analysis revealed markedly elevated mRNA levels of adipogenic marker genes, including aP2, C/EBPα, PPARγ, and Srebf1, in cells during adipogenesis (Fig. 1c). The adipogenesis protocol ensured the successful adipogenic differentiation of 3T3-L1 cells in this study. To determine MEHP effects on glucose uptake, 3T3-L1 adipocytes
Following the standard adipogenesis protocol as described in Fig. 1a, 3T3-L1 cells were treated with 0.1% DMSO (the vehicle controls) or MEHP (30, 100, and 300 μM) from D5 to D11. At D11, 1 ml of conditioned medium samples were collected for measurements of lactate levels with a colorimetric lactate assay (Randox Laboratories Ltd.). Culture medium without cells incubation was used to determine the background lactate concentration. Lactate levels in conditioned media represented the cellular production of lactate from D9 to D11.
248
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
were treated with 30 and 100 μM MEHP for 6 days (D5 to D11), and then both the basal and insulin-induced glucose uptakes were analyzed. Results in Fig. 1d indicated that adipocytes treated with 30 and 100 μM MEHP exhibited a higher activity in both the basal and insulin-induced glucose uptake than the control adipocytes. Moreover, insulin stimulated glucose uptake in the controls (from 1.00 to 2.80 folds, p < 0.001), 30 μM MEHP-treated (from 1.34 to 3.81 folds, p < 0.001), and 100 μM MEHP-treated adipocytes (from 2.41 to 4.03 folds, p < 0.001). Therefore, the results indicate that both the basal and insulin-induced glucose uptake in MEHP-treated adipocytes is significantly higher than that in the control adipocytes.
Cellular glucose uptake mainly involves two major glucose transporters, GLUT1 and GLUT4 (Ebeling et al., 1998). GLUT1 is ubiquitously distributed in different tissues and facilitates the insulin-independent glucose uptake. GLUT4 is responsible for the majority of glucose transport into muscle and adipose cells in an insulin-dependent manner. In this study, expression of Glut1 and Glut4 genes in adipocytes was determined by using qPCR analysis. Results in Fig. 2b (top panel) revealed that mRNA levels of both Glut1 and Glut4 genes were highly expressed in 3T3-L1 cells during adipogenesis. Moreover, mRNA levels of Glut1, but not Glut4, in the adipocytes treated with 30 and 100 μM MEHP for 6 days (D5 to D11) were significantly increased by 1.52 folds (p < 0.05) and 1.94 folds (p < 0.001) (vs. the controls), respectively (Fig. 2b, bottom panel). Results in Fig. 2b suggested the potential contribution of GLUT1 to the higher glucose uptake in MEHP-treated adipocytes.
3.2. The insulin-induced signals are not responsible for the increased glucose uptake in MEHP-treated adipocytes In adipose tissue, insulin induces glucose uptake from the circulation via promoting GLUT4 trafficking to the plasma membrane (Satoh, 2014). In the process, insulin stimulates a signaling cascade composed of insulin receptor β (IRβ), insulin receptor substrate 1 (IRS1), PI3K, and AKT (Satoh, 2014). To determine the involvement of the insulininduced signaling cascade in the increased glucose uptake in MEHPtreated adipocytes, both the control and MEHP-treated adipocytes were treated with 100 nM insulin for 10 min. Then, immunoblot analysis of protein phosphorylation was used to determine the activation of insulin-induced signals, including IRβ, IRS1, and AKT. Results in Fig. 2a revealed that the insulin treatment caused marked increases in phosphorylation of IRβ, IRS1, and AKT in control adipocytes; the insulininduced IRβ phosphorylation was further enhanced in adipocytes treated with MEHP (10 to 300 μM). However, both total protein and the insulin-induced phosphorylation levels of IRS1 and AKT in MEHPtreated adipocytes were lower than that in control adipocytes. The results together suggest that the insulin-induced signals are not responsible for the increased glucose uptake in MEHP-treated adipocytes.
3.3. MEHP-treated adipocytes secrete significantly increased amounts of FGF21 protein In this study, both Glut1 expression (Fig. 2b, bottom panel) and glucose uptake (Fig. 1d) in MEHP-treated adipocytes were up-regulated. FGF21 has been previously demonstrated to enhance glucose uptake in adipocytes by induction of Glut1 expression (Kharitonenkov et al., 2005), via activation of the serum response factor (SRF)/Ets-like protein-1 (Elk-1) (Ge et al., 2011). Because of the potential roles of FGF21 in the increased glucose uptake in MEHP-treated adipocytes, it was critical to know the Fgf21 gene expression patterns in adipocytes. In the present study, time-course of Fgf21 mRNA levels in 3T3-L1 cells during adipogenesis were determined with qPCR. Results in Fig. 3a revealed that Fgf21 mRNA levels in the control adipocytes during adipogenesis at D5, D7, D9, and D11 were increased by 5.6, 5.2, 6.8, and 5.6 folds (vs. D0), respectively. Importantly, Fgf21 mRNA levels in MEHP-treated adipocytes were markedly higher than those in the
Fig. 2. The insulin-induced signals and gene expression of Glut1 and Glut4 in adipocytes. (a) For determining activation of the insulin-induced signals, both the control and MEHP-treated adipocytes (at D11) were starved in serum-free DMEM-HG for 4 h and then treated with 100 nM insulin for 10 min. Then, activation of the insulin-induced signals, including IRβ, IRS1, and AKT, was determined with immunoblot analysis of total protein and their phosphorylation levels, with α-tubulin as the loading controls. (b) mRNA levels of Glut1 and Glut4 in the control cells during adipogenesis (at D0, D5, D8, and D11) (top panel) as well as in the control and MEHP-treated adipocytes after adipogenesis (at D11) (bottom panel) was determined with qPCR. Results are expressed as fold changes (vs. D0 in top panel; vs. the control in bottom panel) and presented as means ± SD (n ≥ 3). *p < 0.05 and ***p < 0.001 (vs. the control). 249
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
2008). In our previous microarray study, Fgf21 expression was significantly up-regulated in adipocytes subjected to both MEHP and the PPARγ agonist rosiglitazone treatments (Chiang et al., 2017). Taken together, these results indicate that MEHP stimulates FGF21 protein secretion from adipocytes via activation of PPARγ. Phthalates are lipophilic and easy to accumulate in adipose tissue. We previously demonstrated that, as treated with MEHP in vitro, adipocytes accumulated MEHP from culture medium (Chiang et al., 2016). In this study, it was important to further clarify whether the MEHPaccumulated adipocytes could sustainably secrete such high levels of FGF21 protein. Following treatments with 100 μM MEHP for 4 days (D5 to D9), 3T3-L1 adipocytes were switched to culture medium without MEHP at D9; then FGF21 protein levels in conditioned medium were monitored at D11, D13, and D15. Results in Fig. 3c revealed that the MEHP withdrawal resulted in a dramatic dropping of FGF21 from 2516.0 (D9) to 343.5 pg/ml (D11). Clearly, the FGF21 level was down to the control level in two days, indicating that the MEHP-accumulated adipocytes could not sustainably secrete such high levels of FGF21 protein. 3.4. Expression of Fgfr family and βKlotho genes in 3T3-L1 adipocytes Results in Fig. 3 provided clear evidence that MEHP-treated adipocytes secreted significantly increased amounts of FGF21 protein in culture medium. FGF21 stimulates glucose uptake in adipocytes via FGFR and the cofactor βKlotho (Ogawa et al., 2007). In this study, gene expression profiles of Fgfr family and βKlotho in both the control and MEHP-treated 3T3-L1 adipocytes were determined. qPCR analysis revealed that mRNA levels of Fgfr family, including Fgfr1, Fgfr2, Fgfr3, and Fgfr4, in the control adipocytes during adipogenesis were increased in varying degrees (Fig. 4a, left panel), where the relative mRNA levels (vs. D0 of each gene) at D11 were Fgfr3 (4.7 folds) > Fgfr1 (2.9 folds) > Fgfr4 (2.4 folds) > Fgfr2 (1.1 folds). At D11, both Fgfr3 and Fgfr4 mRNA levels in adipocytes treated with 100 μM MEHP were significantly higher than that in the controls (Fig. 4a, right panel). Moreover, βKlotho mRNA levels in the control adipocytes during adipogenesis were markedly increased by 158, 1202, and 2252 folds (vs. D0) at D5, D8, and D11, respectively (Fig. 4b, left panel). At D11, no significant difference in βKlotho mRNA levels was detected between the control and MEHP-treated adipocytes (Fig. 4b, right panel). The βKlotho expression results were further confirmed with immunoblot analysis (see Fig. S1 in Supplemental Material). Results in Fig. 4 suggested the involvement of FGFR3 and FGFR4 in the increased glucose uptake in MEHP-treated adipocytes. Our previous microarray study revealed the up-regulation of Fgfr3 and Fgfr4 genes in adipocytes subjected to both MEHP and the PPARγ agonist rosiglitazone treatments (Chiang et al., 2017), suggesting that MEHP stimulates the gene expression via PPARγ.
Fig. 3. MEHP-treated adipocytes secrete increased amounts of FGF21 protein. Following the standard adipogenesis protocol, 3T3-L1 cells were treated 0.1% DMSO (the control) or MEHP (30 and 100 μM) from D5 to D11. (a) Fgf21 mRNA levels in adipocytes during adipogenesis were analyzed with qPCR. Results are expressed as fold change (vs. D0 of the control) and presented as means ± SD (n = 3). (b) Cellular secretion of FGF21 protein in culture medium was analyzed by using ELISA. FGF21 protein levels are presented as means ± SD (n = 4). (c) At D9, culture medium of both the control and MEHP-treated adipocytes was switched to complete medium without MEHP for continuous culture, conditioned medium samples were collected at D9, D11, D13, and D15 for analysis of FGF21 protein. FGF21 protein levels are presented as means ± SD (n = 4). *p < 0.05 and ***p < 0.001 (as indicated).
3.5. MEHP, FGF21, and lactate together enhance Fgf21 expression in MEHP-treated adipocytes MEHP-treated adipocytes secrete significant amounts of FGF21 protein in culture medium (Fig. 3b). In addition to enhancing glucose uptake in adipocytes (Kharitonenkov et al., 2005; Ge et al., 2011), FGF21 also stimulates its own gene expression via activation of mammalian target of rapamycin complex 1 (mTORC1)/ribosomal S6 kinase (RSK) (Minard et al., 2016). In this study, conditioned medium from MEHP-treated adipocytes (CM) was collected to test its effect on Fgf21 expression. 3T3-L1 adipocytes were treated with the CM for 16 h, using fresh medium (FM) and fresh medium with 100 μM MEHP (FM + MEHP) as the controls. Then, Fgf21 mRNA levels were determined with qPCR. As shown in Fig. 5a, the CM and the FM + MEHP caused significant increases in Fgf21 mRNA by 26.0 folds (vs. FM) (p < 0.001) and 9.1 folds (vs. FM) (p < 0.05), respectively. Because both CM and FM + MEHP contained the same level of MEHP (100 μM), results in Fig. 5a were interpreted that the MEHP in CM contributed to
control adipocytes in a MEHP dose-dependent manner. For example at D11, Fgf21 mRNA levels in adipocytes treated with 30 and 100 μM MEHP rose dramatically from 5.6 folds to 21.0 folds (p < 0.05) and 241.0 folds (p < 0.001), respectively (Fig. 3a). FGF21 is a secreted protein. Cellular secretion of FGF21 protein in culture medium were also analyzed with ELISA. Results in Fig. 3b showed that during adipogenesis the pattern of FGF21 protein secretion was very similar to that of Fgf21 mRNA. At D11, FGF21 protein levels from adipocytes treated with 30 and 100 μM MEHP rose significantly from 174 pg/ml to 444 pg/ml (p < 0.05) and 1872 pg/ml (p < 0.001), respectively (Fig. 3b). Therefore, it is clear that MEHPtreated adipocytes secrete significantly increased amounts of FGF21 protein in vitro. Previous study found that adipose Fgf21 was upregulated by PPARγ and thus altered metabolic status (Muise et al., 250
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 4. Expression of Fgfr family and βKlotho genes in 3T3-L1 adipocytes. During adipogenesis, 3T3-L1 cells were treated with 0.1% DMSO (the control) or MEHP (30 and 100 μM) from D5 to D11. Gene expression of (a) Fgfr family and (b) βKlotho was determined by qPCR analysis of mRNA levels. Gene expression in the control was monitored at D0, D5, D8, and D11 (left panels, a and b). At D11, mRNA levels in the control and MEHP-treated adipocytes were compared (right panels, a and b). Results are expressed as fold change (vs. D0 in left panels) (vs. the control in right panels) and presented as means ± SD (n ≥ 3). *p < 0.05 and ***p < 0.001 (vs. D0) in left panels; *p < 0.05 and **p < 0.01 (vs. the control) in right panels.
inhibited in the presence of the protein synthesis inhibitor cycloheximide. The study suggests that FGF21 effect on glucose uptake requires transcriptional activation. In the present study, the involvement of the two signal axes in activation of glucose uptake in adipocytes was determined by blocking the pathways with specific inhibitors, i.e., PD161570 (an FGFR tyrosine kinase inhibitor), LY294002 (a PI3K inhibitor), and AKTi (an AKT1 and AKT2 dual kinase inhibitor). Both the control and 100 μM MEHPtreated adipocytes were pretreated with PD16570 (10 μM) for 4 h, LY294002 (50 μM) for 10 min, or AKTi (30 μM) for 10 min and then both the basal and insulin-induced glucose uptakes in adipocytes were determined. Results in Fig. 6a revealed that inhibition of PI3K (with LY294002) or AKT (with AKTi) completely blocked the insulin-induced glucose uptake in both the control and MEHP-treated adipocytes. The results confirmed the well-recognized roles of PI3K/AKT in the insulininduced glucose uptake. It was also noted that the inhibition of PI3K or AKT also fully abolished the basal glucose uptake in both the control and MEHP-treated adipocytes (Fig. 6a). Although the insulin-induced pathway was not responsible for the increased glucose uptake in MEHPtreated adipocytes (Fig. 2a), the present results suggested that the functional PI3K/AKT was essential for the basal glucose uptake in adipocytes. FGF21/FGFR signals activate the basal glucose uptake in adipocytes in an insulin-independent manner (Kharitonenkov et al., 2005), mainly via up-regulation of GLUT1 protein expression (Ge et al., 2011). Unexpectedly, inhibition of FGFR by PD16570 attenuated the insulin-induced glucose uptake in the control and MEHP-treated adipocytes by 26% (p > 0.05) and 30% (p < 0.05), respectively (Fig. 6a). The mild inhibitory effect of PD16570 on the insulin-induced glucose uptake implicated the potential crosstalk between FGF21/FGFR and insulin/ PI3K/AKT signal axes. Moreover, it was also noted that FGFR inhibition of by PD16570 pretreatment for 4 h was not able to significantly block the basal glucose uptake in both the control and MEHP-treated adipocytes (Fig. 6a). The reason could be that, before subjected to the PD16570 treatments at D11, the 100 μM MEHP-treated adipocytes had been exposed to MEHP, FGF21, and lactate in culture medium for 6 days (D5 to D11) (Figs. 3b and 5b) and thus already had significantly increased amounts of GLUT1 (Fig. 2b, bottom panel) for the basal glucose uptake. To further determine the roles of FGFR in regulation of the basal glucose uptake in MEHP-treated adipocytes, a long-term inhibition of FGFR was necessary. During adipogenesis, 3T3-L1 cells were treated
32.4% of the Fgf21 mRNA induction by CM. Lactate was recently found to induce Fgf21 gene expression in mouse primary white adipocytes (Jeanson et al., 2016). In our previous study, the MEHP-treated 3T3-L1 adipocytes secreted significant amounts of lactate in culture medium (Chiang et al., 2016). Indeed, results in Fig. 5b revealed that at D11 lactate concentration in conditioned medium of 3T3-L1 adipocytes treated with 30 and 100 μM MEHP was significantly increased from 4.23 mM to 6.93 mM (p < 0.001) and 12.02 mM (p < 0.001), respectively. Reducing cellular lactate import with the monocarboxylate transporter inhibitor CHC was shown to block the lactate-induced Fgf21 expression in adipocytes (Jeanson et al., 2016). In the present study, CHC was adopted to determine the effects of lactate in culture medium on Fgf21 expression. During adipogenesis, the 100 μM-treated 3T3-L1 adipocytes were subjected to 2 mM CHC treatments for 48 h (D9 to D11), and then Fgf21 mRNA levels were analyzed with qPCR. Results in Fig. 5c revealed that the CHC treatments significantly inhibited Fgf21 mRNA levels in MEHP-treated adipocytes by 19.1% (from 24.6 folds to 20.1 folds) (p < 0.05), indicating that lactate in culture medium contributed to 19.1% of the Fgf21 mRNA induction. Taken together, during adipogenesis in the presence of MEHP, 3T3L1 adipocytes secreted significant amounts of FGF21 (Fig. 3b) and lactate (Fig. 5b) in culture medium. Results in Figs. 3 and 5 suggest that MEHP, FGF21, and lactate in culture medium together activate a robust expression of Fgf21 gene in MEHP-treated adipocytes; the possible involvements of other metabolites and/or factors in conditioned medium cannot be excluded. 3.6. Roles of FGF21/FGFR and insulin/PI3K/AKT signal axes in regulation of glucose uptake in adipocytes FGF21/FGFR (Kharitonenkov et al., 2005) and insulin/PI3K/AKT (Kohn et al., 1996; Hernandez et al., 2001) signal axes play critical and differential roles in regulation of glucose uptake in adipocytes, mainly via the basal growth-related glucose transporter GLUT1 and the insulinresponsive glucose transporter GLUT4, respectively. The insulin/PI3K/ AKT signal axis stimulates the rapid glucose uptake via direct translocation of GLUT4 from intracellular membranes to the cell surface (Leney and Tavare, 2009); the process requires no transcriptional activation. In contrast to the rapid response induced by insulin, at least 4 h of FGF21 treatment is requires for adipocytes to exhibit an increased glucose uptake activity (Kharitonenkov et al., 2005), which is markedly 251
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 6. Involvement of PI3K/AKT and FGFR in regulation of glucose uptake in adipocytes. (a) For short-term inhibition, both the control and 100 μM MEHPtreated 3T3-L1 adipocytes were treated with 10 μM PD161570 (an FGFR inhibitor) for 4 h, 50 μM LY294002 (a PI3K inhibitor) for 10 min, or 30 μM AKTi (an AKT inhibitor) for 10 min. Then, the basal and insulin-induced glucose uptake in both the control and MEHP-treated adipocytes was determined. (b) For long-term inhibition of FGFR, both the control and MEHP-treated adipocytes were treated with 10 μM PD161570 for 6 days (D5 to D11). Then, the basal glucose uptake in both the control and MEHP-treated adipocytes was determined. Results are expressed as fold changes (vs. the basal glucose uptake in the controls with no inhibitor). Data are presented as mean ± SD (n = 3). *p < 0.05 and ***p < 0.001.
Fig. 5. FGF21, MEHP, and lactate in culture medium together activate Fgf21 gene expression in MEHP-treated adipocytes. (a) Conditioned medium was collected from the 100 μM MEHP-treated adipocytes at D11. 3T3-L1 adipocytes were treated with conditioned medium (CM) for 16 h, with fresh medium (FM) and fresh medium added with 100 μM MEHP (FM + MEHP) as the negative and MEHP controls, respectively. Fgf21 mRNA levels in adipocytes were analyzed with qPCR. Results are expressed as fold change (vs. FM) and presented as means ± SD (n = 3). *p < 0.05 and ***p < 0.001 (vs. FM). (b) Conditioned medium was collected from the control, 30 μM MEHP-treated, and 100 μM MEHP-treated adipocytes at D11. Lactate levels in conditioned medium were determined and presented as means ± SD (n = 5). ***p < 0.001 (vs. the control). (c) 100 μM MEHP-treated adipocytes were treated with the monocarboxylate transporter inhibitor CHC (2 mM) from D9 to D11 (100 μM MEHP + CHC). At D11, Fgf21 mRNA levels in the control and 100 μM MEHP-treated adipocytes with or without the inhibitor treatment were determined with qPCR. Results are expressed as fold change (vs. the control) and presented as means ± SD (n = 3). *p < 0.05 and ***p < 0.001 (vs. the control or as indicated).
the basal glucose uptake in MEHP-treated adipocytes. 4. Conclusions The study aimed to address the roles of FGF21 in regulation of glucose uptake in MEHP-treated adipocytes. Following adipogenesis in the presence of MEHP, MEHP-treated 3T3-L1 adipocytes exhibited a higher glucose uptake activity than the untreated controls. However, immunoblot analysis revealed that the insulin-induced signals were not responsible for the increased glucose uptake in MEHP-treated adipocytes. The MEHP-treated adipocytes exhibited increased mRNA levels of Glut1, Fgf21, Fgfr3, and Fgfr4; the cells also secreted significantly increased amounts of FGF21 protein. MEHP, FGF21, and lactate in culture medium together enhanced Fgf21 gene expression in MEHPtreated adipocytes. Roles of FGF21/FGFR and PI3K/AKT signal axes in regulation of glucose uptake in adipocytes were determined. We found that FGF21/FGFR signals played the major roles in up-regulation of the basal glucose uptake in MEHP-treated adipocytes. Taken together, on the basis of results of this and previous studies, a proposed model for
with 100 μM MEHP in the presence of 10 μM PD16570 for 6 days (D5 to D11); then, at D11 the basal glucose uptake in both the control and MEHP-treated adipocytes was analyzed. Results in Fig. 6b revealed that inhibition of FGFR with PD16570 for 6 days completely blocked the basal glucose uptake in both the control and MEHP-treated adipocytes, suggesting the critical involvement of FGF21/FGFR in up-regulation of 252
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 7. Proposed schematic diagram for the increased glucose uptake in MEHP-treated 3T3-L1 adipocytes. MEHP binding to PPARγ triggers PPARγ activation and subsequent expression of PPARγ target genes, e.g., Fgf21, Fgfr3, and Fgfr4. In MEHP-treated adipocytes, the higher glucose uptake is accompanied with a proportional lactate secretion (Chiang et al., 2016; 2017); lactate in turn induces Fgf21 expression through a p38-MAPK pathway (Jeanson et al., 2016). FGF21 also stimulates its own gene expression via activation of the mTORC1/RSK pathway (Minard et al., 2016). Following secretion and accumulation in culture medium, FGF21 binds to FGFR3/βKlotho or FGFR4/βKlotho complex and then activates the receptors. The FGF21/FGFR signal axis enhances glucose uptake in adipocytes by induction of Glut1 expression (Kharitonenkov et al., 2005), via activation of the ERK1/2/RSK pathway and then the SRF/Elk-1 pathway (Ge et al., 2011).
References
the increased glucose uptake in MEHP-treated adipocytes is schematically summarized in Fig. 7.
Campioli, E., Batarseh, A., Li, J., Papadopoulos, V., 2011. The endocrine disruptor mono(2-ethylhexyl) phthalate affects the differentiation of human liposarcoma cells (SW 872). PLoS One 6, e28750. Chiang, H.C., Kuo, Y.T., Shen, C.C., Lin, Y.H., Wang, S.L., Tsou, T.C., 2016. Mono(2ethylhexyl)phthalate accumulation disturbs energy metabolism of fat cells. Arch. Toxicol. 90, 589–601. Chiang, H.C., Wang, C.H., Yeh, S.C., Lin, Y.H., Kuo, Y.T., Liao, C.W., et al., 2017. Comparative microarray analyses of mono(2-ethylhexyl)phthalate impacts on fat cell bioenergetics and adipokine network. Cell Biol. Toxicol. 33, 511–526. Ebeling, P., Koistinen, H.A., Koivisto, V.A., 1998. Insulin-independent glucose transport regulates insulin sensitivity. FEBS Lett. 436, 301–303. Ellero-Simatos, S., Claus, S.P., Benelli, C., Forest, C., Letourneur, F., Cagnard, N., et al., 2011. Combined transcriptomic-(1)H NMR metabonomic study reveals that monoethylhexyl phthalate stimulates adipogenesis and glyceroneogenesis in human adipocytes. J. Proteome Res. 10, 5493–5502. Feige, J.N., Gelman, L., Rossi, D., Zoete, V., Metivier, R., Tudor, C., et al., 2007. The endocrine disruptor monoethyl-hexyl-phthalate is a selective peroxisome proliferator-activated receptor gamma modulator that promotes adipogenesis. J. Biol. Chem. 282, 19152–19166. Ge, X., Chen, C., Hui, X., Wang, Y., Lam, K.S., Xu, A., 2011. Fibroblast growth factor 21 induces glucose transporter-1 expression through activation of the serum response factor/Ets-like protein-1 in adipocytes. J. Biol. Chem. 286, 34533–34541. Hao, C., Cheng, X., Xia, H., Ma, X., 2012. The endocrine disruptor mono-(2-ethylhexyl) phthalate promotes adipocyte differentiation and induces obesity in mice. Biosci. Rep. 32, 619–629. Hernandez, R., Teruel, T., Lorenzo, M., 2001. Akt mediates insulin induction of glucose uptake and up-regulation of GLUT4 gene expression in brown adipocytes. FEBS Lett. 494, 225–231. Hsu, H.F., Tsou, T.C., Chao, H.R., Shy, C.G., Kuo, Y.T., Tsai, F.Y., et al., 2010. Effects of arecoline on adipogenesis, lipolysis, and glucose uptake of adipocytes-A possible role of betel-quid chewing in metabolic syndrome. Toxicol. Appl. Pharmacol. 245, 370–377. Huang, H.B., Chuang, C.J., Su, P.H., Sun, C.W., Wang, C.J., Wu, M.T., et al., 2017.
Transparency document The Transparency document associated to this article can be found, in the online version. Conflicts of interest TCT reports grants and personal fees from Ministry of Science and Technology, (Taiwan) R.O.C. during the conduct of the study. JWH and HWC report personal fees from Ministry of Science and Technology, (Taiwan) R.O.C. during the conduct of the study. SCY and FYT have nothing to disclose. Acknowledgements This work was supported by grants from the Ministry of Science and Technology (105-2320-B-400-022, 106-2320-B-400-006, and 1072320-B-400-001) and the National Health Research Institutes (EM-105PP-03, EM-106-PP-03, and EM-107-PP-03) in Taiwan. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.tiv.2019.04.021. 253
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Health Perspect. 125, 615–622. Martinelli, M.I., Mocchiutti, N.O., Bernal, C.A., 2006. Dietary di(2-ethylhexyl)phthalateimpaired glucose metabolism in experimental animals. Hum. Exp. Toxicol. 25, 531–538. Minard, A.Y., Tan, S.X., Yang, P., Fazakerley, D.J., Domanova, W., Parker, B.L., et al., 2016. mTORC1 is a major regulatory node in the FGF21 signaling network in adipocytes. Cell Rep. 17, 29–36. Muise, E.S., Azzolina, B., Kuo, D.W., El-Sherbeini, M., Tan, Y., Yuan, X., et al., 2008. Adipose fibroblast growth factor 21 is up-regulated by peroxisome proliferator-activated receptor gamma and altered metabolic states. Mol. Pharmacol. 74, 403–412. Ogawa, Y., Kurosu, H., Yamamoto, M., Nandi, A., Rosenblatt, K.P., Goetz, R., et al., 2007. BetaKlotho is required for metabolic activity of fibroblast growth factor 21. Proc. Natl. Acad. Sci. U. S. A. 104, 7432–7437. Rosen, E.D., Spiegelman, B.M., 2006. Adipocytes as regulators of energy balance and glucose homeostasis. Nature 444, 847–853. Satoh, T., 2014. Molecular mechanisms for the regulation of insulin-stimulated glucose uptake by small guanosine triphosphatases in skeletal muscle and adipocytes. Int. J. Mol. Sci. 15, 18677–18692. Suzuki, M., Uehara, Y., Motomura-Matsuzaka, K., Oki, J., Koyama, Y., Kimura, M., et al., 2008. betaKlotho is required for fibroblast growth factor (FGF) 21 signaling through FGF receptor (FGFR) 1c and FGFR3c. Mol. Endocrinol. 22, 1006–1014. Wang, H., Zhou, Y., Tang, C., He, Y., Wu, J., Chen, Y., et al., 2013. Urinary phthalate metabolites are associated with body mass index and waist circumference in Chinese school children. PLoS One 8, e56800. Xu, J., Zhou, L., Wang, S., Zhu, J., Liu, T., Jia, Y., et al., 2018. Di-(2-ethylhexyl)-phthalate induces glucose metabolic disorder in adolescent rats. Environ. Sci. Pollut. Res. Int. 25, 3596–3607. Zeng, Q., Wei, C., Wu, Y., Li, K., Ding, S., Yuan, J., et al., 2013. Approach to distribution and accumulation of dibutyl phthalate in rats by immunoassay. Food Chem. Toxicol. 56, 18–27. Zhang, Y.H., Chen, B.H., Zheng, L.X., Wu, X.Y., 2003. Study on the level of phthalates in human biological samples. Zhonghua Yu Fang Yi Xue Za Zhi 37, 429–434. Zhao, J.F., Hsiao, S.H., Hsu, M.H., Pao, K.C., Kou, Y.R., Shyue, S.K., et al., 2016. Di-(2ethylhexyl) phthalate accelerates atherosclerosis in apolipoprotein E-deficient mice. Arch. Toxicol. 90, 181–190.
Prenatal and childhood exposure to phthalate diesters and thyroid function in a 9year follow-up birth cohort study: Taiwan maternal and infant cohort study. Epidemiology 28 (Suppl. 1) (S10-S8). Itoh, N., 2014. FGF21 as a hepatokine, adipokine, and myokine in metabolism and diseases. Front. Endocrinol. 5. James-Todd, T., Stahlhut, R., Meeker, J.D., Powell, S.G., Hauser, R., Huang, T., et al., 2012. Urinary phthalate metabolite concentrations and diabetes among women in the National Health and Nutrition Examination Survey (NHANES) 2001-2008. Environ. Health Perspect. 120, 1307–1313. Jeanson, Y., Ribas, F., Galinier, A., Arnaud, E., Ducos, M., Andre, M., et al., 2016. Lactate induces FGF21 expression in adipocytes through a p38-MAPK pathway. Biochem. J. 473, 685–692. Johnson, C.L., Weston, J.Y., Chadi, S.A., Fazio, E.N., Huff, M.W., Kharitonenkov, A., et al., 2009. Fibroblast growth factor 21 reduces the severity of cerulein-induced pancreatitis in mice. Gastroenterology 137, 1795–1804. Kharitonenkov, A., Shiyanova, T.L., Koester, A., Ford, A.M., Micanovic, R., Galbreath, E.J., et al., 2005. FGF-21 as a novel metabolic regulator. J. Clin. Invest. 115, 1627–1635. Kim, J.H., Park, H.Y., Bae, S., Lim, Y.H., Hong, Y.C., 2013. Diethylhexyl phthalates is associated with insulin resistance via oxidative stress in the elderly: a panel study. PLoS One 8, e71392. Koch, H.M., Christensen, K.L., Harth, V., Lorber, M., Bruning, T., 2012. Di-n-butyl phthalate (DnBP) and diisobutyl phthalate (DiBP) metabolism in a human volunteer after single oral doses. Arch. Toxicol. 86, 1829–1839. Kohn, A.D., Summers, S.A., Birnbaum, M.J., Roth, R.A., 1996. Expression of a constitutively active Akt Ser/Thr kinase in 3T3-L1 adipocytes stimulates glucose uptake and glucose transporter 4 translocation. J. Biol. Chem. 271, 31372–31378. Lee, K.I., Chiang, C.W., Lin, H.C., Zhao, J.F., Li, C.T., Shyue, S.K., et al., 2016. Maternal exposure to di-(2-ethylhexyl) phthalate exposure deregulates blood pressure, adiposity, cholesterol metabolism and social interaction in mouse offspring. Arch. Toxicol. 90, 1211–1224. Leney, S.E., Tavare, J.M., 2009. The molecular basis of insulin-stimulated glucose uptake: signalling, trafficking and potential drug targets. J. Endocrinol. 203, 1–18. Manteiga, S., Lee, K., 2017. Monoethylhexyl phthalate elicits an inflammatory response in adipocytes characterized by alterations in lipid and cytokine pathways. Environ.
254
Contents lists available at ScienceDirect
Toxicology in Vitro journal homepage: www.elsevier.com/locate/toxinvit
Fibroblast growth factor 21 secretion enhances glucose uptake in mono(2ethylhexyl)phthalate-treated adipocytes Jhih-Wei Hsua,b, Szu-Ching Yeha, Feng-Yuan Tsaia, Hsin-Wei Chena, Tsui-Chun Tsoua, a b
T
⁎
National Institute of Environmental Health Sciences, National Health Research Institutes, Zhunan, Miaoli 350, Taiwan Department of Life Science, National Central University, Zhongli, Taoyuan City 320, Taiwan
A R T I C LE I N FO
A B S T R A C T
Keywords: Endocrine disruptor Fat cells FGF21 Glucose metabolism MEHP Plasticizers
Previous studies revealed that cellular accumulation of mono(2-ethylhexyl)phthalate (MEHP) disturbed energy metabolism in adipocytes, where glucose uptake was significantly increased. The present study aimed to determine the mechanisms underlying the increased glucose uptake. MEHP-treated 3T3-L1 adipocytes exhibited a significantly increased glucose uptake activity. Immunoblot analysis suggested that the insulin-induced signals were not responsible for the increased glucose uptake. qPCR analysis revealed that both Glut1 and Glut4 genes were highly expressed during adipogenesis; Glut1 mRNA levels in MEHP-treated adipocytes were significantly increased. Moreover, MEHP-treated adipocytes exhibited significantly increased levels of fibroblast growth factor 21 (FGF21) in both mRNA and secreted protein. FGF21 is a peptide hormone with pleiotropic effects on regulation of insulin sensitivity and glucose/lipid homeostasis. We found that MEHP, FGF21, and lactate in culture medium together enhanced Fgf21 gene expression in MEHP-treated adipocytes. FGF21 signaling requires fibroblast growth factor receptor (FGFR) and βKlotho. Fgfr family and βKlotho genes were actively expressed during adipogenesis; mRNA levels of Fgfr3 and Fgfr4 genes in MEHP-treated adipocytes were significantly increased. Roles of FGF21/FGFR and phosphoinositide 3-kinase (PI3K)/AKT signal axes in regulation of glucose uptake were determined. We demonstrated that FGF21/FGFR signals played the major roles in up-regulation of the basal glucose uptake in MEHP-treated adipocytes. The in vitro evidence suggests that cellular FGF21 secretion enhances the basal glucose uptake in MEHP-treated adipocytes.
1. Introduction Epidemiological studies revealed the association between phthalate exposure and prevalence of metabolic diseases including obesity and its complications, e.g., insulin resistance and type 2 diabetes mellitus (James-Todd et al., 2012; Kim et al., 2013; Wang et al., 2013). In the Taiwan Maternal and Infant Cohort Study (TMICS), the early life exposure to phthalates was associated with decreased thyroid hormone levels in young children (Huang et al., 2017), in which the association is evident especially for di(2-ethylhexyl)phthalate (DEHP). Animal studies showed that maternal exposure to DEHP deregulated blood pressure, adiposity, cholesterol metabolism, and social interaction in mouse offspring (Lee et al., 2016). Dietary DEHP accelerated atherosclerosis in apolipoprotein E-deficient mice (Zhao et al., 2016), and
induced glucose metabolic disorder in rats (Martinelli et al., 2006; Xu et al., 2018). Both epidemiological and animal studies suggests the possible impacts of phthalate exposure on human health, especially metabolic disorder. Analysis of humans ingested with phthalates revealed that the majority of phthalates (> 90%) was excreted rapidly via urine in the first 24 h, where monoesters were the major metabolites (Koch et al., 2012). Adipose tissue is the major storage sites of lipophilic pollutants, e.g., phthalates. Previous studies revealed that phthalate accumulation in adipose tissues in both human (Zhang et al., 2003) and rats (Zeng et al., 2013). Therefore, despite the short biologic half-lives of phthalates, adipose tissue could be an important target of phthalates in vivo. Adipose tissue is a metabolically dynamic organ, acting as a crucial integrator of glucose homeostasis (Rosen and Spiegelman, 2006). Mice
Abbreviations: AIM, adipogenesis-inducing medium; AMM, adipogenesis-maintaining medium; [14C]2-DOG, 2-[1-14C]-deoxy-D-glucose; CHC, 2-cyano-3-(4-hydroxyphenyl)-2-propenoic acid; DEHP, di(2-ethylhexyl)phthalate; Dex, dexamethasone; DMEM-HG, Dulbecco's modified Eagle's medium with high glucose formula; Elk-1, Ets-like protein-1; FBS, fetal bovine serum; FGF21, fibroblast growth factor 21; FGFR, fibroblast growth factor receptor; IBMX, 3-isobutyl-1-methylxanthine; IRβ, insulin receptor β; IRS1, insulin receptor substrate 1; MEHP, mono(2-ethylhexyl)phthalate; mTORC1, mammalian target of rapamycin complex 1; PBS, phosphate-buffered saline; PI3K, phosphoinositide 3-kinase; PPARγ, peroxisome proliferator-activated receptor γ; RSK, ribosomal S6 kinase; SRF, serum response factor ⁎ Corresponding author at: National Institute of Environmental Health Sciences, National Health Research Institutes, 35 Keyan Road, Zhunan, Miaoli 350, Taiwan. E-mail address: [email protected] (T.-C. Tsou). https://doi.org/10.1016/j.tiv.2019.04.021 Received 9 November 2018; Received in revised form 29 March 2019; Accepted 17 April 2019 Available online 19 April 2019 0887-2333/ © 2019 Elsevier Ltd. All rights reserved.
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
were incubated in adipogenesis-inducing medium (AIM) (DMEM-HG with 0.25 μM dexamethasone, 0.5 mM 3-isobutyl-1-methylxanthine, 1.5 μM insulin, and 10% FBS) for 3 days (D3), in adipogenesis-maintaining medium (AMM) (DMEM-HG with 1.5 μM insulin and 10% FBS) for 2 days (D5), and then in DMEM-HG with 10% FBS for 6 days (D11). At D11, > 95% of the cells were fully differentiated into adipocytes (Fig. 1b). For MEHP treatments, the cells were treated with 10 to 300 μM MEHP as indicated or with 0.1% DMSO (as the solvent controls) for 6 days (from D5 to D11). Lipid droplet formation in adipocytes was determined with Oil Red O staining as described (Hsu et al., 2010) with minor modifications (Chiang et al., 2016). Oil Red O stock solution (0.5% of Oil Red O in isopropanol) was prepared and stored at 4 °C for use. Adipocytes were washed with phosphate-buffered saline (PBS) twice and fixed in 10% formalin solution for 1 h at room temperature. Oil Red O working solutions were freshly prepared by mixing 60% of Oil Red O stock solution with 40% of deionized water. The fixed cells were washed twice with PBS and stained with Oil Red O working solution for at least 1 h at room temperature. Following staining, cells were washed with deionized water for photography.
with perinatal DEHP exposure resulted in obesity in offspring (Hao et al., 2012). In vitro studies revealed that mono(2-ethylhexyl)phthalate (MEHP), the primary metabolite of DEHP, promoted adipogenesis of adipocytes via peroxisome proliferator-activated receptor γ (PPARγ) activation (Feige et al., 2007; Campioli et al., 2011). Moreover, MEHP treatments altered lipid metabolism (Ellero-Simatos et al., 2011) and energy metabolism (e.g., glucose uptake) (Chiang et al., 2016; Chiang et al., 2017), and elicited inflammatory responses (Manteiga and Lee, 2017) in adipocytes. The studies clearly suggest that MEHP not only enhances adipogenesis via PPARγ but also exhibits significant effects on various biological activities in adipocytes. Fibroblast growth factor 21 (FGF21), a peptide hormone, is most abundantly expressed in liver and is also detected in adipose tissue, skeletal muscle (Itoh, 2014), and pancreas (Johnson et al., 2009). Animal studies revealed that Fgf21-transgenic mice were resistant to dietinduced obesity, and administration of FGF21 reduced plasma levels of glucose and triglycerides in both ob/ob and db/db mice (Kharitonenkov et al., 2005). Fgf21 was transcriptionally up-regulated in adipose tissue of db/db mice treated with PPARγ agonists (Muise et al., 2008). Moreover, in vitro evidence indicated that FGF21 stimulated glucose uptake and modulated GLUT1 expression in adipocytes (Kharitonenkov et al., 2005) and βKlotho was required for FGF21 signaling through fibroblast growth factor receptors (FGFR) (Suzuki et al., 2008). These studies together suggest FGF21 as a pivotal glucose regulator in adipocytes. We previously demonstrated that 3T3-L1 adipocytes treated with MEHP ≥30 μM for 6 days exhibited a markedly increased glucose uptake activity (Chiang et al., 2016). Moreover, it was noted that significantly increased FGF21 levels were detected in both DEHP-treated mice and MEHP-treated adipocytes (Chiang et al., 2017). The in vivo and in vitro evidence suggested the potential involvement of FGF21 in systemic regulation of glucose homeostasis in response to DEHP/MEHP treatments. The present in vitro study explored how MEHP enhanced glucose uptake in adipocytes via FGF21 and its associated signaling molecules.
2.3. Glucose uptake assay Following MEHP treatments and/or inhibition of PI3K, AKT, or FGFR as indicated, glucose uptake activity in 3T3-L1 adipocytes was determined as previously described (Chiang et al., 2016). Briefly, adipocytes were starved in serum-free DMEM-HG for 4 h, washed twice with Krebs–Ringer phosphate buffer (128 mM NaCl, 4.7 mM KCl, 1.65 mM CaCl2, 2.5 mM MgSO4, and 5 mM Na2HPO4, pH 7.4, 37 °C), and then maintained in the same buffer for glucose uptake assay. The cells were left untreated (for determination of the basal glucose uptake activity) or treated with insulin (100 nM) (for determination of the insulin-induced glucose uptake activity) for 10 min and then glucose uptake was started by addition of [14C]2-DOG (0.1 μCi/well) for 10 min at 37 °C. Cells were washed three times with ice-cold PBS and lysed in 800 μl of lysis solution (0.5 M NaOH and 0.1% SDS). For determining cellular uptake of [14C]2-DOG, lysate samples were assayed with a Topcount NXT Scintillation Counter (Packard Instrument Company, Meriden, CT, USA).
2. Methods and materials 2.1. Chemicals and cells
2.4. qPCR analysis of gene expression MEHP (M542490) was purchased from Toronto Research Chemicals Inc. (North York, Canada). 2-[1-14C]-deoxy-D-glucose ([14C]2-DOG) (NEC495A050UC) was purchased from PerkinElmer Life Sciences (Boston, MA, USA). Insulin (I6634), 3-isobutyl-1-methylxanthine (IBMX) (I5879), dexamethason (Dex) (D4902), Oil Red O (O0625), 10% formalin solution (HT5012), PD161570 (an FGFR tyrosine kinase inhibitor) (PZ0109), LY294002 [a phosphoinositide 3-kinase (PI3K) inhibitor] (L9908), and AKTi (an AKT1/2 inhibitor) (A6730) were purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA). The monocarboxylic acid transport inhibitor 2-cyano-3-(4-hydroxyphenyl)2-propenoic acid (CHC) (ab146008) was from Abcam (Cambridge, MA, USA). Dulbecco's modified Eagle's medium with high glucose formula (DMEM-HG) (12800–017), bovine serum (16170–078), penicillin/ streptomycin (15140–122), and fetal bovine serum (FBS) (10091–148) were from Gibco/Invitrogen (Carlsbad, CA, USA). 3T3-L1 cells (BCRC60159) were purchased from the Bioresources Collection and Research Center, Food Industry Research and Development Institute (Hsinchu, Taiwan). The cells are routinely maintained in DMEM-HG with 10% bovine serum and 1% penicillin/streptomycin in a humidified atmosphere containing 5% CO2 at 37 °C.
3T3-L1 cells during or after adipogenesis were collected as indicated times and gene expression of aP2, C/EBPα, Fgf21, Fgfr1, Fgfr2, Fgfr3, Fgfr4, Glut1, Glut4, βKlotho, PPARγ, Srebf1, and β–actin by using qPCR analysis of mRNA levels. Following RNA isolation and cDNA synthesis, quantification of cDNA with the LightCycler-FastStart DNA Master SYBR Green I system (Roche Diagnostics GmbH) was performed by using a LightCycler 480 instrument (Roche Diagnostics GmbH). DNA sequences of mouse-specific primer sets for qPCR were summarized (Table S1 in Supplementary Material). Melting curve analysis was used to confirm the amplification specificity. All qPCR assays were performed in duplicate for at least three times (n ≥ 3) with β–actin as the internal control. Relative quantification analysis was applied to determine the ratio change of a specific gene expression with the LightCycler analysis software. 2.5. Immunoblot analysis Following treatments, the cells were lysed and cell lysates were subjected to SDS-PAGE and immunoblot analysis according to standard protocols. The blot was incubated with an antibody against pIRβ(Y1150/Y1151) (#04–299) (Millipore), IRβ (#3025) (Cell Signaling), p-IRS1(Y632) (sc-17196) (Santa Cruz), IRS1 (#2382) (Cell Signaling), p-AKT(T308) (#2965) (Cell Signaling), p-AKT(S473) (#9271) (Cell Signaling), AKT (sc-1618) (Santa Cruz), α–tubulin (as the loading controls of total protein) (GTX112141) (GeneTex), βKlotho
2.2. Induction of adipogenesis and phthalate treatments Induction of adipogenesis in 3T3-L1 cells was performed as previously described (Hsu et al., 2010; Chiang et al., 2016) with modifications as shown in Fig. 1a. Two days postconfluence (D0), the cells 247
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 1. MEHP-treated adipocytes exhibit an increased glucose uptake activity. (a) Adipogenesis protocol and cell treatments in this study. Following the standard adipogenesis protocol as described in “Materials and method”, 3T3-L1 cells were treated 0.1% DMSO (the control) or 10 to 300 μM MEHP from D5 to D11. (b) Representative bright field (D0 and D11) and Oil Red O-stained (D11) images of the control adipocytes are presented. (c) mRNA levels of aP2, C/EBPα, PPARγ, and Srebf1 in the control adipocytes during adipogenesis were analyzed with qPCR. Results are expressed as fold changes (vs. D0) and presented as means ± SD (n ≥ 4). (d) At D11, both the basal and 100 nM insulin-induced glucose uptake levels in adipocytes were determined. Results are expressed as fold changes (vs. the basal glucose uptake in the controls) and presented as means ± SD (n = 4). ***p < 0.001. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
2.8. Statistical analysis
(GTX122197) (GeneTex), or PPARγ (sc-7273) (Santa Cruz) at 4 °C overnight. HRP conjugated secondary antibodies were then used to reveal the specific protein bands with Western Lightning Chemiluminescence Reagent Plus (PerkinElmer Life Sciences).
Each experiment was performed independently at least three times. All analyses were performed with GraphPad Prism version 7.0 for Windows (GraphPad Software, La Jolla, CA, USA). All data were presented as means ± SD. Comparisons between groups were performed with ANOVA followed by Bonferroni's multiple comparison test. Differences were considered statistically significant when p value < 0.05.
2.6. ELISA analysis of FGF21 in conditioned medium Following treatments, culture medium was collected and stored at −80 °C until analysis. FGF21 levels in culture medium were determined with the Quantikine ELISA Kit (MF2100) (R&D Systems, Minneapolis, MN, USA) according to the manufacturer's instruction manual.
3. Results and discussion 3.1. MEHP-treated adipocytes exhibit an increased glucose uptake activity
2.7. Measurement of lactate in conditioned media Adipogenesis of 3T3-L1 cells was determined by using oil droplet formation and expression of adipogenic marker genes. Following the standard adipogenesis protocol (Fig. 1a), oil droplet formation in cells at D11 was evident as observed with light microscopy combined with Oil Red O staining (Fig. 1b). qPCR analysis revealed markedly elevated mRNA levels of adipogenic marker genes, including aP2, C/EBPα, PPARγ, and Srebf1, in cells during adipogenesis (Fig. 1c). The adipogenesis protocol ensured the successful adipogenic differentiation of 3T3-L1 cells in this study. To determine MEHP effects on glucose uptake, 3T3-L1 adipocytes
Following the standard adipogenesis protocol as described in Fig. 1a, 3T3-L1 cells were treated with 0.1% DMSO (the vehicle controls) or MEHP (30, 100, and 300 μM) from D5 to D11. At D11, 1 ml of conditioned medium samples were collected for measurements of lactate levels with a colorimetric lactate assay (Randox Laboratories Ltd.). Culture medium without cells incubation was used to determine the background lactate concentration. Lactate levels in conditioned media represented the cellular production of lactate from D9 to D11.
248
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
were treated with 30 and 100 μM MEHP for 6 days (D5 to D11), and then both the basal and insulin-induced glucose uptakes were analyzed. Results in Fig. 1d indicated that adipocytes treated with 30 and 100 μM MEHP exhibited a higher activity in both the basal and insulin-induced glucose uptake than the control adipocytes. Moreover, insulin stimulated glucose uptake in the controls (from 1.00 to 2.80 folds, p < 0.001), 30 μM MEHP-treated (from 1.34 to 3.81 folds, p < 0.001), and 100 μM MEHP-treated adipocytes (from 2.41 to 4.03 folds, p < 0.001). Therefore, the results indicate that both the basal and insulin-induced glucose uptake in MEHP-treated adipocytes is significantly higher than that in the control adipocytes.
Cellular glucose uptake mainly involves two major glucose transporters, GLUT1 and GLUT4 (Ebeling et al., 1998). GLUT1 is ubiquitously distributed in different tissues and facilitates the insulin-independent glucose uptake. GLUT4 is responsible for the majority of glucose transport into muscle and adipose cells in an insulin-dependent manner. In this study, expression of Glut1 and Glut4 genes in adipocytes was determined by using qPCR analysis. Results in Fig. 2b (top panel) revealed that mRNA levels of both Glut1 and Glut4 genes were highly expressed in 3T3-L1 cells during adipogenesis. Moreover, mRNA levels of Glut1, but not Glut4, in the adipocytes treated with 30 and 100 μM MEHP for 6 days (D5 to D11) were significantly increased by 1.52 folds (p < 0.05) and 1.94 folds (p < 0.001) (vs. the controls), respectively (Fig. 2b, bottom panel). Results in Fig. 2b suggested the potential contribution of GLUT1 to the higher glucose uptake in MEHP-treated adipocytes.
3.2. The insulin-induced signals are not responsible for the increased glucose uptake in MEHP-treated adipocytes In adipose tissue, insulin induces glucose uptake from the circulation via promoting GLUT4 trafficking to the plasma membrane (Satoh, 2014). In the process, insulin stimulates a signaling cascade composed of insulin receptor β (IRβ), insulin receptor substrate 1 (IRS1), PI3K, and AKT (Satoh, 2014). To determine the involvement of the insulininduced signaling cascade in the increased glucose uptake in MEHPtreated adipocytes, both the control and MEHP-treated adipocytes were treated with 100 nM insulin for 10 min. Then, immunoblot analysis of protein phosphorylation was used to determine the activation of insulin-induced signals, including IRβ, IRS1, and AKT. Results in Fig. 2a revealed that the insulin treatment caused marked increases in phosphorylation of IRβ, IRS1, and AKT in control adipocytes; the insulininduced IRβ phosphorylation was further enhanced in adipocytes treated with MEHP (10 to 300 μM). However, both total protein and the insulin-induced phosphorylation levels of IRS1 and AKT in MEHPtreated adipocytes were lower than that in control adipocytes. The results together suggest that the insulin-induced signals are not responsible for the increased glucose uptake in MEHP-treated adipocytes.
3.3. MEHP-treated adipocytes secrete significantly increased amounts of FGF21 protein In this study, both Glut1 expression (Fig. 2b, bottom panel) and glucose uptake (Fig. 1d) in MEHP-treated adipocytes were up-regulated. FGF21 has been previously demonstrated to enhance glucose uptake in adipocytes by induction of Glut1 expression (Kharitonenkov et al., 2005), via activation of the serum response factor (SRF)/Ets-like protein-1 (Elk-1) (Ge et al., 2011). Because of the potential roles of FGF21 in the increased glucose uptake in MEHP-treated adipocytes, it was critical to know the Fgf21 gene expression patterns in adipocytes. In the present study, time-course of Fgf21 mRNA levels in 3T3-L1 cells during adipogenesis were determined with qPCR. Results in Fig. 3a revealed that Fgf21 mRNA levels in the control adipocytes during adipogenesis at D5, D7, D9, and D11 were increased by 5.6, 5.2, 6.8, and 5.6 folds (vs. D0), respectively. Importantly, Fgf21 mRNA levels in MEHP-treated adipocytes were markedly higher than those in the
Fig. 2. The insulin-induced signals and gene expression of Glut1 and Glut4 in adipocytes. (a) For determining activation of the insulin-induced signals, both the control and MEHP-treated adipocytes (at D11) were starved in serum-free DMEM-HG for 4 h and then treated with 100 nM insulin for 10 min. Then, activation of the insulin-induced signals, including IRβ, IRS1, and AKT, was determined with immunoblot analysis of total protein and their phosphorylation levels, with α-tubulin as the loading controls. (b) mRNA levels of Glut1 and Glut4 in the control cells during adipogenesis (at D0, D5, D8, and D11) (top panel) as well as in the control and MEHP-treated adipocytes after adipogenesis (at D11) (bottom panel) was determined with qPCR. Results are expressed as fold changes (vs. D0 in top panel; vs. the control in bottom panel) and presented as means ± SD (n ≥ 3). *p < 0.05 and ***p < 0.001 (vs. the control). 249
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
2008). In our previous microarray study, Fgf21 expression was significantly up-regulated in adipocytes subjected to both MEHP and the PPARγ agonist rosiglitazone treatments (Chiang et al., 2017). Taken together, these results indicate that MEHP stimulates FGF21 protein secretion from adipocytes via activation of PPARγ. Phthalates are lipophilic and easy to accumulate in adipose tissue. We previously demonstrated that, as treated with MEHP in vitro, adipocytes accumulated MEHP from culture medium (Chiang et al., 2016). In this study, it was important to further clarify whether the MEHPaccumulated adipocytes could sustainably secrete such high levels of FGF21 protein. Following treatments with 100 μM MEHP for 4 days (D5 to D9), 3T3-L1 adipocytes were switched to culture medium without MEHP at D9; then FGF21 protein levels in conditioned medium were monitored at D11, D13, and D15. Results in Fig. 3c revealed that the MEHP withdrawal resulted in a dramatic dropping of FGF21 from 2516.0 (D9) to 343.5 pg/ml (D11). Clearly, the FGF21 level was down to the control level in two days, indicating that the MEHP-accumulated adipocytes could not sustainably secrete such high levels of FGF21 protein. 3.4. Expression of Fgfr family and βKlotho genes in 3T3-L1 adipocytes Results in Fig. 3 provided clear evidence that MEHP-treated adipocytes secreted significantly increased amounts of FGF21 protein in culture medium. FGF21 stimulates glucose uptake in adipocytes via FGFR and the cofactor βKlotho (Ogawa et al., 2007). In this study, gene expression profiles of Fgfr family and βKlotho in both the control and MEHP-treated 3T3-L1 adipocytes were determined. qPCR analysis revealed that mRNA levels of Fgfr family, including Fgfr1, Fgfr2, Fgfr3, and Fgfr4, in the control adipocytes during adipogenesis were increased in varying degrees (Fig. 4a, left panel), where the relative mRNA levels (vs. D0 of each gene) at D11 were Fgfr3 (4.7 folds) > Fgfr1 (2.9 folds) > Fgfr4 (2.4 folds) > Fgfr2 (1.1 folds). At D11, both Fgfr3 and Fgfr4 mRNA levels in adipocytes treated with 100 μM MEHP were significantly higher than that in the controls (Fig. 4a, right panel). Moreover, βKlotho mRNA levels in the control adipocytes during adipogenesis were markedly increased by 158, 1202, and 2252 folds (vs. D0) at D5, D8, and D11, respectively (Fig. 4b, left panel). At D11, no significant difference in βKlotho mRNA levels was detected between the control and MEHP-treated adipocytes (Fig. 4b, right panel). The βKlotho expression results were further confirmed with immunoblot analysis (see Fig. S1 in Supplemental Material). Results in Fig. 4 suggested the involvement of FGFR3 and FGFR4 in the increased glucose uptake in MEHP-treated adipocytes. Our previous microarray study revealed the up-regulation of Fgfr3 and Fgfr4 genes in adipocytes subjected to both MEHP and the PPARγ agonist rosiglitazone treatments (Chiang et al., 2017), suggesting that MEHP stimulates the gene expression via PPARγ.
Fig. 3. MEHP-treated adipocytes secrete increased amounts of FGF21 protein. Following the standard adipogenesis protocol, 3T3-L1 cells were treated 0.1% DMSO (the control) or MEHP (30 and 100 μM) from D5 to D11. (a) Fgf21 mRNA levels in adipocytes during adipogenesis were analyzed with qPCR. Results are expressed as fold change (vs. D0 of the control) and presented as means ± SD (n = 3). (b) Cellular secretion of FGF21 protein in culture medium was analyzed by using ELISA. FGF21 protein levels are presented as means ± SD (n = 4). (c) At D9, culture medium of both the control and MEHP-treated adipocytes was switched to complete medium without MEHP for continuous culture, conditioned medium samples were collected at D9, D11, D13, and D15 for analysis of FGF21 protein. FGF21 protein levels are presented as means ± SD (n = 4). *p < 0.05 and ***p < 0.001 (as indicated).
3.5. MEHP, FGF21, and lactate together enhance Fgf21 expression in MEHP-treated adipocytes MEHP-treated adipocytes secrete significant amounts of FGF21 protein in culture medium (Fig. 3b). In addition to enhancing glucose uptake in adipocytes (Kharitonenkov et al., 2005; Ge et al., 2011), FGF21 also stimulates its own gene expression via activation of mammalian target of rapamycin complex 1 (mTORC1)/ribosomal S6 kinase (RSK) (Minard et al., 2016). In this study, conditioned medium from MEHP-treated adipocytes (CM) was collected to test its effect on Fgf21 expression. 3T3-L1 adipocytes were treated with the CM for 16 h, using fresh medium (FM) and fresh medium with 100 μM MEHP (FM + MEHP) as the controls. Then, Fgf21 mRNA levels were determined with qPCR. As shown in Fig. 5a, the CM and the FM + MEHP caused significant increases in Fgf21 mRNA by 26.0 folds (vs. FM) (p < 0.001) and 9.1 folds (vs. FM) (p < 0.05), respectively. Because both CM and FM + MEHP contained the same level of MEHP (100 μM), results in Fig. 5a were interpreted that the MEHP in CM contributed to
control adipocytes in a MEHP dose-dependent manner. For example at D11, Fgf21 mRNA levels in adipocytes treated with 30 and 100 μM MEHP rose dramatically from 5.6 folds to 21.0 folds (p < 0.05) and 241.0 folds (p < 0.001), respectively (Fig. 3a). FGF21 is a secreted protein. Cellular secretion of FGF21 protein in culture medium were also analyzed with ELISA. Results in Fig. 3b showed that during adipogenesis the pattern of FGF21 protein secretion was very similar to that of Fgf21 mRNA. At D11, FGF21 protein levels from adipocytes treated with 30 and 100 μM MEHP rose significantly from 174 pg/ml to 444 pg/ml (p < 0.05) and 1872 pg/ml (p < 0.001), respectively (Fig. 3b). Therefore, it is clear that MEHPtreated adipocytes secrete significantly increased amounts of FGF21 protein in vitro. Previous study found that adipose Fgf21 was upregulated by PPARγ and thus altered metabolic status (Muise et al., 250
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 4. Expression of Fgfr family and βKlotho genes in 3T3-L1 adipocytes. During adipogenesis, 3T3-L1 cells were treated with 0.1% DMSO (the control) or MEHP (30 and 100 μM) from D5 to D11. Gene expression of (a) Fgfr family and (b) βKlotho was determined by qPCR analysis of mRNA levels. Gene expression in the control was monitored at D0, D5, D8, and D11 (left panels, a and b). At D11, mRNA levels in the control and MEHP-treated adipocytes were compared (right panels, a and b). Results are expressed as fold change (vs. D0 in left panels) (vs. the control in right panels) and presented as means ± SD (n ≥ 3). *p < 0.05 and ***p < 0.001 (vs. D0) in left panels; *p < 0.05 and **p < 0.01 (vs. the control) in right panels.
inhibited in the presence of the protein synthesis inhibitor cycloheximide. The study suggests that FGF21 effect on glucose uptake requires transcriptional activation. In the present study, the involvement of the two signal axes in activation of glucose uptake in adipocytes was determined by blocking the pathways with specific inhibitors, i.e., PD161570 (an FGFR tyrosine kinase inhibitor), LY294002 (a PI3K inhibitor), and AKTi (an AKT1 and AKT2 dual kinase inhibitor). Both the control and 100 μM MEHPtreated adipocytes were pretreated with PD16570 (10 μM) for 4 h, LY294002 (50 μM) for 10 min, or AKTi (30 μM) for 10 min and then both the basal and insulin-induced glucose uptakes in adipocytes were determined. Results in Fig. 6a revealed that inhibition of PI3K (with LY294002) or AKT (with AKTi) completely blocked the insulin-induced glucose uptake in both the control and MEHP-treated adipocytes. The results confirmed the well-recognized roles of PI3K/AKT in the insulininduced glucose uptake. It was also noted that the inhibition of PI3K or AKT also fully abolished the basal glucose uptake in both the control and MEHP-treated adipocytes (Fig. 6a). Although the insulin-induced pathway was not responsible for the increased glucose uptake in MEHPtreated adipocytes (Fig. 2a), the present results suggested that the functional PI3K/AKT was essential for the basal glucose uptake in adipocytes. FGF21/FGFR signals activate the basal glucose uptake in adipocytes in an insulin-independent manner (Kharitonenkov et al., 2005), mainly via up-regulation of GLUT1 protein expression (Ge et al., 2011). Unexpectedly, inhibition of FGFR by PD16570 attenuated the insulin-induced glucose uptake in the control and MEHP-treated adipocytes by 26% (p > 0.05) and 30% (p < 0.05), respectively (Fig. 6a). The mild inhibitory effect of PD16570 on the insulin-induced glucose uptake implicated the potential crosstalk between FGF21/FGFR and insulin/ PI3K/AKT signal axes. Moreover, it was also noted that FGFR inhibition of by PD16570 pretreatment for 4 h was not able to significantly block the basal glucose uptake in both the control and MEHP-treated adipocytes (Fig. 6a). The reason could be that, before subjected to the PD16570 treatments at D11, the 100 μM MEHP-treated adipocytes had been exposed to MEHP, FGF21, and lactate in culture medium for 6 days (D5 to D11) (Figs. 3b and 5b) and thus already had significantly increased amounts of GLUT1 (Fig. 2b, bottom panel) for the basal glucose uptake. To further determine the roles of FGFR in regulation of the basal glucose uptake in MEHP-treated adipocytes, a long-term inhibition of FGFR was necessary. During adipogenesis, 3T3-L1 cells were treated
32.4% of the Fgf21 mRNA induction by CM. Lactate was recently found to induce Fgf21 gene expression in mouse primary white adipocytes (Jeanson et al., 2016). In our previous study, the MEHP-treated 3T3-L1 adipocytes secreted significant amounts of lactate in culture medium (Chiang et al., 2016). Indeed, results in Fig. 5b revealed that at D11 lactate concentration in conditioned medium of 3T3-L1 adipocytes treated with 30 and 100 μM MEHP was significantly increased from 4.23 mM to 6.93 mM (p < 0.001) and 12.02 mM (p < 0.001), respectively. Reducing cellular lactate import with the monocarboxylate transporter inhibitor CHC was shown to block the lactate-induced Fgf21 expression in adipocytes (Jeanson et al., 2016). In the present study, CHC was adopted to determine the effects of lactate in culture medium on Fgf21 expression. During adipogenesis, the 100 μM-treated 3T3-L1 adipocytes were subjected to 2 mM CHC treatments for 48 h (D9 to D11), and then Fgf21 mRNA levels were analyzed with qPCR. Results in Fig. 5c revealed that the CHC treatments significantly inhibited Fgf21 mRNA levels in MEHP-treated adipocytes by 19.1% (from 24.6 folds to 20.1 folds) (p < 0.05), indicating that lactate in culture medium contributed to 19.1% of the Fgf21 mRNA induction. Taken together, during adipogenesis in the presence of MEHP, 3T3L1 adipocytes secreted significant amounts of FGF21 (Fig. 3b) and lactate (Fig. 5b) in culture medium. Results in Figs. 3 and 5 suggest that MEHP, FGF21, and lactate in culture medium together activate a robust expression of Fgf21 gene in MEHP-treated adipocytes; the possible involvements of other metabolites and/or factors in conditioned medium cannot be excluded. 3.6. Roles of FGF21/FGFR and insulin/PI3K/AKT signal axes in regulation of glucose uptake in adipocytes FGF21/FGFR (Kharitonenkov et al., 2005) and insulin/PI3K/AKT (Kohn et al., 1996; Hernandez et al., 2001) signal axes play critical and differential roles in regulation of glucose uptake in adipocytes, mainly via the basal growth-related glucose transporter GLUT1 and the insulinresponsive glucose transporter GLUT4, respectively. The insulin/PI3K/ AKT signal axis stimulates the rapid glucose uptake via direct translocation of GLUT4 from intracellular membranes to the cell surface (Leney and Tavare, 2009); the process requires no transcriptional activation. In contrast to the rapid response induced by insulin, at least 4 h of FGF21 treatment is requires for adipocytes to exhibit an increased glucose uptake activity (Kharitonenkov et al., 2005), which is markedly 251
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 6. Involvement of PI3K/AKT and FGFR in regulation of glucose uptake in adipocytes. (a) For short-term inhibition, both the control and 100 μM MEHPtreated 3T3-L1 adipocytes were treated with 10 μM PD161570 (an FGFR inhibitor) for 4 h, 50 μM LY294002 (a PI3K inhibitor) for 10 min, or 30 μM AKTi (an AKT inhibitor) for 10 min. Then, the basal and insulin-induced glucose uptake in both the control and MEHP-treated adipocytes was determined. (b) For long-term inhibition of FGFR, both the control and MEHP-treated adipocytes were treated with 10 μM PD161570 for 6 days (D5 to D11). Then, the basal glucose uptake in both the control and MEHP-treated adipocytes was determined. Results are expressed as fold changes (vs. the basal glucose uptake in the controls with no inhibitor). Data are presented as mean ± SD (n = 3). *p < 0.05 and ***p < 0.001.
Fig. 5. FGF21, MEHP, and lactate in culture medium together activate Fgf21 gene expression in MEHP-treated adipocytes. (a) Conditioned medium was collected from the 100 μM MEHP-treated adipocytes at D11. 3T3-L1 adipocytes were treated with conditioned medium (CM) for 16 h, with fresh medium (FM) and fresh medium added with 100 μM MEHP (FM + MEHP) as the negative and MEHP controls, respectively. Fgf21 mRNA levels in adipocytes were analyzed with qPCR. Results are expressed as fold change (vs. FM) and presented as means ± SD (n = 3). *p < 0.05 and ***p < 0.001 (vs. FM). (b) Conditioned medium was collected from the control, 30 μM MEHP-treated, and 100 μM MEHP-treated adipocytes at D11. Lactate levels in conditioned medium were determined and presented as means ± SD (n = 5). ***p < 0.001 (vs. the control). (c) 100 μM MEHP-treated adipocytes were treated with the monocarboxylate transporter inhibitor CHC (2 mM) from D9 to D11 (100 μM MEHP + CHC). At D11, Fgf21 mRNA levels in the control and 100 μM MEHP-treated adipocytes with or without the inhibitor treatment were determined with qPCR. Results are expressed as fold change (vs. the control) and presented as means ± SD (n = 3). *p < 0.05 and ***p < 0.001 (vs. the control or as indicated).
the basal glucose uptake in MEHP-treated adipocytes. 4. Conclusions The study aimed to address the roles of FGF21 in regulation of glucose uptake in MEHP-treated adipocytes. Following adipogenesis in the presence of MEHP, MEHP-treated 3T3-L1 adipocytes exhibited a higher glucose uptake activity than the untreated controls. However, immunoblot analysis revealed that the insulin-induced signals were not responsible for the increased glucose uptake in MEHP-treated adipocytes. The MEHP-treated adipocytes exhibited increased mRNA levels of Glut1, Fgf21, Fgfr3, and Fgfr4; the cells also secreted significantly increased amounts of FGF21 protein. MEHP, FGF21, and lactate in culture medium together enhanced Fgf21 gene expression in MEHPtreated adipocytes. Roles of FGF21/FGFR and PI3K/AKT signal axes in regulation of glucose uptake in adipocytes were determined. We found that FGF21/FGFR signals played the major roles in up-regulation of the basal glucose uptake in MEHP-treated adipocytes. Taken together, on the basis of results of this and previous studies, a proposed model for
with 100 μM MEHP in the presence of 10 μM PD16570 for 6 days (D5 to D11); then, at D11 the basal glucose uptake in both the control and MEHP-treated adipocytes was analyzed. Results in Fig. 6b revealed that inhibition of FGFR with PD16570 for 6 days completely blocked the basal glucose uptake in both the control and MEHP-treated adipocytes, suggesting the critical involvement of FGF21/FGFR in up-regulation of 252
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Fig. 7. Proposed schematic diagram for the increased glucose uptake in MEHP-treated 3T3-L1 adipocytes. MEHP binding to PPARγ triggers PPARγ activation and subsequent expression of PPARγ target genes, e.g., Fgf21, Fgfr3, and Fgfr4. In MEHP-treated adipocytes, the higher glucose uptake is accompanied with a proportional lactate secretion (Chiang et al., 2016; 2017); lactate in turn induces Fgf21 expression through a p38-MAPK pathway (Jeanson et al., 2016). FGF21 also stimulates its own gene expression via activation of the mTORC1/RSK pathway (Minard et al., 2016). Following secretion and accumulation in culture medium, FGF21 binds to FGFR3/βKlotho or FGFR4/βKlotho complex and then activates the receptors. The FGF21/FGFR signal axis enhances glucose uptake in adipocytes by induction of Glut1 expression (Kharitonenkov et al., 2005), via activation of the ERK1/2/RSK pathway and then the SRF/Elk-1 pathway (Ge et al., 2011).
References
the increased glucose uptake in MEHP-treated adipocytes is schematically summarized in Fig. 7.
Campioli, E., Batarseh, A., Li, J., Papadopoulos, V., 2011. The endocrine disruptor mono(2-ethylhexyl) phthalate affects the differentiation of human liposarcoma cells (SW 872). PLoS One 6, e28750. Chiang, H.C., Kuo, Y.T., Shen, C.C., Lin, Y.H., Wang, S.L., Tsou, T.C., 2016. Mono(2ethylhexyl)phthalate accumulation disturbs energy metabolism of fat cells. Arch. Toxicol. 90, 589–601. Chiang, H.C., Wang, C.H., Yeh, S.C., Lin, Y.H., Kuo, Y.T., Liao, C.W., et al., 2017. Comparative microarray analyses of mono(2-ethylhexyl)phthalate impacts on fat cell bioenergetics and adipokine network. Cell Biol. Toxicol. 33, 511–526. Ebeling, P., Koistinen, H.A., Koivisto, V.A., 1998. Insulin-independent glucose transport regulates insulin sensitivity. FEBS Lett. 436, 301–303. Ellero-Simatos, S., Claus, S.P., Benelli, C., Forest, C., Letourneur, F., Cagnard, N., et al., 2011. Combined transcriptomic-(1)H NMR metabonomic study reveals that monoethylhexyl phthalate stimulates adipogenesis and glyceroneogenesis in human adipocytes. J. Proteome Res. 10, 5493–5502. Feige, J.N., Gelman, L., Rossi, D., Zoete, V., Metivier, R., Tudor, C., et al., 2007. The endocrine disruptor monoethyl-hexyl-phthalate is a selective peroxisome proliferator-activated receptor gamma modulator that promotes adipogenesis. J. Biol. Chem. 282, 19152–19166. Ge, X., Chen, C., Hui, X., Wang, Y., Lam, K.S., Xu, A., 2011. Fibroblast growth factor 21 induces glucose transporter-1 expression through activation of the serum response factor/Ets-like protein-1 in adipocytes. J. Biol. Chem. 286, 34533–34541. Hao, C., Cheng, X., Xia, H., Ma, X., 2012. The endocrine disruptor mono-(2-ethylhexyl) phthalate promotes adipocyte differentiation and induces obesity in mice. Biosci. Rep. 32, 619–629. Hernandez, R., Teruel, T., Lorenzo, M., 2001. Akt mediates insulin induction of glucose uptake and up-regulation of GLUT4 gene expression in brown adipocytes. FEBS Lett. 494, 225–231. Hsu, H.F., Tsou, T.C., Chao, H.R., Shy, C.G., Kuo, Y.T., Tsai, F.Y., et al., 2010. Effects of arecoline on adipogenesis, lipolysis, and glucose uptake of adipocytes-A possible role of betel-quid chewing in metabolic syndrome. Toxicol. Appl. Pharmacol. 245, 370–377. Huang, H.B., Chuang, C.J., Su, P.H., Sun, C.W., Wang, C.J., Wu, M.T., et al., 2017.
Transparency document The Transparency document associated to this article can be found, in the online version. Conflicts of interest TCT reports grants and personal fees from Ministry of Science and Technology, (Taiwan) R.O.C. during the conduct of the study. JWH and HWC report personal fees from Ministry of Science and Technology, (Taiwan) R.O.C. during the conduct of the study. SCY and FYT have nothing to disclose. Acknowledgements This work was supported by grants from the Ministry of Science and Technology (105-2320-B-400-022, 106-2320-B-400-006, and 1072320-B-400-001) and the National Health Research Institutes (EM-105PP-03, EM-106-PP-03, and EM-107-PP-03) in Taiwan. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.tiv.2019.04.021. 253
Toxicology in Vitro 59 (2019) 246–254
J.-W. Hsu, et al.
Health Perspect. 125, 615–622. Martinelli, M.I., Mocchiutti, N.O., Bernal, C.A., 2006. Dietary di(2-ethylhexyl)phthalateimpaired glucose metabolism in experimental animals. Hum. Exp. Toxicol. 25, 531–538. Minard, A.Y., Tan, S.X., Yang, P., Fazakerley, D.J., Domanova, W., Parker, B.L., et al., 2016. mTORC1 is a major regulatory node in the FGF21 signaling network in adipocytes. Cell Rep. 17, 29–36. Muise, E.S., Azzolina, B., Kuo, D.W., El-Sherbeini, M., Tan, Y., Yuan, X., et al., 2008. Adipose fibroblast growth factor 21 is up-regulated by peroxisome proliferator-activated receptor gamma and altered metabolic states. Mol. Pharmacol. 74, 403–412. Ogawa, Y., Kurosu, H., Yamamoto, M., Nandi, A., Rosenblatt, K.P., Goetz, R., et al., 2007. BetaKlotho is required for metabolic activity of fibroblast growth factor 21. Proc. Natl. Acad. Sci. U. S. A. 104, 7432–7437. Rosen, E.D., Spiegelman, B.M., 2006. Adipocytes as regulators of energy balance and glucose homeostasis. Nature 444, 847–853. Satoh, T., 2014. Molecular mechanisms for the regulation of insulin-stimulated glucose uptake by small guanosine triphosphatases in skeletal muscle and adipocytes. Int. J. Mol. Sci. 15, 18677–18692. Suzuki, M., Uehara, Y., Motomura-Matsuzaka, K., Oki, J., Koyama, Y., Kimura, M., et al., 2008. betaKlotho is required for fibroblast growth factor (FGF) 21 signaling through FGF receptor (FGFR) 1c and FGFR3c. Mol. Endocrinol. 22, 1006–1014. Wang, H., Zhou, Y., Tang, C., He, Y., Wu, J., Chen, Y., et al., 2013. Urinary phthalate metabolites are associated with body mass index and waist circumference in Chinese school children. PLoS One 8, e56800. Xu, J., Zhou, L., Wang, S., Zhu, J., Liu, T., Jia, Y., et al., 2018. Di-(2-ethylhexyl)-phthalate induces glucose metabolic disorder in adolescent rats. Environ. Sci. Pollut. Res. Int. 25, 3596–3607. Zeng, Q., Wei, C., Wu, Y., Li, K., Ding, S., Yuan, J., et al., 2013. Approach to distribution and accumulation of dibutyl phthalate in rats by immunoassay. Food Chem. Toxicol. 56, 18–27. Zhang, Y.H., Chen, B.H., Zheng, L.X., Wu, X.Y., 2003. Study on the level of phthalates in human biological samples. Zhonghua Yu Fang Yi Xue Za Zhi 37, 429–434. Zhao, J.F., Hsiao, S.H., Hsu, M.H., Pao, K.C., Kou, Y.R., Shyue, S.K., et al., 2016. Di-(2ethylhexyl) phthalate accelerates atherosclerosis in apolipoprotein E-deficient mice. Arch. Toxicol. 90, 181–190.
Prenatal and childhood exposure to phthalate diesters and thyroid function in a 9year follow-up birth cohort study: Taiwan maternal and infant cohort study. Epidemiology 28 (Suppl. 1) (S10-S8). Itoh, N., 2014. FGF21 as a hepatokine, adipokine, and myokine in metabolism and diseases. Front. Endocrinol. 5. James-Todd, T., Stahlhut, R., Meeker, J.D., Powell, S.G., Hauser, R., Huang, T., et al., 2012. Urinary phthalate metabolite concentrations and diabetes among women in the National Health and Nutrition Examination Survey (NHANES) 2001-2008. Environ. Health Perspect. 120, 1307–1313. Jeanson, Y., Ribas, F., Galinier, A., Arnaud, E., Ducos, M., Andre, M., et al., 2016. Lactate induces FGF21 expression in adipocytes through a p38-MAPK pathway. Biochem. J. 473, 685–692. Johnson, C.L., Weston, J.Y., Chadi, S.A., Fazio, E.N., Huff, M.W., Kharitonenkov, A., et al., 2009. Fibroblast growth factor 21 reduces the severity of cerulein-induced pancreatitis in mice. Gastroenterology 137, 1795–1804. Kharitonenkov, A., Shiyanova, T.L., Koester, A., Ford, A.M., Micanovic, R., Galbreath, E.J., et al., 2005. FGF-21 as a novel metabolic regulator. J. Clin. Invest. 115, 1627–1635. Kim, J.H., Park, H.Y., Bae, S., Lim, Y.H., Hong, Y.C., 2013. Diethylhexyl phthalates is associated with insulin resistance via oxidative stress in the elderly: a panel study. PLoS One 8, e71392. Koch, H.M., Christensen, K.L., Harth, V., Lorber, M., Bruning, T., 2012. Di-n-butyl phthalate (DnBP) and diisobutyl phthalate (DiBP) metabolism in a human volunteer after single oral doses. Arch. Toxicol. 86, 1829–1839. Kohn, A.D., Summers, S.A., Birnbaum, M.J., Roth, R.A., 1996. Expression of a constitutively active Akt Ser/Thr kinase in 3T3-L1 adipocytes stimulates glucose uptake and glucose transporter 4 translocation. J. Biol. Chem. 271, 31372–31378. Lee, K.I., Chiang, C.W., Lin, H.C., Zhao, J.F., Li, C.T., Shyue, S.K., et al., 2016. Maternal exposure to di-(2-ethylhexyl) phthalate exposure deregulates blood pressure, adiposity, cholesterol metabolism and social interaction in mouse offspring. Arch. Toxicol. 90, 1211–1224. Leney, S.E., Tavare, J.M., 2009. The molecular basis of insulin-stimulated glucose uptake: signalling, trafficking and potential drug targets. J. Endocrinol. 203, 1–18. Manteiga, S., Lee, K., 2017. Monoethylhexyl phthalate elicits an inflammatory response in adipocytes characterized by alterations in lipid and cytokine pathways. Environ.
254