Only a fraction of cells are transduced, whereas in the chemically defined media protocol, the entire culture is uniformly exposed to the requisite signaling molecules. Nonetheless, our observations indicate that expression of Pax6 or Six3 in undifferentiated ES cells is sufficient to direct a subset of the cells to differentiate towards a lens fate. These findings hold relevance for two reasons. First, this system may allow the study of lens differentiation mechanisms in vitro. Such knowledge could help delineate the underlying genetic circuitry used in endogenous lens development and also needed to generate lens cells from undifferentiated ES cells for future cellbased therapies. Second, an in vitro model for lens development could allow study of the pathological mechanisms that underlie congenital lens defects. For example, recently Lachke et al. found that mutations in the gene encoding the RNA granule protein Tdrd7 cause cataracts and an associated glaucoma. The presence of Tdrd7 granules in these cultures provides a potential system to further analyze their composition and function. In addition, this system could allow functional tests of lens associated candidate genes identified by bioinformatics tools such as iSyTE. Mechanistically, the idea that a single-gene manipulation can initiate the development of a complex tissue is highly appealing and can be understood in the context of scale free networks in which certain highly connected nodes function as “hubs”. In this case, key upstream regulatory genes such as Pax6 and Six3 may function as hubs and serve to initiate a series of distinct downstream transcriptional events and TWS119 cellular interactions that lead to the emergence of lens cell types. Previous studies have shown that co-culture of primate and mouse ES cells on PA6 stromal feeders can direct ES cell differentiation along the lens pathway, the latter in a Pax6-dependent process. These results suggest an important role for signaling interactions between feeder and ES cells. An important role for signaling interactions is also indicated by the efficient induction of lens cell fate in chemically defined ES cell induction protocols. By tracing the mES cells transduced with a lentiviral vector constitutively expressing either Pax6 or Six3 along with GFP under a constitutive EF1a promoter, we were able to track the fate and location of the Pax6 or Six3- expressing cells relative to the lens marker expressing cells in these cultures. Interestingly, we found that while,1–5% of GFP expressing cells co-express lens markers, the majority of lens-marker-expressing cells reside near Pax6-GFP expressing cells. This observation is consistent with results from the aforementioned co-culture experiments, as Pax6 expressing cells appear able to recruit nearby undifferentiated cells into the lens differentiation program. We therefore suspect that individual Pax6 expressing cells recruit other cells to the lens pathway via non-cell autonomous mechanisms, and that the expression of Pax6 suffices to initiate this differentiation cascade.