Recombination signal sequence-binding protein Jkappa alters mesodermal cell fate decisions by suppressing cardiomyogenesis

Abstract

The transcription factor recombination signal sequence-binding protein Jkappa (RBP-J) is a key downstream element in the signaling pathway of all four mammalian Notch receptors that are critically involved in the control of embryonic and adult development. RBP-J-deficient mice display complex defects and die around day 9.5 postcoitum. Here, we investigate the function of RBP-J in the development of mesodermal cell lineages by using the OP9 stroma coculture system. RBP-J-deficient embryonic stem (ES) cells gave rise to cardiomyocytes, endothelial cells, and primitive and definitive hematopoietic cells. Thus, RBP-J-mediated signals are not required for generation of these cell types. However, when compared with parental RBP-J-expressing ES cells, cardiomyogenesis derived from RBP-J-deficient ES cells was increased. Repression over the cardiogenic pathway was restored by expressing RBP-J in RBP-J-deficient ES cells. Our data indicate that Notch signaling via RBP-J plays an important role for the correct specification of myocardial cell fates.

Figure 1

Confirmation of the absence of functional RBP-J in RBP-J^−/− ES cell lines. (A) Absence of RBP-J protein in RBP-J^−/− ES cells. Western blot analysis of RBP-J^+/+, RBP-J^+/−, and RBP-J^−/− ES cells is shown. The unspecific band shows equal loading of all lanes. (B) Absence of functional, RBP-J-dependent Notch signaling in RBP-J^−/− ES cells. Transient reporter transfections were done by using a luciferase reporter under the control of 12 RBP-J-binding sites [(RBP-J RE)₁₂-Luc]. Transient cotransfection of RBP-J restored RBP-J-dependent signaling of activated Notch (N^IC).

Figure 2

Generation of RBP-J^−/−-derived ES cell clones expressing RBP-J or VP16-RBP-J. (A) Expression of the RBP-J and VP16-RBP-J proteins in rescued RBP-J^−/− ES clones. The unspecific band shows equal loading of all lanes. (B) Detection of functional RBP-J in RBP-J rescued clones. Rescue of RBP-J-dependent signaling of activated Notch (N^IC) was confirmed by using a luciferase reporter under the control of 12 RBP-J-binding sites. (C) Detection of functional VP16-RBP-J in VP16-RBP-J rescued clones. Constitutive activation of RBP-J-dependent signaling was detected by using luciferase reporters under the control of a minimal promoter with or without additional 12 RBP-J-binding sites.

Figure 3

Increased cardiomyogenesis from RBP-J^−/− ES cells. Flk1⁺ cells derived from ES cells were cultured on OP9 cells for 6 days and analyzed for spontaneously contracting colonies and expression of cardiac-specific markers. (A) ES cells can be differentiated into cardiomyocytes coexpressing the cardiac-specific markers ventricular myosin and GATA4 or Nkx-2.5. The double-positive colonies showed the typical morphology of colonies spontaneously contracting in living cultures. Magnification is ×630. (B) Increased generation of cells expressing the cardiac marker ventricular myosin in the absence of RBP-J. Expression of ventricular myosin is visualized by a dark-purple staining. Unstained compact ES cell-derived colonies show a light-brown color. Three independent RBP-J^+/− and RBP-J^−/− clones were analyzed in triplicate or quadruplicate cultures. Photomicrographs of one representative RBP-J^+/− clone and one representative RBP-J^−/− clone are shown. Magnification is ×25. (C) Increased generation of cardiomyocyte colonies in the absence of RBP-J. Three independent RBP-J^+/− and RBP-J^−/− clones were analyzed in triplicate or quadruplicate cultures. Mean values ± SEM of all replicates of the three clones are shown as one bar, respectively. Synchronously beating colonies were counted as one cardiomyocyte colony. The experiment was repeated six times with virtually identical results. One representative experiment is shown. The increase in cardiomyocyte colonies derived from RBP-J-deficient ES cells is statistically significant (P < 0.001). (D) The increased cardiomyogenesis of RBP-J^−/− ES cells can be reverted by exogenous RBP-J expression. Two independent mock- and RBP-J-rescued clones (for the three original RBP-J^−/−clones) were analyzed in quadruplicate cultures. The experiment was repeated four times with virtually identical results. One representative experiment is shown. The decrease in cardiomyocyte colonies derived from RBP-J-rescued ES cells is statistically significant (P < 0.001).

Figure 4

Hematopoietic development from RBP-J^−/− ES cells in vitro. Flk1⁺ cells derived from three independent RBP-J^+/− and RBP-J^−/− clones were cultured on OP9 cells for 10 days. The experiment was repeated six times with virtually identical results. (A) Photomicrographs of hematopoietic cells derived from RBP-J^−/− ES cells. Cytospin preparations were stained with May–Grünwald–Giemsa. Magnification is ×1,000. (i) Primitive erythroid cells. (ii) Neutrophil granulocytic cells. (iii) Megakaryocyte. (iv and v) Definitive erythroid cells. (vi) Macrophages. (B) Detection of hematopoietic progenitors and B lymphoid cells derived from RBP-J^−/− ES cells by expression of cell surface markers. Expression of surface markers on cultured cells derived from RBP-J^−/− J ES cells was analyzed by FACS on days 8 and 10 of differentiation on OP9 stroma. Results of one representative clone are shown.

Figure 5

Reduced but not RBP-J-rescued generation of hematopoietic colony-forming cells from RBP-J^−/− endothelial–hematopoietic progenitors. ES cells differentiated into PECAM-1⁺ cells on OP9 stromal cells were cloned in methylcellulose in the presence of hematopoietic cytokines in triplicate. One RBP-J^+/−, one RBP-J^+/+, two different RBP-J^−/−, and two independent mock- and RBP-J-rescued subclones (for two original RBP-J^−/− clones) were analyzed in triplicate cultures. Mean values ± SEM of the colony number generated from the different genotypes after 8 days are shown as one bar. Mixed, colonies consisting of erythroid and myeloid cells; BFU-e, erythroid bursts; GM, colonies consisting of granulocytes and/or macrophages. The experiment was repeated twice with virtually identical results. One representative experiment is shown. No obvious differences in the colony types were apparent between RBP-J^+/− and RBP-J^−/− cultures. The decrease in hematopoietic colonies derived from RBP-J^−/− ES cells is statistically significant (P < 0.001), but the numbers of hematopoietic colonies derived from mock- and RBP-J-rescued clones are not significantly different (P > 0.2).