Chemical biology: probes and therapeutics

Abstract

Listen

Translate article

In the Chemical Biology Minisymposium, cochaired by Alice Ting and Lisa Belmont, the first three talks focused on the development of novel probes for cell biology, while the last three presentations highlighted inhibitors of proteins previously believed to be “undruggable.” Daniel Hochbaum (Cohen lab, Harvard University) described an optical probe for imaging single action potentials using the fluorescence of a rhodopsin protein, Archaerhodopsin 3 (Arch), expressed in cultured rat hippocampal neurons. This voltage indicator exhibited a 10-fold improvement in sensitivity and speed over existing protein-based voltage indicators, with a twofold increase in brightness between −150 mV and +150 mV and a submillisecond response time. Arch detected single electrically triggered action potentials with a signal-to-noise ratio > 10. The mutant ArchD95N lacked endogenous proton pumping and showed 50% greater sensitivity than wild-type. Although it had a slower response (41 ms), ArchD95N resolved individual action potentials. Alice Ting (MIT) presented a novel approach to determining the proteomic composition of subcellular compartments by targeting a promiscuous biotin-conjugating enzyme to subcellular regions. The labeled proteins were enriched and identified by mass spectrometry. This approach was used to determine the proteomes of mitochondria and the endoplasmic reticulum of live mammalian cells without using subcellular fractionation. Katie White, a graduate student in Ting's lab, then described improvements to in vivo fluorophore labeling using mutants of lipoic acid ligase (LplA). She engineered LplA to accept a blue coumarin fluorophore and then used yeast display evolution to evolve LplA into a probe ligase with high activity in the secretory pathway. The LplA variants allowed imaging of intra- or intercellular protein–protein contacts. To expand the scope of bioluminescence imaging, Stephen Miller (University of Massachusetts Medical School) developed new aminoluciferin substrates for firefly luciferase that emit light at longer wavelengths than d-luciferin. Although these substrates were initially limited by product inhibition, this could be ameliorated by mutation of luciferase. Moreover, mutant luciferases were identified that displayed selectivity for these synthetic substrates over d-luciferin. This system has two advantages: it is potentially better suited for in vivo imaging, because tissue is more transparent to light at longer wavelengths, and orthogonal luciferase–luciferin pairs could allow multiplexed bioluminescence imaging. An NMR-based fragment screen for inhibitors of the Ras oncoprotein presented by Guowei Fang (Genentech) identified 25 compounds that inhibit Ras with two distinct mechanisms of action. One class of compounds binds to a small pocket between Switch I and Switch II that expands to accommodate the inhibitor. These compounds competitively inhibit nucleotide exchange by blocking the interaction of RasGDP with its nucleotide exchange factor, SOS. The second class of compounds binds in a pocket created by the interface of the Ras–SOS complex and acts by accelerating nucleotide release. Corey Nislow (University of Toronto) highlighted the power of yeast genetics by screening cell-active, structurally diverse compounds against 1100 heterozygous strains (haploid for essential genes) and 5000 homozygous deletion strains. He identified 55 novel drug targets, including Sec14 and septin. Sec14 coordinates lipid biosynthesis with signaling pathways and was thought to be undruggable. He then showed how chemical genetic profiling revealed that elesclomol, a compound demonstrating efficacy in metastatic melanoma, inhibits the electron transport chain. These approaches will continue to bear fruit as additional targets are confirmed and similar technology is applied to Candida albicans and other organisms. Lisa Belmont described a model for synergy between paclitaxel and navitoclax, a Bcl-2/Bcl-xL inhibitor, in which cells in mitotic arrest slowly degrade Mcl-1, while navitoclax causes acute inhibition of Bcl-xL. Across 50 cancer cell lines, cells with high levels of Bcl-xL relative to Mcl-1 exhibited lowered paclitaxel response and higher paclitaxel/navitoclax synergy. The cell line synergy translated to xenograft studies, and analysis of ovarian cancer tissue from paclitaxel-treated patients demonstrated that high Bcl-xL predicted poor response to paclitaxel. Taken together, the data suggest the paclitaxel–navitoclax combination might be effective in cancers expressing high levels of Bcl-xL.