Our laboratory is interested in understanding how intercellular signals control embryogenesis and organogenesis. From our work and that of many others, it has emerged that molecules of the Wnt family regulate numerous decisions that cells make during embryogenesis. It is now also clear that Wnt signals influence how stem cells divide during the regeneration and renewal of adult tissues. Unrestrained Wnt signaling, after mutations in Wnt signaling components, is implicated in cancer, including human colon cancer.
Our current research is based on methods we developed to purify active Wnt proteins. During this work, we established that Wnt proteins are unusual, in that they are modified by fatty acids. This modification makes the Wnt protein very hydrophobic, raising questions about how it can act as a secreted factor in vivo. In part, this may be explained by specific Wnt-binding proteins, and it is therefore interesting that we found one such partner, a protein that we termed SWIM (secreted Wnt-interacting molecule). When added to purified Wnt, SWIM promotes the solubility and activity of the Wnt signal. SWIM is particularly important when Wnt proteins have to be transported to act at a distance from where they are produced.
With the purified active Wnt in hand, we tested how the protein can be used to manipulate the behavior of stem cells in culture. These studies have borne fruit in several different stem cell areas, including in embryonic stem (ES) cells and neural stem cells. In all cases, we established that we can expand stem cells in an undifferentiated state, with full retention of specific differentiation capacities. The latter is revealed when the Wnt protein is withdrawn and the stem cells are transplanted in vivo. Despite the different origins of these stem cells, there is a common theme: the Wnt protein works in combination with other signals to promote stem cell expansion, and often these other molecules are growth factors that activate a tyrosine kinase pathway (Wnt itself does not use this pathway). These findings have significant fundamental and practical implications, as one of the key questions in the stem cell field is how to control the decisions that these cells make to stay undifferentiated or to become committed.
We found, for example, that mouse ES cells can be grown with purified Wnt protein, plus LIF (leukemia inhibitory factor). When implanted into a mouse blastocyst, the Wnt-LIF–exposed ES cells are fully able to contribute to all the embryonic lineages. We have also established that Wnt protein is specifically required for preventing the differentiation that ES cells would undergo. This important result suggests that human ES cells respond similarly to Wnt, something we are currently testing.
The multiple roles of Wnt signals during development are illustrated by our findings that at slightly later stages of embryogenesis and ES cell differentiation in culture, Wnt signals provide other important cues. When differentiated as aggregates, called embryoid bodies, ES cells form descendants of all three germ layers. We found that such embryoid bodies also establish anterior-posterior polarity in vitro and that this process is dependent on local activation of the Wnt pathway. Adding more Wnt protein posteriorizes the embryoid body, resulting in mesendodermal differentiation; inhibiting Wnt signaling promotes anterior character and results in neurectodermal differentiation. When we grow the embryoid bodies in the absence of inducing factors, the Wnt signals are not activated, and we showed that factors like Wnt itself or bone morphogenetic protein in the medium are required for its initiation. Once established, the Wnt signaling center is self-sustaining. Our findings show that the Wnt pathway mediates the local execution of a gastrulation-like process in the embryoid body, which displays an unexpected degree of self-organization (see Figure).
The third area of stem cell biology we have explored is neural stem cells. Using a Wnt reporter mouse, we identified Wnt-responsive cells in the subventricular zone of the developing mouse brain, as well as in known neurogenic zones in the adult brain such as the dentate gyrus of the hippocampus. When we added purified Wnt protein to neural cells in culture to test whether Wnt signals act as a stem cell factor, we found that the protein causes a clonogenic outgrowth of neural stem cells. These cells self-renew in culture for several passages, each time starting from single cells, in a Wnt protein–dependent manner. By itself, the Wnt protein is poorly mitogenic, but appears to mainly act by blocking the differentiation of the cells, while FGF (fibroblast growth factor) has to be added as a mitogen. When Wnt is withdrawn, neural stem cell colonies are multipotential and can form the three cell types of the central nervous system. Blocking the Wnt signaling pathway with the soluble Wnt inhibitor results in a depletion of stem cell populations. To demonstrate that the Wnt reporter–positive cells from mouse embryos have stem cell properties, we isolated the cells and cultured them in vitro. Wnt-responsive cells from these embryos exhibit enhanced colony-forming potential similar to cultures of whole populations of neural stem cells treated with Wnt. Our data show that Wnt proteins are not only important regulators of neurogenesis in vivo but also can be used in vitro for the clonal expansion of neural stem cells in an undifferentiated state.
We are also interested in Wnt signaling during the repair of damaged tissue. This follows the evidence that the same pathways that control growth of embryonic cells also govern regeneration of adult tissue. In the lung, several mutant phenotypes have revealed that the Wnt pathway is required for lung development, but its function in adult tissues is not well understood. We have used various Wnt reporter mice, in which signaling can be visualized in vivo, to examine the activation of Wnt signaling in adult lungs. These reporter lines reveal Wnt signaling in Clara cells, suggesting that Wnts may play a role in Clara cell formation or maintenance. To test this hypothesis, we use a naphthalene-mediated injury mode to stimulate Clara cell formation and follow Wnt reporter activation during lung repair. Naphthalene administration selectively ablates Clara cells and is followed by Clara cell regeneration. New Clara cells arise from an injury-resistant progenitor population called variant Clara cells, which reside at specific locations within the lung and proliferate in response to Clara cell loss. The Wnt reporter genes are expressed in Clara cells within neuroepithelial bodies, which harbor variant Clara cells. BrdU-labeling experiments have shown that TOP-Gal–positive Clara cells are proliferative, suggesting that the TOP-Gal reporter may mark progenitors for the bronchiolar epithelium and that Wnt signaling is activated during Clara cell regeneration.
