The brain uses neuronal oscillations to organize information processing, coding and retention. Different neuronal populations and brain structures produce rhythmic fluctuations of the neuronal membrane potential, which serve as a temporal window for communication. The hypothesis ‘‘communication through coherence’’ (CTC) (Fries, 2005) postulates that only coherently oscillating (at the same frequency) neuronal groups can interact effectively, because their communication windows for input and for output are open at the same time. It has been shown that phase coherence reflects various cognitive processes in humans (Canolty et al., 2006, Axmacher et al., 2010) and animals (Montgomery and Buzsáki, 2007, Tort et al., 2008, Tort et al., 2009, Wulff et al., 2009, Canolty and Knight, 2010). Moreover, in these and many other studies rhythms of different frequencies coexisted and were often synchronized to each other or nested into each other. The cross-frequency coupling may represent a mechanism for the regulation of communications between different spatiotemporal scales (Palva et al., 2005, Holz et al., 2010). The phase coupling between theta and gamma oscillations, namely, the modulation of the gamma amplitude by theta phase, is the most studied phenomena of phase coherence (Buzsáki et al., 2003, Bragin et al., 1995, Mormann et al., 2005, Canolty et al., 2006, Sirota et al., 2008, Tort et al., 2008, Schomburg et al., 2014).

Theta-rhythm (4–12 Hz), the principal oscillatory pattern generated by the hippocampus, plays a fundamental role in cognitive processes such as learning, spatial navigation, memory consolidation, reaction to novelty, selective attention and many more (Green and Arduini, 1954, Vinogradova, 1995, Hasselmo et al., 2002). A detailed analysis of the commonality of all these functions suggests that the theta rhythm serves as a temporal filter that provides admittance to the output of hippocampal neurons only those signals that come in certain phase relation to the theta wave (Vinogradova, 2001). All neuronal populations of the hippocampal formation, to a greater or lesser extent, show binding of their activity to the phase of the theta-rhythm, in other words, they generate action potentials predominantly in a certain phase of the theta cycle (Klausberger et al., 2003, Mizuseki et al., 2009; Somogyi et al.,2014), thus forming the nested high-frequency oscillations. A bulk of evidence suggests that successful memory performance requires coupling of 30–100 Hz gamma rhythms to particular phases of the theta cycle (Tort et al., 2009, Igarashi et al., 2014, Takahashi et al., 2014, Colgin et al., 2009). It is assumed that the mechanisms of organization of slow (∼25–50 Hz) and fast (∼55–100 Hz) gamma-oscillations are different since they are generated by different networks involving some particular classes of GABAergic interneurons (for review, see Colgin, 2016, Mably and Colgin, 2018). Gamma patterns segregate around theta phase: slow gamma dominated CA1 local field potentials (LFPs) on the descending phase of CA1 theta waves during navigation, and fast gamma was bound to the theta peak (Hasselmo et al., 2002, Schomburg et al., 2014). These signals corresponded to CA3 and entorhinal input, respectively. Long-term potentiation in CA1 is most easily induced at the ascending phase of theta when EC input is maximal. This indicates that the theta phase when EC inputs preferentially arrive may coincide with the time when memory encoding occurs optimally and raises the possibility that the EC related CA1 fast gamma facilitates memory encoding. Retrieval of information is thought to occur at a different theta phase than memory encoding, during which time CA3 input to CA1 is maximal and incoming signals from EC are suppressed (Colgin et al.,2009).

While the role of entorhinal afferents in the modulation of hippocampal TGC is widely accepted, the influence of other main input to the hippocampus, from the medial septal area (MSA) is poorly understood.

Hippocampus is considered as the main and the only generator of the theta rhythm; however, its activity shows a strong dependence on the connection with MSA. Lesion of MSA or its projections leads to the disruption of the theta activity in the hippocampus (Gray, 1971, Mitchell et al., 1982). A less structurally organized MSA, whose neurons fire in bursts at the theta frequency plays a role of a pacemaker of the hippocampal theta rhythm.

MSA contains three main types of neurons: GABAergic, cholinergic, and glutamatergic neurons (Manseau et al., 2005, Sotty et al., 2003). One population of GABAergic neurons expressing calretinin innervates other cells within the MSA only (Kiss et al., 1997). The second group of neurons projects to the hippocampus and expresses parvalbumin (Freund, 1989). There is a consensus that it is projections from parvalbumin-containing GABAergic neurons terminating on hippocampal interneurons that are involved in synchronization of the septo-hippocampal network at theta frequency (Hangya et al., 2009, Varga et al., 2008). Silencing of MSA GABAergic neurons during REM sleep suppresses theta-gamma coupling and theta phase coherence (Bandarabadi et al., 2017).

MSA forms the main cholinergic input to the hippocampus. The role of cholinergic neurons in the generation of theta rhythm and cross-frequency coherence in the hippocampus was demonstrated using an optogenetic approach that allows selective stimulation of various neurochemical populations of MSA. Medial septal cholinergic activation can both enhance theta rhythm and suppress peri-theta frequency bands, allowing theta oscillations to dominate (Vandecasteele et al., 2014). In addition, activation of cholinergic MSA neurons led to a synergistic suppression of the activity of CA3 pyramids due to direct metabotropic activation of hippocampal interneurons and, at the same time, an indirect effect due to excitation of septal parvalbumin projections. As a result the suppression of the slow gamma input by CA3 inhibition created conditions for fast gamma coupling and information encoding in CA1 (Dannenberg, 2015).

Glutamatergic MSA neurons synaptically drive hippocampal principal cells (Huh et al., 2010). Robinson and coauthors (2016) showed that while hippocampal oscillations strictly followed to the frequency of optogenetic stimulation of MSA glutamatergic neurons in theta range, while activation of septo-hippocampal glutamatergic projections had no effect on theta rhythm. Fuhrmann et al. (2015) also showed that optogenetically activated septal glutamatergic neurons entrain hippocampal theta oscillations and also pace and control the velocity of locomotion. However, it remains unknown how hippocampal activity evoked by rhythmic depolarization of glutamatergic MSA neurons is associated with information encoding, and in particular with theta-gamma coupling.

Here we investigate how phasic activation of MSA glutamatergic neurons in different frequencies influences theta-gamma coupling in the hippocampus.