The Impact of Developmental and Genetic Influences on Neural Circuits

Abstract

Background

Neural circuits form the basis for brain function and behavior. However, solely focusing on adult brains can sometimes overlook prior events that may have shaped this anatomy. These other perspectives can provide further insight into brain circuit organization.

Summary

To support this suggestion, specific examples of developmental and genetic events are presented that might not be appreciated if only adult brains were investigated. Examples include the following. One is the origin of nuclei from the same developmental field that divides into major subdivisions. Each of the resulting areas has neurons with different morphology, connections, and molecular signatures. The other is new circuit formation that is produced in three ways. One is the result of peripheral sensory receptor reduction. Another is the consequence of genetic mutation. The third is due to axonal pruning or perinatal injury.

Key Message

In certain instances, developmental and genetic features provide additional perspectives to better understand how circuit formation in the brains of adult animals came to be.

Introduction

Brain circuits have been investigated by countless reports. Nevertheless, many questions remain unanswered. When adult brains are the sole focus of study, only the endpoints resulting from a variety of processes are seen. How prior events may have shaped these adult brain circuits can sometimes be better understood through additional approaches. This communication provides examples that illustrate how developmental and genetic changes can influence the morphology of neural circuits in the brains of a variety of vertebrates. Specific instances are presented in detail and are supplemented by shorter descriptions of similar types of studies. For all of these examples, the final, adult appearance tells only a part of the story.

Structures Arising from a Single Developmental Field

Development, Neuronal Morphology, and Circuitry of the Habenula

The habenula arises from a single developmental field. In amniotes (reptiles, birds, mammals), this habenular anlage subsequently divides into two, a medial and a lateral nucleus. These divisions in amniotes correspond to the dorsal and ventral nuclei, respectively, in fish and amphibians [1]. Further subdivisions of the medial and lateral habenular nuclei have been described in a variety of vertebrates [2, 3]. The following summarizes some of the major observations. Despite their common origin, the medial and lateral habenula exhibit striking differences in both cellular anatomy and circuitry (Fig. 1). The dendritic morphology of neurons in the medial and lateral habenula differ markedly [4–6]. Dendrites of the medial habenula have unusual distal appendages considered to be unique among neurons in the brain [4]. Lateral habenular dendrites have a beaded appearance and lack the unusual distal dendritic appendages seen in the medial habenula [6]. While considerable details of habenular circuitry have been described for the medial and lateral habenula, their respective inputs, outputs, and types of information transmitted differ significantly [7–9].

Fig. 1.

Habenular development and circuitry. A simplified schematic shows the origin of the habenula from a single developmental field (anlage) that divides into medial and lateral nuclei. The markedly different morphology of neurons in the lateral and medial habenula is shown. The major inputs and outputs of each subdivision as well as the major transmitter systems are indicated. Circuitry is simplified from [7] and the drawn cell morphology is modified from [6]. a, axon; Hb, habenula; HbL, lateral habenular nucleus; HbM, medial habenular nucleus; IPN, interpeduncular nucleus; RMTg, rostromedial tegmental nucleus; SM, stria medullaris.

The development of the habenula provides a more complete picture for understanding its final adult form. Signaling center pathways influence both axes of the habenula anlage. The dorsal-ventral axis is affected by opposing gradients of bone morphogenetic proteins and Sonic Hedgehog (Shh) whereas the anterior-posterior axis is influenced by signaling centers. The latter include Wingless-INT proteins and fibroblast growth factors [1, 10]. To further understand how subtypes of habenular neurons develop and connect with their respective targets, additional methods have been used. These include single-cell RNA sequencing, molecular, functional, and anatomical approaches. Together, these techniques have a provided a more complete understanding of habenular subnuclei development [11].

Other Examples

The habenula represents just one of many central nervous system structures in which a single developmental field gives rise to multiple divisions with differing properties. Two other prominent examples are the dorsal thalamus and the spinal cord. Each provides a window to understand not only how these structures develop but also how errors can result in functional and behavioral effects.

Dorsal Thalamus

The dorsal thalamus represents that part of the diencephalon connected with the telencephalon [12]. It originates from a single entity located in the alar portion of prosomere 2, which is the source of the epithalamus and the dorsal thalamus [13]. From its initial homogeneous organization, the dorsal thalamus subsequently becomes transformed into multiple nuclei [12]. Investigations of the dorsal thalamus relevant to the present discussion have mainly focused on two areas. One is its intrinsic development and differentiation from the surrounding areas of the diencephalon. The other is its interconnections with the cerebral cortex.

The differentiation of the diencephalon, including the dorsal thalamus, has been understood through the unfolding of molecular events. Among these are the formation of the mid-diencephalic organizer (zona limitans intrathalamica) which contains a band of Shh expressing cells. This morphogen determines the expression of transcriptional programs of region specific transcription factors that define the identity of thalamic neurons. Other crucial molecules, such as Wnt1 and fibroblast growth factor 8, establish gradients that regionalize the diencephalon including the dorsal thalamus [14].

Studies on the interconnections between the dorsal thalamus and cortex have examined its development, the effects of genetic mutations, and its evolution. Most of these observations have investigated mammals with limited studies examining sauropsids (reptiles and birds) [15]. Sauropsids possess several sensory thalamo-telencephalic circuits that are similar to mammals. However, these particular thalamic projections in sauropsids terminate in a telencephalic structure known as the dorsal ventricular ridge, which is not present in mammals [16]. Whether thalamo-telencephalic connections with the dorsal ventricular ridge in sauropsids will follow a development pattern similar to that found in mammals remains to be seen.

Spinal Cord

The transformation of the spinal cord from a homogeneous structure into its adult layered morphology provides another example of how development shapes its adult form. Unlike the dorsal thalamus, which is transformed from a unitary structure into multiple nuclei, the spinal cord becomes layered. The basis of this adult cytoarchitectural pattern of individual lamina has been unraveled through a series of molecular developmental studies. Gene regulatory networks act on transcriptionally distinct neural progenitors. These neural progenitors, which are located in discrete regions along both the anterior-posterior and dorsal-ventral axes, produce different neuronal cell types. The resulting neuronal cell types diversify further in response to multiple morphogens and cell-cell interactions occurring at specific time periods [17]. Subsequent observations identified genetic tiers that divide neurons into several subtypes that characterize distinct functional subclasses related to birthdate [18]. Recently, the importance of temporal factors was examined in one part of the spinal cord, the dorsal horn. Here, six waves of neurogenesis occurred at different time intervals. Each wave produced an inhibitory and excitatory family. These individual cell families were influenced by a dorsal-ventral gradient of Zlc transcription factors present in the dorsal horn progenitors [19].

New Circuits

New brain circuits are ones that are not commonly present in a given species. These new connections have been produced in several ways and comprise three general categories.

New Circuit Formation Secondary to Peripheral Sensory Receptor Reduction

The blind mole rat has a subterranean existence with micro-ophthalmic eyes that are covered by fur and skin [20]. Yet, retinal projections to visual targets in the brain of the blind mole rat are similar to those found in non-burrowing, above ground rodents. However, these terminal visual structures in the blind mole rat are sparsely innervated and have an abnormal cytoarchitecture [21]. What is unusual about the blind mole rat is a projection from the auditory midbrain, the inferior colliculus, to the visual thalamus, the dorsal lateral geniculate nucleus [22]. This anomalous circuit likely explains the auditory responses recorded in two visual areas: the dorsal lateral geniculate nucleus and the visual cortex [20].

A similar type of re-organization occurs in humans that are deaf or blind. The auditory cortex of deaf individuals is active in response to somatosensory and visual stimuli. Likewise, visual cortices are responsive when blind adults hear auditory motion, discriminate tactile patterns, and localize sounds (summarized in [23]). Even spoken language in blind children can activate visual cortices [24].

Genetic Mutations Produce Abnormal Circuits

Visual abnormalities are found in albino animals, that is, species that lack eye pigmentation. Albino animals have a specific genetic mutation. As a consequence, they display a variety of abnormalities in visual circuits as well as in visual brain areas [25]. Of the albino animals studied, none have captured more attention than Siamese cats. One of the most striking abnormalities in Siamese cats is the temporal retinal projection to the contralateral lateral geniculate nucleus as opposed to the ipsilateral termination found in normal cats. From an anatomical and physiological perspective, this feature results in an abnormal map in the visual cortex. Yet, Siamese cats are able to suppress much of this anomalous cortical map and have excellent visually guided behavior [26].

Subsequent studies in albino mice have provided additional information to unravel the mechanisms underlying these visual circuit abnormalities. Albino mice share the same, following two visual circuit abnormalities found in Siamese cats. One is a reduced retinal ganglion cell projection to the ipsilateral dorsal lateral geniculate nucleus. The other is an additional area of ventrotemporal retinal ganglion axons that end in the region of the dorsal lateral geniculate nucleus that would normally have received the contralateral input from the ventrolateral retina. Together, these circuits produce a visual cortical organization similar to that found in Siamese cats. Furthermore, the molecular mechanisms that underlie the misrouting of axons at the optic chiasm have been uncovered. In addition, reduced melanin production in the retinal pigment epithelium of albino mice influences retinal ganglion cell production [27].

Changes in Circuits due to Pruning and Perinatal Injury

New brain connections have been produced by a variety of perinatal insults in a variety of animals [28]. The following summary of experiments in hamsters provides one example (Fig. 2). At birth, the retina projects to several structures that normally do not receive direct visual input in adults. One of these circuits present at birth is a contralateral retinal projection to a somatosensory thalamic nucleus, the ventrobasal nucleus. This pathway is absent in adults [29]. However, this anomalous contralateral visual circuit can be maintained in adulthood by a combination of damage to the following structures at birth. One is the superior colliculus, which receives direct contralateral retinal input. A second is damage to the visual cortex. This insult resulted in retrograde degeneration in the dorsal lateral geniculate nucleus, which normally receives contralateral retinal input. The third site was injury to a variety somatosensory structures that normally project to the contralateral ventrobasal nucleus. These included the spinal cord, the main sensory nucleus of the trigeminal, and the dorsal column nuclei. In the absence of input from these damaged structures, contralateral retinal projections to the ventrobasal nucleus are preserved in adult animals [30]. In these circumstances, maintained circuits depend on competition with other connections for the available synaptic space in the target structure [28].

Fig. 2.

New brain circuit after newborn brain injury. Schematic diagrams of certain contralateral retinal projections in newborn, adult, and newborn brain injured hamsters are shown. After damage either to structures or their main tracts in newborn animals, contralateral retinal projections to the ventrobasal complex that are present in newborn animals are maintained in adult animals. This figure is and based on data from [29, 30] and modified from Fig. 1 [30]. DCN, dorsal column nuclei; dLGN, dorsal lateral geniculate nucleus; SC, superior colliculus; SpC, spinal cord; VB, ventrobasal nucleus; Vs, main sensory nucleus of the trigeminal.

Changes in Circuits due to Pruning and Mutations

The following example illustrates how a circuit can change during development and be maintained by genetic alterations (Fig. 3). Motor circuits involved in hand and digit movements in mammals are complex. These behaviors are controlled by connections between motor cortex and the cervical spinal cord in a variety of mammals [31]. However, differences in corticospinal circuits among mammals are present. One of the most significant differences is direct corticospinal tract termination on motor neurons of the cervical spinal cord. These neurons innervate hand muscles which allow for individual digit movement. This direct connection between motor cortex and motor neurons in the cervical spinal cord is present in monkeys and humans but absent in other mammals [31]. Corticospinal synapses on cervical spinal cord interneurons do occur in mammals. However, the most common explanation for individual hand digit movement is the direct corticospinal connection with cervical spinal cord motor neurons [32]. This direct circuit to cervical motor neurons is missing in adult mice but is present in postnatal day 2 mice. This fiber path from motor cortex to motor neurons of the spinal cord can be maintained by genetic alterations produced in PlexA1^fl/fl;Emx1-Cre mice (Fig. 3, modified mouse). Associated with this circuit in these mice is the ability for individualized movement of the forepaw digits [33].

Fig. 3.

Corticospinal tract circuits. Schematic, simplified corticospinal tract circuits in a variety of mice and macaque monkeys are illustrated. Note the difference between the postnatal and adult mouse. The direct connection between motor cortex and spinal cord motor neurons found in postnatal mice is lost during development when adulthood is reached. This circuit is maintained in the genetically engineered mouse (modified mouse). This direct circuit from motor cortex to motor neurons in the cervical spinal cord is similar to the one found in monkeys.

The expression of PlexA1 in layer V of motor cortex, the source of efferents to the spinal cord, differs between mice and humans. Its expression is strong in wild type mice but weak in humans which may explain why fine motor control is present in higher primates [33].

Corticomotor connections with the spinal cord and their influence on skilled hand movements involve other factors as well. Among these are the connection between corticospinal neurons and muscle [34]; certain brainstem neurons, propriospinal neurons, and intersegmental neurons [35]; and corticospinal collaterals that end in the spinal cord [36]. Additional studies that investigated molecular mechanisms of corticospinal tract development have added to the understanding of hand movement dexterity [37].

Summary and Unanswered Questions

This report examines a variety of developmental and genetic influences on neural circuit organization in adult animals. Each category of examples emphasizes that solely focusing on adult animals would overlook the impact of changes that occurred during development or were the result of genetic mutations or perinatal injury. These additional perspectives provide a more complete explanation for the observations seen in adult animals.

Why a given anlage, exemplified by the habenula, dorsal thalamus, and spinal cord, can give rise to such different morphology and circuits after its initial subdivision remains incompletely known. As shown by these examples, a single developmental field can give rise to multiple nuclei with radically different characters. Most likely, an understanding will come from further gene and transcription factor experiments [1, 10, 11, 14, 17–19] and new techniques for visualizing neural circuits [38].

An explanation for new circuit formation will build upon the voluminous data that have investigated the formation of normal brain connections. Using this knowledge, similar approaches can then be applied to studying new circuits. Two general areas will likely yield further insights.

One is the development of a new brain circuit that occurs “normally” while the other is an “artificial” connection produced by genetic changes. An example of the former is the corpus callosum. The corpus callosum is absent in marsupials but is present in placental mammals, where it represents a new mammalian circuit [39]. Recent investigations comparing marsupials and placental mammals have used these gene and transcription expression techniques to suggest how the corpus callosum may have developed in placentals [40]. Some of the genetic changes that produced new brain circuits have been discussed previously. Other examples include the many mutations, almost exclusively studied in mammals, especially mice that have been detailed above [15, 33]. Additional insights are likely to come from studies of development in these mutant animals together with gene and transcription expression analyses.

The importance of neural connections in sculpting brain function and behavior cannot be overemphasized. A better understanding of the production of new as well variations in a given neural circuit would likely have been overlooked if only adult animals were examined. Other perspectives, such as the ones described in this commentary, will provide some of this missing information. These additional data will be key to understanding circuit variation and changes in vertebrate brains and how these alterations in connections have influenced brain evolution.

Acknowledgments

Suggestions and comments by Prof. G. F. Striedter and two anonymous reviewers greatly improved this manuscript.

Conflict of Interest Statement

The author has no conflict of interest to declare.

Funding Sources

No funding was received for this study.

Author Contributions

M.B.P. analyzed the data, wrote the paper, and made the figures.

Funding Statement

No funding was received for this study.