Fibroblasts keep melanoma safe from harm.

The history of intravital microscopy is rooted in the study of cellular interactions in specific tissue environments. Imaging cells on glass coverslips is useful for mapping out basic features of signal transduction pathways, such as the regulation of the actin cytoskeleton by Rho-family GTPases. But the answers to many biological questions lie in the fine spatial and temporal details of signal transduction: “where” and “when” questions which can only be investigated in situ. This applies to diverse processes from T-cell / B-cell interactions within lymph nodes, to cancer-stromal cell interactions within tumors, to virtually all of embryonic development. More recently, intravital microscopy has found an important role to play in drug discovery, answering questions of where, when, and for how long drugs hit their targets at the tissue, cellular, and sub-cellular levels. This trend has been accelerated by the development of FRET biosensors which allow signal transduction to be imaged with high spatial and temporal resolution in pre-clinical cancer models. The development of B-Raf inhibitors provides an example of the promise and peril of targeted therapies, i.e. drugs designed to specifically interfere with only cancer cells. Around 50% of melanoma patients carry a mutation at V600, with the majority of these being V600E. Early clinical trials showed unprecedented improvements in overall and progression free survival of B-Raf V600E metastatic melanoma patients treated with the B-Raf inhibitor vemurafenib. These reports were accompanied by astonishing images of cancer patients riddled by metastatic melanoma being apparently cleared of their disease. Unfortunately, the benefits were short-lived and in most cases a form of melanoma returned which was completely resistant to the effects of the inhibitor. Intensive research has since uncovered several different mechanisms of acquired vemurafenib resistance, which generally involve B-Raf independent re-activation of the MAP-kinase pathway. Now, the Sahai group have used a combination of intravital microscopy and 3-dimensional culture systems to uncover a new type of drug resistance which emerges through tumor-stroma interaction. They used a FRET biosensor for ERK kinase, the terminal kinase of the MAP kinase cascade (Fig. 1) to study the response of both cancer and stromal cells to inhibition of B-Raf, the first kinase of the cascade. Surprisingly, their work shows that an off-target effect of B-Raf inhibition is activation of melanoma-associated fibroblasts (MAFs), which maintain ERK activation within the melanoma cells despite B-Raf inhibition. The MAFs do this by increasing production of extra-cellular matrix, especially fibronectin, which re-activates Erk though melanoma signaling pathways downstream of b1 integrin. The study begins with the simple observation that the B-Raf inhibitor PLX4720 impaired the growth of 2 mouse melanoma cell lines in vitro (5555 and 4434 cells), but did not retard the growth of the same cells grown as subcutaneous tumors. These melanoma cell lines were subsequently transfected with a nuclear version of the EKAREV FRET reporter, and the authors used intravital microscopy to assess the time course of Erk activation in subcutaneous tumors following daily PLX4720 treatment. They could show that Erk was effectively inhibited at 4 hours following the first treatment, however by 24 hours Erk activity had returned to pre-treatment levels and these “re-activated” cells were no longer responsive to drug treatment. Interestingly, small