Introduction Due to their potential

Introduction Due to their potential to differentiate into all cell types of the three primary germ layers, embryonic stem diltiazem hcl (ESCs) are perfectly suited to investigate early developmental processes such as heart development (Wobus, 2001). The investigation of cardiac cell subtype specification in differentiating ESCs provides a promising source for cell-based heart therapies including biological pacemakers. The rhythmic beating of the heart is initiated by a population of pacemaker cells located in the sinoatrial node (SAN) (Opthof, 1988). The development of the SAN involves a multistep differentiation process of specialized cells originating from multipotent stem cells (Christoffels et al., 2010). Functional impairment of these cells leads to arrhythmogenic heart diseases with increased mortality risk (Choudhury et al., 2015). The elucidation of molecular mechanisms in normal and impaired SAN development and function is therefore of crucial clinical relevance. Previously, we and others have demonstrated that a gene regulatory network involving the homeodomain transcription factor Shox2 controls cardiac pacemaker development and specification in mouse and zebrafish (Blaschke et al., 2007; Espinoza-Lewis et al., 2009; Puskaric et al., 2010; Hoffmann et al., 2013; Ye et al., 2015). Homozygous Shox2 deletion leads to early embryonic lethality due to developmental defects of the SAN, while heterozygous mice are viable and fertile (Blaschke et al., 2007). In zebrafish embryos, the loss of Shox2 substantially impairs pacemaker function with severe bradycardia and irregular heartbeat (Blaschke et al., 2007; Hoffmann et al., 2013). In line with this, we recently showed for the first time that SHOX2 mutations associate with atrial fibrillation, the most common cardiac arrhythmia in humans (Hoffmann et al., 2016). Shox2 activates the SAN genetic program either by direct regulation of its target Isl1, which controls cardiac pacemaker subtype identity (Hoffmann et al., 2013; Dorn et al., 2015; Vedantham et al., 2015), or by antagonizing the transcriptional output of Nkx2.5 (Espinoza-Lewis et al., 2009, 2011; Ye et al., 2015). Shox2-deficient mice used in the current study recapitulate this molecular pathway by diminished expression of SAN-specific genes and ectopic expression of chamber-specific genes (Blaschke et al., 2007; Puskaric et al., 2010; Hoffmann et al., 2013). Furthermore, the SHOX2 promoter is suited for the isolation of mouse ESC-derived SAN-like cells and as lineage-specific promoter to drive the expression of a voltage-sensitive fluorescent protein in nodal-like human induced pluripotent stem cell-derived cardiomyocytes (Hashem and Claycomb, 2013; Hashem et al., 2013; Chen et al., 2017). In addition, Shox2 overexpression during embryonic and mesenchymal stem cell differentiation strongly favours pacemaker cell specification (Ionta et al., 2015; Feng et al., 2016). Shox2 among other embryonic transcription factors such as Tbx3, Tbx18 or Isl1 (Bakker et al., 2012; Kapoor et al., 2013; Dorn et al., 2015), is able to direct pacemaker cell type determination. Taken together, Shox2 represents one of the major genes in the developing SAN and proper function is of crucial relevance regarding the origin of arrhythmogenic heart diseases. The elucidation of the underlying molecular mechanisms is therefore mandatory. To provide a cell-model for unravelling these mechanisms in health and disease, we established a stem cell-based cardiac differentiation model using Shox2 as a molecular tool.
Material and methods
Results
Discussion We have previously demonstrated that a gene regulatory network involving the homeodomain transcription factor Shox2 controls the development and function of the cardiac pacemaker with a highly restricted expression pattern in the SAN (Blaschke et al., 2007; Puskaric et al., 2010; Hoffmann et al., 2013). Shox2-deficiency is associated with impaired SAN function and increased susceptibility to atrial fibrillation (Blaschke et al., 2007; Hoffmann et al., 2016). The molecular mechanisms underlying Shox2 function in health and disease are barely understood. This prompted us to develop a murine ESC-based cardiac differentiation model using Shox2-deficient mice, which may contribute fundamentally to the current knowledge of the intrinsic genetic program. The enrichment of an ESC-derived SAN-like cell population is a prerequisite to specifically investigate signaling pathways, mandatory for differentiation of pacemaker cells. The differentiation of ESC-derived cardiomyocytes is influenced by many parameters, including the initial number of cells per EB, media composition and specific additives (Wobus et al., 2002). We applied and modified in total five differentiation protocols using pharmacological treatments or FACS-based selection (Wobus et al., 2002; Wiese et al., 2011; Chen et al., 2013; Hashem et al., 2013; Scavone et al., 2013) and compared the expression levels of SAN-specific marker genes (Shox2 and Hcn4). Most efficient enrichment of SAN-like cells was observed with a previously described FACS-based differentiation protocol (Scavone et al., 2013). The media composition and the predetermined number of ESCs were adapted to our standard differentiation protocol. For FACS analysis the surface marker CD166 was applied, as it is specifically but transiently expressed in the prospective SAN during embryonic development (Hirata et al., 2006). FACS-based selection does not influence molecular mechanisms compared to chemical-based enrichment procedures, which is a major advantage of this protocol. However, a limitation of this methodology is the restricted amount of CD166+ cells. Only 5% of the sorted cells were CD166+, consequently an enormous number of differentiated ESCs are required to obtain a sufficient cell number for RNA isolation and subsequent expression analysis. To overcome this challenge, nCounter technology was used for comparative expression analysis, as very low amounts of input material (100ng) are sufficient. We observed a significant downregulation of most SAN-specific genes (6/9 targets) in the Shox2 cells. This confirms the specificity of our established ESC-based cardiac differentiation model and recapitulates the data observed in the mouse model. The Shox2 SAN-like cells showed highly increased transcript levels for the pluripotency markers Pou5f1 and Nanog, suggesting that Shox2-deficiency may influence the differentiation potential of ESCs and thereby explain the hypoplastic SAN phenotype observed in the KO mouse.