Roles of Klf5 Acetylation in the Self-Renewal and the Differentiation of Mouse Embryonic Stem Cells

Abstract

Transcription factor Krüppel-like factor 5 (Klf5) plays important roles in the formation of the inner cell mass (ICM) and the trophectoderm during embryogenesis, as well as the self-renewal and the differentiation of mouse embryonic stem cells (ESCs). Acetylation of KLF5 has been shown to reverse the transcriptional activity of KLF5 in human epidermal cells and prostate cancer cells. Whether Klf5 acetylation contributes to the lineage specification in the blastocyst and pluripotency maintenance in ESCs remains unexplored. Here, we showed the ubiquitous expression of acetylated Klf5 in the ICM and the trophectoderm, ruling out the possibility that differential acetylation status of Klf5 leads to the lineage specification in the blastocyst. We found that K358Q mutation, mimicking acetylation, enhances the transcriptional activity of Klf5 for pluripotency genes in ESCs, and that K358Q Klf5 is more potent in pluripotency maintenance and in somatic cell reprogramming, compared to K358R Klf5. In ESCs, Klf5 acetylation, stimulated by TGF-β signaling, is involved in enhancing Sox2 expression. Moreover, upon ESC differentiation, acetylation of Klf5 facilitates the suppression of many differentiation genes, except for that K358Q Klf5 activates Cdx2, promoting trophectodermal differentiation. In summary, our results revealed the regulatory functions of Klf5 acetylation in ESC self-renewal and differentiation.

Fig 1. Klf5 expression in the blastocyst, ESCs and TSCs.

(A) Comparison of different species Klf5 protein sequences around the human KLF5 acetylation site. The numbers mark the coordination of the first shown residues “P” in Klf5 proteins of different species. The triangle highlights the conserved lysine residue 369 that is acetylated in human KLF5. (B) Immunofluorescence staining of total, acetylated (Ac-), and unacetylated (unAc-) Klf5 in the blastocyst. Cdx2 expression was shown to mark TE cells, and nuclei were stained by Hoechst. (C) Western blot of total, Ac-, and unAc-Klf5 in mouse ESCs and TSCs. Tubulin was used as a loading control.

Fig 2. Functions of Ac-Klf5 in self-renewal and pluripotency regulation.

(A) The effect of WT, KR, and KQ Klf5 overexpression on the expression of pluripotency genes in mouse ESCs. Empty pCAGIPuro, Flag-tagged WT, KR and KQ Klf5 overexpression plasmids were transfected into V6.5 ESCs with Lipofectamine 3000. Two days after transfection, cells were harvested and subjected to RNA purification or Western blot. The top panel shows RNA quantification results (averages and standard deviations from three independent experiments were plotted), and the bottom panel shows Western blots to demonstrate the similar expression levels of Flag-tagged WT, KR and KQ Klf5. (B) The effect of WT, KR, and KQ Klf5 overexpression on the expression of pluripotency genes during mouse ESC differentiation. Experiments were carried out as described in (A), except for that LIF was withdrawn from the medium upon transfection. Averages and standard deviations from three independent experiments were plotted. (C) Relative expression level of total Klf5 mRNA in control, WT, KR, and KQ Klf5 overexpression ESC lines. Averages and standard deviations from three independent experiments were plotted. (D) Proliferation rates of control, WT, KQ, and KR Klf5 overexpression ESCs. 1.5×10⁵ Cells were seeded in one well of a 6-well plate, and 48 hours later, total cell number was counted to measure the proliferation rate. Averages and standard deviations from three independent experiments were plotted. P values were calculated by reusable two-factor analysis of variance. (E) Colony forming assays for control, WT, KR, and KQ Klf5 overexpression ESCs growing without LIF and feeder cells. The left panel shows the image of AP stained colonies of each ESC line, and the quantification result of colony number is plotted in the right panel. Averages and standard deviations from three independent experiments were plotted. P values were calculated by reusable two-factor analysis of variance. (F) Reprogramming ability of WT, KR, and KQ Klf5 in combination with Oct4, Sox2 and c-Myc. MEFs were reprogrammed by WT, KQ, or KR Klf5, together with Oct4, Sox2, and c-Myc. Accumulated iPS colony numbers of three independent experiments (labeled with three different colors) are shown in the plot.

Fig 3. Klf5 acetylation, stimulated by TGF-β signaling, facilitates pluripotency maintenance in mouse ESCs.

(A) TGF-β signaling promotes Klf5 acetylation. V6.5 ESCs were treated with 10 μg/ml TGF-β or 5 μM SB525334 (SB) for 48 hours, and then harvested for subsequent Western blot experiments. The nonspecific band in the anti-Klf5 blot is marked by an asterisk. The expression levels of Ac-Klf5 were quantified and normalized to Klf5. Averages and standard deviations of Ac-Klf5 expression from three experiments are shown. (B) Overexpression of KQ Klf5 mutant antagonizes the suppression effect of TGF-β signaling on Sox2 protein expression. Empty pCAGIPuro, Flag-tagged WT and KQ Klf5 overexpression plasmids were transfected into V6.5 ESCs. Cells were cultured in mouse ESC medium with or without 5 μM SB525334 for 2 days, and then harvested for subsequent Western blot experiments. The nonspecific band in the anti-Klf5 blot is marked by an asterisk. (C) The effect of WT and KQ Klf5 overexpression on the repression of pluripotency gene transcription by TGF-β signaling. Cells were treated as described in (B), and harvested for RNA purification and quantitative RT-PCR analysis. Averages and standard deviations from three independent experiments were plotted.

Fig 4. Functions of acetylated Klf5 in mouse ESC differentiation.

(A) Expression of total, Ac-, and unAc-Klf5 during ESC differentiation. The nonspecific band in the anti-Klf5 blot is marked by an asterisk. V6.5 ESCs were cultured without LIF and feeder cells. Cells were harvested on day 0, 2, 4, and 6, and subjected to Western blot assay. (B) Klf5 knockdown affects the expression of pluripotency genes and differentiation genes in day 4 EBs. Day 4 EBs from shGFP control ESCs and two independent stable Klf5 knockdown ESC clones were harvested for quantitative RT-PCR analysis. Averages and standard deviations from three independent experiments were plotted. (C) Rescue effect of WT, KR, and KQ Klf5 in Klf5 knockdown ESCs. shKlf5 targeting sequences in WT, KR, and KQ Klf5 were mutated to render them resistant to shKlf5 knockdown. Stable ESC lines overexpressing shKlf5-immune WT, KR, and KQ Klf5 were established in the Klf5 knockdown ESCs. Day 4 EBs from these ESCs were harvested for quantitative RT-PCR analysis. Averages and standard deviations from three independent experiments were plotted. P values were calculated by reusable two-factor analysis of variance.