Research Express@NCKU - Articles Digest

Introduction

New neurons are continuously generated from adult neural stem/progenitor cells throughout life in discrete regions of the mammalian central nervous system. In the hippocampus, a portion of newly generated dentate granule cells (DGCs) become synaptically integrated into the existing neural network and are then maintained in the adult brain. A central question in the field of adult neurogenesis relates to the physiological significance of the continuous addition of a small population of new neurons in the mammalian brain. Are these new neurons merely replacing dying mature neurons, or are they making unique contributions to specific brain functions that could not be achieved by existing mature neurons? Accumulating evidence from behavioral analysis suggests that adult neurogenesis is essential for some forms of learning, memory, and mood regulation. Recent studies using c-Fos and Arc expression as indicators for neuronal activation suggest an interesting possibility that new DGCs may be preferentially recruited into circuits in the adult brain that mediate spatial information processing and memory formation. How such preferential incorporation of adult-born neurons is achieved remains largely unknown.

Synaptic plasticity is widely regarded as a substrate for many brain functions, including learning and memory. Comparing properties of synaptic plasticity between adult-born neurons and existing mature neurons may provide important insights into the potential function of adult neurogenesis. Previous studies using field recordings from a heterogeneous neuronal population, consisting of mature DGCs and newborn DGCs of mixed ages, provided indirect evidence suggesting that new neurons exhibit an enhanced synaptic plasticity. More recently, immature newborn DGCs in adult rats, identified by their expression of TOAD-64 or PSA-NCAM, were shown to exhibit a lower threshold for the induction of glutamatergic long-term potentiation (LTP) when compared with mature DGCs. In the adult rodent brain, new neurons express these immature neuronal makers transiently only during 2–3 weeks after they are born, a period when glutamatergic synaptic inputs are just beginning to form. A key question remains: are special physiological properties maintained in adult-born neurons after fully integrating into the existing circuitry, or are they transient in nature? Recent development of the retroviral approach permits permanent labeling and single-cell analysis of new neurons with a defined birth date. Aided by electrophysiological and morphological analysis, studies using this approach have delineated the sequential synaptic integration process of new neurons in the adult brain. The properties and underlying mechanisms of synaptic plasticity in fully integrated adult-born neurons, however, remain largely uncharacterized.

While the adult brain exhibits significant plasticity for life-long learning, the magnitude of synaptic plasticity and anatomical changes in response to experience distinguish the adult form of plasticity from developmental plasticity. It has been well established that there exists a critical period when neuronal properties are particularly susceptible to modification by experience, which is concurrent with large-scale anatomical changes that become irreversible after the closure of the critical period. The classic critical period of enhanced plasticity occurs mostly in juvenile animals and has been considered as a central mechanism for establishing fine-tuned neuronal circuits in the developing brain. Whether the adult mammalian hippocampus retains such developmental plasticity through continuous neurogenesis is unknown. Thus, the primary aim of this study is to examine the temporal regulation and underlying mechanism of synaptic plasticity of adult-born DGCs along their maturation by using retrovirus-mediated birth dating and labeling.

Results

Excitatory postsynaptic potentials (EPSPs) were monitored under the current-clamp in response to a low-frequency stimulation (every 30 s) of the medial perforant pathway (Figure 1). To examine the synaptic plasticity of newborn DGCs, we used a paradigm of theta-burst stimulation (TBS). Significant LTP of EPSPs was reliably induced with TBS in GFP+ DGCs examined at 1 month postinfection (mpi; n = 7; Figure 1A). When the same stimulation paradigm was used to induce LTP in GFP− mature DGCs (Figure 1B), however, only 64% of neurons recorded exhibited significant LTP (Figure 1D). Thus, adult-born DGCs at 1 mpi in mice also exhibit a lower threshold for LTP induction, similar to those new DGCs at about 2–3 weeks of the cell age as reported in adult rats. We then examined whether adult-born DGCs maintain this characteristic during their lifespans (Figure 1C). We recorded from GFP+ DGCs at 4 mpi when new neurons already reach both morphological and physiological maturation. Interestingly, only 67% of these DGCs exhibited significant LTP (Figure 1D), similar to that of existing mature DGCs. Thus, the property of a lower threshold for LTP induction in adult-born neurons is not maintained once new neurons reach maturation. We also quantified the amplitude of LTP, an important physiological characteristic of synaptic plasticity. Interestingly, the mean LTP amplitude for GFP+ DGCs at 1 mpi was significantly larger than that of GFP− mature DGCs (Figure 1E). GFP+ DGCs at 4 mpi, however, exhibited very similar LTP amplitude to that of mature DGCs (Figure 1E). Taken together, these results demonstrate that adult-born DGCs transiently exhibit enhanced synaptic plasticity, as reflected by both an increase in the LTP amplitude and a decrease in the LTP induction threshold.

Figure 1. Adult-born neurons exhibit enhanced synaptic plasticity with a critical period. LTP recorded from newborn neurons at 1 mpi (A) or 4 mpi (C) in the adult brain. (B) LTP recorded from mature DGCs. (D and E) Percentage and potentiation of LTP recorded from newborn neurons at different developmental stages.

We next examined the molecular mechanism underlying the critical period plasticity of adult-born neurons. Application of APV (50 μM), a specific antagonist of NMDARs, abolished LTP from both GFP− mature DGCs and GFP+ newborn DGCs at all stages examined (Figure 1A–1C). Thus, activation of NMDARs is required for LTP of DGCs regardless of the cell age. The expression of NR2B subtypes has been shown to be associated with an enhanced synaptic plasticity during early postnatal development; we therefore examined the contribution of NR2B-containing NMDARs to evoked excitatory postsynaptic currents (EPSCs) in adult-born DGCs during their maturation (Figure 2). We recorded pharmacologically isolated NMDAR-mediated EPSCs from GFP+ DGCs under the whole-cell voltage-clamp. Application of ifenprodil (3 μM), an NR2B-subtype-specific antagonist, reduced the NMDAR-mediated EPSCs in GFP+ DGCs at 1 mpi by 72.1% ± 3.6% (n = 6; Figure 2A). In contrast, the same treatment led to only 25.1% ± 4.2% (n = 5) and 26.2% ± 4.2% (n = 4) reduction in GFP+ DGCs at 2 mpi and in mature DGCs, respectively (Figure 2B). Thus, there exists a developmental shift in the contribution of NR2B subtypes to the total NMDAR-mediated EPSCs during the maturation of adult-born neurons.

Figure 2. Developmental regulation of synaptic expression of NR2B-containing NMDA receptors in adult-born neurons during their maturation. (A) Blockade of NR2B-containg NMDA receptor-mediated EPSCs by ifenprodil. (B) Contribution of NR2B-containg NMDA receptors to the total NMDA receptor-mediated EPSCs in newborn and mature DGCs.

To determine the physiological role of NR2B subtypes in regulating synaptic plasticity of DGCs in the adult brain, we examined LTP from adult-born DGCs at 1 mpi, 2 mpi, and from mature DGCs in the presence of pharmacological inhibition of NR2B-containing NMDARs. Interestingly, treatment of ifenprodil (3 μM) almost completely abolished LTP of GFP+ DGCs at 1 mpi (Figure 3A). In contrast, the same treatment only slightly affected LTP of GFP− mature DGCs or GFP+ DGCs at 2 mpi (Figure 3A). Similar results were also obtained in the presence of Ro25-6981 (0.5 μM; Figure 3B), another NR2B-subtype-specific antagonist (Barria et al., 2005, Barth et al., 2001). Thus, LTP of adult-born DGCs exhibit differential dependence on NR2B subtypes during the maturation of these new neurons in the adult brain.

Figure 3. Different requirement of NR2B-signaling for LTP of mature and newborn DGCs in the adult brain. (A) Summary of LTP of EPSPs in newborn and mature DGCs with or without addition of ifenprodil. (B) Summary of the modulation of the mean enhancement of EPSP by ifenprodil or Ro25-6981.

Conclusion

In conclusion, our systematic analysis of synaptic plasticity of newborn DGCs along their maturation in the adult brain demonstrates the existence of a critical period when adult-born neurons exhibit enhanced synaptic plasticity. Such critical period plasticity is associated with developmentally regulated synaptic expression of NR2B subtypes in adult-born neurons, which, in turn, plays an instructive role in the enhanced plasticity. This adult form of critical period plasticity resembles features of the classic early postnatal critical period plasticity in juvenile animals, which is normally coincident with a high volume of information processing to establish fine-tuned neuronal circuits. Thus, adult-born neurons within the critical period may serve as major mediators for experience-driven plasticity and therefore function as special units in the adult circuitry to contribute to specific brain functions throughout life. The transient nature of such critical period plasticity, on the other hand, may also allow maintenance of stability of the mature circuitry that is essential for proper functions of the adult brain.