MSCs and Twist
We are interested in the roles of Twist (and other HLH proteins) in maintaining uncommitted progenitor populations and their interactions with targets and signaling pathways which promote their differentiation into mesenchymal lineages. Our collaborators, Stan Gronthos (IMVS), Karen Lyons (UCLA), Sandra Sharp (CSULA) and I are currently investigating these processes in order to gain insights that will enable us to develop therapies to regulate human stem cell growth and development for the efficient repair of skeletal tissues. The use of human stem cells – whether derived from a variety of adult tissues or from human embryonic cells – is of paramount importance to this research. Two CSULA Master’s students, Anna Munoz and Joyce Ho, supported by City of Hope’s Cancer Collaborative Program, are initiating these studies. They have successfully cloned the Twist, Id-1 and Id-2 promoters into pGL4-luciferase, and are currently examining the differentiation of three MSC lineages (muscle, bone and fat) over time to determine the functional activity of Twist and Id-1 and Id-2 in this process. Their results will be combined with real-time PCR and immunohistochemical biomarker expression studies completed last year.
Twist in Breast Cancer Metastasis
In the past few years, our research group has also found Twist to be at higher levels in breast cancer cells when compared to normal breast tissues. Furthermore, over-expression of Twist is associated with highly metastatic disease. But what is the mechanism responsible here? The laboratory is currently conducting a series of experiments to understand how signaling pathways are affected by Twist over-expression in breast epithelial cells.
A pivotal event is a process known as EMT (epithelial-mesenchymal transition). Certain epithelial-associated proteins, such as E-cadherin (the “E” means “epithelial”) and keratin, are considered crucial to cell-cell adhesion. When they predominate, the cells “stick together”, and the motility required for wayward cells to break free and become invasive is hindered. In EMT, however, expression of E-cadherin and keratin is repressed, while expression of mesenchymal-associated proteins, most notably N-cadherin (“N” means “neural”) and vimentin, is activated. This decreases cell-cell adhesion, increases motility/invasiveness, and so promotes metastasis. We demonstrated that Twist directly regulates this transition. In fact, we and other groups have convincing evidence that Twist is required for breast tumor cell invasion.
Graduate student Shan Li is determined to understand in detail how Twist mediates metastasis. To that end, she began by stably over-expressing Twist (MCF10ATw, 10ATw) in a human breast epithelial cell line (MCF10A, 10A) to study its function. Twist over-expression resulted in distinctive morphological changes in 10A cells, converting them from stereotypic epithelial to a disseminated structure with membrane protrusions. Migration and invasion assays confirmed that 10ATw cells were highly motile and invasive. This was in line with expectations.
However, when we further examined the mechanism of the changes, we found that the morphology that 10ATw cells displayed was dependent on the presence of epidermal growth factor (EGF), which was used to supplement the culture medium. When EGF was withdrawn from the medium, 10ATw cells lost their motile/invasive morphology and acquired a 10A-like phenotype. Moreover, no detectable migration was seen when EGF was absent. To dissect the molecular basis of this phenomenon, we performed quantitative PCR and found a 2.2-fold increase of EGF receptors (EGFR) in 10ATw cells compared to 10A cells. In addition, inhibition of MEK kinase with the EGF inhibitor U0126 led to an epithelial-like phenotype, similar to that of 10ATw cells cultured in the absence of EGF. However, no morphologic changes were observed in the MCF-10ATw cells with the addition of a PI3K inhibitor, LY294002.
Shan’s results therefore suggest that the EGFR/MEK/ERK signaling pathway mediates the Twist-induced changes in 10ATw cells. By demonstrating the correlation between Twist and EGFR, we show that Twist can not only change the physical connection between cells, but also alter cellular responses to the extracellular environment. When placed in a microenvironment coupled with activated EGFR ligands, Twist over-expressing cells can become more motile and invasive, which allows them to disseminate from the original tumor bed and metastasize to other sites in the body. Her manuscript, titled “Human breast stem cell transformation and invasiveness is mediated by upregulation of IL-8 by TWIST interaction with NF-kB,” is in progress.
Because Twist increases the invasiveness of breast tumor cells, we hope to introduce proteins that can interfere with Twist by blocking the affected cytokine signaling pathways to restore normal cellular activity. The function of cytokines in breast tumorigenesis has been well described and provides a molecular framework in which to devise novel anti-Twist therapies for highly invasive breast cancer.
Our laboratory aims to develop a series of soluble Twist inhibitors (STIs) and selectively deliver them (using neural stem cells, as described below) to metastatic breast cancer cells. Therapeutic efficacy in a murine model will be assessed. This data will be used to refine drug development and eventually ready suitable candidates for human trials.
Raquel Raices, a postdoctoral fellow who joined our laboratory in August 2008, has commenced research in this area. We are currently seeking funding to proceed with this important and exciting project. Our objective is to design soluble Twist inhibitors (STIs). These are non-functional, dominant-negative Twist mutants that:
- Inhibit activity of wild-type Twist
- Reduce Twist-mediated migration/invasion of breast cancer cell lines in vitro
- Prevent metastasis in a murine invasive breast cancer model
Raquel has engineered three non-functional STIs (S144K/R145E, ∆183-202, and S144K/R145E/∆183-202) that act as dominant negatives of endogenous Twist function, as suggested by their lack of transcriptional activity as well as decreased activity of wild-type Twist on the IL-8 promoter (a Twist target) when co-expressed with STIs. STIs were fused to the N- and C-termini of the protein transduction domain, VP22, to facilitate transfer into the nucleus of target cells. VP22:STI fused genes are under the T7 promoter to facilitate expression in E. coli, and were tagged with 6XHis and c-myc to facilitate purification and detection of recombinant protein.
For her experimental strategy, Raquel will perform an ex vivo incubation of recombinant VP22:STIs with the highly invasive SUM1315 breast cancer human cell line (VP22-fused recombinant proteins are taken up by target cells within 20 minutes). STI-mediated inhibition of endogenous Twist function in target SUM1315 cells will be assessed by Western blotting of EMT markers and migration/invasion assays. These results will be compared to those from SUM1315 cells treated with anti-Twist siRNA.
Neural stem cells (NSCs) possess the inherent ability to specifically migrate to primary and metastatic tumor sites. Because of this property, fetal HB1.F3.CD (a NSC cell line) will be transduced with AdVP22:STIs and used to deliver STIs to breast tumors and metastatic sites. The ability of transduced HB1.F3.CD cells to migrate towards SUM1315 will be evaluated in vitro using migration/invasion assays. Once HB1.F3.CD cells have migrated towards the cancer cells, expression of STIs will be induced by addition of doxycycline. VP22:STI should be released from NSCs and taken up by target cancer cells. The effect of STIs on SUM1315 expression of EMT markers will be assessed by Western blotting. Subsequently, SUM1315 cells will be genetically modified to express the firefly luciferase gene and used to establish a xenograft metastatic breast cancer model in nude mice. Tumor progression and metastasis will be assessed using a xenogen in vivo imaging system. Once NSC biodistribution is determined, transgene expression will be turned “on” with intravenous doxycycline administration only when NSCs have arrived to tumor sites, thereby avoiding complications of systemic delivery and minimizing any potential effect of loss of Twist function on NSC migration.
Given the role of Twist as a master regulator of breast cancer metastasis, NSC-mediated STI delivery to tumor sites represents a promising therapeutic strategy (combined with surgery or radiation to remove primary tumors) to reduce mortality due to breast cancer metastasis.
Shan Li’s Ph.D. thesis project is evaluating the STIs for their capacity to selectively induce apoptosis and/or transformation of metastatic mammary cells to a normal differentiated or apoptotic phenotype.
Neural Stem Cells as Targeted Drug Delivery Vehicles
Capitalizing on the inherent migratory and tumor targeting abilities of human neural stem cells (NSCs) provides our rationale for developing sophisticated novel therapeutic modalities for the treatment of invasive neoplastic disease. In collaboration with Karen Aboody, our laboratory is investigating NSC-mediated delivery of three therapeutics for the treatment of metastatic breast cancer.
We therefore propose to use genetically engineered NSCs as a delivery vehicle to secrete:
- Rabbit carboxylesterase (rCE), followed by systemic administration of CPT-11 [pro-drug therapy] Note: Donghong Zhou is the lead postdoctoral fellow on this project; she has just received a DoD concept award to pursue this innovative project.
- anti-Her2/neu immunoglobulin [immunotherapy] Note: Rik Frank, a third-year student in the Aboody laboratory, has successfully cloned the full-length Her2/neu IgG in hNSCs and is testing its ability of the hNSCs to secrete Her2/neu in vitro and is evaluating its function using chemotoxicity assays. His first manuscript, “Stem cell-mediated targeted HER2 immunotherapy,” is in the process of being submitted.
- Soluble TWIST inhibitors (STIs) [Transcription Factor (TF) targeted therapy] STIs are delivered to tumor sites to inhibit tumor growth and formation of metastases. We hypothesize that NSCs can effectively deliver these therapeutic agents selectively to breast tumor cells and significantly improve therapeutic efficacy in metastatic breast cancer mouse models.