3.1. Construction and verification of the c-Jun-ko mESC cell line
To study the role of c-Jun in early cardiac development, c-Jun exon was knocked out in mESCs, in accordance to the model shown in Supplemental Fig. 1A. c-Jun-ko mESCs were compared to wt mESCs, through PCR and agarose gel electrophoresis, to verify the deletion of the c-Jun exon (Supplemental Fig. 1B). Both mESC types were induced to differentiate into ectodermal stem cells (EpiSCs) [21], and no morphological difference found between wt and c-Jun knockout mESCs, confirming successful differentiation (Supplemental Fig. 1C). c-Jun mRNA expression in those EpiSCs were also compared, in which no expression was present in the ko group, contrasting with the wt control (Supplemental Fig. 1D). This finding was further confirmed at the protein level with Western blot, where c-Jun protein was found to be absent in ko EpiSCs (Supplemental Fig. 1E). To determine if c-Jun ko affected the pluripotency of mESCs, RT-qPCR was performed to detect the expression of mouse pluripotency genes (Oct4, Sox2, Nanong, Esrrb, Tfcp2l1) [30, 31] in the 2 mESC groups, and no significant difference for expression of any of those genes were found between the 2 groups (Supplemental Fig. 1F). Therefore, we confirmed that c-Jun was successfully knocked out in mESCs, without altering their differentiation and pluripotency capabilities.
To further confirm that no off-targets occurred with the CRISPR-Cas9 system, 2 clones of c-Jun-ko mESCs, #27 and #34, were tested. EBs formed from the 2 clones were found to share the same morphological phenotype (Supplemental Fig. 2A) and percentage of differentiated beating cardiomyocytes in Day 6 of differentiation (Supplemental Fig. 2B).
3.2. Bulge formation within EBs is associated with cardiomyocyte differentiation
To investigate the regulatory role of c-Jun in cardiac development, EBs were formed from both wt and c-Jun-ko mESCs through the hanging drop method, under suspension culture, for 6 days. Afterwards, between Days 7–9, EBs were subjected to adherent cell culturing (Fig. 1A). Cell aggregates forming EBs were observed in most hanging drops on Day 6 for both groups, and no significant differences were found for the numbers of successfully formed EBs (Fig. 1B). No significant morphological differences were found between wt and c-Jun-ko EBs on Day 3; however, almost all c-Jun-ko EBs demonstrated a bulging portion in an otherwise smooth sphere on Day 6, which was not present in the wt group (Fig. 1C). The cells comprising that bulge in c-Jun-ko EBs appeared to rhythmically beat when observed under the microscope, which was not present in wt EBs (Fig. 1D, Supplemental Videos 1–2). Based on those findings, we postulated that knocking out c-Jun promoted cardiomyocyte differentiation as part of the EB formation process, and that the spontaneous-contracting bulge in c-Jun-ko EBs might be comprised of induced cardiomyocytes. Indeed, a difference in cell differentiation between the 2 groups, after adherence culturing during days 7–9, was observed. Most c-Jun-ko cells continued to adhere to the well and beat like cardiomyocytes under the microscope on Day 7, while beating cardiomyocytes only appeared from Day 8 in the wt. The proportion of beating cardiomyocytes and their beating frequency in c-Jun-ko were significantly higher than for the wt group (Fig. 1E). These findings indicated that c-Jun ko promoted cardiomyocyte differentiation.
3.3. c-Jun knockout promoted EB development into mesoderm and endoderm layers
To further verify whether the c-Jun-ko beating cells observed in EBs on Day 6 were cardiomyocytes, cells from Days 0 (mESCs), 3, 6 (EBs), and 9 (adherent cells) were collected and analyzed by RT-qPCR for expression of specific myocardial markers Actc1, Actn2, Tnnt2, Myh6, and Gata4 [29, 32], where they were revealed to be expressed in a time-dependent manner, with expression levels peaking on Day 6 for the c-Jun-ko group. The expression of those markers on that day was also significantly higher for c-Jun-Ko than for the wt group, indicating that differentiated cardiomyocytes comprised the spontaneously contracting cells within the EB bulge (Fig. 2A). Next, FACS was used to detect the percentage of cTnT + cells, a marker of early cardiomyocyte differentiation, in the Day 6 EBs, and c-Jun-ko EBs had more cTnT + cells present (Fig. 2B). Immunostaining was then performed for cells from both groups on Day 9, revealing significantly higher cTnT in c-Jun-ko compared to the wt group. Additionally, under higher magnification, c-Jun-ko ESCs had obvious sarcomere structures (Fig. 2C), suggesting that knocking out c-Jun promoted cardiomyocyte development in mESCs.
This promotion of cardiomyocyte differentiation upon knocking out c-Jun was surprising, considering that the normal developmental pathway involves mESCs forming EBs, which then form the 3 germ layers: endoderm, mesoderm, and ectoderm [33]. In light of our finding that Gata4 had significantly higher expression levels on Day 3 in c-Jun-ko ESCs compared to wt, we hypothesized that knocking out c-Jun would promote the development of the mesoderm layer from EBs, followed by cardiomyocyte differentiation. This was owed to Gata4 being found to be upregulated in visceral mesoderm and foregut endoderm during mouse embryonic development [34, 35], as well as the fact that cardiomyocytes are derived from the mesoderm cell layer [13]. To test this hypothesis, EBs from both groups were collected on Days 0, 3, and 6, followed by detection of genes associated with those 3 germ layers. The results showed upregulation of mesodermal markers T, Flk-1, and Fgf8 following c-Jun ko on Day 3 (Fig. 3A), as well as for endoderm markers Sox17 and Tm4sf2 (Fig. 3B). This finding of endoderm-associated gene upregulation upon c-Jun ko is supported by studies showing that the earliest stages of mouse heart formation were dependent on signals from the adjacent endoderm[36]. By contrast, expression of ectoderm markers Nestin and Sox1 were downregulated on Day 3 in c-Jun-ko compared to the wt group (Fig. 3C). Our study therefore suggests that c-Jun-ko promoted early-stage EB differentiation into mesoderm and endoderm, along with inhibiting ectoderm differentiation.
3.4. Co-culture of c-Jun-ko with wt cells proved c-Jun-ko being responsible for further mESC differentiation into cardiomyocytes
To further validate our finding that knocking out c-Jun promoted cardiomyocyte differentiation, c-Jun-ko and wt mESCs were co-cultured. To distinguish between the 2 cell types, a plasmid containing mCherry and puromycin resistance genes was transfected into c-Jun-ko mESCs (Fig. 4A). Transfected cells, stably expressing mCherry, were then screened via puromycin administration (Fig. 4B), and those mCherry-expressing c-Jun-ko cells were mixed with varying proportions of wt cells to form EBs. Those EBs were observed under the microscope on Day 6, where increasing c-Jun-ko cell proportions corresponded with increasingly prominent convex bulge structures (Fig. 4C). Higher c-Jun-ko cell proportions also correlated with increased percentages of spontaneous beating cells within the co-cultured EBs, to the point where EBs purely comprised of c-Jun-ko mESCs had similar proportions of beating cells as EBs with 9/1 c-Jun-ko/wt cell ratios (Fig. 4D). All of these increases with respect to EB bulges and beating cell percentages corresponded with increased levels of mCherry signaling, in which the strongest signaling was in region with the strongest beating action (Supplemental Videos 3–5). These findings further confirmed that c-Jun ko is responsible for mESC differentiation into cardiomyocytes, and that the bulging convex structure formed in EBs is related to cardiomyocyte formation.