Chemically inducible inactivation of protein synthesis in genetically targeted neurons.

Introduction Synaptic plasticity mediates a variety of physiological functions, including emotional regulation, drug addiction, and learning and memory (Goelet et al., 1986). There are two temporally distinct forms of synaptic plasticity: short-term plasticity, which lasts seconds to minutes, and long-term plasticity, which lasts hours and days (Davis and Squire, 1984). Long-term plasticity differs from the short-term one in its dependence on gene transcription and protein translation. This conclusion, however, is solely based on the use of pharmacological inhibitors, which have various side effects with limited specificity (Sidhu and Omiecinski, 1998; Xiong et al., 2006). For example, the most potent protein synthesis inhibitor, anisomycin, has been shown to elicit relatively specific and strong activation of stress-related p46/54 JNK and p38 MAPK (Cano and Mahadevan, 1995), which are themselves implicated in synaptic plasticity (Bolshakov et al., 2000). In addition, cycloheximide, one of the most commonly used protein synthesis inhibitors, has significant toxic side effects, including DNA damage, teratogenesis, and other nonspecific effects (EmmanouilNikoloussi et al., 2007). Furthermore, these pharmacological approaches cannot inhibit protein synthesis in a celltypeor region-specific manner. Thus, there is a demand for developing chemical genetic approach to inhibit protein synthesis in neurons. Here, we describe a chemically inducible system that allows rapid and reversible inhibition of protein synthesis in specific cells. This system uses the double-stranded (ds) RNA-dependent protein kinase [or protein kinase R (PKR)], which reversibly phosphorylates the subunit of eukaryotic initiation factor-2 (eIF2 ) to control protein synthesis in eukaryotic cells (Klann and Dever, 2004). The kinase activity of PKR is very low at rest, but is significantly induced upon binding of its dsRNAbinding domains to dsRNAs during viral infection, leading to dimerization and activation of the kinase, and subsequent protein synthesis inhibition (Wu and Kaufman, 1997). To establish a genetically engineered protein synthesis inhibition system, we used a FKBP12based, chemically inducible dimerization system (id-PKR) to control the dimerization of the PKR kinase domain (Inoue et al., 2005). This system allows specific inhibition of eIF2 -mediated protein synthesis selectively in presynaptic or postsynaptic neurons that express the system, in an inducible and reversible manner. We have tested the utility of the system in two different models: dendritic spines and filopodia of hippocampal neurons, and late-phase long-term potentiation (L-LTP) in acute hippocampal slices. Our results suggest that in each particular model, protein synthesis in either presynaptic or postsynaptic neuron is uniquely involved in specific aspects of synaptic regulation. The inducible PKR system may serve as a useful tool to study the role of protein synthesis in neuronal functions in vitro and in cognitive behaviors in vivo.