|
Figure 1: Developmental gene expression profiles during Xenopus laevis spontaneous metamorphosis. mRNA expression levels of thyroid hormone receptor (TRα) (red square), TRβ (green triangle), RXRα (blue diamond), D1 (yellow circle), D2 (closed circle), and D3 (open circle) are shown. Total RNA was extracted from hind limbs (A), and tail (B and C) of 3–6 tadpole siblings at the indicated stages. mRNA levels of metamorphosis-related genes were quantified by RT-PCR and normalized to EF mRNA. Data are shown as means ± SE. (n = 3–7). EF, elongation factor.
|
|
Figure 2: Thyroid hormone receptor (TR)-transfected tail muscle cells die in response to low levels of TH. (A) The schema shows the injection protocol for the experiments in (C–F) and Fig. S1A,C in Supporting Information. The empty vector, TRα, TRβ, and/or RXRα expression constructs (25 ng each) were injected with a GFP reporter gene (25 ng) into the third tail myomere of NF-stage 53–54 methimazole-treated tadpoles, and the GFP and dominant negative thyroid hormone receptor α (DNTRα) genes (25 ng each) were injected into the seventh myomere. The numbers of myomeres are indicated. (B) The injection protocol for the experiment in (G). (C) Representative photographs of the hind limbs (HL) and the GFP fluorescence of tadpole tails injected with a vector or the TRα expression construct. T4 treatment induced the reduction in GFP fluorescence in the TRα-injected myomeres. The white lines trace the contours of tails other than fins. Scale bars, 1 mm. (D) The response of hind limbs to 0 nm (circle, solid line), 0.3 nm (triangle, thin dotted line), or 1 nm T4 (square, dotted line) treatment. The ratio of the length between the first and the fifth digits measured on the indicated day to the length on day 0 was evaluated. Data are expressed as means ± SE. (n = 59–110). **Significance difference between control and T4-treated hind limbs, determined by Dunnett's test (P < 0.001). (E–G) The ratio of the fluorescence intensity of the third myomere to that of the DNTRα-injected myomere was calculated. The value at the starting point was set to 100%. Data are expressed as means ± SE. (n = 4–24). (E) GFP expression in the tail myomeres injected with the vector (black circle), TRα (red square), TRβ (green triangle), or RXRα (blue diamond) expression construct in the absence (solid line) or presence of 1 nm T3 (dotted line). Statistical significance by comparison with control tail myomeres was assessed using Student's t-test. *P < 0.05; **P < 0.001. (F) GFP expression in the muscle cells transfected with the vector (black circle), TRα (red square), or TRα-RXRα (blue diamond) expression constructs in the presence of 0 nm (solid line), 0.3 nm (thin dotted line) or 1 nm (dotted line) T4. Statistical significance by comparison with control tail myomeres was assessed using Dunnett's test. *P < 0.05; **P < 0.001. (G) The xbcl-XL co-injection suppressed the reduction in GFP fluorescence that was induced by T4 in the TRα-transfected tail muscle cells. The GFP construct (25 ng) and various combinations of expression constructs encoding no protein, TRα, and xbcl-XL (25 ng each) were injected into the third myomere, and the GFP and DNTRα constructs (25 ng each) were injected into the seventh myomere. The combinations of the injected genes and T4 treatments are indicated in the figure. The fluorescence intensity of the GFP signal was examined on days 0 and 10. Statistical significance by comparison with bcl-XL-injected myomeres was assessed using Student's t-test. *P < 0.05; **P < 0.001. NF, Nieuwkoop and Faber; TH, thyroid hormone.
|
|
Figure 4: D2 mRNA is induced in the thyroid hormone receptor (TRα)-transfected tail muscle cells in response to a low level of T4. (A) The schema shows the experimental protocol. (B–D) The TRα construct was co-injected with the GFP construct (25 ng each) into both the fifth and sixth myomeres of methimazole-treated tadpole tails. One week after injection, the tadpoles were treated with 1 nm T4 for 0 (open circle) or 4 days (closed circle). Afterward, the TRα-injected and contralateral tail myomeres were dissected out, and the mRNA expression levels of D2 (B), D3 (C), thyroid hormone receptor (TRβ) (D), and TRα in the TRα-overexpressing muscle cells were estimated as described in Experimental procedures. Each point represents a value from more than three tadpoles. The average mRNA expression levels in the contralateral myomeres that were treated for 0 (an open square) and 4 days (a closed square) are plotted at similar positions. Pearson correlation analysis showed significant correlations between expression levels of D2 and TRα in T4-treated tadpoles. *P < 0.05.
|
|
Figure 5: D2 mRNA is induced by thyroid hormone (TH) in hind limbs and directly in a cultured cell line. (A–C) RT-PCR was performed using the total RNA that was isolated from hind limbs (more than five tadpoles per group). Nieuwkoop and Faber (NF)-stage 51–52 tadpoles were treated with 1 mm methimazole for more than 24 days and were reared in the absence (no TH) or presence of 1 nm TH (1 nm T4 or 1 nm T3) for 7 days. Hind limbs of NF-stage 54 tadpoles were used as a control. The copy numbers of D2 (A), D3 (B), and TRβ (C) mRNAs per 10 000 copies of elongation factor (EF) mRNA are shown as means ± SE. (n = 5–12). Statistical significance by comparison with control (no TH) hind limbs was assessed using Kruskal–Wallis followed by Dunnett's test. *P < 0.05; **P < 0.001. (D) TH directly induced D2 mRNA expression. RT-PCR was performed using the total RNA that was isolated from XLT-15 cells treated with or without 10 nm T3 for 8 h in the absence or presence of a protein synthesis inhibitor (10 μg/mL cycloheximide [CHX]). The copy number of D2 mRNA per 10 000 copies of EF mRNA is shown as the mean ± SE. (n = 3). Statistical significance by comparison with control (T3-) cells was assessed using Student's t-test. *P < 0.05. TR, thyroid hormone receptor.
|
|
Figure 6: D2 TH-response element (TREs) have a lower affinity for thyroid hormone receptor (TR) compared with the TRβ TRE. (A) The sequences of the consensus TRE, Xenopus laevis TRβ TRE, D2 TREs, and mutated D2 TREs. Bold nucleotides indicate identity with the consensus motif. The homology of each TRE to the consensus TRE is shown in parentheses. (B) Gel mobility shift assays were performed using the D2A TRE (open circle) or D2B TRE (closed circle) as a radiolabeled probe with cold TREs. The cold TRβ TRE (solid line) was a better competitor than either of the cold X. laevis D2 TREs (dotted line) for TR/RXR binding of labeled TREs. The complex of TR/RXR with radiolabeled D2A and D2B TREs was not reduced by the addition of a 100-fold excess of cold mD2A (open square) and mD2B (closed square) TREs, respectively. (C,D) D2 mRNA was more effectively induced by TH in the TRα-overexpressing cells. XLT15-9 cells were transfected with 250 ng of xbcl-XL and 125 ng of RXRα and TRα (closed circle, solid line) or 250 ng of the vector (closed triangle, dotted line); 2 days later, they were treated with 1 nm T3 for the indicated times. The untransfected cells were also treated with T3 (open square, solid line). The copy numbers of D2 (C) and TRβ (D) mRNA per 10 000 copies of EF mRNA are shown as the means ± SE. of three experiments. EF, elongation factor; TH, thyroid hormone.
|
|
Figure 7: D2 TH-response element (TRE) is functional in the oocyte system. (A) Sequence comparison of the TREs of the expression constructs TRβTRE2, D2TRE2, and mD2TRE2 with the consensus TRE. Bold nucleotides indicate identity with the consensus motif. (B–D) The mRNAs for thyroid hormone receptor (TRα) and RXRα (3, 30 or 300 pg/oocyte each) were injected into the cytoplasm of oocytes. The firefly luciferase reporter vector TRβTRE2 (B,D) or D2TRE2 (C,D) was injected into the nucleus together with the control Renilla luciferase plasmid. The oocytes were treated without or with 100 nm T3. After an overnight incubation, the oocytes were harvested for the dual-luciferase assay. Each oocyte was used for a luciferase assay. The ratio of the firefly luciferase activity in the TRβTRE2- or D2TRE2-injected oocytes to the control Renilla luciferase activity was determined (F/R) (B,C). The relative luciferase activities of the T3-treated samples were calculated as the fold luciferase activity, with onefold activity defined as the luciferase activity in the samples that were not subjected to the T3 treatment (D). Data are expressed as means ± SE. (n = 4–8). (E) A mutant version of D2TRE2 was not functional. The oocytes were injected without (−) or with (+) 30 pg/oocyte TRα and RXRα mRNAs followed by an injection of the luciferase reporter TRβTRE2, D2TRE2, mD2TRE2, or the expression vector together with the Renilla luciferase plasmid. After an overnight incubation in the absence or presence of 100 nm T3, the oocytes were harvested for the dual-luciferase assay. The firefly luciferase activity was normalized to the Renilla luciferase activity (F/R). Data are expressed as means ± SE. (n = 5–9). C and T indicate 0 and 100 nm T3, respectively. Statistical significance by comparison with control oocytes was assessed using Student's t-test. *P < 0.05; **P < 0.001.
|
|
Figure 8: Model of differential tissue sensitivity to thyroid hormone (TH). (A) Thyroid hormone receptor (TRα) protein is not sufficient in tail muscle cells to occupy a low-affinity TH-response element (TRE) of D2 gene promoter at the beginning of prometamorphosis when T4 level is low and T3 is not detectable. (B) Abundant TR can bind D2 TRE in hind limbs and TR-transfected tail muscle cells at the beginning of prometamorphosis. Once a part of TR on D2 TRE makes a complex with T4, D2 gene is stimulated and a small amount of D2 protein is synthesized, leading to the activation of T4 to T3. (C) T3 efficiently binds TR on D2 TRE and forms the stable transcriptional complex in D2 gene promoter to produce more D2 mRNA. (D) High levels of D2 protein convert T4 to T3 and augment TH signaling.
|
|
Figure 3: D2 activity is required for the death of thyroid hormone receptor (TRα)-transfected tail muscle cells induced by a low level of thyroid hormone (TH). (A) The empty vector or TRα expression construct was co-injected with the GFP gene (25 ng each) into the third tail myomere, and the GFP and DNTRα constructs (25 ng each) were injected into the seventh myomere as in Fig. 2A. One week after injection, the tadpoles were immersed in the rearing water containing 10 μm iopanoic acid (IOP) or ethanol (EtOH) with or without 0.6 nm TH. GFP fluorescence was quantified in the tail myomeres injected with the vector (closed column) or TRα (open column) on days 0 and 10. Data are expressed as means ± SE. (n = 5–13). Statistical significance by comparison of TH-treated tail myomeres in the absence and presence of IOP was assessed using Student's t-test. *P < 0.05. (B) The size of hind limbs in tadpoles used in A was measured on days 0 and 10. Data are expressed as means ± SE. (n = 17–25). Statistical significance by comparison of TH-treated hind limbs in the absence and presence of IOP was assessed using Student's t-test. **P < 0.001.
|
