Tetrabromobisphenol-A Promotes Early Adipogenesis and Lipogenesis in 3T3-L1 Cells

Chappell VA, Janesick A, Blumberg B, Fenton S
Toxicological Sciences (2018) DOI: https://doi.org/10.1093/toxsci/kfy209 PMID: 30496566
DOI: https://doi.org/10.22427/NTP-DATA-021-00003-0001-0000-7

Publication

Abstract

Tetrabromobisphenol A (TBBPA) is the most common flame retardant used in electrical housings, circuit boards and automobiles. High throughput screening and binding assays have identified TBBPA as an agonist for human peroxisome proliferator-activated receptor gamma (PPARγ), the master regulator of adipogenesis. TBBPA has been suggested to be an obesogen based on in vitro cellular assays and zebrafish data. We hypothesized that exposing preadipocytes to TBBPA could influence adipogenesis via genes other than those in the PPARγ pathway due to its structural similarity to bisphenol A, which demonstrates varied endocrine disrupting activities. Mouse-derived 3T3-L1 preadipocytes were induced to differentiate and continually treated with TBBPA for 8 days. High-content imaging of adipocytes displayed increased adipocyte number and lipid accumulation when treated with TBBPA. TBBPA exhibited weak induction of mPPARγ, with an AC50 of 397µM. Quantitative PCR revealed that TBBPA exposure increased early expression of genes involved in glucocorticoid receptor (GR) signaling and PPARγ transcriptional activation, as well as upregulating downstream genes needed for adipocyte maintenance and non-traditional ER signaling, such as Gpr30. Additionally, Pref1 and Thy1, inhibitors of differentiation, were downregulated by some concentrations of TBBPA. Furthermore, proliferating preadipocytes treated with TBBPA, only prior to differentiation, exhibited increased adipocyte number and lipid accumulation after 8 days in normal culture conditions. In conclusion, TBBPA influenced gene expression changes in GR, non-traditional ER, and known adipogenic regulatory genes, prior to PPARγ expression; effects suggesting early programming of adipogenic pathways.

Figures

Figure 1. TBBPA and receptor activation assays

Mouse PPARγ receptor activation by (A) TBBPA (AC₅₀ = 397µM) and rosiglitazone (AC₅₀ = 1.05µM). Human RXRα receptor activation by (B) TBBPA and AGN194204 (AC₅₀ = 2.45nM). Human glucocorticoid receptor activation by (C) TBBPA and dexamethasone (AC₅₀ = 3.30nM). Results are expressed as fold induction over DMSO control per well (mean ± SEM; n = 3). Solid black line: non-linear variable slope curve fit analysis.

Figure 1 (1 MB)
Figure 1 Graphpad Raw Data (68 KB)

Figure 2. TBBPA effects on adipocyte differentiation and lipid accumulation in 8 day assays

3T3-L1 cells were either grown in adipocyte maintenance media (uninduced) (A) or treated (induced) with MDI (B) for 8 days with and without TBBPA. DMSO: vehicle control, ROSI :1µM rosiglitazone positive control, and 10µM TBBPA. Cells were co-stained with Hoechst (nuclei) and Nile red (lipid). High content capture of images using DAPI and FITC filters provided data on (C) mean percent adipocytes per well and (D) mean lipid area (µm2) per adipocyte per well (mean ± SEM, n = 4). Dash line: induced control mean. Dotted line: uninduced control mean. One-way ANOVA with multiple comparisons Dunnett’s post hoc test p<0.05, * statistically significant over induced control, p<0.05, and † statistically significant over uninduced vehicle control, p<0.05.

Figure 2 (2 MB)
Figure 2 Graphpad Raw Data (37 KB)

Figure 3. Critical adipogenic transcription factor mRNA expression profiles during differentiation

Two day post-confluent 3T3-L1 cells were treated for 8 days with induction agents (MDI) and exposed to TBBPA. MDI: adipogenesis positive control, ROSI: rosiglitazone and MDI (PPARγ agonist positive control). Messenger RNA expression was quantified by RT-qPCR (A) Cebpa mRNA levels (B) Pparg mRNA levels. Relative gene expression data are presented as mean fold change over vehicle control ± SEM, n = 3. Two-way ANOVA with multiple comparisons Dunnett’s post hoc test was performed on ΔCt data, * statistically significant over MDI control, p<0.05. ‡ statistically significant over MDI control, p<0.1 (Pparg day 8 10nM TBBPA p = 0.06).

Fig 3_4_5_6 qPCR Graphpad Raw Data (224 KB)
Figure 3 (654 KB)

Figure 4. TBBPA effects on adipogenesis and adipocyte-specific mRNAs by day 8 of differentiation

Two-day post-confluent 3T3-L1 cells were treated for 2 days with induction agents (MDI) and TBBPA for 8 days. MDI: adipogenesis positive control, ROSI: rosiglitazone and MDI (PPARγ agonist positive control). Messenger RNA expression was quantified by RT-qPCR (A & B) Adipogenesis genes, (C & D) adipocyte specific genes, and (E) lipogenic genes. Relative gene expression data are presented as mean fold change over vehicle control ± SEM, n=3. One-way ANOVA with multiple comparisons Dunnett’s post hoc was performed on ΔCt data, * statistically significant over MDI control, p<0.05. ‡ statistically significant over MDI control, p<0.1 (Fasn 10nM TBBPA p = 0.052).

Figure 4 (2 MB)

Figure 5. Preadipocyte and adipogenic mRNAs altered early in differentiation by TBBPA

Two-day post-confluent 3T3-L1 cells were treated for 2 days with induction agents (MDI) and TBBPA. MDI: adipogenesis positive control, ROSI: rosiglitazone and MDI (PPARγ agonist positive control). Messenger RNA expression was quantified by RT-qPCR. (A & B) Early adipogenesis transcription factors on day 2, (C) preadipocyte genes on day 2, (D) selected krüppel–like factors on day 2, and (E) Klf15 on day 8 of the differentiation assay. Relative gene expression data are presented as mean fold change over vehicle control ± SEM, n=3. One-way ANOVA with multiple comparisons Dunnett’s post hoc test was performed on ΔCt data, * statistically significant over MDI control, p<0.05. ‡ statistically significant over MDI control, p<0.1 (Klf5 1pM TBBPA p = 0.08, Klf15 10µM TBBPA p = 0.08).

Figure 5 (2 MB)

Figure 6. Receptor mRNA levels 8 days after induction and TBBPA exposure

Two-day post-confluent 3T3-L1 cells were treated for 2 days with induction agents MDI and exposed to TBBPA for 8 days. MDI: adipogenesis positive control, ROSI: rosiglitazone and MDI (PPARγ agonist positive control). Messenger RNA expression was quantified by RT-qPCR (A) Estrogen receptors –alpha and -beta, (B) estrogen-related receptors, (C) glucocorticoid receptor (Nr3c1) and a downstream gene (Pdk4), and (D) insulin receptors and insulin receptor substrate-1. Relative gene expression data are presented as mean fold change over vehicle control ± SEM, n=3. One-way ANOVA with multiple comparisons Dunnett’s post hoc test was performed on ΔCt data, * statistically significant over MDI control, p<0.05. ‡ statistically significant over MDI control, p<0.1 (Insr 100pM TBBPA p = 0.1, Irs1 100pM TBBPA p = 0.06).

Figure 6 (1 MB)

Figure 7. Effects of TBBPA exposure during 3T3-L1 cell proliferation on adipogenesis and lipid accumulation

3T3-L1 cells were treated with TBBPA during cell proliferation prior to monolayer confluence and then induced to differentiate two days later. High content imaging analysis of cells provided (A) Mean percent adipocytes per well and (B) mean lipid area (µm2) per adipocyte per well (mean ± SEM, n = 4). Dash line: induced control mean. Dotted line: uninduced control mean. Two-way AVOVA with multiple comparisons Dunnett’s post hoc test, * statistically significant over induced control, p<0.05, and † statistically significant over uninduced vehicle control, p<0.05.

Figure 7 (633 KB)
Figure 7 Graphpad Raw Data (37 KB)

Figure 8. Regulators of adipogenesis and adipocyte specific genes

Transcription factors and nuclear receptors involved in early and late phases of differentiation; mature adipocyte specific genes and their upstream regulators. Solid lines denote reported direct interactions. Dashed lines indicate proposed indirect interactions. Color-coded up and down arrows indicate TBBPA-induced or -repressed effects on gene expression. Image created using Ingenuity Pathway Analysis software version: #31813283. Adapted from (Wu, et al., 2013).

Figure 8 (4 MB)

Supplemental Materials

Figure S1. TBBPA alters numerous adipogenic gene expression patterns during differentiation

Two-day post-confluent 3T3-L1 cells were treated for 2 days with MDI and exposed to TBBPA for 8 days. MDI: adipogenesis positive control, ROSI: rosiglitazone and MDI (PPARγ agonist positive control). Each assay was run in triplicate. Messenger RNA expression was quantified by RT-qPCR. Relative gene expression data are presented over a time course of 1, 2, 4, 6, and 8 days post MDI induction as mean fold change over vehicle control ± SEM, n=3 biological replicates