March 1, 2008;
IRE1beta is required for mesoderm formation in Xenopus embryos.
is an atypical serine/threonine kinase transmembrane protein with RNase activity. In the unfolded protein response (UPR), they function as proximal
sensor of the unfolded proteins in the endoplasmic reticulum (ER). Upon activation by ER stress, IRE1
performs an unconventional cytoplasmic splicing of XBP1
pre-mRNA and thus allows the synthesis of active XBP1
, which activates UPR target genes to restore the homeostasis of the ER. IRE1
signaling is hence essential for UPR but its function during embryogenesis is yet unknown. The transcripts of the two isoforms of IRE1
in Xenopus, xIRE1alpha and xIRE1beta are differentially expressed during embryogenesis. We found that xIRE1beta is sufficient for cytoplasmic splicing of xXBP1
pre-mRNA. Although gain of xIRE1beta function had no significant effect on Xenopus embryogenesis, overexpression of both, xIRE1beta and xXBP1
pre-mRNA, inhibits activin A induced mesoderm
formation, suggesting that an enhanced activity of the IRE1
pathway represses mesoderm
formation. Surprisingly, while loss of XBP1
function promotes mesoderm
formation, the loss of IRE1beta
function led to a reduction of mesoderm
formation, probably by action of IRE1
being different from the IRE1
pathway. Therefore, both gain and loss of function studies demonstrate that IRE1
is required for mesoderm
formation in Xenopus embryos.
[+] show captions
Fig. 2. Tempo-spatial expression patterns of xIRE1α and xIRE1β. (A) Temporal expression patterns of xIRE1α and xIRE1β revealed by real-time RT-PCR. (B–G) Spatial expression pattern of xIRE1α. ba, branchial arch; ey, eye; fb, forebrain; hb, hindbrain; mb, midbrain; nt, neural tube; ov, otic vesicle; pa, pancreas anlage; pn, pronephros. (H–M) Spatial expression pattern of xIRE1β. cg, cement gland; hg, hatching gland.
Fig. 3. Overexpression of xIRE1β does not affect mesoderm and endoderm formation but enhances cleavage of xXBP1 pre-mRNA. (A) Uninjected control embryo at tailbud stage. (B) Injection of xIRE1β at 1.5 ng. (C and D) Overexpression of xIRE1β did not change the expression of Xbra (C) and Xsox17α (D) at stage 10.5. (E) Overexpression of xIRE1β resulted in increase of cytoplasmic splicing of xXBP1. ODC (ornithine decarboxylase) was used as loading control. xXBP1(U), unspliced xXBP1; xXBP1(N), nuclear spliced xXBP1.
Fig. 9. ER-luminal domain and cytoplasmic domain of xIRE1β show different effects on embryogenesis and xXBP1 splicing and have different subcellular localizations. (A) Diagram depicting the construction of deletion mutants. S and TM denote the signal peptide and the transmembrane domain. (B) An uninjected control embryo at tailbud stage. (C and D) Embryos injected with 1 ng xIRE1βδC (C) and 500 pg xIRE1βδN (D) caused phenotypic changes. (E) Expression of Xbra was repressed by injection of xIRE1βδC (middle) but not by injection of xIRE1βδN (right). (F) Xsox17α was not affected significantly by injection of either xIRE1βδC (middle) or xIRE1βδN (right). (G) Injection of xIRE1βδC resulted in decrease of cytoplasmic variant of xXBP1 in embryos collected at indicated stages while injection of xIRE1βδN caused an increase. (H) Monitoring of subcellular localization and splicing activity of the xIRE1β-specific EGFP fusion constructs xIRE1β (1–958, upper panel), xIRE1βδC (1–549, middle panel) and xIRE1βδN (439–958, lower panel) in HeLa cells. Both, xIRE1β and xIRE1βδC show a similar subcellular localization but xIRE1βδC the did not activate the xXBP1(U)-RFP sensor (middle panel). The xIRE1βδN mutant is active but shows a different subcellular localization (lower panel).