Depletion of retinoic acid receptors initiates a novel positive feedback mechanism that promotes teratogenic increases in retinoic acid

Abstract

Normal embryonic development and tissue homeostasis require precise levels of retinoic acid (RA) signaling. Despite the importance of appropriate embryonic RA signaling levels, the mechanisms underlying congenital defects due to perturbations of RA signaling are not completely understood. Here, we report that zebrafish embryos deficient for RA receptor αb1 (RARαb1), a conserved RAR splice variant, have enlarged hearts with increased cardiomyocyte (CM) specification, which are surprisingly the consequence of increased RA signaling. Importantly, depletion of RARαb2 or concurrent depletion of RARαb1 and RARαb2 also results in increased RA signaling, suggesting this effect is a broader consequence of RAR depletion. Concurrent depletion of RARαb1 and Cyp26a1, an enzyme that facilitates degradation of RA, and employment of a novel transgenic RA sensor line support the hypothesis that the increases in RA signaling in RAR deficient embryos are the result of increased embryonic RA coupled with compensatory RAR expression. Our results support an intriguing novel mechanism by which depletion of RARs elicits a previously unrecognized positive feedback loop that can result in developmental defects due to teratogenic increases in embryonic RA.

Figure 1. RARαb1 and RARαb2 sequences and expression.

(A) Schematic representation of RAR domains. Blue box indicates the variable A domain, which is different between RARαb1 and the previously identified RARαb2 splice variant. (B) Schematic representation of RARαb1 and RARαb2 genomic organization (adapted from Ensemble_v9). Blue bars represent the first exon, which encodes the respective A domains. White bars represent the 5′ UTRs. Black bars represent the exons that are common to the two variants. Green bars represent the target of the antisense probes used for ISH. Red arrows indicate the position of the primers used to perform RT-PCR. Purple bars indicate the position of the morpholino target sequences. (C) Alignments of the A domains of human (Hs) RARα1, mouse (Mm) RARα1, and zebrafish (Dr) RARαb1. The presence of this previously unrecognized splice variant was recently confirmed in the latest zebrafish genome assembly (Ensemble Zv9). There is no RARαa splice variant 1 ortholog in the zebrafish genome. (D) Alignments of the A domains of Hs RARα2, Mm RARα2, Dr RARαa2, and Dr RARαb2. (E) Reverse transcriptase PCR (RT-PCR) for the zebrafish rarαb isoforms. max was used as the control. -RT control did not reveal genomic contamination (data not shown). (F) Rarαb1 is expressed in the ventral anterior of the embryo and the presomitic paraxial mesoderm (arrow) at the 8 somite (s) stage. (G) Rarαb2 is expressed in rhombomeres 5 and 6, the spinal cord and the posterior lateral plate mesoderm (LPM). Arrows indicate the space between the posterior spinal cord and LPM expression domains. (H) Together, the expression patterns recapitulate a previously reported rarαb probe (referred to as rarαb1/2), which detects both isoforms . In F–H, embryos are flatmounted and are dorsal views with anterior up.

Figure 2. RARαb1 deficient embryos have enlarged hearts with increased CM number.

(A, B) Hearts from control sibling and RARαb1 deficient Tg(-5.1myl7:DsRed-NLS)^f2 embryos. Images are frontal views. Red indicates ventricle. Green indicates atrium. (C–H) ISH for CM differentiation marker genes. (I–L) ISH for CM progenitor marker genes. Brackets in I and J indicate length of nkx2.5 expression. Arrows in K indicate posterior and anterior limits of the hand2 expression domains in the LPM. In C–L, views are dorsal with anterior up. (M) Mean CM number at 48 hpf. (N) qPCR for CM differentiation marker gene expression at 48 hpf. (O) Areas of the amount of cells expressing the CM differentiation marker genes at the 20 s and 22 s stages. (P) qPCR for CM differentiation marker gene and nkx2.5 expression at 24 hpf. (Q) qPCR for CM progenitor gene expression at the 8 s stage. Asterisk in all graphs indicate a statistically significant difference compared to controls (p<0.05). Error bars in all graphs indicate standard deviation.

Figure 3. RARαb1 deficient embryos have increased expression of RA signaling responsive genes.

(A) qPCR for RA signaling responsive gene expression at the 8 s stage. (B, C) ISH for hoxb5b expression at the 8 s stage. Bracket indicates length of expression in the LPM. Views are dorsal with anterior up. (D, E) ISH for cyp26a1 expression at the 8 s stage. Arrows in E indicate increased expression in the tailbud and spinal cord. Views are lateral with anterior up and dorsal right. (F–H) Fronto-lateral views of Tg(-5.1myl7:GFP)^f2 embryos at 48 hpf of control sibling, RARαb1 deficient embryos, and RARαb1+Hoxb5b deficient embryos. (I–K) Hearts from control sibling, RARαb1 deficient embryos and RARαb1+Hoxb5b deficient Tg(-5.1myl7:DsRed-NLS)^f2 embryos. Images are frontal views. Red indicates ventricle. Green indicates atrium. (L) Percentage of control+rarαb1 MOs (n = 60), hoxb5b+rarαb1 MOs (n = 68) showing enlarged and normal hearts. (M) qPCR for CM differentiation gene expression at 48 hpf. (N) Mean CM number at 48 hpf.

Figure 4. Concurrent depletion of RARαb1 and RARαb2 promotes increased RA signaling and atrial CM number.

qPCR for (A) RA signaling responsive gene, (B) RA metabolizing gene, and (C) zebrafish rar expression in control sibling, RARαb1 deficient, RARαb2 deficient, RARαb1+RARαb2 (suboptimal doses) deficient, RA treated, and DEAB treated embryos at the 8 s stage. (D–G) ISH for egfp expression in Tg(12XRARE-ef1a:EGFP)^sk72 embryos. Brackets indicate the length of egfp expression in the spinal cord. (H) Measurements of the length in arbitrary units (AU) of egfp expression in the spinal cord of Tg(12XRARE-ef1a:EGFP)^sk72 embryos. (I–L) Hearts from control and RARαb depleted Tg(-5.1myl7:DsRed-NLS)^f2 embryos. Images are frontal views. Red indicates ventricle. Green indicates atrium. (M) Mean CM number from Tg(-5.1myl7:DsRed-NLS)^f2 hearts at 48 hpf. (N) qPCR for CM marker gene expression at 48 hpf. While modest increases in vmhc expression in RARαb1+RARαb2 deficient embryos were observed relative to RARαb1 (suboptimal dose) deficient embryos, corresponding increases in ventricular CM number were not observed.

Figure 5. Concurrent depletion of RARαb1 and Cyp26a1 results in phenotypes resembling RA treatment.

(A–D) Control sibling, RARαb1 deficient, Cyp26a1 deficient, and RARαb1+Cyp26a1 deficient embryos. A suboptimal dose of the cyp26a1 MOs was used that did not cause ostensible defects for these experiments. In D, arrow indicates loss of the MHB and line indicates shortened tail. Images are lateral views with dorsal right and anterior up. (E–H) ISH for eng2a, which marks the MHB. 100% of (E) control sibling (n = 11), (F) RARαb1 deficient (n = 7), and (G) Cyp26a1 deficient (n = 7) had eng2a expression. 85% of (H) RARαb1+Cyp26a1 deficient embryos (n = 7) had a complete absence of eng2a expression (arrow in H). Equivalent results were obtained using pax2a, which also marks the MHB (data not shown). (I–L) Hearts from control sibling, RARαb1 deficient, Cyp26a1 deficient, and RARαb1+Cyp26a1 deficient Tg(-5.1myl7:DsRed-NLS)^f2 embryos. Images are frontal views. Red indicates ventricle. Green indicates atrium. (M–P) ISH for egfp in Tg(β-actin:GDBD-RLBD)^cch1;Tg(UAS:EGFP) embryos. Lateral views with dorsal right and anterior up. (Q) Mean CM number at 48 hpf and (R) qPCR for egfp expression at 15 s in control sibling, RARαb1 deficient, Cyp26a1 deficient, and RARαb1+Cyp26a1 deficient embryos. Double asterisks in Q indicate a statistically significant difference relative to control and RARαb1 deficient embryos. Pound sign in Q indicates a statistically significant difference relative to RARαb1 deficient embryos.

Figure 6. Reduction of RA in RARαb1 deficient embryos can rescue developmental defects.

(A–H) Control sibling, RARαb1+Cyp26a1 deficient, control sibling treated with DEAB, and RARαb1+Cyp26a1 treated with DEAB embryos. In B and F, arrows indicates loss of the MHB and eng2a expression. Images are lateral views with dorsal right and anterior up. (I) Percentage of control sibling (n = 16), RARαb1+Cyp26a1 deficient embryos (n = 16), control sibling embryos treated with DEAB (n = 13), and RARαb1+Cyp26a1 deficient embryos treated with DEAB (n = 14) that had a MHB based on morphology. (J) Percentage of control sibling (n = 17), RARαb1+Cyp26a1 deficient embryos (n = 14), control sibling embryos treated with DEAB (n = 15), and RARαb1+Cyp26a1 deficient embryos treated with DEAB (n = 12) that had eng2a expression at the MHB. (K–N) Hearts from Tg(-5.1myl7:DsRed-NLS)^f2 control sibling, RARαb1 deficient, DEAB treated, and DEAB+RARαb1deficient embryos. Images are frontal views. Red indicates ventricle. Green indicates atrium. (O) Mean CM number at 48 hpf. (P,Q) Hearts from Tg(-5.1myl7:DsRed-NLS)^f2 control sibling embryos and Tg(-5.1myl7:DsRed-NLS)^f2 embryos treated with a low concentration of RA. Images are frontal views. Red indicates ventricle. Green indicates atrium. (R) Mean CM number at 48 hpf.

Figure 7. Models of the effects of RA signaling on heart patterning and the RA feedback mechanism.

(A) Model depicting the consequences on atrial and ventricular CM specification at different levels of RA signaling. (B) Model of the previously unrecognized feedback mechanism that triggers increased RA signaling when depleting RARs. ROL = retinol. RAL = retinal. Red and green arrows indicate the effects on gene expression. Blue arrow indicates the effect on RA levels.