Genetic Basis of Metastatic Behavior

Genetic scheme for making benign tumor cells and detecting metastatic mutations. A) Clones of GFP-labeled cells in the developing eye disc also express a Ras^V12 transgene. Overproliferation of these cells results in localized noninvasive tumors that never move from the head region. B) Second-site genetic alterations may cause the development of metastatic behaviors, resulting in the presence of GFP-marked cells in ectopic sites. C) Example showing metastasis-promoting mutations such as scrib-/- cooperates with RasV12 in inducing GFP-positive cells to migrate.

Tumor progression and metastasis is the major cause of mortality for cancer patients. Because multiple genetic alterations contribute to tumor progression and metastasis involves multiple tissues, it has been difficult to decipher mechanisms underlying the growth and spread of tumor cells. We have designed and performed genetic screens in Drosophila to interrogate its genome in somatic cells for mutations that can promote tumor growth and/or cause metastatic behavior. The cellular composition of these flies resembles that of cancer patients who are chimeric individuals carrying a small number of mutated somatic cells.

Furthermore, this approach allows us to identify and examine the effects of mutations, which will otherwise cause lethality. The Drosophila model also offers unique opportunity to uncover developmental mechanisms, which contribute to tumor growth and metastasis.

A genome-wide screen for mutations affecting tumor growth and metastasis. Three main classes of mutants are recovered that promote metastasis or affect the growth of Ras^V12 tumors.

Our genome-wide genetic screens have identified more than 200 mutations that can cause otherwise noninvasive RasV12tumors of the developing eye to exhibit metastatic behaviors. The screens also recovered more than 500 mutations which can promote tumor growth, and 1900 mutations that suppress tumor growth.

Fly tumors, for example, exhibit a full range of metastatic behavior similar to that observed in human malignant tumors including accelerated tumor growth, loss of cell adhesion, basement membrane degradation, tumor cell migration and invasion, as well as secondary tumor formation. Although notable physiological differences exist between flies and humans (e.g. the lack of angiogenesis in flies), our fly model has distinct advantages for studying the basic biology underlying tumor progression and metastasis.

First, forward genetic screens can be used to systematically interrogate the genome for relevant alterations, whose role can be clearly elucidated without the complication of background mutations that are acquired during the extended latency of mammalian tumor progression. Second, many biological processes can be monitored, such as basement membrane degradation, that are difficult to examine in tissue culture, or in patients and mice. Finally, an inherent advantage of our approach over mammalian models is the shorter experimental cycle and the large number of animals.

Oncogenic Ras and cell polarity defect collaborate in promoting tumor growth and metastasis. InRas^V12 cells, cell polarity defects contribute to tumor growth and metastasis by activating JNK signaling and down regulating E-Cad. CYLD positively regulates JNK signaling by deubiquitinating fly TRAF2. JNK induces MMP for basement membrane degradation.

Mutations recovered in our screen have not only identified many fly homologs of known human tumor suppressors (e.g. PTEN), but also new genes affecting unexpected biology and pathways. For example, a class of mutations comprises apicobasal polarity genes. Indeed, apparent defects in cell polarity are often seen in many human cancers; however, the underlying mechanism is unknown. We showed that cell polarity mutations activate c-Jun N-terminal kinase (JNK) signaling and down-regulate the E-cadherin/ß-catenin adhesion complex, both of which are necessary and sufficient to cause metastatic behavior.

We further showed that JNK signaling promotes metastasis by activating MMP expression, highlighting the ability of tumor cells to acquire invasive properties by hijacking the JNK/MMP pathway normally used for developmental invasion process such as disc eversion.

We have used flies genetics to map new components of the JNK pathway and have recently determined that the cylindromatosis (CYLD) tumor suppressor, positively regulates JNK signaling by deubiquitinating DrosophilaTRAF2. Our investigation of JNK activation in these tumor cells has also led us to the discovery of a new tumor suppressor, Eiger, and an unexpected cellular mechanism for TNF/JNK activation.

Our fly model of metastasis has also led us to develop a methodology to produce stable tumor cell lines with defined genetic constitution from single flies. This advancement makes full circle in utilizing the fly as a model organism for both genetics and biochemical analyses.

Finally, we extending our genetic approach to address similar questions in mammalian models of metastasis, namely, using the piggyBac transposon in a genome-wide mutagenesis to identify mutations contributing to the metastatic process.