UC Berkeley

The survival and growth of cells can be influenced by the properties of adjacent cells. This reflects the ability of cells to communicate with each other and can result in either cooperation or competition for limited resources. Cell competition is an example of a situation in which cells exert a non-autonomous influence over the survival of their neighbors. The phenomenon was first discovered over forty years ago during studies of genetically mosaic Drosophila melanogaster. Clones of slow-growing, though viable, "Minute" cells were found to be eliminated from tissues in which they were surrounded by wild-type cells. It was later discovered that signaling between the two cell types at their clone borders causes the apoptosis and elimination of the Minute cells. Conversely, faster growing cells known as "supercompetitors", such as cells overexpressing the transcription factor, dMyc, can induce the elimination of wild-type cells. Thus, cell competition is a short- range interaction between cells with different growth rates in the same tissue and leads to the elimination of the slower-growing cells. The nature of this interaction is still unknown. Specifically, there are two important, unanswered questions regarding the mechanism of cell competition: 1) How do cells at clonal boundaries compare their properties and designate a "winner" and "loser"? 2) What are the signals that then induce the death of the designated losers?

Understanding the mechanism of cell competition is important from several standpoints. First, cell competition has been proposed to play a role in regulating organ size and tissue composition during normal development. Thus, understanding the mechanism of cell competition may advance our understanding of basic developmental biology. Second, many of the genes involved in cell competition are also misregulated in certain cancers. Thus, cell competition would be expected to occur at the borders between such tumors and their host tissue. A competitive advantage could make the tumor more malignant, while a competitive disadvantage might make the tumor more manageable. Therefore, a better understanding of the mechanisms involved in cell competition could also advance our understanding of tumor progression and improve our ability to treat tumors more effectively. Finally, a thorough understanding of cell competition and the ability to manipulate it could eventually be useful in therapies. For example, it may become possible in regenerative medicine to give grafts the ability to replace injured or dysfunctional tissue.

The goal of my thesis research has been to identify and characterize novel regulators of cell competition in order to gain a better understanding of the underlying mechanism. Accordingly, I devised an assay that would allow me to readily identify mutations that make cells into supercompetitors. The "supercompetitor assay" described here takes advantage of established mosaic analysis techniques in Drosophila. Small, marked clones of cells that are heterozygous for a mutation are created in an eye primordium that is primarily composed of homozygous-mutant cells. If the mutant cells are supercompetitors, the marked heterozygous clones are eliminated and cannot be recovered in the adult eye. Using this method I was able to test candidates from a collection of mutations in tumor suppressor genes from several different pathway that regulate growth. While mutations in components of some growth-regulating pathways, notably the Hippo pathway, caused cells to become supercompetitors, mutations in other pathways did not. Thus, only a subset of the pathways that regulate growth appear to be involved in cell competition.

The supercompetitor assay was also used to conduct an unbiased genetic screen for novel mutations that make cells supercompetitors. Among the mutations found in the screen were several alleles of crumbs (crb), a gene that regulates apicobasal polarity. This gene had no previously defined role in cell competition or growth. I found that while loss of Crb causes cells to become supercompetitors, overexpression of Crb causes cells to be eliminated from wild-type epithelia. Furthermore, as expected in instances of cell competition, high levels of apoptosis were observed preferentially at boundaries between wild-type and crb mutant cells, as well as at boundaries between Crb-overexpressing and wild-type cells. Thus, cells that express higher levels of Crb appear to be eliminated through a mechanism that resembles cell competition when they are near cells that express lower levels of Crb.

It is still unclear how cells compare their Crb levels. Cells may compare the levels of a molecule that is downstream of Crb. One candidate is the Hippo pathway, which has been repeatedly linked to cell competition. I found that crb mutant cells upregulate some of the transcriptional targets of the Hippo pathway suggesting that Crb impinges upon the Hippo signaling pathway. Alternatively, cells may directly compare their levels of Crb, as it is a transmembrane protein with a large extracellular domain of unknown function. Interestingly, the extracellular domain of Crb appears to be required to elicit some of the heterotypic interactions that I observe. For example, cells that express the intracellular domain of Crb alone are not eliminated. Furthermore, I see evidence of interactions between Crb molecules on adjacent cells that could be the basis of a direct comparison mechanism. Future work aimed at testing such models may yield important insights into a mechanism of cell competition.