Adult-born granule cells mature through two functionally distinct states

Two different integrative states of ABGCs

We addressed the maturation process of ABGCs after their integration into the hippocampal circuit. From the third week of their maturation ABGCs are already capable of firing action potentials, they receive synaptic inputs and provide synaptic output to the CA3, thus, they are already functional neurons. Our experiments demonstrated that 3-10 weeks old ABGCs form two functionally distinct subpopulations. These separate populations show highly specific input integration properties that are potentially important for their function in dentate gyrus-dependent computation. There are fundamental differences in how they translate excitatory drive into action potential output, in a manner that is not directly predicted by the cellular age alone. The groups were termed as S- standing for sensitive, and L-group referring to their linear input-output transformation. Around the third postmitotic week, ABGCs represent a functionally homogeneous population (S-group) characterized by highly variable and sensitive input-output transformation, as indicated by remarkably large increase in the response spike counts for sinusoidal current injections (that mimicked temporally organized input patterns), at certain frequency and input intensity ranges. This variability and sensitivity is characterized by large VAR and ASL value of the individual ABGCs. This characteristic integrative property of young cells potentially underlies the effective disambiguation of input patterns which proposed as a major function of ABGCs (Clelland et al., 2009; Deng et al., 2010; Sahay et al., 2011; Nakashiba et al., 2012). Importantly, most ABGCs retain quantitatively similar integrative capabilities for an extended time period, some of them even until the 9^th postnatal week. However, from the fifth week, individual ABGCs start to switch function by losing their sensitivity to a particular input-strength as their output incrementally reports a wide input range (L-group). Thus, the presence of two integrative states of ABGCs facilitate the maturation stage-dependent distinct transformation of DG input to its downstream network (Aimone et al., 2010a;

Sahay et al., 2011). Moreover, cells in S-group are tuned to serve similar integrative

74

function, however being substantially heterogeneous. This heterogeneity might have important functional consequences. For instance, ABGCs could not be divided into two populations based on the offset of the input-output function and these values increased with cellular age among S-group cells. This heterogeneity in the offset values broadens the dynamic range of the input-strengths that the population can process and transform to pattern separated signals.

Due to the sudden transition of individual cells from S- to L-state in an extended maturation period (5-9^th postnatal weeks), two overlapping populations with potentially different function are concomitantly maintained. Importantly, our data indicate that

“classmate” cells (born during the same period) can contribute to the network with fundamentally different functions; and conversely, similar functions can be served by ABGCs, which were born during different periods of the animal’s life. This observation extends previous hypotheses on the plasticity provided by adult neurogenesis, from the maturation of ABGCs being predominantly determined by their post mitotic cellular age.

Furthermore, the discovery of functional heterogeneity among ABGCs that are born during the same time challenges previously held concepts that largely relied on predicting similar functions of ABGCs born at the same time. (Aimone et al., 2006, 2010a; Ge et al., 2007; Sahay et al., 2011).

Strikingly, the two functionally distinct populations are maintained in spite of the continuous and gradual changes in their other intrinsic biophysical parameters. Despite their highly variable biophysical properties, the spiking output of the S-group cells is tuned to similar function by being particularly sensitive to certain input ranges. Cluster analysis performed on these multiple parameters was not sufficient to predict the S- and L-functionality probably, because this method considers parameters linearly based on their arithmetical values. It is generally accepted however, that the underlying biophysical parameters contribute nonlinearly to the input-output properties. Therefore, our results suggest, together with the non-concomitant changes in cellular properties along the age of the cells (Mongiat et al., 2009), that precise homeostatic tuning and the complex interactions of the biophysical properties (Marder and Goaillard, 2006) can be underlying factors of the balanced input-output functions of the two functional groups of ABGCs.

For instance, decreased spike threshold can be compensated by decrease of the resting membrane potential.

75 Developmentally generated granule cells

An additional unexpected observation was found when we recorded developmentally generated (not adult born) GCs in young, 17-18 days old animals. We found that their integrative properties were homogenous. Most surprisingly, they performed linear input-output conversion i.e. showed L-group properties. These results raised several questions, however, to unveil them was out of the scope of our study. For example, it raised the possibility that at the level of specific intrinsic cellular properties the developmentally generated and adult-born granule cells may mature differentially or the S-functionality might be a privilege of ABGCs. Nevertheless, these experiments provided a good control for those that addressed the functional properties of ABGCs in adult animals.

Future perspectives

Whether the quick transition between S- and L- integrative states is triggered by intrinsic (e.g. distinct molecular signaling events) or extrinsic signals (e.g. certain activity patterns) remains to be addressed in the future. In addition to the differences in the specific intrinsic cellular properties of ABGCs revealed in our study, their synaptic inputs and outputs also show changes during maturation (Laplagne et al., 2006; Toni et al., 2008;

Markwardt et al., 2009, 2011; Marin-Burgin et al., 2012; Vivar et al., 2012). Three to four weeks after cells are born, the amplitudes of unitary glutamatergic synaptic currents are similar in ABGCs (Mongiat et al., 2009). However, later, maturation-state-dependent short-term plasticity (Marin-Burgin et al., 2012), synaptic plasticity (Ge et al., 2007; Gu et al., 2012) and GABAergic inhibition (Markwardt et al., 2009, 2011; Dieni et al., 2013) can potentially either enhance or diminish the separation of the ABGC pool into two functional cell populations. Moreover, environmental conditions that increase or reduce hippocampal neurogenesis may have specific effects on the relative contribution of newly generated granule cells to the S- or L-groups that may explain distinct behavioral consequences of altered neurogenesis (Kempermann et al., 1997; Gould and Tanapat, 1999; van Praag et al., 1999).

76

5.2. Single bursts of single mossy fibers functionally reorganize