The specific roles of the thalamus and the cortex in the generation and propagation of slow oscillations are still a matter of debate (Chauvette et al., 2010; McCormick et al., 2003; Wu et al., 2008). Early results point to the neocortex as generator, as the thalamic slow oscillations do not survive decortication (Timofeev and Steriade, 1996). Moreover, cortical slow oscillations persist both upon thalamic lesions as well as in cortical find protocol slice preparations (Constantinople and Bruno, 2011; Sanchez-Vives and McCormick, 2000; Steriade et al., 1993c). In thalamocortical slice preparations, thalamic stimulation

can trigger cortical slow-wave-associated Up states; yet, the thalamus is not required for their generation (MacLean et al., 2005; Rigas and Castro-Alamancos, 2007). However, a recent study suggests a re-evaluation of the role of the thalamus, providing

Romidepsin evidence for a critical role of two intrinsic thalamic oscillators, which may interact with a synaptically based cortical oscillator (Crunelli and Hughes, 2010). This work challenges the view that the cortex is causally involved in the generation of slow oscillations in vivo. A way of causally probing the distinct roles of cortex and the thalamus involves the targeted manipulations of cortical and thalamic networks. Optogenetics can provide the tools necessary for a local and

specific interrogation of neuronal circuitry (Gradinaru et al., 2010; Zhang et al., 2007). The use of optogenetics provided critical insights into the cell type-specific induction of gamma oscillations and its consequences on information flow (Sohal et al., 2009). However, in order to investigate initiation and long-range propagation of slow oscillatory activity, optogenetics needs to be combined with an effective technique to record network activity with sufficient temporal resolution and spatial specificity. In the present study, we monitored the Ca2+ transients associated with slow-wave activity by using mainly optic fiber-based fluorometric Urease Ca2+ recordings (Adelsberger et al., 2005; Grienberger et al., 2012). For this purpose, we developed a fluorescence detection and stimulation system consisting of a multimode optical fiber used both for delivering the excitation light and for collecting the emitted fluorescence signals. For the detection of slow oscillation-associated Ca2+ network spikes, we devised an optical fiber-based approach, allowing for the excitation of Ca2+ indicator Oregon green 488 BAPTA-1 (OGB-1), the collection of emission light, and the stimulation of ChR2-expressing neurons (Figure 1A).