RESEARCH ARTICLE Bidirectional Global Spontaneous Network Activity Precedes the Canonical Unidirectional Circuit Organization in the Developing Hippocampus Yulin Shi,1 Taruna Ikrar,1 Nicholas D. Olivas,1 and Xiangmin Xu1,2* 1Department of Anatomy and Neurobiology, School of Medicine, University of California, Irvine, California 92697-1275 2Department of Biomedical Engineering, University of California, Irvine, California 92697-2715 ABSTRACT to DG. Photostimulation-evoked GNA also shows promi- Spontaneous network activity is believed to sculpt devel- nent backward propagation in the developing hippocam- oping neural circuits. Spontaneous giant depolarizing pus from CA3 to DG. Mouse GNA is strongly correlated potentials (GDPs) were first identified with single-cell to electrophysiological recordings of highly localized recordings from rat CA3 pyramidal neurons, but here we single-cell and local field potential events. Photostimula- identify and characterize a large-scale spontaneous net- tion mapping of neural circuitry demonstrates that the work activity we term global network activation (GNA) in enhancement of local circuit connections to excitatory the developing mouse hippocampal slices, which is meas- pyramidalneuronsoccursoverthesametimecourseas ured macroscopically by fast voltage-sensitive dye imag- GNA and reveals the underlying pathways accounting for ing. The initiation and propagation of GNA in the mouse GNA backward propagation from CA3 to DG. The disap- is largely GABA-independent and dominated by glutama- pearance of GNA coincides with a transition to the adult- tergic transmission via AMPA receptors. Despite the fact like unidirectional circuit organization at about 2 weeks that signal propagation in the adult hippocampus is of age. Taken together, our findings strongly suggest a strongly unidirectional through the canonical trisynaptic critical link between GNA activity and maturation of func- circuit (dentate gyrus [DG] to CA3 to CA1), spontaneous tional circuit connections in the developing hippocampus. GNA in the developing hippocampus originates in distal J. Comp. Neurol. 522:2191–2208, 2014. CA3 and propagates both forward to CA1 and backward VC 2013 Wiley Periodicals, Inc. INDEXING TERMS: imaging; photostimulation; development The hippocampal circuit underlies learning and mem- rat hippocampal preparations with single-cell and local ory processes and mediates pathological conditions such field electrophysiological recordings and Ca21 imaging as temporal lobe epilepsy. The hippocampus has a dis- methods (Ben-Ari et al., 1989; Bonifazi et al., 2009; tinctive trisynaptic organization that is strongly feedfor- Garaschuk et al., 1998; Leinekugel et al., 1997, 1998). ward in the directionality of its information flow (i.e., dentate gyrus, DG!CA3!CA1), which contributes to both normal function and pathology. There have been Additional Supporting Information may be found in the online version considerable efforts to understand the developmental of this article. processes that lead to the adult hippocampal circuit This is an open access article under the terms of the Creative Com- mons Attribution-NonCommercial-NoDerivs License, which permits use organization. Spontaneous activity in many developing and distribution in any medium, provided the original work is properly neural circuits, including the hippocampus, contributes cited, the use is non-commercial and no modifications or adaptations are made. to circuit formation. In the early developing hippocam- Y. Shi and T. Ikrar contributed equally to this work. pus, GABA actions are depolarizing and excitatory, coin- Grant sponsor: National Institutes of Health; Grant numbers: ciding with spontaneous-recurring network events DA023700; NS078434. termed giant depolarizing potentials (GDPs; Ben-Ari et al., *CORRESPONDENCE TO: Dr. Xiangmin Xu, Department of Anatomy and Neurobiology, School of Medicine, University of California, Irvine, CA 1989, 2007). GDPs were initially identified with single- 92697-1275. E-mail: [email protected] cell recordings from CA3 pyramidal neurons in slices of Received August 4, 2013; Revised November 22, 2013; neonatal rats; further studies have continued mostly in Accepted December 17, 2013. DOI 10.1002/cne.23528 Published online December 20, 2013 in Wiley Online Library VC 2013 Wiley Periodicals, Inc. (wileyonlinelibrary.com) The Journal of Comparative Neurology | Research in Systems Neuroscience 522:2191–2208 (2014) 2191 Y. Shi et al. Propagating network events related to GDPs also have nal projections (Bischofberger et al., 2006), in well- been examined by using fast voltage-sensitive dye imag- oxygenated (95% O225% CO2), ice-cold sucrose-con- ing in rat hippocampal slices (Bolea et al., 2006). GDPs taining cutting solution (in mM: 85 NaCl, 75 sucrose, are believed to be GABA dependent events mediated by 2.5 KCl, 25 glucose, 1.25 NaH2PO4, 4 MgCl2, 0.5 GABAA receptor transmission (Ben-Ari, 2002; Ben-Ari CaCl2, and 24 NaHCO3) or in artificial cerebrospinal et al., 1997, 2007; but see Bolea et al., 1999; Sipila fluid (ACSF; in mM: 126 NaCl, 2.5 KCl, 26 NaHCO3,2 et al., 2005; Xie et al., 1994). As with GDPs observed in CaCl2, 2 MgCl2, 1.25 NaH2PO4, and 10 glucose) with a vitro, spontaneous sharp waves (SPWs) visible in extrac- broad-spectrum excitatory amino acid antagonist, kynur- ellular field potential recordings have been observed in enic acid (0.2 mM). On average, one or two morphologi- the hippocampus of neonatal and adult rats in vivo (Buhl cal intact slices (as illustrated at the horizontal plates and Buzsaki, 2005; Buzsaki, 1986; Leinekugel et al., 147–150 of Paxinos and Franklin, 2001) between dorsal 2002); some of these researchers suggest that SPWs and ventral hippocampus from each hemisphere was and GDPs are essentially the same phenomena measured used for experiments. For VSD imaging experiments, sli- under different experimental conditions. GDPs or SPWs ces were first incubated in the cutting solution for 30 recorded at the microscopic level indicate that spontane- minutes at 32C and then transferred for dye staining ous activity is important for developing hippocampus; at room temperature (22C) for 1 hour in oxygenated however, studies to date have not provided a macro- ACSF containing 0.12 mg/ml of the absorption voltage- scopic network-level view of the spatial and temporal sensitive dye NK3630 (Kankoh-Shikiso Kenkyusho Co., dynamics of spontaneous activity or addressed the rela- Japan), then maintained in regular ACSF before use. We tive timing compared with the development of mature cir- used standard open recording chambers, which main- cuit connectivity. tained slice health and viability well, as evidenced by To provide a macroscopic global view of these proc- measurement of GNA activity for periods lasting more esses in developing hippocampus, we combined fast than 6 hours. Most imaging experiments were con- voltage-sensitive dye (VSD) imaging of neuronal activity ducted at room temperature. Further experiments were and laser photostimulation by the uncaging of neuro- conducted at 32C with an in-line solution heater transmitters to examine the developing circuit activity (Warner Instruments, Hamden, CT). The characteristics and connections in slice preparations of mouse hippo- of spontaneous and evoked network activity did not dif- campus. This approach allows for high spatiotemporal- fer at different temperatures. The data obtained were resolution imaging of the entire circuit, including the den- pooled together for analysis. tate gyrus (DG), CA3, and CA1, along with functional mapping of circuit connections. Here we describe spon- Voltage-sensitive dye imaging taneous global network activation (GNA) measured by Our overall system of electrophysiological recordings, VSD imaging, which propagates from distal CA3 to DG photostimulation, and imaging was described previously as well as to CA1 in the trisynaptic circuitry. Spontane- (Xu et al., 2010). Unless otherwise specified, optical ous and evoked mouse GNAs have similar spatiotempo- recording of VSD signals was performed under the 43 ral properties as assessed by VSD imaging and laser objective with a sampling rate of 4.4 msec per frame photostimulation in the developing circuitry. Bidirectional (frame resolution 88 [w] 3 60 [h] pixels). The field of GNA precedes the maturation of the mouse hippocampal view covered the area of 1.28 3 1.07 mm2 with a spa- circuit, insofar as GNA disappears immediately before tial resolution of 14.6 3 17.9 lm2/pixel. To image the emergence of the unidirectional trisynaptic circuit spontaneous network activity, five to 10 sessions (nine organization that characterizes adult hippocampus. trials per session) were conducted with 3 minutes of off-illumination intervals between sessions. For each MATERIALS AND METHODS trial, the VSD imaging duration was 2,000 frames (i.e., Slice Preparation and Experimental 8.8 seconds) with an intertrial interval of 12 seconds Conditions (i.e., 8.8 seconds for data recording plus 3.2 seconds All animals were handled and experiments were con- for saving data). Each session lasted for 9 3 12 sec- ducted in accordance with procedures approved by the onds (108 seconds total). With the known concern of Institutional Animal Care and Use Committee at the dye phototoxicity from prolonged illumination, we chose University of California, Irvine. Seventy-five C57/B6- not to image stained slices continuously and instead background mouse pups (either sex) of postnatal day 1 used spaced sessions to detect GNA events