Dynamic interaction between nature and nurture in the neural wiring of the brain

In this series of three lectures, Prof. Carla J. Shatz considered how neural activity, first spontaneous and then resulting from sensory stimulation, contributes to the elaboration and optimization of neural circuits during critical periods of brain development. These lectures focused on the development of the mammalian visual system, and more specifically on the connections between the retina, the lateral geniculate body (LGB) of the thalamus, and the primary visual cortex. These connections begin to form in utero in many species, and in any case well before the onset of vision. Initial neuronal wiring is established between the eye and the brain, using axonal guidance signals. This is followed by a prolonged phase of activity-dependent development, in which the initially diffuse synaptic connections are refined into the precise circuits of the adult brain. This process probably occurs throughout the brain during development, endowing it with a formidable capacity for adaptation to the environment and lifelong learning.

In the visual system, retinal ganglion cells connect to LMC neurons, forming alternating layers specific to each eye. The GLC neurons representing each eye in turn project to neurons in layer IV of the primary visual cortex to form the alternating columns of ocular dominance (OD). But during development, neuronal projections from both eyes are intermingled. The appearance of alternating layers in the LMC, or columns in the cortex, is accompanied by a remodeling of neuronal connections. This remodeling requires signals from retinal ganglion cells: blocking the action potentials of these cells prevents the formation of layers in the LMC (Shatz and Stryker, 1988; Sretavan et al, 1988), and also disrupts the formation of OD columns.

1. Brain waves and synaptic restructuring in the development of the visual system

This conference presented evidence of spontaneous retinal activity long before the onset of vision. Neighboring ganglion cells produce synchronous action potentials, constituting "waves" of regional excitation in the retina (Meister et al., 1991; Wong et al., 1993; Feller et al., 1996). These retinal waves are necessary for ganglion cell axons to segregate into alternating eye-specific layers in the thalamic LMC: blocking these waves of neuronal activation prevents segregation, and altering their spatio-temporal profile disrupts this segregation (Penn et al., 1998; Stellwagen and Shatz, 2001). It's as if the eye is intermittently "probing" the brain, several weeks before the onset of vision, to ensure that neuronal connections are correctly organized. Thus, early in development, the brain is the site of coordinated neuronal activity, prior to any sensory stimulation.

2. Transient scaffolding for neuronal circuit construction: neurons of the cortical subplate and cerebral cortex

The conference addressed the concept that the development of connections between the thalamus and cortex involves an intermediate stage in which a complete neural circuit is first built in the cortical subplate, then enters operation, and is finally dismantled, leaving few traces in the adult brain. Before the formation of connections between axons originating in the GLC and neurons in layer IV of the visual cortex, there is a long period of development during which thalamic axons interact with a transient contingent of neurons called cortical subplate neurons. These are the first post-mitotic neurons of the neocortex, whose axons inaugurate the connection between cortex and thalamus (McConnell et al., 1989). These neurons then serve as temporary targets for thalamo-cortical projections, and finally disappear through cell death. Experiments involving the ablation of these neurons at different stages of development, together with an analysis of the consequences for the formation of thalamo-cortical connections, have revealed their essential role in the structuring and functioning of the cerebral cortex. Early deletion of subplate neurons prevents axons originating from the LMC from invading the visual cortex, implying that these neurons are required for normal axon target selection (Ghosh et al., 1990). Deletion at a later stage prevents the segregation of these axons into OD columns (Ghosh and Shatz, 1992), and alters the expression of genes involved in synaptic plasticity, such as the gene encoding BDNF, a growth factor known to play an essential role in OD column formation (Lein et al., 1999). More recent experiments show that subplate neurons are essential for strengthening the synaptic connections needed to establish columnar organization in the cerebral cortex (Kanold et al., 2003), as well as for controlling synaptic plasticity triggered by vision deprivation in one eye (Kanold and Shatz, 2006). These experiments show that subplate neurons play a crucial role at various early stages in the formation of connections between the thalamus and cortex. It is currently thought that damage to the fetal brain leading to the destruction of subplate neurons may be responsible for postnatal diseases such as infantile cerebral palsy, autism, or certain learning disorders in children.

3. Overcoming the inhibition of synaptic plasticity: immune system genes acting in the brain

This latest conference introduced the notion that there are signaling pathways that oppose synaptic plasticity, in addition to those already known to enable activity-dependent plasticity. Such a "brake" was discovered in a differential PCR screen for genes whose expression in the thalamus GLC is modulated by "waves" of endogenous retinal neuronal activity. Unexpectedly, members of the major histocompatibility complex (MHC) class I gene family (the HLA genes in humans) were identified (Corriveau et al., 1998; Goddard et al., 2007). The discovery is particularly surprising, as neurons were previously thought not to express MHC-I genes under normal conditions, a notion described by the term "immune privilege" of the brain (review by Boulanger and Shatz, 2004). To assess the role of MHC-I genes in the brain, the GLC was examined in mice lacking MHC-I. In these mice, the specific layers of each eye do not form, and the synapse regression step does not occur. What's more, increased synaptic reinforcement is observed (in contrast to the usual situation in the inactivation of other genes, where synaptic plasticity is abolished). In particular, DO column plasticity is increased in the visual cortex (Datwani et al., 2009), and long-term synaptic strengthening(long term potentiation, LTP) in the adult hippocampus is 150% higher than in control mice, while the opposite phenomenon(long term depression, LTD) has disappeared (Huh et al., 2000). These observations suggest that MHC-I molecules may act as suppressors of synaptic plasticity, acting as a kind of "molecular brake".

In the immune system, certain members of the MHC-I family function in cell-mediated immunity by interacting with a variety of receptors, the best known of which is undoubtedly the T-cell receptor. Similar receptors located on neurons could interact with neuronal MHC-I proteins, underpinning activity-dependent synaptic processes. In a systematic search for receptors known to bind MHC-I proteins in the innate immune system, we found that PirB messenger RNA, a transmembrane receptor composed of "immunoglobulin-like" domains, is highly expressed by neurons in several regions of the mouse central nervous system, in particular the cerebral cortex, olfactory bulbs and cerebellum. We have produced mutant mice in which PirB is inactivated, and found that in these mice, OD columns occupied a larger area in the visual cortex than in non-mutant mice (Syken et al., 2006). Thus, the function of PirB, like that of its MHC-I ligands, appears to be to limit the extent of synaptic plasticity in the central nervous system. The results obtained indicate that this family of molecules, previously thought to be restricted to the immune system, also acts in neurons to limit the intensity, or perhaps the speed, of change in the "strength" of synapses in response to new experiences (Shatz, 2009). These molecules could play a crucial role in controlling circuit excitability and stability, both in the developing and adult brain. Dysfunction of this newly identified signaling in neurons could contribute to diseases such as autism and schizophrenia.