During the initial stages of carcinogenesis, transformation events occur in a single cell within an epithelial monolayer. However, it remains unknown what happens at the interface between normal and transformed epithelial cells during this process. In Drosophila, it has been recently shown that normal and transformed cells compete with each other for survival in an epithelial tissue; however the molecular mechanisms whereby “loser cells” undergo apoptosis are not clearly understood. Lgl (lethal giant larvae) is a tumor suppressor protein and plays a crucial role in oncogenesis in flies and mammals. Here we have examined the involvement of Lgl in cell competition and shown that a novel Lgl-binding protein is involved in Lgl-mediated cell competition. Using biochemical immunoprecipitation methods, we first identified Mahjong as a novel binding partner of Lgl in both flies and mammals. In Drosophila, Mahjong is an essential gene, but zygotic mahjong mutants (mahj^−/−) do not have obvious patterning defects during embryonic or larval development. However, mahj^−/− cells undergo apoptosis when surrounded by wild-type cells in the wing disc epithelium. Importantly, comparable phenomena also occur in Mahjong-knockdown mammalian cells; Mahjong-knockdown Madin-Darby canine kidney epithelial cells undergo apoptosis, only when surrounded by non-transformed cells. Similarly, apoptosis of lgl^−/− cells is induced when they are surrounded by wild-type cells in Drosophila wing discs. Phosphorylation of the c-Jun N-terminal kinase (JNK) is increased in mahj^−/− or lgl^−/− mutant cells, and expression of Puckered (Puc), an inhibitor of the JNK pathway, suppresses apoptosis of these mutant cells surrounded by wild-type cells, suggesting that the JNK pathway is involved in mahj- or lgl-mediated cell competition. Finally, we have shown that overexpression of Mahj in lgl^−/− cells strongly suppresses JNK activation and blocks apoptosis of lgl^−/− cells in the wild-type wing disc epithelium. These data indicate that Mahjong interacts with Lgl biochemically and genetically and that Mahjong and Lgl function in the same pathway to regulate cellular competitiveness. As far as we are aware, this is the first report that cell competition can occur in a mammalian cell culture system. Cell transformation arises from the activation of oncoproteins and/or inactivation of tumor suppressor proteins. During the initial stage of carcinogenesis, transformation occurs in a single epithelial cell that grows within an epithelial monolayer. However, it remains unclear what happens at the interface between normal and transformed epithelial cells during this process. In Drosophila, it has been shown that normal and transformed cells often compete with each other for survival in an epithelial tissue, in a process called “cell competition.” Lethal giant larvae (Lgl) is a tumor suppressor protein in flies and mammals. Using biochemical methods, we identified Mahjong as a novel binding partner of Lgl in flies and mammals. Furthermore, we have demonstrated that Mahjong is involved in cell competition in both flies and mammals. In particular, we found that canine kidney epithelial cells depleted for Mahjong undergo apoptosis, but only when surrounded by non-transformed cells. This represents the first evidence that cell competition can occur in a mammalian cell culture system. Although it is not clear at present what molecules/signaling pathways are regulated by Lgl/Mahjong during cell competition, future studies might reveal important pathway components that could be targeted therapeutically to prevent tumor cells from “winning” in their race against normal tissue cells.

Malignant tumors are caused by uncontrolled proliferation of transformed mutant cells that have lost the ability to maintain tissue integrity. Although a number of causative genetic backgrounds for tumor development have been discovered, the initial steps mutant cells take to escape tissue integrity and trigger tumorigenesis remain elusive. Here, we show through analysis of conserved neoplastic tumor-suppressor genes (nTSGs) in Drosophila wing imaginal disc epithelia that tumor initiation depends on tissue-intrinsic local cytoarchitectures, causing tumors to consistently originate in a specific region of the tissue. In this “tumor hotspot” where cells constitute a network of robust structures on their basal side, nTSG-deficient cells delaminate from the apical side of the epithelium and begin tumorigenic overgrowth by exploiting endogenous Janus kinase/signal transducer and activator of transcription (JAK/STAT) signaling activity. Conversely, in other regions, the “tumor coldspot” nTSG-deficient cells are extruded toward the basal side and undergo apoptosis. When the direction of delamination is reversed through suppression of RhoGEF2, an activator of the Rho family small GTPases, and JAK/STAT is activated ectopically in these coldspot nTSG-deficient cells, tumorigenesis is induced. These data indicate that two independent processes, apical delamination and JAK/STAT activation, are concurrently required for the initiation of nTSG-deficient-induced tumorigenesis. Given the conservation of the epithelial cytoarchitecture, tumorigenesis may be generally initiated from tumor hotspots by a similar mechanism. Transformed mutant cells (pro-tumor cells) can evolve through a multistep process in which they become tumorigenic and invasive. Many genes that are involved in the different steps towards cancer development have been identified; however, how certain mutant cells destroy normal tissue organization and undergo uncontrolled proliferation during the initial stages of this process remains largely unclear. Using the epithelial tissue of the wing imaginal discs of the fruit fly (Drosophila melanogaster) larvae as a model system, we have analyzed these initial stages of inducing tumors by depletion of a neoplastic tumor suppressor gene (nTSG). We discovered that these tumors always originate from specific regions of the epithelial tissue of the wing disc. We show that in other regions that we dubbed “tumor coldspots” and that lack specific cellular structures, pro-tumor cells are eliminated from the epithelial tissue by the surrounding cells. However, in “tumor hotspots,” cells constitute specific structures in their basal side, and we found that pro-tumor cells successfully avoid potential elimination and deviate from the apical side of the tissue, initiating tumorous overgrowth. Our findings reveal the molecular and cellular mechanisms underlying the initial steps of tumorigenesis at tumor hotspots in the Drosophila imaginal wing discs.

The elucidation of principles governing evolution of gene regulatory sequence is critical to the study of metazoan diversification. We are therefore exploring the structure and organizational constraints of regulatory sequences by studying functionally equivalent cis-regulatory modules (CRMs) that have been evolving in parallel across several loci. Such an independent dataset allows a multi-locus study that is not hampered by nonfunctional or constrained homology. The neurogenic ectoderm enhancers (NEEs) of Drosophila melanogaster are one such class of coordinately regulated CRMs. The NEEs share a common organization of binding sites and as a set would be useful to study the relationship between CRM organization and CRM activity across evolving lineages. We used the D. melanogaster transgenic system to screen for functional adaptations in the NEEs from divergent drosophilid species. We show that the individual NEE modules across a genome in any one lineage have independently evolved adaptations to compensate for lineage-specific developmental and/or genomic changes. Specifically, we show that both the site composition and the site organization of NEEs have been finely tuned by distinct, lineage-specific selection pressures in each of the three divergent species that we have examined: D. melanogaster, D. pseudoobscura, and D. virilis. Furthermore, by precisely altering the organization of NEEs with different morphogen gradient threshold readouts, we show that CRM organizational evolution is sufficient for explaining changes in enhancer activity. Thus, evolution can act on CRM organization to fine-tune morphogen gradient threshold readouts over a wide dynamic range. Our study demonstrates that equivalence classes of CRMs are powerful tools for detecting lineage-specific adaptations by gene regulatory sequences. The regulatory control of genes allows an organism to generate a diversity of cell types throughout its body. Gene regulation involves specialized DNA sequences called transcriptional enhancers that increase the expression of genes in specific places and times. Enhancers contain clusters of specific DNA sequences that are uniquely recognized by DNA binding proteins, whose activities are also regulated in space and time. The critical role that DNA enhancers play in generating the diversity of cell types within a single organism suggests that changes in these DNA sequences may also underlie the diversity of organismal forms produced by evolution. However, few examples linking specific changes in enhancer sequences to functional adaptations have been documented. We studied a group of neuro-embryonic enhancers that turn on a certain group of genes in different fruit fly species that have been diverging from each other for ∼50 million years. Each species has experienced unique changes in its protein-coding sequences, gene regulatory sequences, egg morphology, and developmental timing. We found that the organizational spacing between the protein binding sites in these enhancers has evolved in a manner that is consistent with functional adaptations compensating for the dynamic and idiosyncratic evolutionary history of each species.