Precious GEMMs: Mouse Models Simulate Metastatic Disease for Tomorrow’s Cancer Therapeutics
Sunnybrook Health Sciences Centre ♦ Published in Nature Reviews Cancer, Feb., 2011
Before cancer therapeutics are moved to the clinic for testing in humans, they must first be assessed in laboratory animals for both safety and efficacy. Developing efficacious therapeutics for cancer treatment is a challenge, by any standard of the word. The process requires animal models of disease that closely simulate similar disease conditions in humans. The introduction of the xenograft mouse model was a large step forward in the development of new treatment regimes for cancer. When carrying out xenograft experiments in mice, researchers transplant human cancer cells into a specific location in the mouse, often subcutaneously, to observe tumour formation and how the tumour responds to treatment with different therapeutics.
Dr. Robert Kerbel and his colleagues at the Sunnybrook Health Sciences Centre reiterate the importance of improving the mouse model for the generation of more effective cancer treatments. With respect to the selection of new drugs for cancer treatment, the predictive abilities of xenograft models suffer from the fact that they fail to adequately simulate the corresponding disease in humans. Human cancer cells, when transplanted subcutaneously into the mouse, grow considerably faster than they do in humans. As a result, xenograft tumours are somewhat hyper-sensitive to chemotherapeutics, which have been designed to target rapidly dividing cells. The location of transplantation plays an important role in how a tumour will respond to treatment. As an example, a tumour found in a human breast will respond differently to cancer treatment than a breast cancer tumour beneath the skin of a mouse. The biological niches in which these tumours reside are quite different, and given the importance the cellular niche plays in cell processes, a difference in niche will translate to a difference in response to therapeutics.
Genetically engineered mouse models (GEMMs) provide a solution to at least some of the problems encountered with the more simple xenograft model. The genetic code of these mice can be altered through the deletion or over-expression of genes that are involved in the tumorigenic processes of specific cancer types. These genetic alterations give rise to tumours in selected tissues such as the breast and lung that are composed of cells and vasculature that are the host’s own. Although GEMMs are a one-up on xenograft models, and have proven valuable for the study of formation and early onset of different cancers, they fall short on one cylinder — primary tumours found in GEMMs rarely metastasize — a hallmark of many cancers. Thus, the clinical predictive power of GEMMs is higher than xenograft models, but is still fundamentally limited.
Dr. Kerbel and his team have addressed this issue using an in vivo selection technique to generate melanoma and breast cancer cells lines that exhibit extensive metastatic capacities following their transplantation into the skin or breast respectively. The selection protocol is rather logical. To create a breast cancer cell line with metastatic abilities a breast cancer cell line is transplanted into the mammary fat pads of an immune compromised mouse. Roughly 4-6 months later some mice will have tumour metastases in the lung. These tumours are dissected, a new cell line is established in vitro, and these cells are then transplanted into the mammary tissue of a second mouse. After two rounds of this selection, mice exhibit extensive metastases to the lung and in some cases, at later stages, the brain. So why is this clinically relevant?
Well, beyond mimicking the normal “metastatic cascade”, treatment of mice that have primary tumours and distant metastases has actually mirrored observations we have made in the clinic for some time. Dr. Kerbel provides a nice example using two monoclonal antibodies. Treating mice with trastuzumab as a monotherapy potently inhibited primary tumours but had very little effect on metastases. Similarly, treatment of mice with a VEGF receptor 2 antibody (DC101) inhibited primary tumour growth but failed to elicit a reduction in the growth of metastases. On the contrary, mice treated with the chemotherapeutic CTX in combination with DC101 inhibited primary tumour growth and the appearance of metastases. This finding is in keeping with the clinical observation that treatment with trastuzumab or DC101 alone as monotherapies has very little clinical benefit for cancer patients, and a combination of antibody and chemotherapy is required for successful treatment in a metastatic setting. We have come a long way with animal disease models and as their predictive powers continue to rise, so should our ability to chose the most effective cancer therapies for the clinic.
TrkC & PTPσ, the Velcro at Neural Junctions
University of British Columbia ♦ Published in Neuron, Jan. 27, 2011
The development of healthy synapses requires a confluence of biological events to occur in harmony at the site where the axon and dendrite meet. The axon of one neuron extends from its cell body towards a dendrite, a short projection radiating from the cell body of an adjacent neuron. Once these two neural components are in close proximity, two key events must occur to drive synaptogenesis. Firstly, “synapse organizing” proteins must help locally recruit pre-synaptic and post-synaptic elements to the ends of the axon and dendrite respectively, and secondly, the axon and dendrite must come into physical contact, a process that is mediated by cell-adhesion molecules. Researchers at the University of British Columbia have discovered a new complex that spans the synapse to bridge axon and dendrite. Dr. Ann Marie Craig and her team used a functional expression screen to identify TrkC, a post-synaptic adhesion molecule, and PTPσ, a high-affinity pre-synaptic receptor of TrkC, which when bound maintain tight synaptic junctions. Neurotrophin receptor tyrosine kinases (Trks) have been known to contribute to nervous system development by interacting with soluble neurotrophins at the post-synaptic membrane. Activation of Trks by neurotrophins leads to signaling cascades that modulate synaptic development. The finding that TrkC interacts with PTPσ on the pre-synaptic membrane is currently the best explanation for why Trks, which are typically catalytic proteins, have cell-adhesion domains and non-catalytic isoforms.