Lack of sik1 in mouse embryonic stem cells impairs cardiomyogenesis by down-regulating the cyclin-dependent kinase inhibitor p57kip2

Abstract

Sik1 (salt inducible kinase 1) is a serine/threonine kinase that belongs to the stress- and energy-sensing AMP-activated protein kinase family. During murine embryogenesis, sik1 marks the monolayer of future myocardial cells that will populate first the primitive ventricle, and later the primitive atrium suggesting its involvement in cardiac cell differentiation and/or heart development. Despite that observation, the involvement of sik1 in cardiac differentiation is still unknown. We examined the sik1 function during cardiomyocyte differentiation using the ES-derived embryoid bodies. We produced a null embryonic stem cell using a gene-trap cell line carrying an insertion in the sik1 locus. In absence of the sik1 protein, the temporal appearance of cardiomyocytes is delayed. Expression profile analysis revealed sik1 as part of a genetic network that controls the cell cycle, where the cyclin-dependent kinase inhibitor p57(Kip2) is directly involved. Collectively, we provided evidence that sik1-mediated effects are specific for cardiomyogenesis regulating cardiomyoblast cell cycle exit toward terminal differentiation.

Figure 1. Effects of gene trapping insertion and screening of homozygous mutant ES cells.

(A) Schematic representation of the sik1 gene encoding for a protein of 779 amino acids. The SIK1 protein structure is visualized: the kinase domain (27 to 278), the UBA motif (303–343), the PKA motif (567–584) and the nuclear localization signal (586 to 612). The insertion of the gene-trap vector occurred in the seventh intron of the wild-type sik1 gene, causing a deletion of the protein segment from residue 250 to 779. The corresponding fusion transcript and the resulting protein are depicted. (B) Identification of homozygous ES clones for the trapped allele with RT-PCR analysis. Schematic representation of the oligonucleotides used to discriminate between the wild-type and fusion transcripts. X, Y and Z correspond to primers sik1UP, sik1LW and L232, respectively. The amplification of wild-type and mutant cDNA results in two products of different molecular weight, 682 bp and 378 bp respectively. Lanes: 1,2 and 3, amplification of sik1^wt/wt cDNA; 4,5 and 6, amplification of sik1^wt/flp cDNA; 7,8 and 9 amplification of sik1^flp/flp cDNA; 10, sik1^wt/wt total RNA; 11 and 12, sik1^wt/flp total RNA; 13, sik1^flp/flp total RNA. The arrow indicates the 500-bp band of the DNA molecular weight marker. The arrow indicates the 500-bp band of the DNA molecular weight marker. (C) Western blot on proteins extracted from wild-type and sik1^flp/flp ES cells using an antibody against the β-galactosidase. In sik1^flp/flpES cells is possible to detect a band corresponding to the fusion protein that is absent in wild-type ES cells. Different amounts of proteins were used on western blot. (D) Expression plasmids (6 µg) for FLAG-tagged sik1 and its truncated version (1–249aa) were transfected into HEK293T cells and subjected to immunoprecipitation with FLAG-M2 resin. Aliquots of FLAG-SIK1 wt and truncated IPs were subjected to western blotting (WB) using an anti-FLAG antibody (left panel) and to in vitro kinase assays using [γ-³²P]- ATP with AMARA peptide as a substrate (right panel). The relative activation was calculated by subtracting counts incorporated in the assay of non-immune sample from gross counts of immunoprecipitated samples to determine net cpm. (E) Intracellular localization of HA-tagged wild-type and mutant SIK1 proteins in HEK293T cells.

Figure 2. In vitro differentiation and proliferation potential of sik1^flp/flp ES cells.

(A) Percentages of EBs differentiating to beating cardiomyocytes. EBs derived from wild-type, sik1^wt/flp, sik1^flp/flp and sik1^flp/flp+HA-sik1 ES cells were used to monitor differentiation into cardiomyocytes. Sik1^flp/flp + HAsik1 used corresponds to HAsik1 B6 clone expressing the highest level of sik1 (B). Three independent experiments were performed, each plating an equal number of EBs (n = 120). The statistical significance of the observed differences between the percentage of wild-type and mutant beating EBs were analyzed and the time points in which major differences were observed (days 6 and 7) showed significant p-values of 1.109e⁻⁰⁸ and 3.271e⁻⁰⁸, respectively. P-values were calculated using a two-sample test for equality of proportions with continuity correction. (B) RT- and Q-PCR analysis on wild-type ES cells and three representative clones carrying the HA-sik1 transgene. Data represent fold change of transgene expression vs. sik1 endogenous level. (C) Proliferation assay measured by BrdU incorporation. (D). Cell-cycle distribution of the wild-type and sik1^flp/flp populations by FACS analysis.

Figure 3. sik1 deficiency affects terminal cardiogenic differentiation.

(A) Q-PCR of a pluripotency gene (Oct3/4), of an early mesodermal marker (brachyury), of pre-cardiac mesodermal markers (mesp1, mesp2, Isl1), early cardiogenic markers (Mef2c, Tbx5, Nkx2.5) and terminal cardiac markers (α-MyHC, Mlc2V, αCach and cTnI) during in vitro differentiation of wild-type and sik1^flpflp ES cells. (B) Q-PCR on hematopoietic/endothelial markers (runx1, vegfC and flk1). n = 3, *, P<0.05, **, P<0.001 vs. control cells. P values were calculated using a two-tailed, unpaired t test.

Figure 4. Measurement of myosin positive cells in wild-type and sik1^flp/flp EBs by FACS.

(A) EBs were analyzed by FACS at days 0, 5, 6, 7, 9, 11 and 12 of differentiation with the antibody MF20. Data represent results of one out of three independent experiments. (B) Graphic representation of FACS analysis. Data represent mean ± S.E. of three biologically independent experiments (*, p<0.05 compared with wild-type EBs. P-values were calculated using a two-tailed, unpaired t test). (C) Graphic representation of low fluorescence vs high fluorescence MF20 positive cells distribution, as by FACS analysis (n = 3). Wild-type and sik1^flpflp cells were used throughout the analysis. (D) Costaining experiment using anti-troponin and anti-BrdU to detect proliferating cells only in cardiomyocytes population.

Figure 5. Neuronal differentiation of wild-type and sik1^flp/flp ES cells.

(A) Immunofluorescence analysis of wild-type and sik1^flp/flp EBs at day 16 of differentiation using an anti-βIII-tubulin antibody reveals the presence of neurons. (B) Quantitative real-time PCR analysis to measure the expression of the neuron- and glial-specific markers NFM and GFAP, respectively, in wild-type and sik1^flp/flp EBs at different time points during differentiation, n = 3.

Figure 6. Expression profile of CDKIs during both cardiomyocyte and neuron differentiation.

(A) Expression profile of genes encoding for CDK inhibitors belonging to the KIP/CIP family (Cdkn1a, Cdkn1b and Cdkn1c) measured by quantitative real-time PCR at different time points during cardiomyocyte differentiation of wild-type and sik1^flpflp ES cells, n = 3, *, P<0.05, **, P<0.001 vs. control cells. P values were calculated using a two-tailed, unpaired t test. (B) Western blots analysis of CDKIs protein level during in vitro cardiomyocyte differentiation of wild-type and sik1^flpflp ES cells. (C) p21, p27 and p57 protein levels were expressed as the ratio between their arbitrary densitometric units (ADU) and b-actin. Densitometry analyses were performed using the ImageQuant 5.2 software (GE Healthcare). (D) Expression profile of Cdkn1c gene measured by quantitative real-time PCR at different time points during neuronal differentiation of wild-type and sik1^flpflp ES cells.

Figure 7. Over-expression of p57 in sik1^flp/fl cells.

(A) Western blot analysis with the anti-p57 antibodies to detect the expression of transgene in three different clones in sik1^flpflp ES cells. (B) Measurement of myosin positive cells in wild-type, sik1^flp/flp and sik1^flp/flp overexpressing p57 EBs by FACS, analyzed at days 0, 5, 6 and 7 of differentiation with the antibody MF20. Data represent results of one out of two independent experiments (C) Q-PCR analysis of terminal cardiac markers (α-MyHC, Mlc2V, αCach and cTnI).