Role of Thyroid Hormone Receptor in Amphibian Development

Abstract

The amphibian Xenopus laevis has long been used as a model for studying vertebrate cell and developmental biology largely due to the easiness to manipulate this system in vivo and in vitro. While most of the developmental studies have been on Xenopus embryogenesis, considerable efforts have been made to understand its metamorphosis, a process mimicking postembryonic development in mammals when many organs mature into their adult forms in the presence of high levels of thyroid hormone (T3). Amphibian metamorphosis is totally dependent on T3 and offers a number of advantages for experimental analyses compared to the late stage, uterus-enclosed mammalian embryos. Earlier studies on metamorphosis in Xenopus laevis have revealed dual functions of T3 receptors (TR) during premetamorphic development and metamorphosis as well as important roles of TR-interacting corepressors and coactivators during these two periods, respectively. The development of gene-editing technologies that functions in amphibians in recent years has made possible for the first time to study function of endogenous TRs, especially in the highly related diploid anuran species Xenopus tropicalis. Here, we first review the current mechanistic understanding of the regulation of metamorphosis by T3 and TR, and then describe a detailed method to use TALEN to knock out TRα for studying its role in gene regulation by T3 in vivo and Xenopus development.

1. Introduction

Amphibian undergoes biphasic development with its embryogenesis producing a free-living larva, which later metamorphoses into the adult form [1]. Anuran amphibians undergo the most dramatic metamorphosis, with essentially every single organ/tissue changed during this process. This process bears strong similarity to the so-called postembryonic development in mammals, a period around birth when many organs/tissues are matured into the adult form, resembling the organ remodeling during anuran metamorphosis [1, 2]. Furthermore, both postembryonic development in mammals and anuran metamorphosis are characterized by peak levels of plasma thyroid hormone (T3) [1, 2]. Importantly, T3 is the causative agent of amphibian metamorphosis. Blocking synthesis of endogenous T3 leads to the formation of giant tadpoles that cannot metamorphose while exogenous T3 can induce premetamorphic tadpoles to undergo precocious metamorphosis prior to the synthesis of endogenous T3 [1, 2]. Furthermore, most, if not all, organs are genetically predetermined to undergo specific organ autonomous changes, making it possible to study anuran metamorphosis even in organ cultures. These properties together with the recent advancement in genetic tools for amphibian studies make anuran metamorphosis one of the best models to study the in vivo function of T3 and its underlying molecular mechanisms. Below, we summarize some of the studies on Xenopus laevis and Xenopus tropicalis, two highly related and best studied anuran species.

1.1. Dual Roles of Thyroid Hormone Receptors (TRs) in Xenopus Metamorphosis

T3 can exert its effect through both genomic and non-genomic pathways. It can regulate gene transcription through its nuclear receptors, TRs, while its non-genomic effects are believed to be mediated by cell surface and cytoplasmic binding proteins, including the small fractions of TRs present in the cytoplasm [3, 4, 5, 6, 7, 8, 9, 10, 11, 12]. TRs are known to function mainly as heterodimers with 9-cis retinoic acid receptors (RXRs) on T3-inducible genes. TR/RXR heterodimers are nuclear proteins that bind constitutively to thyroid hormone response elements (TREs) in chromatin to repress or activate target gene expression in the absence or presence of T3, respectively.

Earlier studies have shown that zygotic transcription of TR, especially TRα , occurs well before the synthesis and secretion of endogenous T3. TRα and RXRα genes are expressed at high levels by stage 45 when tadpole feeding begins while plasma T3 is detectible only by stage 54, the onset of metamorphosis [13, 14, 15, 16]. This and the ability of TR to repress or activate target gene in a T3-dependent manner have led to a dual function model for TR during frog development [17, 18]. That is, between stage 45, when tadpole feeding begins, and stage 54, the onset of metamorphosis, TRs are present mainly in the unliganded form due to only very low levels of T3, if any, in the tadpole plasma . The unliganded TR/RXR heterodimers repress T3 inducible genes to prevent precocious metamorphosis. After stage 54, the rising concentration of endogenous T3 leads to the binding of T3 to TR, and the liganded TR/RXR then activate the target genes to promote metamorphosis.

1.2. Analyses of TR Function via Overexpression during Xenopus laevis Development

Functional analysis of frog TRs initially came from microinjection studies. Overexpression of TR and RXR by microinjecting their mRNAs into fertilized eggs led to repression of endogenous target genes in the resulting embryos, and when the embryos were treated with exogenous T3, the target genes were activated, leading to embryonic defects and lethality [19]. RXR was found to be important for these effects, suggesting that TR functions as heterodimers with RXR in vivo during Xenopus development.

The development of the reliable sperm-mediated transgenic approach [20] made it possible to study the effect of overexpressed mutant TRs during Xenopus development. Several different laboratories analyzed the effect of overexpressing a dominant negative TR that had a C-terminal deletion to prevent T3 binding, leading to a constitutively unliganded TR that represses T3-inducible genes. Overexpression of such a dominant negative TR by either transgenesis in whole animals or in vivo gene transfer in the tail inhibited target gene induction by T3 and metamorphic transformation in Xenopus laevis [21, 22, 23, 24, 25]. Mechanistically, the transgenic dominant negative TR was found to compete against endogenous TR for binding to endogenous target genes, consequently preventing the regulation of the genes by T3 [25]. Thus, TR is required to mediate the effects of T3 on target gene induction and metamorphosis of different organs/tissues.

As indicated above, T3 has both genomic and non-genomic effects. To determine whether TR is sufficient to mediate the metamorphic effects of T3, a transgenic study with a dominant positive TR was carried out. This dominant positive TR consisted of the dominant negative TR above fused to the N-terminus a strong viral transcription activation protein, VP16 [26]. This mutant TR could not bind to T3 but functioned as a constitutive activator on T3 target genes due to the presence of the VP16 moiety. To prevent likely precocious induction of metamorphosis by the expression of the dominant positive TR, it was placed under the control of a heat shock-inducible promoter for transgenesis. When transgenic tadpoles and their wild-type siblings reached premetamorphic stages, they were subjected to heat shock treatment to induce the transgene expression. This enabled the transgenic but not the wild-type tadpoles to metamorphose just like premetamorphic tadpoles after T3 treatment [26]. Mechanistically, the dominant positive TR was shown to bind to endogenous target genes and specifically activate their expression [26]. Furthermore, when the intestine of such transgenic premetamorphic tadpoles was cultured in vitro , heat shock treatment also led to intestinal metamorphosis [27]. Thus, TR is sufficient to mediate the effects of T3 on individual organs during metamorphosis. These findings further demonstrate that the non-genomic effects of T3 are not necessary for amphibian metamorphosis.

1.3. Determining the Roles of Endogenous TR in Xenopus tropicalis

While molecular and transgenic studies have provided strong evidence to support the dual functions of TR during amphibian development, the roles of endogenous TRs had eluded researchers until recently. With the development of first TALEN and then CRISPR-nuclease mediated knockout/knockdown technologies that can effectively edit endogenous genes in both the pseudo-tetraploid Xenopus laevis and diploid Xenopus tropicalis, it becomes possible to study endogenous genes during Xenopus development [28, 29, 30, 31, 32, 33]. Using the TALEN technology, we and the Buchholz’s laboratory initially independently knocked down the endogenous TRα [34, 35, 36, 37]. Xenopus tropicalis was chosen for such studies due to its diploid nature. Both groups microinjected mRNAs encoding a TALEN nuclease targeting TRα into fertilized eggs and studied the effect on the resulting animals, the Fo generation knockdown animals, with similar conclusions [34, 35, 36, 37]. In our study, we generated a TALEN targeting exon 3, encoding part of the DNA binding domain of Xenopus tropicalis TRα (Fig. 1) [35]. Microinjecting the mRNA encoding the left and right arms of the TALEN into fertilized egg prior to the first cell division led to the expression of the TALEN in all cells of the early stage embryos and the resulting Fo tadpoles had genetic alterations at the TRα target site with about 90% efficiency by the time when feeding began (stage 45, 3 days of age) [35]. Such high efficiency suggested that we had nearly a complete knockout of the TRα gene and could study the phenotype in the Fo animals. Indeed, the findings in the Fo knockdown animals were subsequently confirmed by studies on complete TRα knockout animals derived from germ line breeding of the knockdown animals [38, 39].

Fig. 1.

Targeting TRα with TALEN in Xenopus tropicalis embryos. (a) Schematic diagram of the TRα gene showing the TALEN target site. The protein coding exons were shown as numbered, colored boxes. The TALEN-recognized sequences were shown as red boxes (top) or bold letters (below). (b) Schematic representation of the TRα TALEN and the targeted DNA sequence in the TRα gene. The four types of RVDs (repeat variable di-residues) recognizing nucleotide A, G, T, or C were depicted in different colors, respectively. The left arm contained the FokI-ELD nuclease and the right arm contained the FokI-KKR nuclease, which, when both arms bind to their respective binding sites, heterodimerize to form the functional nuclease to make a double-stranded break in the intervening sequence. See [35] for more details

Using either Fo knockdown or complete knockout animals, both the groups found that TRα knockdown/knockout led to precocious limb development (Fig. 2) [34, 35, 36, 37, 38, 39]. Stage 54 is considered to be the onset of metamorphosis and is determined based on hindlimb morphology. Based on this, both groups observed that the TRα knockdown/knockout tadpoles initiated metamorphosis at younger age, with concurrent upregulation (derepression) of several well-known T3-inducible genes [35, 36, 38, 39]. In addition, the TRα knockdown/knockout tadpoles had reduced response to exogenous T3 treatment in terms of both gene regulation and morphological changes, indicating that TRα is important both for controlling the timing of the onset of metamorphosis and for mediating the response to T3. Consistently, when the knockdown/knockout animals were allowed to undergo natural metamorphosis, it was found that TRα knockdown/knockout animals took much longer time to develop from stage 54, the onset of metamorphosis, to stage 58, the early metamorphic climax. Furthermore, the studies on total TRα knockout animals also led to several novel observations [38, 39]. First, TRα knockout appears to have no gross effect on embryogenesis. Second and surprisingly, TRα knockout does not prevent the animal from completing metamorphosis or affect the total developmental time from fertilization to the end of metamorphosis as judged based on total resorption of the tail. Finally, TRα is important for temporal coordination of organ-specific transformations.

Fig. 2.

TRα regulates metamorphic timing. (a) TRα knockout tadpoles develop faster. One representative animal is shown for each of the three genotypes: wild type, heterozygous, and homozygous TRα knockout siblings. Dashed boxes in the left panel were enlarged and shown in the right panel. Note that the homozygous TRα knockout tadpole had significantly larger and more advanced hindlimb buds. (b) TRα knockout tadpoles develop into more advanced stages. The stage of individual sibling tadpoles in the three genotypes was plotted with the median (darker solid line) for each genotype shown in the figure. Note that homozygous TRα knockout tadpole reached a median stage of 51 while the wild-type and TRα heterozygous knockout tadpoles reached a median stage of 47.5. (c) TRα knockout tadpoles initiate metamorphosis, i.e., reaching stage 54, at a younger age. The age in days for each tadpole to reach stage 54 was plotted with the median (darker solid line) for each genotype shown in the figure. “**” indicates significant difference between homozygous knockout and the other groups (P < 0.01). See [38] for more details

Thus, the gene-editing technologies make it possible to investigate TR function during postembryonic development. Furthermore, the ability to introduce TALEN into fertilized eggs and the high efficiency of TALEN-mediated mutagenesis also enable one to determine the function of TR in Fo generation animals within a few weeks after micro-injecting of the TALEN mRNAs, providing a fast and powerful tool to study gene function in vivo.

In addition, TALEN-mediated mutations in certain genes may lead to lethal phenotype during early embryogenesis. This makes it difficult to study the functions of such genes during postembryonic development. However, it is possible to microinject mRNAs/proteins into fertilized eggs to rescue the embryonic defect due to the gene alteration, leading to the formation of free-feeding tadpoles. As the mRNA or protein injected into the eggs only last a few days, one can use the resulting rescued animals to study the effect of the knockout on subsequent tadpole development and metamorphosis.

2. Materials

2.1. Animal

1. Adult Xenopus tropicalis, Nasco (Wisconsin).

2.2. Devices for Microinjections

1. Needle puller, World precision Instruments, PUL-1 (see Note 1).

2. Needles, Drummond Microcaps 1-000-0300 (see Note 1).

2.3. Plasmids and Constructs

1. Golden Gate TALEN and TAL Effector Kit (Addgene, Kit#1000000024) [40] (see Note 2).

2. pDP(RFP)HG construct ([41] and unpublished).

3. pDP(GFP)HG construct ([41] and unpublished).

4. pCS2+ vector-based TAL- KKR and TAL- ELD constructs [30].

5. pCR TOPO-TA vector (Invitrogen).

2.4. Enzymes

1. T4 DNA ligase (include 10× T4 DNA ligase buffer).

2. Restriction endonuclease BsaI (New England Biolabs).

3. Restriction endonuclease Esp31 (Fermentas or Fisher).

4. Restriction endonuclease AgeI (New England Biolabs).

5. Restriction endonuclease EcoRI (New England Biolabs).

6. Restriction endonuclease NotI (New England Biolabs).

7. PrimeSTAR HS DNA Polymerase (Clontech) (Note that any high fidelity Taq DNA polymerase would be fine).

8. GoTaq® Green Master Mix (Promega).

2.5. Chemicals and Reagents

1. 10 mM ATP.

2. SOC medium: 3.6 g/l glucose, 0.19 g/l potassium chloride, 0.58 g/l sodium chloride, 20 g/l tryptone, and 5 g/L yeast extract (Quality Biologicals, MD).

3. Human chorionic gonadotropin (HCG, Novarel).

4. 20 mg/ml X-gal in DMSO.

5. 10 mg/ml IPTG.

6. LB broth: 10 g/l sodium chloride, 10 g/l tryptone, and 5 g/l yeast extract (Quality Biologicals, MD).

7. 10 mg/ml tetracycline stock solution.

8. 50 mg/ml spectinomycin stock solution.

9. 100 mg/ml ampicillin stock solution.

10. 1× Marc’s Modified Ringers (MMR): 100 mM NaCl, 2 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 5 mM HEPES, pH 7.5.

11. 0.1× MMR.

12. 0.1× MMR/6% Ficoll.

13. Ficoll (Sigma).

14. Cysteine.

15. Agarose.

16. 1× TAE buffer for DNA electrophoresis.

17. dNTPs, individual solutions or mix of equal concentration for each, for PCR reactions.

18. Mineral oil (for microinjectors).

2.6. Primers

1. Primers for sequencing during TALEN assembly:

   • ○pCR8_F1: 5′-TTGATGCCTGGCAGTTCCCT-3′
   • ○pCR8_R1: 5′-CGAACCGAACAGGCTTATGT-3′
   • ○TAL_F1: 5′-TTGGCGTCGGCAAACAGTGG-3′
   • ○TAL_R2: 5′-GGCGACGAGGTGGTCGTTGG-3′
   • ○TAL_Seq: 5′-CATCGCGCAATGCACTGAC-3′
   • ○TAL_R3: 5′-GGCTCAGCTGGGCCACAATG-3′

2. Primers for modifying TALEN backbone constructs:

   • ○Forward: 5′-CTTGGTCTCACCGGCACCATGGCTCCAAAGAAGAAG-3′ (AgeI-compatible end by BsaI digestion)
   • ○Reverse: 5′- CTTGGTCTCCAATTCGTAATACGACTCACTATAGTTC-3′ (EcoRI-compatible end by BsaI digestion)

3. Primers for detecting TRα mutations due to TALEN expression [35]:

   • ○Forward primer F: 5′-ATTGGGTTGGTTGTGGGTCG-3′
   • ○Reverse primer R: 5′-TAAGGTGGGGGCAGTACAGG-3′
   • ○Forward primer f: 5′-ATGTTCCATGAGACGCCCCT-3′
   • ○Forward primer f1: 5′-CCAGCTATCTGGACAAAGAC-3′

2.7. Other Materials

1. RNA transcription kit, Ambion mMessage mMachine.

2. Plasmid DNA Miniprep kit (Qiagen).

3. PCR Purification kit (Qiagen).

4. DNA Gel Extraction kit (Qiagen).

5. pCR TOPO-TA Cloning kit (Invitrogen).

6. Chemically Competent cells TOP10 (Invitrogen).

7. LB plates with 10 mg/l tetracycline, 50 mg/l spectinomycin, or 50 mg/l ampicillin, respectively.

3. Methods

3.1. TALEN Assembly

• 1.Design of TALEN target

Obtain Xenopus tropicalis TRα gene sequence and intron/exon structure from frog genome database (http://www.xenbase.org/entry/ (http://www.xenbase.org/entry/)).

Choose the sequence encompassing the exon 2 to exon 4 for TALEN target screening.

Screen for TALEN targets via a web-based TALEN-targeter tool (https://talent.cac.cornell.edu/node/add/talen (https://talent.cac.cornell.edu/node/add/talen)).

Select a pair of TALEN targets within the exon 3 [35] (see Note 3).

• 2.Modification on TALEN backbone constructs

Golden Gate TALEN and TAL Effector Kit assemble both arms of a TALEN into a backbone vector which expresses TALEN containing the same FokI domain [40] (see Note 2). For assembling the two arms of a TALEN into a set of backbones such that one TALEN construct expresses TALEN containing a mutant FokI-ELD and the other TALEN construct expresses TALEN containing a mutant FokI-KKR in order to further increase the specificity [29, 30], the TALEN backbone constructs are modified to be compatible with the Golden Gate TALEN and TAL Effector Kit. The modified TALEN backbone constructs are based on two double-promoter transgenesis constructs, pDP(GFP)HG (with a green fluorescent protein gene driven by a γ-crystallin promoter) and the pDP(RFP)HG (with a red fluorescent protein gene driven by a γ-crystallin promoter), both of which can produce the TALEN mRNA from a SP6 promoter in vitro for microinjection or express the TALEN protein in transgenic animals from a heat shock-inducible promoter if desired [41] (see Note 4).

PCR-amplify the TAL- KKR and TAL- ELD coding regions from the corresponding pCS2+ vector-based constructs [29, 30] with the same set of primers (forward primer 5′-CTTGGTCTCACCGGCACCATGGCTCCAAAGAAGAAG-3′ and reverse primer 5′- CTTGGTCTCCAATTCGTAATACGACTCACTATAGTTC-3′).

Gel-purify the PCR fragments and digest with BsaI to produce AgeI and EcoRI compatible ends, respectively.

Digest the double-promoter transgenesis constructs pDP(GFP)HG (with a green fluorescent protein gene driven by a γ-crystallin promoter) and the pDP(RFP)HG (with a red fluorescent protein gene driven by a γ-crystallin promoter) with AgeI and EcoRI, respectively, and gel-purify the vector part.

Ligate the PCR amplicon of TAL-KKR containing AgeI and EcoRI-compatible ends with the pDP(RFP)HG vector bearing AgeI and EcoRI ends to make pDP(RFP)-TAL-KKR. Simultaneously, ligate the PCR amplicon of TAL-ELD containing AgeI and EcoRI-compatible ends with the pDP(GFP)HG vector bearing AgeI and EcoRI ends to make pDP(GFP)-TAL-ELD.

Transform the ligated products into competent E. coli TOP10 cells, plate the cells on LB plates containing 100 μg/ml ampicillin, and incubate at 37 °C overnight.

Pick up a couple of colonies each for bacteria growth in 5 ml of LB broth supplemental with ampicillin overnight and extract plasmid DNA from them.

Confirm the cloned TAL-KKR and TAL-ELD sequences by DNA sequencing.

• 3.TALEN assembly into pDP(GFP)-TAL-ELD and pDP(RFP)-TAL-KKR constructs

Assemble the TALEN arms essentially following the instructions for Golden Gate TALEN and TAL Effector Kit (https://www.addgene.org/taleffector/goldengatev2/ (https://www.addgene.org/taleffector/goldengatev2/)) [40] except for the use of the TALEN backbone vectors.

Assemble the TALEN left arm (TALEN-L) and right arm (TALEN-R) into the double-promoter transgenesis constructs-based TALEN backbones pDP(GFP)-TAL-ELD and pDP(RFP)-TAL-KKR, respectively.

Confirm the TALEN-L and TALEN-R by DNA sequencing.

3.2. In Vitro Transcription

1. Linearize plasmid DNA with restriction enzyme NotI so that the SP6 RNA polymerase transcription would produce sense mRNA with a 3′-untranslated region followed by a poly(A)₃₀ tail suitable for TALEN translation (see Note 5). Prepare enough linearized DNA (e.g., digest about 20μg) for at least several transcription reactions at 1 μg per reaction.

2. Confirm linearization and DNA purity by running a small aliquot on an agarose gel (see Note 6).

3. Purify the linearized TALEN constructs following the manufacturer’s instructions of a PCR Purification kit (see Note 7).

4. Transcribe RNA following directions of Ambion’s mMessage mMachine kit (see Note 8). One μg of the DNA was transcribed in vitro with SP6 RNA polymerase.

5. Remove template DNA, purify RNA and quantitate it (see Note 9).

6. RNA can be aliquoted and stored at −80 °C for months.

3.3. In Vitro Fertilization of Xenopus Eggs

1. Xenopus tropicalis males and females are primed with 20 U of chorionic gonadotropin (HCG) the evening prior to the day that ovulation is desired (see Note 10).

2. The primed frogs are then boost with a 200 U of HCG injection early next morning. The females can be injected at different intervals of 1~2 h apart so they begin to ovulate at different times to ensure constant supply of eggs within the whole day.

3. A Xenopus male frog is sacrificed when the first batch of females begin to ovulate, the testes are removed, and keep in 1× MMR on ice.

4. Macerate 1 testis in 0.3 ml of 1 × MMR to make sperm solution, transfer into a 1.5 ml microcentrifuge tube, and keep on ice. The sperm solution can be used throughout the day.

5. Eggs are expelled from ovulating females onto petri dishes with gentle squeezing. Remove the water from the petri dish and add 0.1 ml of the sperm solution for about 200 eggs and 0.9 ml of ddH₂O.

6. After gentle splaying out of eggs into the sperm solution and 2-4 min incubation, the dishes are flooded with 0.1× MMR and incubated for an additional 15 min.

7. To remove the jelly coat, the fertilized eggs are washed with 3% cysteine, pH 8.0, until they started to touch each other (about 3-5 min), and then washed 4-6 times with 0.1× MMR before being transferred into fresh 0.1× MMR/6% Ficoll for microinjection (see Note 10).

3.4. Microinjection of TALEN mRNAs and Mutation Analysis

1. Place the fertilized eggs in an agar-coated plate and inject the mRNA into the cytoplasm of the fertilized eggs prior to the first division with a Drummond Nanojet II Auto-Nanoliter Microinjector.

2. For TALEN-targeted mutagenesis of TRα, equal amounts of TALEN-L and TALEN-R mRNA were mixed together and injected at 400 pg for each mRNA per egg (see Note 11).

3. The TALEN mRNA-injected eggs along with some uninjected eggs from the same batch of in vitro fertilization were incubated side by side in a 25 °C incubator overnight.

4. Examine the embryos for potential phenotypes following TALEN mRNA injection (see Note 12).

5. Collect embryos for mutation rates evaluation starting from day 3 after the TALEN mRNA injection (see Note 13). To do so, three or more embryos from the same group are pooled together and lysed with the lysis buffer (10 mM Tris–HCl, 100 mM NaCl, 10 mM EDTA, and 0.5% SDS, pH 8.0) containing 8 μg/ml proteinase K at 56 °C overnight.

6. Precipitate the genomic DNA with isopropanol, wash with 70% ethanol, dry in air for 10 min., and dissolve the DNA in 10 mM Tris–HCl, pH 8.0.

7. Subject the genomic DNA to PCR amplification for a DNA fragment encompassing the 2 TALEN-targeted regions by using primers F and R [35].

8. Clone the PCR fragments into pCR TOPO-TA vector (see Note 14), transform into E. coli competent cells, plate the cells on LB plates containing 100 μg/ml ampicillin, and incubate at 37 °C overnight.

9. Screen the colonies for mutations by colony PCR using semi-nested primers f or f1 paired with R [35] (see Note 15).

10. Grow up the colonies containing mutant DNA as identified by PCR above for plasmid DNA purification and sequencing to determine the exact mutations (see Note 16)

11. If the mutation rate is high, grow up the remaining injected embryos for phenotypic analysis of the Fo knockdown animals, or use the Fo animals to produce total knockout animals for analysis (see [35, 38].