«Distribution Agreement In presenting this thesis or dissertation as a partial fulfillment of the requirements for an advanced degree from Emory ...»
1.3.1 A Critical Period for Amygdala Influence on Emotional Development The importance of understanding the normative trajectory of brain development follows logically from a neuroconstructivist approach to studying anxiety disorders. Simply put, neuroconstructivism posits that brain function, genes, and environmental factors actively interact to shape brain development (Karmiloff-Smith, 2009). As such, understanding the function and physiology of a region in the mature brain is insufficient for interpreting that region’s contribution to the function of the immature brain or the etiology of developmental disorders.
Section 1.3 therefore serves to identify putative CPs in amygdala development, windows of profound change when the contribution of this brain region to emotional processing is in flux.
These CPs constitute the most likely periods when perturbations to amygdala function emerge in terms of behavioral deficits or influence the wiring of the limbic system.
A number of studies of amygdala lesions support a neuroconstructivist approach to emotional development, suggesting the function of the immature amygdala coordinates some aspects of emotional development. Specifically, these studies highlight differences in the effects of lesions of the amygdala during development and in adulthood. Comparing amygdala lesions in early development and adulthood illustrates that humans with early amygdala damage have worse memory for emotionally arousing stimuli (Shaw et al., 2005). In addition, amygdala lesions later in life interfere minimally with recognition of emotional facial expressions (Hamann and Adolphs, 1999), contrasted with the profound effect of congenital amygdala lesions (Adolphs et al., 1994). Similar findings exist for nonhuman primates: lesions of the macaque amygdala in infancy impact emotional and social behavior distinctly from lesions in adulthood (Prather et al., 2001). Furthermore, adolescent lesions of the macaque amygdala have milder effects than neonatal lesions, which promote lower social dominance, reduced aggression, and enhanced social fear (Bachevalier and Malkova, 2006). Finally, while ablating the amygdala of a rat at P21 impacts its social behavior, if ablation occurs at an earlier point in development, at P7, the deficits are more severe and also impact the animal’s stress response (Wolterink et al., 2001). These studies suggest the amygdala functions during brain development to organize the formation of emotional circuits in the brain, and early life changes to amygdala function are therefore of great importance for emotional outcomes.
1.3.2 Development of the Human Amygdala Human imaging studies have identified a number of gross developmental changes to amygdala structure and function, in general identifying childhood and adolescence as a period of accelerated amygdala growth. The basic structure of the amygdala can be detected at birth (Humphrey, 1968; Ulfig et al., 2003). Nonhuman primate studies suggest that the amygdala develops most quickly during infancy (Payne et al., 2010), but its development is protracted. The amygdala enlarges through adolescence relative to the rest of the temporal lobe (Giedd et al., 1996). Late maturation of the amygdala is evidenced by studies of white matter density; the frontal and temporal lobes gain white matter relatively late in human development (Deoni et al., 2011).
In terms of function, the amygdala is active in childhood and adolescence but its specific function changes. The amygdala is activated in response to emotional faces during childhood and adolescence (Baird et al., 1999; Thomas et al., 2001a), and also contributes to fear conditioning as early as adolescence (Monk et al., 2003b). However, with age comes increased integration of the amygdala into limbic circuitry. Compared to adults, children show significantly weaker functional connectivity of the amygdala with limbic structures, the vmPFC, and polymodal association cortices. Interestingly, the role of the BLA seems to become more specified with age, as children show greater intrinsic connectivity between the BLA and CeA, as well as greater overlap in their functional connectivity with target regions (Qin et al., 2012).
As described in Section 1.2.1, a wealth of studies have identified greater responsiveness of the adolescent amygdala to social stimuli with a negative emotional valence (Baird et al., 1999;
Monk et al., 2003a; Killgore and Yurgelun-Todd, 2004; Guyer et al., 2008) and diminished functional connectivity of the PFC and amygdala. This increased amygdala activity during adolescence may reflect development of the amygdala before the emergence of top-down control from higher-order cortical areas, especially the prefrontal cortex, that serve to inhibit the amygdala in adulthood (for review, see Casey et al., 2008).
1.3.3 Development of the Rodent Amygdala 220.127.116.11 Development of Amygdala Neuron Morphology and Physiology Studies of the rodent amygdala have provided a much finer resolution description of amygdala development, identifying changes to the structure and function of individual neurons throughout infancy and into adolescence. Morphological studies have revealed the structure of the BLA in rodents does not begin to stabilize until at least P28. The basolateral complex of the amygdala emerges by E17 in rats (Berdel et al., 1997b), and the majority of neurogenesis there occurs between E14 and E16 (Bayer et al., 1993). The BLA specifically increases in volume until the third postnatal week (Berdel et al., 1997a; Chareyron et al., 2012). Neuronal density is reduced in half between P7 and P14 (Berdel et al., 1997a). From birth to P7, the cross-sectional area of BLA neurons doubles, but at P7 the majority of neurons are still small and have only one or two main dendrites (ibid.). By P14, the cross-sectional area of neurons is the same as in the adult BLA. A Golgi-Cox study of developing BLA neurons found an expansion of dendrites during the first few postnatal weeks (Escobar and Salas, 1993).
Concomitant with changes to the structure of the BLA as a whole and its constituent neurons are a variety of changes to the connectivity of BLA neurons. There is a three-fold increase in total synapses in the BLA from P7 to P28, as measured by synaptophysin staining (Morys et al., 1998), reflecting increased intrinsic connectivity as well as maturation of inputs to the amygdala. Tract-tracing studies have shown that putative glutamatergic inputs to the BLA from the PFC and thalamus mature between P7 and P13 (Bouwmeester et al., 2002b) and stabilize by P25, before undergoing pruning in late adolescence (Cressman et al., 2010). Dopaminergic and noradrenergic inputs to the BLA become more dense between P14 and P20 (Brummelte and Teuchert-Noodt, 2006), and vesicular monoamine transporter 2 (VMAT2) is present in the BLA as early as E17, supporting a role for these monoamines in the early development of the BLA (Lebrand et al., 1998). Amygdala efferents also become refined postnatally; projections from the BLA to PFC seem to develop particularly late, during the second and third postnatal weeks, while projections to the thalamus and nucleus accumbens have mature morphology and density as early as P7 (Verwer et al., 1996; Bouwmeester et al., 2002a).
Furthermore, studies with a trans-synaptic tracer injected into the stomach showed that efferents of the amygdala mature during the first postnatal week, as labeling of the amygdala was much greater at P8 than P4 (Rinaman et al., 2000). These studies suggest the wiring of the BLA changes postnatally, but no studies have characterized the development of dendritic spines.
Dendritic spines provide a means of compartmentalization of biochemical and electrical signals related to neurotransmission, and should therefore reflect the maturation of amygdala connectivity (Shepherd, 1996; Lee et al., 2012). In Chapter 3, we address the knowledge gap regarding the morphology of BLA principal neurons, as none of these studies addressed specific neuronal subtypes, both in terms of dendritic arborization and spine expression.
While much attention has been paid to the development of amygdala morphology, very little is known concerning amygdala neuron physiology throughout development. No study has yet to characterize the intrinsic physiology of BLA principal neurons during development. We address this knowledge gap in Chapter 2. The few electrophysiological studies performed in the developing BLA have identified changes to synaptic plasticity. Thalamic inputs to the BLA exhibit long-term potentiation (LTP) following high-frequency stimulation at P28, but not at P60 (Pan et al., 2009). Similarly, before P10, high-frequency stimulation of cortical input to the BLA results in LTP, while after P10 this same protocol elicits long-term depression (LTD; Thompson et al., 2008). In adulthood, GABAergic inhibition limits amygdala excitability and prevents LTP;
therefore, this switch is likely driven by developmental changes to GABAergic transmission in the BLA, which are described in the following section and further characterized in Chapter 4.
18.104.22.168 Emergence of Amygdala Inhibition and Developmental Critical Periods The structure and function of the GABAergic system undergo protracted development that lasts well into postnatal life. As described in Section 1.2.3, GABA is the sole mediator of fast synaptic inhibition in the adult BLA, and it regulates amygdala function in a variety of important ways. However, the function of this neurotransmitter system is classically regulated throughout development (Ben-Ari et al., 2012; Kilb, 2012). The extent of maturation of GABAergic transmission in the amygdala is largely unknown, despite early GABA dysregulation being implicated in the pathophysiology of neurodevelopmental disorders (Chattopadhyaya and Cristo, 2012; King et al., 2013). This section details what is known regarding GABAergic development in the BLA, as well as in the brain at large, to identify knowledge gaps related to the etiology of anxiety disorders.
GABAergic transmission is present very early, but plays distinct roles in the immature and mature brain. GABAergic neurons arise from the medial and caudal ganglionic eminences and migrate throughout the brain during embryogenesis (Miyoshi et al., 2007; Miyoshi et al., 2010). GABAergic synapses have been identified in the brain as early as E16 by some measures (Konig et al., 1975; De Felipe et al., 1997). In the embryo and early postnatally, GABA performs distinct functions from in the adult brain; GABA is implicated in a variety of neurodevelopmental processes including cell proliferation, migration, and differentiation, synapse maturation and stabilization, and circuit wiring (Owens and Kriegstein, 2002; Huang and Scheiffele, 2008; Le Magueresse and Monyer, 2013). GABAergic transmission has also been shown to influence the function of neuronal stem cells and neuroblasts (LoTurco et al., 1995; Owens et al., 1996; Manent et al., 2005).
Postnatally, a number of structural and functional changes occur to GABAergic synapses and neurons. Refinement of structures and processes in postsynaptic neurons leads to refinement of GABAergic transmission (Le Magueresse et al., 2011). These synapses increase in density and have release events at higher frequencies across the first postnatal month (Luhmann and Prince, 1991). The postsynaptic currents elicited by GABAergic synapses become faster with age, influencing the duration of inhibition as well as the entrainment of action potentials and oscillations (Pouille and Scanziani, 2001; Tamas et al., 2004). Specifically in the BLA, there is a significant increase in the density of GABAergic fibers from P14 to P21, while the density of GABAergic cell bodies decreases (Brummelte et al., 2007). As discussed below, there are a number of postnatal changes to parvalbumin-expressing interneurons, which form perisomatic basket synapses and regulate network oscillations; these neurons only emerge in the BLA around P17, and do not reach mature levels until P30 (Berdel and Morys, 2000; Davila et al., 2008).
In general, GABAergic terminals early in development have high release probability and high neurotransmitter output, but revert to lower release probability and output in adulthood;
these high output, immature synapses exhibit short-term depression, when the response becomes weaker upon rapid, repeated synapse activation (for review, see Zucker and Regehr, 2002).
Variations in short-term plasticity alter temporal filtering mechanisms of synapses (Buonomano, 2000; Fortune and Rose, 2001; Pfister et al., 2010), and short-term depression of GABAergic synapses specifically promotes high-pass filtering of excitatory transmission and increases the information transmitted by bursts of action potentials (Abbott and Regehr, 2004; George et al., 2011).
The ionotropic receptor for GABA, the GABAA receptor, which mediates all fast inhibition in the adult amygdala, is not inhibitory early in development. GABAA receptors are excitatory at birth and assume their mature function postnatally (for review, see Ben-Ari et al., 2012). The switch from excitatory to inhibitory GABA results from changes to the concentration gradient of chloride, the major ion mediating GABAA currents. At birth, there is greater expression of sodium-potassium-chloride cotransporter 1 (NKCC1), which accumulates intracellular chloride and renders GABAA receptors excitatory. In adulthood, potassium-chloride cotransporter 2 (KCC2) is expressed more highly, extruding chloride from the cell and rendering GABAA receptors inhibitory (Ben-Ari et al., 2012). Excitatory GABA early in development is thought to promote calcium influx and modulate neuronal growth and synapse formation (BenAri et al., 1997).
GABAergic transmission also plays a role in coordinating brain development. Activation of GABAergic synapses directly influences the development of those synapses (Akerman and Cline, 2007; Chattopadhyaya et al., 2007; Huang, 2009), but also organizes circuit maturation during CPs. One well-established trigger for the onset of CPs, or developmental windows of high plasticity, is activity at GABAA receptors (for review, see Hensch, 2005). Transplantation of interneurons restores CP plasticity, suggesting the development of inhibitory neurons specifically triggers this plasticity (Southwell et al., 2010).