3,3′,5-Triiodothyroacetic acid (TRIAC) induces embryonic ζ-globin expression via thyroid hormone receptor α

To the editor,

The inherited hemoglobin disorders, including thalassemia and sickle-cell disease, are an emerging global health burden. It is estimated that in excess of 330,000 affected infants are born annually [1]. There is an urgent need to identify new types of agents for these hemoglobinopathies.

Embryonic ζ-globin gene expression is normally limited to the early stages of primitive erythropoiesis and transcriptionally silenced at 6–7 weeks of gestation [2]. Relatively little attention has been paid to understanding the processes that control ζ-globin expression in the past few decades. Intriguingly, recent reports have shown that continued expression of human ζ-globin is not only able to rescue a lethal α-thalassemia mouse model [3], but also efficiently inhibits sickle hemoglobin polymerization in a transgenic mouse model of sickle-cell disease [4], suggesting induction of this embryonic globin may act as a novel therapeutic for both α-thalassemia and sickle-cell disease. However, pharmacologic compounds capable of activating ζ-globin gene expression have not yet been available so far.

3,3′,5-triiodothyroacetic acid (TRIAC, also known as Tiratricol) is a naturally occurring thyroid hormone metabolite, with high affinity for thyroid hormone receptors. It has been used on an empirical basis to treat patients with thyroid hormone resistance [5]. More recently, TRIAC has also displayed great therapeutic potential for the treatment of Allan–Herndon–Dudley syndrome [6]. Although the relevance and use of TRIAC have been extensively explored over the last decades, its role in the regulation of globin gene expression has not previously been elucidated.

As a first step in seeking whether TRIAC affects globin gene expression, we used zebrafish as a model organism, which is an ideal system for modeling erythropoiesis of humans [7]. Zebrafish larvae were incubated with TRIAC for up to 24 h, and then, globin gene expression was assessed by quantitative real-time PCR (qPCR). The data showed that TRIAC administration strikingly increased hbae5 mRNA levels, an ortholog of human HBZ, while had little effect on other embryonic globin genes expression (Fig. 1A). To further confirm this, we examined the effect of TRIAC on hbae5 expression by whole-mount mRNA in situ hybridization (WISH). In line with the qPCR result, we found that the expression of hbae5 was dramatically induced by TRIAC treatment (Fig. 1C).

Since the biological actions of TRIAC closely resemble those of the bioactive hormone 3,3′,5-triiodothyronine (T3) [8], we wonder whether T3 or the pro-hormone thyroxine (T4) has a similar effect. Zebrafish larvae were treated with T3 or T4 for up to 24 h, and then, qPCR and WISH assay were performed. Expectedly, the results showed that both T3 and T4 administration also significantly increase hbae5 transcripts (Fig. 1B, D). Taken together, these data suggest that thyroid hormones have the abilities to selectively induce embryonic ζ-globin gene expression in zebrafish.

In order to examine whether thyroid hormone could also induce ζ-globin production in human cells, we used K562-derived erythroid cells. During hemin-induced erythroid differentiation of K562 cells, TRIAC was added and incubated for up to 48 h, and then, globin gene expression was assessed by qPCR assay and western blot. The results showed that TRIAC sharply increased HBZ mRNA and protein levels in hemin-treated K562 cells (Fig. 1E, F), although it also has a comparatively weak stimulatory effect on other globin genes expression. To evaluate genome-wide gene expression changes promoted by TRIAC, we performed RNA sequencing (RNA-seq) analysis. As expected, TRIAC-treated K562 cells showed significant induction of HBZ, with a mild induction in other globin levels (Fig. 1G). Thus, these experiments indicate that TRIAC preferentially induces ζ-globin expression in differentiated K562 cells.

We next determined whether thyroid hormone was able to induce ζ-globin expression in primary human erythroid cells. To this end, we used human CD34⁺ hematopoietic stem and progenitor cell (HSPC)-derived primary erythroblasts. During erythroid differentiation, CD34⁺ cells were incubated with TRIAC, T3 or T4, respectively. The total RNA samples were collected on day 7 of differentiation and then subjected to qPCR analysis. Again, we found that TRIAC, as well as T3 or T4, triggered an increase in HBZ mRNA levels, as compared with control cells (Fig. 1H). Collectively, these data suggest that thyroid hormones could also induce ζ-globin expression in human cells.

The biological effect of thyroid hormone is predominantly mediated by thyroid hormone receptors, which are encoded by the thyroid hormone receptor α (THRA) and thyroid hormone receptor β (THRB) genes. An increasing number of reports have implicated that THRA, but not THRB, is required for erythroid development [9‐11], implying that ζ-globin expression might be positively regulated by THRA in erythroid cells. To determine whether THRA was essential for TRIAC-induced ζ-globin production, we first knocked down THRA expression in K562 cells by lentiviral-mediated short hairpin RNA (shRNA). As shown in Fig. 2A, the THRA shRNA efficiently reduced the THRA mRNA levels to 10% after 5 days of infection. Western blot using an anti-THRA antibody further confirmed that the THRA protein level was indeed decreased after knockdown of the THRA mRNA (Fig. 2B). Then, these shRNA-transduced K562 cells were treated with TRIAC for up to 48 h, and globin gene expression was assessed by qPCR. The result showed that THRA depletion reduced the level of HBZ mRNA by 40% compared to the control infected cells (Fig. 2C), indicating that THRA was involved in TRIAC-induced ζ-globin expression. We next examined TRIAC-induced ζ-globin expression level upon morpholino oligonucleotide (MO)-mediated thyroid hormone receptor α (thraa) knockdown in zebrafish. Zebrafish embryos were microinjected with specific translation-blocking morpholino against the thraa, homologous to human THRA, and exposed to TRIAC at 48 hpf. Then, embryos were collected for WISH assay after 24 h of TRIAC treatment. We found that Thraa deficiency reversed the increased expression of hbae5 triggered by TRIAC (Fig. 2D). Taken together, these data indicated that THRA is the major effector responsible for TRIAC-induced ζ-globin expression.

THRA is a member of the nuclear receptor superfamily and functions as a ligand-inducible transcription factor, which binds to thyroid hormone-responsive elements (TREs) to regulate gene transcription [12]. To determine whether THRA regulates ζ-globin expression directly or indirectly, we performed chromatin immunoprecipitation and sequencing (ChIP-seq) with an antibody to THRA in K562 cells. We observed that there was a moderate THRA ChIP-seq peak at ~ 40 kb upstream of the ζ-globin gene locus (Fig. 2E, red), which was known as HS-40, the most critical cis-regulatory element for α-like globin genes expression (Fig. 2E, blue) [13]. This finding is consistent with previously reported ChIP-seq data for THRA in K562 cells (Fig. 2E, blue) [14]. The ChIP-seq result was further validated by chromatin immunoprecipitation followed by quantitative PCR (ChIP-qPCR) (Fig. 2F). Thus, these data are highly likely to reflect authentic, direct binding sites of THRA in distal enhancer regulatory element.

In summary, this study shows that TRIAC emerges as a potent inducer of ζ-globin expression, which may allow development of new therapies for α-thalassemia or sickle-cell disease. Further studies need to investigate the in vivo potential of TRIAC in the treatment of these hemoglobin disorders.