Nontranscriptional modulation of intracellular Ca2+ signaling by ligand stimulated thyroid hormone receptor

Abstract

Thyroid hormone 3,5,3'-tri-iodothyronine (T3) binds and activates thyroid hormone receptors (TRs). Here, we present evidence for a nontranscriptional regulation of Ca2+ signaling by T3-bound TRs. Treatment of Xenopus thyroid hormone receptor beta subtype A1 (xTRbetaA1) expressing oocytes with T3 for 10 min increased inositol 1,4,5-trisphosphate (IP3)-mediated Ca2+ wave periodicity. Coexpression of TRbetaA1 with retinoid X receptor did not enhance regulation. Deletion of the DNA binding domain and the nuclear localization signal of the TRbetaA1 eliminated transcriptional activity but did not affect the ability to regulate Ca2+ signaling. T3-bound TRbetaA1 regulation of Ca2+ signaling could be inhibited by ruthenium red treatment, suggesting that mitochondrial Ca2+ uptake was required for the mechanism of action. Both xTRbetaA1 and the homologous shortened form of rat TRalpha1 (rTRalphaDeltaF1) localized to the mitochondria and increased O2 consumption, whereas the full-length rat TRalpha1 did neither. Furthermore, only T3-bound xTRbetaA1 and rTRalphaDeltaF1 affected Ca2+ wave activity. We conclude that T3-bound mitochondrial targeted TRs acutely modulate IP3-mediated Ca2+ signaling by increasing mitochondrial metabolism independently of transcriptional activity.

Figure 1.

T₃-bound TR _β A1 increases IP₃-induced Ca^{2 +} wave period. (a) Spatial-temporal stacks of IP₃ (∼300 nM)-induced Ca²⁺ wave activity in a representative control (water injected) oocyte, a T₃-treated (100 nM) oocyte expressing TR_βA1 and a T₃ (100 nM) treated oocyte. Each image is 745 × 745 μm. (b) Western blot showing expression of TR_βA1. Protein extracts from all groups were collected and loaded at 0.5 oocytes per lane onto 10% SDS-PAGE. The membrane was probed with a monoclonal mouse anti–human TRs antibody (MA1-215) and labeled with an HRP-conjugated secondary antibody. (c) Histogram of average interwave period for each group of oocytes. n values in parentheses represent the total number of oocytes pooled from at least two frogs. Error bars correspond to the mean ± SEM. The asterisks (**) denote a statistic significant difference (ANOVA single factor, P < 0.0001).

Figure 2.

Transcriptional activity of TR _β A1 requires xRXR _α and both cognate ligands. Transcriptional activity was monitored with the TRE-reporter vector, pSEAP ^(TRE). (a) Lanes 1 and 2 are negative (pSEAP^(−ve)) and positive (pSEAP^(+ve)) vector controls. Oocytes expressing TR_βA1 or TR_βA1 plus xRXR_α were incubated with 100 nM T₃ (lanes 3–5) plus 100 nM RA (lane 5) for 3 d. Cytosolic extracts from each group of oocytes was prepared and loaded onto a 10% SDS-PAGE at 2.5 oocytes equivalents per lane. SEAP was detected with the polyclonal rabbit anti–human SEAP antibody and an HRP-conjugated secondary antibody. The SP labeled arrow indicates SEAP immunoreactivity, which was present only in oocytes expressing TR_βA1 and xRXR_α exposed to both T₃ and RA. (b) Transcriptional activity of TR_βA1 requires the pBOX within the DBD and the NLS. Oocytes expressing xTR_βA1ΔpBox-NLS and xRXR_α show no SEAP immunoreactivity when incubated with T₃ (lane 6) or T₃ plus RA (lane 7). Western blot analysis shows that xRXR_α, TR_βA1, and xTR_βA1ΔpBox-NLS are expressed at comparable levels (Western blots below lanes 4–7). TR_βA1 and xTR_βA1ΔpBox-NLS were detected with the monoclonal mouse anti–human TRs antibody (MA1-215). xRXR_α was detected with a polyclonal rabbit anti–human RXR antibody (Sc-774).

Figure 3.

Acute modulation of Ca ²⁺ signaling does not require heterodimerization of TR _β A1 with xRXR _α . (a) Spatial-temporal stacks of IP₃-induced Ca²⁺ wave activity in control oocytes compared with oocytes expressing TR_βA1 or TR_βA1 with xRXR_α. T₃ (100 nM) and RA (100 nM) were added as indicated 10–15 min before injection with IP₃ (∼300 nM). Scale is the same as Fig. 1. (b) Western blots of oocytes expressing TR_βA1 and xRXR_α. Primary and secondary antibodies were identical to those used in Figs. 1 and 2. (c). Histogram of average interwave period (second) of each group of oocytes. The asterisks (**) denote a statistic significance using ANOVA single factor (P < 0.0001). Values in parentheses represent the number of oocytes.

Figure 4.

The pBOX and NLSs of TR _β A1 are not required for the acute regulation of Ca^{2 +} signaling. (a) Schematic figure depicting the position of the pBOX deletion in the DBD and the NLS modification within TR_βA1. (b) Spatial-temporal stack of IP₃-induced Ca²⁺ wave activity in control oocytes compared with oocytes expressing TR mutants ΔpBox-NLS and ΔNLS. Oocytes expressing the TR mutants were incubated with T₃ (100 nM) 10–15 min before IP₃ (∼300 nM) injections. (c) Western blot analysis confirming comparable levels of protein expression for both wild-type and mutant TR_βA1. (d) Histogram of the average Ca²⁺ wave periods for each group of oocytes (n values are in parentheses). Statistic significance over control oocytes is indicated by the asterisks (**; ANOVA single factor, P < 0.0001).

Figure 5.

T₃ stimulation of oocytes expressing TR _β A1 increases the ΔΨ. (a) Images of mitochondria labeled with the potential sensitive dye TMRE. The oocytes are expressing TR_βA1 and have been exposed to T₃ for the indicated amount of time. Images are 50 × 100 μm. (b) Histogram of the log of mitochondrial TMRE fluorescence (F_mito) divided by the cytosolic fluorescence (F_cyto) at the indicated times of T₃ exposure. Values in parentheses refers to the number of mitochondrion analyzed. Statistical significance is indicated by the asterisks (**; ANOVA single factor, P < 0.001).

Figure 6.

Ru ₃₆₀ blocks T₃-bound TR _β A1 increases in IP₃-induced Ca^{2 +} wave period. (a) Spatial-temporal stacks of the effect of Ru360 treatment on Ca²⁺ wave activity in control oocytes compared with oocytes expressing TR_βA1as labeled. (b) Histogram of average interwave period (seconds) of each group of oocytes shown in a. The asterisk (*) denotes a statistic significance using ANOVA single factor (P < 0.01). Values in parentheses represent the number of oocytes.

Figure 7.

T₃ stimulation of oocytes expressing TR _β A1 increases O₂ consumption. (a) Plots of O₂ levels in oocytes as labeled (n = 200 oocytes per group). (b) Histogram represents average change of O₂ consumption rates (before and after T₃ exposure) in control and xTR_βA1 groups. Statistical significance is indicated by the asterisk (*; t test, P < 0.05).

Figure 8.

Xenopus TR _β A1 and NH₂-terminal truncated rat TR _α 1 (rTR _α 1ΔF) localize to mitochondria. (a) Schematic diagram of TRs showing that rTR_α1ΔF and xTR_βA1 have a similar NH₂ terminus. (b and c) Western blots of TR_α1, rTR_α1ΔF, and xTR_βA1 expression in whole oocytes and mitochondrial extracts respectively. FL, full-length receptor; SH, shortened form of the receptor. Extracts were prepared from 300 oocytes in each group. All oocytes were exposed to 100 nM T₃ for at least 15 min before organelle extraction. TRs were immunoprecipitated with a monoclonal mouse anti–human TRs antibody (MA1-215), captured with immobilized protein G, concentrated, and loaded onto a 10% SDS-PAGE. An HRP-conjugated secondary antibody was used for visualization.

Figure 9.

NH ₂ -terminal truncated rat TR _α 1 (rTR _α 1ΔF) stimulates O₂ consumption. (a) Plots of O₂ levels for oocytes expressing full-length rTR_α1 with and without T₃ compared with oocytes expressing the NH₂-terminal truncated rTR_α1ΔF with or without T₃. Protocols used were identical to those described in Fig. 7. (b) Histogram of the average change of the O₂ consumption rates after T₃ exposure in rTR_α1 versus rTR_α1ΔF groups. The asterisk (*) indicates statistical significance (t test, P < 0.05).

Figure 10.

The truncated rTR _α 1ΔF regulates intracellular Ca ²⁺ release. (a) Spatio-temporal stacks of IP₃-induced Ca²⁺ wave activity in control oocytes compared with oocytes expressing rTR_α1ΔF or rTR_α1ΔF. TR expressing oocytes were treated with 100 nM T₃ 10–15 min before IP₃ (∼300 nM) injections and confocal imaging. (b) Western blots of rTR_α1 and rTR_α1ΔF expression levels in experimental oocytes. (c) Histogram of the average interwave periods for each group (n values in parentheses). Note that rTR_α1ΔF has significantly longer periods even though its expression levels are lower than those of full-length rTR_α1. The asterisks (**) indicate statistical significance with P < 0.01 using ANOVA single factor.