Thyroid Hormone Signaling in Embryonic Stem Cells: Crosstalk with the Retinoic Acid Pathway

Abstract

While the role of thyroid hormones (THs) during fetal and postnatal life is well-established, their role at preimplantation and during blastocyst development remains unclear. In this study, we used an embryonic stem cell line isolated from rat (RESC) to study the effects of THs and retinoic acid (RA) on early embryonic development during the pre-implantation stage. The results showed that THs play an important role in the differentiation/maturation processes of cells obtained from embryoid bodies (EB), with thyroid hormone nuclear receptors (TR) (TRα and TRβ), metabolic enzymes (deiodinases 1, 2, 3) and membrane transporters (Monocarboxylate transporters -MCT- 8 and 10) being expressed throughout in vitro differentiation until the Embryoid body (EB) stage. Moreover, thyroid hormone receptor antagonist TR (1-850) impaired RA-induced neuroectodermal lineage specification. This effect was significantly higher when cells were treated with retinoic acid (RA) to induce neuroectodermal lineage, studied through the gene and protein expression of nestin, an undifferentiated progenitor marker from the neuroectoderm lineage, as established by nestin mRNA and protein regulation. These results demonstrate the contribution of the two nuclear receptors, TR and RA, to the process of neuroectoderm maturation of the in vitro model embryonic stem cells obtained from rat.

Keywords: embryonic stem cells; nestin; neuroectoderm; nuclear receptors; retinoic acid; thyroid hormone.

Figure 1

RESC populations and culture conditions. (A). Experimental and temporal approach to obtain the different RESC populations. (B–D). Representative images of cells cultured in different medium compositions and culture conditions are shown in the figure. When RESC grow in SCML-new medium (+LIF/+bFGF) on a ULA surface, cells proliferate as CL, increasing in size over the days in culture (B). Cells resulting from split CL can be either cultured in SCML-new medium (+LIF/+bFGF) on a standard culture surface (Nuclon Delta treated), giving a mixed population of floating clusters (mCL, black arrow in the picture) and attached single cells (mSC, black arrow in the picture) (C), or cultured in SCM medium -LIF/+bFGF on a ULA surface, thus inducing the formation of EB (D). Single cells obtained from the split of EB and cultured on a coated surface spontaneously differentiated (E). Scale bar 100 μm. Abbreviations: bFGF, basic fibroblast growth factor; EB, embryoid bodies; CL, clusters; LIF, leukemia inhibitory factor; mCL, mixed population clusters; mSC, mixed population single cells; SC, single cells; SCM, stem cell medium; SCML, SCM with LIF; ULA, ultra-low attachment.

Figure 2

RESC pluripotency and differentiation capabilities. The expression of pluripotency and inner cell mass differentiation markers was studied in RESC: Oct 4 mRNA (A) and protein (D,G); Stella mRNA (B); Alpha fetoprotein mRNA (C); Laminin (E,H) and F-actin (F,I) proteins. Gene expression was assessed by semiquantitative QPCR and the 2^{(−∆∆Ct)} method, using CL as reference group. The graphs show mean values ± SEM of n = 3 independent experiments. Samples were processed in duplicate. Statistics: One-way ANOVA and Tukey’s post test were performed. Significance was set at p < 0.05. Representative images of CL and EB cryosections stained with the specified antibodies and the nuclear colorant Hoechst 33258 are included in the figure. Scale bar 25 μm. Abbreviations: Afp, alpha fetoprotein; CL, clusters; EB, embryoid bodies; F-actin, filamentous actin; HO, Hoechst; Oct 4, octamer-binding transcription factor 4; mCL, mixed population clusters; mSC, mixed population single cells.

Figure 3

Nuclear receptor and co-regulator expression. The expression of all 47 NRs and 37 co-regulators were analyzed by RT2 PCR array technique and represented as a ΔCq heatmap for each gene in a clustergram showing the maximum (red) and the minimum (green) expression for each gene in the various groups (A). The fold of regulation value for each gene was calculated using CL as a control group, and genes showing more than 2-fold regulations are shown (B). Red color is for genes that are up-regulated whereas green color is for genes that are down-regulated. Bold numbers indicate fold of changes ≥ 2, both as positive and negative value. These genes were used as inputs for the pathway enrichment analysis using the GeneCodis software, and the results of the analysis is shown (C). Abbreviations: CL, clusters; Cq, threshold cycle; EB, embryoid bodies; mCL, mixed population clusters; mSC, mixed population single cells.

Figure 4

RESC and Thyroid Hormone. RESC gene expression of TH converting enzymes D1, D2, D3 (A–C), transporters MCT8, MCT10 (D,H) and receptors TRα1, TRα2 and TRβ (E–G) studied by semiquantitative QPCR and the 2^{(−∆∆Ct)} method, using CL as reference group. A total of n = 3 independent experiments were performed; samples were processed in duplicate. Mean values ± SEM have been included in the graphs. Statistical analysis performed: 1- way ANOVA and Tukey’s post test (A–F,H) and Student’s t-test (G), results were significant when p < 0.05. In the figure representative images of D3 enzyme as well as TRα1- and TRβ1- receptor positive cells Hoechst 33258 stained are included (I–N). Scale bar 25 μm. Abbreviations: CL, clusters; D1, deiodinase enzyme type 1; D2, deiodinase enzyme type 2; D3, deiodinase enzyme type 3; EB, embryoid bodies; HO, Hoechst; MCT8, monocarboxylate transporter 8; MCT10, monocarboxylate transporter 10; TRα1, thyroid hormone receptor alpha 1; TRα2, thyroid hormone receptor alpha 2; TRβ, thyroid hormone receptor beta; mCL, mixed population clusters; mSC, mixed population single cells.

Figure 5

RESC and neuroectoderm lineage. Single cells obtained from the EB split were cultured onto gelatin coated coverslip in SCM medium without mitogens and allowed to differentiate. Representative images of spontaneously differentiated cells (A,C) or 1 μM RA-treated cells for 5 and 10 DIV (B,D) nestin-immunostained (green) and Hoechst 33,258 nuclear-stained cells (blue) are shown. The number of nestin-positive cells was significantly higher in the 1 μM RA-treated group of cells at both time points studied (E,F). Statistical analysis performed: Student’s t-test. Scale bar 100 μm. Abbreviations: DIV, days in vitro; RA, retinoic acid.

Figure 6

(A). TR antagonist 1-850 and RA treatment experimental design. The diagram shows the experimental conditions used to block T3 binding to its receptors TRα and TRβ in the different RESC culture conditions. Experimental groups: (A) TR antagonist 1-850 treatment for the duration of the experiment; (B) cells treated when growing as CL; (C) cells treated during the EB induction period; (D) cells treated during the differentiation phase as single cells obtained from EB; (E) control group of cells (not TR antagonist 1-850 treated). (B–E): Effect of TR antagonist 1-850 treatment in the RA-induced differentiation. (B–D): TRα and TRβ inhibition induced a reduction in the number of nestin-positive cells after 10DIV, EB and differentiating single cells obtained from EB being the RESC populations most affected. (E): Nestin RNA expression in single cells. RA induced an increase in the mRNA expression of nestin, which was downregulated by the TR antagonist 1-850. Graphs contain mean values ± SEM from 2–3 independent experiments performed in duplicate. Statistical analysis performed: Student’s t-test. Abbreviations: bFGF, basic fibroblast growth factor; CL, clusters; DIV, days in vitro; EB, embryoid bodies; LIF, leukemia inhibitory factor; RA, retinoic acid; SCML, stem cell medium with LIF; TR, thyroid hormone receptor.