Kodjabachian L et al. (2001), A study of Xlim1 function in the Spemann-Mangol...

XB-ART-9243

Int J Dev Biol January 1, 2001; 45 (1): 209-18.

A study of Xlim1 function in the Spemann-Mangold organizer.

Kodjabachian L , Karavanov AA , Hikasa H , Hukriede NA , Aoki T , Taira M , Dawid IB .

Abstract
The Spemann-Mangold organizer is required in amphibian embryos to coordinate cell fate specification, differentiation of dorsal cell types and morphogenetic movements at early stages of development. A great number of genes are specifically expressed within the organizer, most of them encoding secreted proteins and transcription factors. The challenge is now to uncover genetic cascades and networks of interactions between these genes, in order to understand how the organizer functions. The task is immense and requires loss-of-function approaches to test the requirement for a given factor in a specific process. For transcription factors, it is possible to generate inhibitory molecules by fusing the DNA binding region to a repressor or activator domain, which should in principle antagonize the activity of the endogenous protein at the level of the DNA targets. We used this strategy to design activated and inhibitory forms of the LIM homeodomain transcription factor Lim1, which is encoded by an organizer gene involved in head development, as revealed by analyses of knockout mice. We found that Lim1 is a transcriptional activator, and can trigger dorso-anterior development upon ventral expression of hyperactive forms, in which Ldb1 is fused to Lim1. Using inhibitory Lim1 fusion proteins, we found that Lim1, or genes closely related to it, is required for head formation as well as for notochord development. Co-expression experiments revealed that Lim1 is required downstream of the early organizer factor Siamois, first, to establish the genetic program of the organizer and second, to mediate the action of organizer agents that are responsible for blocking ventralizing activities in the gastrula.

PubMed ID: 11291848
Article link: Int J Dev Biol

Genes referenced: cer1 dkk1 egr2 en2 frzb ldb1 lhx1 otx2 sia1 ventx1.1

Article Images: [+] show captions

	Fig. 2. Phenotypes induced by activated and repressive forms of Lim1. (A) Partial secondary axis induced by ventral co-injection of 400 pg lim1 and 400 pg ldb1 RNAs at the 4-cell stage. (B) Partial ectopic axis induced by ventral injection of 1 ng ldb1-lim1 chimeric RNA at the 4-cell stage. (C) Induction of a single eye by ventral injection of 6ng ldb1-lim1 chimera RNA at the 4-cell stage. Note that this embryo shows deficient secondary axis formation. (D) Dorsal injection at the 4-cell stage of 1 ng ldb1-lim1 chimera RNA leads to suppression of trunk/tail structures. (E) Partial ectopic axis induced by ventral injection of 100 pg ldb1-lim1-VP16 chimera RNA at the 4-cell stage. (F) Injection of 500 pg ldb1-lim1-VP16 chimera RNA at the 4-cell stage stimulates cement gland formation (CG) but not axis development.
	Figure 2 (continued) (G) Ventral injection at the 4-cell stage of 40 pg enR-HD chimera RNA does not provoke any visible phenotype. (H) Secondary axis formation mediated by ldb1-lim1-VP16 chimera RNA (200 pg) is suppressed by the repressive enR-HD chimera RNA (40 pg) upon ventral co-injection at the 4-cell stage.
	Fig. 4. Phenotypes elicited by expression of repressive forms of Lim1. Dorsal injection at the 4-cell stage of 250 pg ∆C Xlim1-enR mRNA provokes anterior truncation (B,D) compared to uninjected siblings (A,C). Embryos were subjected to two-color whole-mount in situ hybridization analysis with two head markers, krox20 (rhombomeres 3 and 5; dark blue) and en2 (isthmus; red) at the tailbud stage. The head in ∆C Xlim1-enR injected embryos is truncated anterior to rhombomere 5 (D). Dorsal injection at the 4-cell stage of 100 pg HD-enR RNA also leads to head truncation and deficiency in notochord, as revealed by staining with the MZ15 antibody (E) or histological section (F). (G) Control section of an uninjected embryo. Note that somitic tissue is present in headless embryos and actually expands in the region normally occupied by the notochord (F). mu, muscle; not, notochord.
	Fig. 6. Repressive forms of Lim1 do not antagonize activity of Otx2 and Siamois in animal tissues. (A,B) Embryos were injected in ventralanimal position at the 4-cell stage with (A) 800pg otx2 RNA or (B) a mixture of 800pg otx2 and 100pg HD-enR RNAs. Over-expression of otx2 leads to the formation of cement gland tissue in the epidermis, and this effect is not antagonized by coexpression of HD-enR. (C,D) Embryos were injected animally at the 2-cell stage with (C) 20 pg Siamois RNA, or (D) a mixture of 20 pg Siamois and 100 pg HD-enR RNAs. Animal caps were dissected at stage 9 and cultured until siblings reached tadpole stages. Siamois expression in animal cells stimulates cement gland formation, and this effect is not suppressed by co-expression of HD-enR. Arrowheads point at cement glands.
	Fig. 7. Repressive form of Lim1 inhibits expression of organizer genes except Siamois. Embryos were injected at the 4-cell stage with a mixture of 100 pg enR-HD and 400 pg lacZ RNAs. (A) After revealing β-galactosidase activity with a red substrate at stage 10.25/10.5, embryos were processed for in situ hybridization with the indicated probes. In the cases of dkk1, frzb and cerberus, embryos were bisected before hybridization to enhance staining in the deep layers of the embryo. All dorsal genes examined are repressed by enR-HD. However, the ventral-posterior gene PV.1 is not ectopically expressed dorsally. Dorsal is right for dkk1, frzb and cerb panels, and dorsal is up in all other panels. (B,C) RT-PCR experiments on embryos injected at the 4-cell stage in a dorsal vegetal position with 100 pg enR-HD RNA. This site of injection targets cells that express the organizer gene Siamois. Panel B shows that expression of Siamois is unchanged in enR-HD injected embryos between stages 9 and 10.25. Panel C shows that the same embryos exhibit reduced levels of expression of cerberus and otx2, confirming that enR-HD was active in this experiment. FGF-R is used as a loading control
	Fig. 8. Position of Lim-1 in the genetic cascade of the Spemann-Mangold organizer. (A,B) Embryos were injected ventrally at the 4-cell stage with (A) 20 pg Siamois RNA or(B) with a mixture of 20 pg Siamois and 100 pg HD-enR RNAs. Expression of Siamois leads to formation of a complete secondary axis, and co-expression of the repressive form of Lim1 suppresses this effect. (C,D) Embryos were injected ventrally at the 4-cell stage with (C) 500 pg truncated BMP receptor (tBR) and 400 pg frzb RNAs or (D) with a mixture of 500 pg tBR, 400 pg frzb and 100 pg HDenR RNAs. Co-expression of the BMP and Wnt antagonist’s tBR and frzb promotes formation of a complete secondary axis, and this effect is inhibited by coexpression of the repressive form of Lim1. (E,F) in situ hybridization with XLim1 probe on (E) stage 10.5 uninjected embryos or (F) embryos injected as in C, reveals activation of XLim1 by co-expression of BMP and Wnt antagonists. (G) Model for XLim1 position within the organizer genetic cascade based on its requirement downstream of known regulators of organizer function. Lim1 activity is critical downstream of Siamois in establishing the organizer (phase 1), and is needed to relay the activity of organizer’s inhibitors to allow dorso-anterior development (phase 2).