Sinoatrial node-specific differentiation of cardiomyocytes from ES cells over-expressing TGFbeta-Activated Kinase1 (TAK1). My group has established the first protocol for the in vitro differentiation of a highly purified population of sinoatrial node (SAN) cells from embryonic stem (ES) cells. This is accomplished by genetically modifying ES cells to overexpress TGFβ-Activated Kinase (TAK1/Map3K7). This differentiation protocol has broad implications for the study of SAN differentiation and physiology and in developing bioengineered grafts for regenerative medicine.

Future Directions:

a) Biofabrication: Within the heart, cardiomyocytes comprise a heterogeneous population of contractile cells each with unique markers, morphologies, physiologies and electrophysiologies. A major challenge for regenerative medicine will be the development of protocols for the differentiation of particular sub-types of cardiomyocytes. The ability to differentiate different subtypes of cardiac cells will be necessary for biofabrication of tissues for regenerative medicine and the method that we have developed, or a similar method can be adapted to this purpose. This could take the form of 3-D organ printing or be accomplished by seeding cells of desired lineages into permanent or biodegradable scaffolds to patch areas of cardiac damage while allowing cells time to form physical and electrophysical connections to host cells. We hope that these studies could be carried out in collaboration with the groups of Drs. Yin Mei and Roger Markwald.

b) Electrophysiological studies: In collaboration with Martin Morad, we are planning to carry out detailed electrophysiological studies of TAK1 overexpressing cardiomyocytes and, in the near future, plan to carry out both in vivo and in vitro studies to determine if these cells can indeed pace ventricular myocardial cells and/or rescue pacing in mammalian (mouse or rabbit) models in which the sinus node has been ablated. These studies will require either dual patch clamp techniques or imaging of fields of cells loaded with calcium or voltage sensitive dyes to determine the extent of cell-cell coupling or coupling of cells across a field of beating cardiomyocytes. I am also collaborating with the Morad lab to create a Tbx18 overexpressing lentivirus to determine if expression of Tbx18 is sufficient to transcriptionally reprogram ventricular cardiomyocytes to the SAN fate.

c) Large scale amplification of cell culture for screening: A near future goal of this work is the identification of the protein interactome that mediates SAN differentiation. To do this we need to adapt our current protocol for large-scale culture of SAN cells. This is required because current technologies for intereactome identification require several million cells of each specific lineage in order to be effective. We are currently working with Ruchi Bajpai, PhD at the University of Southern California to carry out epigenetic profiling of these cells to identify genomic enhancers that are activated during SAN differentiation. For epigenetic profiling, we will perform ChIP-seq to identify genomic regions marked by p300, H3K4me1, H3K27ac histone modifications but lacking H3K4Me3 and H3K27Me3 histone modification. Dr. Bajpai has previously shown that these particular genomic modifications serve as markers for enhancers that are active within a given cell type (See: (Rada-Iglesias et al. 2011). To functionally annotate active SAN enhancers identified by ChIP, we will use GREAT, a software tool designed to assign functional significance to non-coding genomic regions (McLean et al., 2010). This analysis determines the likely targets of an enhancer by nearest neighbor and likely sphere of influence calculations. The estimated gene list is then analyzed for known gene expression patterns, mutant mouse phenotypes and gene ontology enrichment classes. Data from our epigenetic profiling will be validated and compared to microarray data for the same cells.

Enhancers identified in screens of SAN cells will be used to make enhancer reporters (Enhancer::GFP or Enhancer::RFP). These can be validated both in ES cell culture by transiently transfecting them into both wild type and TAK1 overexpressing ES cells, and by establishing enhancer/reporter mouse lines. These studies will not only help us to establish an SAN-specific interactome but will also serve as tools for high throughput screens for factors that enhance SAN differentiation in vitro.

In addition to this, we are working to create a series of lentiviral constructs encoding biosensors for important cell-signaling pathways. We have already produced ES cells expressing a biosensor for active signaling through the Wnt signaling pathway. These cells encode a nuclear-localized fluorescent reporter that is activated when Wnt signaling is active. We are also working to create a sensor for active BMP-signaling. In theory, such sensors can be produced for any signaling pathway whose transcriptional targets are known. In addition because these biosensors will be encoded into lentiviruses, they can easily be transduced into any mammalian cell type for use as a live readout of signaling pathway activation, for example following biomechanical stimulation.

Another major focus of my lab has been to establish in vitro models to study the early mesoderm/endoderm interactions that mediate heart development. We have found that extraembryonic endoderm (XEN) stem cell lines are an excellent model system for these studies. Unlike other endodermal cell lines that have been used in these types of studies, XEN can easily be isolated from both wild type and mutant mouse embryos and ES cells. In addition my group recently showed that XEN cells express markers for the heart inducing endoderm (Brown et al., 2010b) and possess cardiogenic potential similar to END2 cells (Brown et al., 2010a). We have also begun to generate mutant XEN cell lines and have demonstrated proof of principle that these can add new insights into our understanding of early heart development. For example, we showed that Nodal mutant XEN cells both upregulate markers for the AVE and have altered cardiogenic potential (Liu et al. 2012).

Finally, in addition to the enhancer reporters mentioned above, we have developed a number of novel reagents to study early mesoderm/endoderm interactions in vitro including, an ES cell line in which endoderm can be selectively killed by the addition of gangciclovir to our growth medium, as well as ES and XEN cell lines expressing multiple endodermal reporters simultaneously.