Development of the Primary Auditory Cortex in the Ferret

UNIVERSITY OF CINCINNATI

DATE: August 13, 2003

I, , Bethany A. Adler hereby submit this as part of the requirements for the degree of:

Master of Arts in:

Communication Sciences and Disorders It is entitled:

Development of the Primary Auditory Cortex in the

Ferret

Approved by: Susan Stanton, Ph.D. Linda Graeter, Ph.D. Gina Montuoro, M.A.

Development of the Primary Auditory Cortex in the Ferret

A thesis submitted to the

Division of Research and Advanced Studies of the University of Cincinnati

In partial fulfillment of the Requirements for the degree of

MASTER OF ARTS

in the Department of Communication Sciences and Disorders of the College of Allied Health Sciences

2003

Bethany A. Adler

B.S., Liberty University, 2001

Committee Chair: Susan Stanton, Ph.D.

ABSTRACT

The purpose of this study was to examine the development of the primary auditory cortex (AI) in

the ferret. This study was conducted using five sable ferrets (Mustela putorius furo) of the

following ages: (1) postnatal age 11 (2) postnatal age 18 (3) postnatal age 25 (4) postnatal age 32

and (5) postnatal age 63. Each subject was perfused transcardially with physiological saline,

followed by tissue fixation using 4% paraformaldehyde in 0.1 M sodium phosphate buffer (pH

7.4). Each ferret was then decapitated, and the skull was dissected to allow removal of the brain,

which was stored in the same fixative. The auditory cortex of each ferret was dissected from the

whole brain and subsequently embedded in a gelatin and egg yolk solution. The embedded

tissue blocks were then immersed in 4% paraformaldehyde for 72 hours and then sectioned using

a vibrotome. Sections cut in 70 µm increments were stained using cresyl violet, and qualitatively

analyzed using a light microscope. The results of this study supported previous findings, indicating that the laminar structure of the primary auditory cortex is in place by postnatal day 18 in the ferret. Additional studies investigating the laminar structure in younger ferrets and utilizing cell counts is recommended.

iii

ACKNOWLEDGEMENTS

I would like to thank my committee for their guidance, suggestions, and feedback during

the production of this thesis. Dr. Susan Stanton, Dr. Linda Graeter, and Gina Montuoro were

gracious enough to lend their expertise to this investigation and I greatly appreciated their time.

I considered it a privilege to study under Dr. Stanton, who turned out to be even more brilliant than I had garnered during my first year of graduate school, when I attended her lectures. Her vision and direction made this project possible and I owe her a great debt of gratitude. I was especially grateful for the vast quantity of laboratory supplies she allowed me to expend during this learning experience.

I would also like to thank my family for their support and encouragement, particularly my sister, who provided late-night chemistry phone support, free of charge. She has always been my academic inspiration, and as usual, she turned out to have the answers I needed. She also served as a part-time counselor as the pressure mounted, which I appreciated more than anything

else.

Finally, I must thank my classmates, and my favorite practicum supervisor, Dr. Thomas

Goldman, who provided invaluable compassion and empathy throughout this project. Their

insightful comments helped me maintain a realistic perspective, and their humor helped me keep

my sanity. I am sure that God put these special people in my path just when I needed them most.

TABLE OF CONTENTS

Page

LIST OF FIGURES AND TABLES...... 2 LIST OF ABBREVIATIONS ...... 3 CHAPTER 1...... 4 INTRODUCTION ...... 4 CHAPTER 2...... 6 LITERATURE REVIEW...... 6 Anatomy of Neocortex...... 6 Laminar Structure of Neocortex ...... 9 Specific Layers of Neocortex ...... 10 Development of the Neocortical Layers in Humans ...... 12 Summary of the Neural Migration Process in Adult Mammals...... 14 Development of the Human Primary Auditory Cortex ...... 17 Development of the Ferret Sensory Cortex...... 19 Visual Cortex ...... 19 Somatosensory Cortex ...... 21 Auditory Cortex ...... 21 Histological Techniques...... 23 CHAPTER 3...... 25 MATERIALS AND METHODS ...... 25 Subjects ...... 25 Materials...... 25 Procedures ...... 25 Tissue Preparation...... 25 Qualitative Observation...... 26 CHAPTER 4...... 28 RESULTS ...... 28 Development of the Neocortical Layers...... 28 CHAPTER 5...... 33 DISCUSSION, LIMITATIONS, AND CONCLUSIONS ...... 33 Discussion...... 33 Limitations ...... 34 Conclusions...... 35 REFERENCES...... 37

LIST OF FIGURES AND TABLES

FIGURES Figure 1: The six layers of neocortex ...... 6 Figure 2: The lobes of neocortex ...... 8 Figure 3: Brodmann’s map of human cerebral cortex ...... 8 Figure 4: Tonotopic organization of human primary auditory cortex ...... 9 Figure 5: Migration of neurons to create new layers of cells in human cortex ...... 16 Figure 6: Neural migration from the ventricular zone to the CP in primates ...... 17 Figure 7: Lateral view of the auditory cortex in the ferret...... 22 Figure 8: P18, Section 25...... 29 Figure 9: P25, Section 64...... 29 Figure 10: P32, Section 35...... 30 Figure 11: P32, Section 35...... 31 Figure 12: P63, Section 44...... 32

TABLES Table 1: Ferrets vs. humans: neocortical development in the posterior cortex...... 20 Table 2: A comparison of time necessary for neurodevelopmental events in the posterior cortex of ferrets and humans...... 21 Table 3: Somatic diameter (microns) of Nissl-stained neurons in ferrets at different postnatal ages ...... 24 Table 4: Number of sections cut from tissue under investigation in the present study ...... 26 Table 5: Development of laminar structure in ferret AI ...... 28

LIST OF ABBREVIATIONS

AI ...... primary auditory cortex CP ...... cortical plate E ...... embryonic IZ ...... intermediate zone ME ...... middle ectosylvian area MZ ...... marginal zone P ...... prenatal PP ...... preplate PS ...... pseudosylvian sulcus SP ...... subplate SS ...... suprasylvian sulcus SV ...... subventricular zone V ...... ventricular zone V1 ...... primary visual cortex VA ...... ventroanterior area VP ...... ventroposterior area WM ...... white matter µm ...... microns

CHAPTER 1

Introduction

The following study examined the development of the auditory cortical layers in the sable ferret (Mustela putorius furo). The primary auditory cortex (AI) from ferrets of various ages was sectioned, stained for Nissl and examined using light microscopy. The objective of this study was to document maturation of the ferret auditory cortex, in an attempt to identify patterns that may be cross-compared to the development of the human auditory cortex.

A foundational understanding of neocortical structure, function, and development is necessary for a discussion of AI. Neocortical structure is largely consistent across cortical regions and thus a basic understanding of its anatomy will assist in the description of the primary auditory cortex. The auditory cortex consists of six layers. Cell density and type vary by layer, and these variations are what make it possible to identify the specific layers of AI.

Neuronal migration is the process by which neurons travel to their permanent destinations in the central nervous system. This process is also an essential concept for the purposes of the present study. Neuronal cells become established in their final destinations of the neocortex in an inside-out process of development, by which means the innermost layers of AI develop early, and the outermost layers develop late.

Human and ferret AI develop on very different timetables, although the progression of events is quite similar. Unlike humans, ferrets are born with an immature neocortex and thus a very high level of neuronal migratory activity occurring in the sensory cortex (Luskin & Shatz,

1985; Jackson, Peduzzi & Hickey, 1989; Herrmann, Antonini & Shatz, 1994). This postnatal development yields an advantageous model for the study of cortical maturation. Since there is a shortage of published research on ferret AI, supplementary data on the development of

somatosensory and visual cortex in the ferret will also be included.

This study was conducted by making qualitative observations of cell density and types at various cortical depths. The results of this study supported previous findings, indicating that the six-layer structure of the primary auditory cortex is in place by postnatal day 18 in the ferret.

CHAPTER 2

Literature Review

Anatomy of Neocortex

The cerebral cortex is commonly recognized as the brain’s surface layer of “gray matter”.

It is primarily in this region that sensory and motor functions are coordinated. The neocortex is a

convoluted, multi-layered type of cerebral cortex which is found only in mammals (Fig. 1).

Figure 1 The six layers of neocortex (modified from Carlson, 1986).

Interspecies comparisons of the neocortex have revealed close similarities in thickness

(Rakic, 1988); in adult humans, the neocortex is approximately 3 to 4 mm thick (Mountcastle,

1997). However, marked differences across species have been demonstrated in terms of greater or lesser surface area. In simple terms, the surface area of the neocortex may be observed in the degree to which gyri and sulci are pronounced upon visual inspection. Neocortex in humans

transitions from a state of relative smoothness to obvious convolution during gestation (Cowan,

1979). Greater or lesser surface area has been attributed to the number of neuronal columns of

which the cortex is comprised. Thus, although neuronal columns seem to be characteristically

similar among mammalian species, it is believed that it is the actual quantity of the columns that

makes greater cognitive potential possible in more complex species. In some cases, this greater

surface area contributes to the capacity for unique functional or anatomical areas (i.e., Broca’s

language area, which is not found across species).

The various regions of neocortex are known as the frontal lobe, parietal lobe, occipital

lobe, and temporal lobe (Fig. 2). The more specific regions of the human neocortex were

mapped and numbered by Korbinian Brodmann at the beginning of the twentieth century. In

humans, AI corresponds with Brodmann’s area 41, which lies on the upper ridge of the temporal

gyrus (Fig. 3). AI is organized tonotopically (Fig. 4), with sound frequency representation increasing in a rostral to caudal direction. AI is distinct from the secondary auditory areas

because it is the primary destination of the axonal projections from the ventral division of the

medial geniculate nucleus of the thalamus (Bear et al, 2001).

Figure 2 The lobes of neocortex (modified from Carlson, 1986).

Figure 3 Brodmann’s map of human cerebral cortex (modified from Bear et al., 2001).

Figure 4 Tonotopic organization of human primary auditory cortex (modified from Bear et al., 2001).

Laminar Structure of Neocortex

Among primates, the cell bodies of cortical neurons are positioned in layers, which generally lie parallel to the brain’s surface. The actual number of layers varies by species and cortical region, but layer numbers always increase with increasing cortical depth (i.e., from the pia mater toward the ventricles; Fig. 1). There are six layers in most neocortical tissue, which are customarily referred to with Roman numerals.

The neural cells found in neocortex can be generally categorized as pyramidal or stellate.

The patterns of neural distribution based upon this dichotomy and the relative density of cells constitute the basis for the identification of separate layers. Hence, a general familiarity with these two cell types is essential for neocortical analysis.

Pyramidal cells are multi-polar, easily identified by their triangular shape, and emit elongated apical dendrites, many of which extend up to layer I and form numerous branches.

Shorter and less prominent dendrites also arise from the base of the pyramidal cell. These

dendrites are typically horizontally oriented.

When compared to pyramidal cells, stellate cells are generally smaller. Also, though not

round, stellate cells appear more spherical than pyramidal cells. The dendritic projections of

stellate cells are also distinctive; while pyramidal cells emit one prominent apical dendrite,

stellate cells emit a radial distribution of numerous dendrites, none of which tends to be pronounced. In the next section, the typical orientation of the neocortical layers will be described.

Specific Layers of Neocortex

The neocortex is usually composed of six layers, labeled I – VI (Fig. 1). Neocortical

layer I, which is also known as the molecular layer, lacks neuronal cell bodies. Instead, it is comprised of horizontal neurites from cell bodies in the deeper cortical layers.

Layer II is known as the dysfibrous layer, or the external granular layer. This region is

dense and contains many small cells, many of which are pyramidal. The cells in layer II emit

dendrites to layer I and axons to the deeper layers.

Layer III is the suprastriate layer, or the external pyramidal layer. This layer is made up

of two sub-layers of pyramidal cells; the more superficial layer consists of medium cells and the

deeper layer contains large cells. The dendrites from these cell bodies extend up to layer I. The

axons from the superficially oriented cells extend to ipsilateral cortical areas, and the axons from

the cells of layer III’s deeper layer spread to contralateral cortical areas.

Layer IV is known as the internal granular layer. This layer is comprised of stellate cells which intercede between connections from other cortical areas to pyramidal neurites. Also contained within this layer is the external band of Baillarger, which is a horizontal network of myelinated fibers. The density of these fibers results in a visible white line, making the region

prominent upon visual inspection.

Layer V is called the interstriate layer, or the internal pyramidal layer. This region contains medium and large pyramidal cells, along with some sparsely distributed stellate cells and the internal band of Baillarger. The dendrites from the large pyramidal cells extend up to layer I, while the dendrites of the small pyramidal cells only extend up to layer IV, or, in some cases, remain contained within layer V. The large pyramidal cells in layer V are some of the largest in the neocortex.

Layer VI is known as the infrastriate layer, and is sometimes called the multiform or fusiform layer. This layer includes varying sizes of pyramidal cells, which emit axons perpendicular to the cortical surface. The larger cells have dendrites extending up to layer I, while the smaller cells’ dendrites terminate at layer IV. Beneath layer VI is the white matter

(WM) of the brain (Aitkin, 1990; Bear et al, 2001; Chusid, 1982; Parent, 1996).

Neuronal Migration

Neuronal migration is the process by which millions of neurons travel from their birthplaces in either the ventricular zone (V) or the subventricular zone (SV), to their permanent locations in the central nervous system. The peak timeframe for this migratory activity is from approximately the third to the fifth months of gestation. This delicate and complex process is subject to complications introduced by either genetic errors or prenatal environmental factors.

Neuronal cells are generated in proliferative zones, and it is in this region that they divide mitotically and subsequently migrate toward their permanent destinations. In the neocortex, the proliferative zones include the V and, later in gestation, the SV. During neurogenesis, precursor cells divide at the ventricular surface, then the two daughter cells move away from the ventricular surface to engage in DNA replication. The precursor cells then return to the

ventricular surface to begin another mitotic cycle. By means which seem to be genetically

programmed, these precursor cells will eventually either undergo apoptosis or migrate toward the cortical plate (CP). Thus, proliferative activity ends at a time that varies by species. The length of time encompassing proliferative activity is believed to be directly related to neocortical surface area. Rather than functioning to result in an increase in cortical depth, an elongated proliferative period results in an increased number of radial columns, whereby surface area and resulting cortical convolutions are increased.

Two main variations of neuronal migration have been described: radial and tangential.

These types of movement are named for the direction of the neuroblasts’ migration (i.e., vertical or horizontal movement). For the purposes of the present study, an emphasis will be placed on radial migration.

Radial migration seems to be highly dependent upon radial glial cells. Radial glial cells are currently believed to have three major functions in neocortical development: guidance of

neural cells to their permanent destinations, the establishment of the columnar organization of

neocortex, and, later in prenatal development, germination and differentiation into astrocytes.

Astrocytes are relatively large glial cells with multiple processes (as reviewed by Rakic, 2000,

and Volpe, 2001).

Development of the Neocortical Layers in Humans

Early in development, at approximately 7 weeks’ gestation, the cortex is characterized by

only two layers: the VZ and the preplate (PP) (Fig. 5). At this point, the neurogenesis of early-

generated cells is occurring within the VZ. These early neuroblasts will come to comprise the

CP. However, before the neuroblasts may migrate to form the CP, fasciculated radial glial fibers

are sent out from the VZ. These will expand their foot processes to form a glial membrane at the

pial surface of the neocortex. Once an early network of glial fibers has been established, migration of early cells to form the PP begins.

As development continues into the 10th to 13th weeks of gestation, the SZ appears, and the PP is replaced by the intermediate zone (IZ), the subplate (SP), the CP and the marginal zone

(MZ). The five zones appearing at this juncture are all transient, disappearing as maturation progresses. Each of these layers has a special significance in the development of the neocortex:

• The SZ is formed by the migration of early-generated cells out of the VZ.

• The IZ will eventually become the white matter.

• The SP neurons will constitute a highly interconnected network, facilitating afferent

and interhemispheric neural communication.

• The CP will give rise to neocortical layers II – VI.

• The MZ will give rise to neocortical layer I.

The establishment of the neural regions appearing during this stage is carried out by waves of migrating neurons. New cell migration continues away from the VZ and through the

IZ and to the developing CP. As the precursor to the WM of the brain, the IZ is mainly populated by neural projections. The resulting density of myelinated structures in this area eventually produces the white color of this region. The CP becomes thicker, more compact, and unlike the IZ, contains few fasciculated bundles of glial fibers. The VZ becomes increasingly thinner during this time and the SZ remains wide by comparison. As neurons in the CP mature, they become noticeably larger as a function of differentiation.

At approximately 17 weeks’ gestation, the layers have yet to transition into their final states. Neuronal development of the SP continues, and this region becomes wider. Neuronal development of the CP also continues.

At 19 to 25 weeks’ gestation, neuronal development of the SP and the CP continues.

Neurogenesis in the VZ is markedly reduced, although migratory activity toward the CP

continues. Most of the neocortical neurons have been generated by this point.

At approximately the 26th to 29th weeks of gestation, changes are noticeable in the laminar structure of the neocortex. At this point, the complement of layers includes the following: the VZ, the SZ, the SP, and neocortical layers I – VI. The CP and MZ disappear, and have been replaced by layers I – VI. Neuronal development of the CP continues during this

stage.

At 37 – 38 weeks’ gestation (the time of full-term human pregnancy), another change in

the laminar structure is observed: the transition of the IZ into the WM is complete. At some time

near term, the VZ will disappear. After all cortical neurogenesis has ceased and the cortical

neurons have reached their final destinations, their differentiation and establishment of synapses

continues, peaking much later, around the second postnatal year (as reviewed by Rakic, 2000,

and Volpe, 2001).

Summary of the Neural Migration Process in Adult Mammals

It is important to understand the foundations of neuronal migration in the development of the neocortex. Immature neocortical neurons (neuroblasts) arise from the V, from which they migrate along radial glia fibers to reach their respective destinations. Collectively, neural migrations to a given depth away from the VZ will constitute a component of a neocortical layer.

The SV is composed of the first neurons to migrate from the VZ. The second series of neurons to migrate make up the CP. Throughout this developmental process, the SP, IZ, and VZ remain stationary. However, the CP and MZ move increasingly further from these other early appearing cortical layers as growth continues. As neurons continue to migrate from the VZ, previously

developed layers do not become increasingly distal from the VZ. Instead, early migrating neurons are surpassed by earlier migrating neurons, resulting in an inside-out course of

development (Fig. 6). Thus, the cortical layers which develop earliest are essentially “buried” beneath later appearing layers, growing increasingly distant from the pia mater until all of the cortical layers are in place.

Neurons generated in the VZ first migrate into the IZ and then into the subplate. As neurons move out of the SP, they must pass through any previously developed layers in order to reach the boundary between the developing CP and the MZ (Rakic, 2001). This region of terminating migrations gives rise to cortical layers II – VI. Layer I contains very few cell bodies

and largely consists of neural projections.

7 wk 10-13 wk 17 wk 19-25 wk 26-29 wk Newborn

Figure 5 Migration of neurons to create new layers of cells in human cortex. CP, cortical plate; IZ, intermediate zone; MZ, marginal zone; PP, preplate; SP, subplate; SV, subventricular zone; V, ventricular zone; WM, white matter (modified from Volpe, 2001).

Figure 6 Neural migration from the ventricular zone to the CP in primates (modified from Rakic, 2001).

Development of the Human Primary Auditory Cortex

The human brainstem auditory pathway is considerably developed by the time of full

gestation, and adult-like maturation is accomplished by age 1 to 3 years. However, the human

auditory cortex has been shown to require a lengthier period of development. From the 16th week of gestation to the 4th postnatal month, the cortex is transformed from a simple layer to a

uniform CP, and then to initial lamination. Coinciding with this development from the 22nd fetal week to the 4th postnatal month, a two-layered band of neurofilament-immunoreactive axons appears in layer I, but before the 4th month is reached, a great reduction in the number of

immunopositive axons is observed.

Between postnatal ages 6 months and 3 years, the cytoarchitecture develops a mature

laminar pattern and a complex of axons grows in layers VI, V, IV, and IIIc. This axonal network of the deep cortical layers continues to become increasingly complex until age 5 years. At age 5, a system of neurofilament-positive axons develops in the superficial layers IIIb, IIIa, and II. Full maturation of the human AI was found to occur by age 11-12 years, by which time the axonal density of the region approximates the level found in young adults (Moore & Guan, 2001).

A model has been produced for the location and boundaries of the human primary auditory cortex, its subdivisions, and its anatomical intersubject variability. The development of this model was based upon a quantitative cytoarchitectonic analysis of 10 postmortem brains, which were compared with an image of a living brain, obtained by magnetic resonance imaging.

This study yielded the production of a spatial reference system that allows the comparison of cytoarchitectonic maps of AI with neural activation centers (Morosan et al., 2001).

While there are some variations in the architectonic organization of the core areas of AI across mammalian species, there are sufficient similarities to facilitate the identification of AI.

The location of AI was found to be consistent across species in a study on the brains of post mortem primates. Neural matter from macaque monkeys, chimpanzees, and humans was examined; in each brain, the auditory core region was found on the dorsal surface of the temporal lobe, contained within the supratemporal plane. AI in primates was described as including a centrally located core, surrounded by a number of belts. The core was characterized by koniocortical cytoarchitecture, which indicates a granular appearance and well-developed layer

IV, often found in sensory areas of cerebral cortex. In contrast, the outlying belts of AI exhibited characteristics commonly associated with para- or pro-koniocortex, which is distinguished by a less-developed layer IV (Hackett, Preuss, and Kaas, 2001).

Development of the Ferret Sensory Cortex

Unlike humans, ferrets are born with an immature neocortex and thus a very high level of

neuronal migratory activity occurring in the sensory cortex (Luskin & Shatz, 1985; Jackson,

Peduzzi & Hickey, 1989; Herrmann, Antonini & Shatz, 1994). This postnatal development yields an advantageous model for the study of cortical maturation. By assessing development between ferrets of various young ages, the cortex may be observed increasing in overall depth, undergoing neocortical layer development, and differentiating among morphological regions.

The majority of studies of ferret cortical development have focused on visual cortex. In

general, somatosensory cortex matures first, followed by auditory cortex. Visual cortex develops

last of these three cortical regions. Due to the relative paucity of past ferret studies of

development in the auditory cortex, data from investigations of visual and somatosensory

cortical development are included as models of neocortical development.

Visual Cortex

The ferret visual cortex (V1) may be found in the occipital lobe areas 17 and 18. In V1,

the production of cortical neurons begins on or about embryonic day 20 (E20), which is

approximately mid-gestation in the 41-day prenatal term. Neurogenesis then continues past E41

until approximately postnatal day 14 (P14) (Jackson, Peduzzi & Hickey, 1989).

The cortical neurons that will ultimately be contained within a particular cortical layer are produced over a span of several days. However, this does not mean that particular days are

devoted solely to neurogenesis for just one layer. On any given day between E20 and P14 in the ferret, neurons are being generated for two or more cortical layers. V1 neurons produced before

E41 generally require one week to migrate, while V1 neurons produced postnatally need about

two weeks to migrate.

Neuronal density in the visual cortex of the ferret CP peaks on P1. This milestone is then

followed by an acute drop in neuronal density. This decline has been attributed to apoptosis and

the expanding volume of the cortex as growth progressed.

Table 1 provides summary data from a meta analysis detailing the chronology of 95

neurodevelopmental events in the posterior cortex, common to nine mammalian species. It is

important to note that this meta analysis was reporting on the timing of neurogenesis necessary for the ultimate development of specific neural structures, rather than the establishment of the final structures themselves. Although the study assessed the development of the posterior cortex,

which is a region separate from that which is being investigated by the present study, the results

are relevant due to the general consistency of structures across neocortex. The outcome data

revealed striking similarities in the development of neocortex among mammalian species.

Although there were significant interspecies timing differences reported, the uniformity of the

total structures which ultimately develop across species was apparent. This data adds credence

to the use of such studies to produce models of human development. Table 2 provides a

comparison of time necessary for the completion of specific neurodevelopmental events in

ferrets and humans (Clancy, Darlington, & Finlay, 2001).

Post-conception Day Developmental Event (Completed) Ferret Human Neurogenesis of Cortical Layer VI 31.8 79.1 Neurogenesis of Cortical Layer V 34.5 86.4 Neurogenesis of Cortical Layer IV 39.3 99.6 Neurogenesis of Cortical Layer II/III 42.6 108.6 Birth 41.0 270.0

Table 1 Ferrets vs. humans: neocortical development in the posterior cortex (Clancy, Darlington & Finlay, 2001).

Percentage of Human Ferret: Duration Human: Duration Developmental Time Neural Structure of Neurogenesis of Neurogenesis Required by Ferret for (days) (days) Same Neural Event Cortical Layer VI 8.1 22.2 0.364864865 Cortical Layer V 7.1 19.3 0.367875648 Cortical Lamina IV 7.2 19.7 0.365482234 Cortical Layer II/III 8.4 22.8 0.368421053

Table 2 A comparison of time necessary for neurodevelopmental events in the posterior cortex of ferrets and humans (Clancy et al, 2001).

Somatosensory Cortex

The same columnar organization found in V1 and A1 has been observed in cortical samples obtained from ferret kits ranging in age from P1 to P62. While the principal neural projections in these young ferrets were found to be longitudinal, some tangential projections were found at the deeper layers, specifically, at the outer borders of the proliferative zone. It has been suggested that these horizontally arranged fibers might provide a supportive structure for the lateral disribution of neurons within the neocortex. Deeper layers of the somatosensory neocortex were found to be more likely to include horizontal projections than the CP, which was characterized by very few radial projections. Interestingly, this pattern seems to reflect the neural maturity of the ferret. Older ferrets were found to have greater numbers of radial projections than younger ferrets (Juliano, Palmer, Sonty, Noctor, & Hill II, 1996).

Although the majority of the neural projections in the somatosensory cortex of young ferrets were longitudinal in nature, a clear radial pattern could be seen. This layered characteristic was defined more obviously by neighboring concentrations of cellular structures, rather than by horizontal connectivity. These layers of varying cellular density appeared to be stacked upon each other in a concentric nature.

Auditory Cortex

Ferret AI is located on the middle ectosylvian gyrus and is organized tonotopically (Fig.

7). Cells most sharply tuned to low frequencies are located ventrally, and those most sharply

tuned to high frequencies are located dorsally (Phillips, Judge & Kelly, 1988). Like human AI, ferret AI is koniocortical in nature, having a granular appearance and a well-developed layer IV.

Figure 7 Lateral view of the auditory cortex in the ferret. Abbreviations: ME: middle ectosylvian area; VA: ventroanterior area; VP: ventroposterior area; SS: suprasylvian sulcus; PS: pseudosylvian sulcus (modified from Wallace, Roeda & Harper, 1997).

A literature review performed by Gao, Newman, Wormington & Pallas in 1999 included

a review of a study published in Germany (Apfelbach & Kruska, 1979), which reported on

neocortical development in another species of the genus Mustela: the American mink (Mustela

vison). Neocortical neuronal density was assessed and was reported to decline sharply from P7

to P20, and then to continue to decline gradually until P60. This reduction in neuronal density

was attributed to apoptosis. The results from the Gao et al. (1999) study of ferret neocortical

development reported similar findings, though the exact species investigated was not indicated.

While human AI becomes responsive in utero, ferrets do not begin hearing until about

P32. This delayed onset has been assessed via behavioral, physiological, and anatomical

measures. The ear canals do not open until the end of the first postnatal month in the ferret, and

this developmental milestone coincides with the appearance of a startle response to loud hand

claps and the recording of acoustically activated neurons in the midbrain. The late onset of

hearing in the ferret might be expected, considering the comparative timetable for development

of the neocortex in relation to humans (Moore, 1982).

Histological Techniques

Previous studies have employed Nissl staining in histochemical data collection, and that

technique will also be applied in the present study. Nissl granules are contained within the

cytoplasm of neuronal cell bodies and are believed to be centers of protein production. They are

also basophilic in nature, which means that they have a tendency to stain readily (Chusid, 1982).

The basophilic Nissl granules represent rough endoplasmic reticulum, or the RNA of ribosomes. In cases of neuronal disease, changes in the distribution patterns of Nissl granules

may be observed. In the early stages of neuropathology, Nissl substance tends to move away

from the cell nucleus to the outlying cell periphery.

While axons do not contain Nissl substance, it is found within neuronal dendrites and

within the cytoplasm of neuronal cell bodies. Nissl staining has been commonly used in order to

identify cellular patterns, much more so than to locate Nissl granules, per se. Adjustments in

chosen dye, pH and length of differentiation facilitate user control over staining patterns. For

example, it is possible to use Nissl stains to identify Nissl substance alone, or to include the

nuclei of neural cell bodies.

Nissl granules may be stained with a number of dyes, such as neutral red, methylene blue,

azur, pyronin, thionin, toluidine blue and cresyl fast violet. It has been demonstrated that first

fixing tissue in alcohol enhances the Nissl staining process (Bancroft & Stevens, 1990). The

present study will employ a series of rinses and staining involving alcohol, distilled water and

cresyl violet. Table 3 has been included below as a practical example of previous data obtained via Nissl staining.

Somatic Diameter (microns) of Nissl-Stained Neurons in Ferrets at Different Postnatal Ages Postnatal Age (days) V1 Nissl AI Nissl 1 5.2 +/- 0.15 5.5 +/- 0.08 7 62. +/- 0.18 5.7 +/- 0.11 14 9.4 +/- 0.22 9.8 +/- 0.17 20 8.1 +/- 0.19 11.1 +/- 0.29 40 11.2 +/- 0.28 13.3 +/- 0.36 60 13.8 +/- 0.53 14.1 +/- 0.65 120 9.7 +/- 0.32 12.0 +/- 0.19 Table 3 Somatic diameter (microns) of Nissl-stained neurons in ferrets at different postnatal ages (Gao et al., 1999).

CHAPTER 3

Materials and Methods

Subjects

This study was conducted using five sable ferrets (Mustela putorius furo) of the following

ages: (1) postnatal age 11 (2) postnatal age 18 (3) postnatal age 25 (4) postnatal age 32 and (5)

postnatal age 63. The ferret brain tissue was fixed and examined in order to study the

cytoarchitectural development of the various neocortical structures.

Materials

4% paraformaldehyde, an Electron Microscopy Sciences OTS-4000 vibrotome, 100%

alcohol, 95% alcohol, distilled water, xylene, gelatin, egg yolk, a solution of 1 ml of 2% cresyl

fast violet and 99 ml of sodium acetate buffer, super glue, positive charged microscope slides,

permount mounting medium, glass coverslips, steel weights, an Olympus B50 microscope and a

DVC 1310 digital camera were used.

Procedures

Once a deep level of anesthesia was achieved, the viscera of thorax were exposed. Each

ferret was then perfused transcardially with physiological saline, followed by tissue fixation

using 4% paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.4). Each ferret was then decapitated, and the skull was dissected to allow removal of the brain, which was stored in the

same fixative.

Tissue Preparation

The tissue was incubated in 4% paraformaldehyde for 5 weeks and then was rinsed in

distilled water and embedded in a gelatin and egg yolk solution, in preparation for subsequent

sectioning. The embedded tissue was refrigerated to allow solidification of the embedding

material. After three days, the embedded tissue blocks were immersed in 4% paraformaldehyde

for 24 hours. Each fixed tissue block was then separately affixed to the vibrotome stage with non-water-soluble glue. Once the glue had dried, each embedded tissue block was immersed in a distilled water bath and sectioned using a vibrotome. The sections were cut in 70 µm increments and were then individually mounted on slides in a distilled water bath. Table 4 indicates the number of sections cut from each tissue block. The mounted sections were then allowed to dry for 12 hours. The tissue was then cleared and counterstained using cresyl fast violet using the following method. Once the slide-mounted tissue had dried, each tissue section was dehydrated in 100% ethanol, rehydrated with distilled water, stained with cresyl violet, rinsed with distilled water, differentiated first in 95% ethanol and then in 100% ethanol, cleared in xylene, and coverslipped using permount mounting medium. The slides were then allowed to dry for 18 hours.

Ferret Age Sections Cut (70 µm) P63 88 P32 89 P25 102 P18 80 P11 60 Table 4 Number of sections cut from tissue under investigation in the present study.

Qualitative Observation

Sectioned material was examined using microscopy. Digital photographs were taken of

each slide. Cell counts were not taken; rather, general trends regarding cell density and cell

types at various cortical depths were observed and noted.

Qualitative analysis was performed using an Olympus B50 microscope and a DVC 1310

digital camera. Selected sections from each tissue block were included in the final analysis.

Sections were studied at magnifications of 40x, 100x and 400x, and then digital images were

acquired. These images were then analyzed visually to identify neocortical layers present at the

various ages under investigation. Images were cropped using Corel Photo House v5.0 to eliminate embedding material and superfluous tissue from each of the images.

CHAPTER 4

Results

Development of the Neocortical Layers

Qualitative analysis revealed that neocortical layers I – VI were present in each age

studied, beginning with P18. Data from P11 could not be used, due to technical problems with

sectioning and mounting of the tissue sections. These findings supported previous studies (Gao

et al., 1999; Gao, Wormington, Newman, & Pallas, 2000). The authors of the aforementioned investigations demonstrated that all six neocortical layers may be found in the ferret at P20.

Rapid neural migration occurs in the first three postnatal weeks of the ferret; the cortical layers present at P20 are a marked contrast to the laminar structure at birth. Table 5 summarizes the findings from the Gao et al. studies (1999; 2000), which were performed via Nissl staining.

Age Neocortical Structures (Postnatal Day) Present P1 CP, SP, IZ P7 CP, V, VI, SP P14 SP, IV, V, VI P20 I, II, III, IV, V, VI Table 5 Development of laminar structure in ferret AI (Gao et al., 1999, 2000).

In tissue sections analyzed in the present study, the six neocortical layers were observed

at each age. The following images provide examples of the sections included in the analysis.

Figure 8 Figure 9 P18, Section 25 P25, Section 64

In Figures 8 and 9, layers I – VI have been labeled. In both of these sections, the most obvious variations are between the first and second layers. The remaining variations are much more subtle and somewhat difficult to see in these images. However, it may be observed that layer 2/3 is darker than layer 1; layer 4 is less dense than layer 2/3; layer 5 is less dense than layer 4; layer

6 is denser than layer 5.

Figure 10 P32, Section 35

In the more greatly magnified image depicted in Figure 10, it is easier to see that there are density differences between layers 2/3 and 4. These are the kinds of variations being assessed in the present study. However, the variations are so slight that high magnification is generally required to identify the boundaries between layers.

Figure 11 P32, Section 35

Figure 11 provides a good example of a neocortical pyramidal cell. The three-sided

shape is obvious, with the apex oriented toward the pia mater. True to the Nissl staining

protocol, the staining process has highlighted the cytoplasm, nuclei, and base of the neuronal dendrites.

Figure 12 P63, Section 44

Figure 12 is an image of a tissue section obtained from the oldest animal used in the study, and therefore the cytoarchitecture should be most mature and additionally, less populous as a result of apoptosis. Unfortunately, the clarity of the sections obtained from this animal was not optimal. However, it was possible to label the six neocortical layers.

CHAPTER 5

Discussion, Limitations, and Conclusions

Discussion

The findings from the present study were consistent with previous investigations and

generally yielded the expected outcomes. This study represented the first investigation within

the University of Cincinnati College of Allied Health Sciences in which an egg yolk and gelatin

embedding material had been used. Additionally, limited information on this technique could be

obtained from the literature, so a great deal of test tissue was embedded, allowed to solidify, and

immersed in fixative using various experimental schedules. Once a good, predictable outcome

could be obtained from the embedding process, the study advanced into embedding the actual

subject tissue.

All test tissue blocks had been oriented in the embedding solution such that sections were

cut in the horizontal plane. However, when the subject tissue was oriented in the embedding

solution, it became clear that the dimensions of some of the tissue blocks would necessitate an orientation that would result in sectioning through the vertical plane. Vertical sectioning had

never been attempted during test sectioning, and the outcome was unexpected. Vertical

sectioning resulted in an unpredicted breakdown of the embedding material. During the

solidification process, the egg yolk and gelatin solution goes through a partial separation, during

which air rises to the top of the embedding material, resulting in something of a “Swiss cheese”

appearance through the upper half of the embedding material, while the lower half develops the

expected consistency of a combination of gelatin and cooked egg. Sectioning in the vertical

plane made it apparent that the two mediums are not integrated well enough to withstand

sectioning and remain attached. This discovery was made while sectioning the tissue from the

P25 ferret and resulted in great difficulty sectioning that material.

A separate issue regarding the embedding material involved the degree of precision associated with placing the tissue blocks in the embedding solution and the subsequent sectioning of the tissue. Because of the properties of the embedding solution, it was difficult to feel a sense of confidence that the sections were being cut through a specific anatomical plane.

The embedding solution contracts as it solidifies, undoubtedly shifting the placement of the tissue. However, this was a factor which the author could not control. It is highly likely that the planes through which sections were cut were far from perpendicular to the surface of the cortex, but this could not be confirmed. It is suspected that this less than perpendicular sectioning method resulted in an exaggerated representation of the depth of each neocortical layer.

Finally, while the egg yolk and gelatin embedding material greatly eased the sectioning process, difficulties were still encountered obtaining intact sections. It was suspected that this was mainly related to the age of the ferrets. Generally, the younger the ferret, the more difficult it became to obtain intact sections.

Limitations

Microscopic images of the tissue from the youngest ferret in this study would have been

very interesting; indeed, if any pronounced differences in cortical development might have been

obtained, they would have been found in the youngest ferret. An analysis of tissue from ferrets

even younger than P11 would have also been helpful. However, sectioning tissue from ferrets

even younger than P11 may be outside the sectioning potential of the vibrotome used in this

study, so not attempting to work with such young tissue satisfied practical concerns.

Another aspect of the study which might have been adjusted would have been to conduct cell counts for each of the tissue sections obtained. This would have advanced the study from a qualitative to a quantitative level, which would have added clarity to the actual progression of neocortical development. However, cell counts were outside the scope of this limited study.

Still, it bears noting that cell counts would have served a valuable purpose; the number of neuronal cell bodies is known to increase as the neocortical layers are developing, and then to decrease as a function of subsequent apoptosis.

Conclusions

The qualitative results of this study supported findings from previous studies conducted by Gao et al. (1999; 2000). Their studies revealed that on P1, only the deep cortical layers were present in V1 and AI. However, it was difficult to differentiate these layers from the CP. Also, the SP was found to be very thick. On P7, layers V & VI were found, and migration of the neurons destined to comprise layer 4 was detected By P14, layer IV had become clearly established, distinguishable from the CP. On P20, all cortical layers were present, and the cortex had become noticeably thicker, though still not as thick as it would become in adulthood. Also, an increase in cortical thickness was reported between P20 and P40.

The present study was limited in comparison to the 1999 and 2000 studies conducted by

Gao et al. The youngest tissue from which data could be collected in this study was P18. Thus, the early establishment of the cortical layers could not be observed. However, this study did find that the six layers of neocortex are in place by P18, a milestone previously identified as occurring by P20. This finding was thus consistent with previous reports and was only unique in the sense that it more precisely identified the point by which the cortical layers are established.

Gao et al.’s 1999 report included several findings which should be particularly noted.

Their study indicated that by P60, the cortex could be characterized as adult-like. Also, soma

diameter of neurons was reported to increase from birth until P60. In addition, the development

of AI was slightly more advanced than that found in V1, which supports the previously identified

anteroposterior progression of cortical development (reviewed by Gao et al., 1999).

Cell counts taken between P1 and P60 revealed a decline in neuronal density in both AI

and V1. Neuronal density was found to decline sharply from P7 through P20, when the effect

was found to begin to taper until cell counts became relatively stable and adult-like. This rapid

decline has been attributed to the expansion of the brain and the subsequent “spreading out” of

the neurons found mainly in the cortical plate earlier in development. Apoptosis is a less significant but certainly noteworthy factor affecting the decline in cell density during postnatal development.

The present study provided the author with a great opportunity to review the literature and gain a clearer perspective on the course of neocortical development. The author discovered that there is a startling lack of published data regarding the actual development of the primary auditory cortex. Furthermore, although the concept of neurogenesis and neural migration has been studied repeatedly, the actual processes involved seem to remain largely theoretical, with little consensus. There is a great deal of room left for future studies.

REFERENCES

Aitkin, L. (1990). The Auditory Cortex. New York: Chapman and Hall.

Bancroft, J.D. & Stevens, A. (Eds) (1990). Theory and Practice of Histological Techniques (3rd

ed.). New York: Churchill Livingstone.

Bear, M.F., Connors, B.W., & Paradiso, M.A. (2001). Neuroscience: Exploring the Brain,

Second Edition. Baltimore: Lippincott Williams & Wilkins.

Carlson, N. R. (1986). Physiology of Behavior (3rd ed.) . Boston : Allyn and Bacon.

Chusid, J.G. (1982). Correlative Neuroanatomy and Functional Neurology (18th ed.). Los

Altos, California: Lange Medical Publications.

Clancy, B., Darlington, B. & Finlay, B. L. (2001). Translating developmental time across

mammalian species. Neuroscience 105, 7-17.

Cowan, W.M. (1979). The development of the brain. Scientific American 241(3), 113-33.

Gao, W.J., Newman, D.E., Wormington, A.B., & Pallas, S.L. (1999). Development of inhibitory

circuitry in visual and auditory cortex of postnatal ferrets: immunocytochemical localization

of GABAergic neurons. The Journal of Comparative Neurology 409, 261-73.

Gao, W.J., Wormington, A.B., Newman, D.E., Pallas, S.L. (2000). Development of inhibitory

circuitry in visual and auditory cortex of postnatal ferrets: immunocytochemical localization

of calbindin- and paravalbumin- containing neurons. The Journal of Comparative

Neurology 422, 140-157.

Hackett, T.A., Preuss, T.M. & Kaas, J.H. (2001). Architectonic identification of the core region

in auditory cortex of macaques, chimpanzees, and humans. Journal of Comparative

Neurology 441, 197-222.

Herrmann, K., Antonini, A., & Shatz, C.J. (1994). Ultrastructural evidence for synaptic

interactions between thalamocortical axons and subplate neurons. European Journal of

Neuroscience 6(11), 1729-1742.

Jackson, C.A., Peduzzi, J.D., & Hickey, T.L. (1989). Visual cortex development in the ferret. I.

Genesis and migration of visual cortical neurons. The Journal of Neuroscience 9(4), 1242-

1253.

Juliano, S.L., Palmer, S.L., Sonty, R.V., Noctor, S., & Hill II, G.F. (1996). Development of local

connections in ferret somatosensory cortex. The Journal of Comparative Neurology 374,

259-277.

Luskin, M.B., & Shatz, C.J. (1985). Studies of the earliest generated cells of the cat's visual

cortex: cogeneration of subplate and marginal zones. The Journal of Neuroscience 5(4),

1062-1075.

Moore, D.R. (1982). Late onset of hearing in the ferret. Brain Research 253, 309-311.

Moore, J.K. & Guan, Y.L. (2001). Cytoarchitectural and axonal maturation in human auditory

cortex. Journal of the Association for Research in Otolaryngology 2(4), 297-311.

Morosan, P., Rademacher, J., Schleicher, A., Amunts, K., Schormann, T., & Zilles, K. (2001).

Human primary auditory cortex: cytoarchitectonic subdivisions and mapping into a spatial

reference system. NeuroImage 13(4), 684-701.

Mountcastle, V. B. (1997). The columnar organization of the neocortex. Brain 120, 701-722.

Parent, A. (1996). Carpenter’s Human Neuroanatomy (9th ed.). Baltimore: Williams and

Wilkins.

Phillips, D.P., Judge, P.W., & Kelly, J.B. (1988). Primary auditory cortex in the ferret (Mustela

putorius): neural response properties and topographic organization. Brain Research 443,

281-294.

Rakic, P. (1988). Specification of cerebral cortical areas. Science 241(4862), 170-176.

Rakic, P. (2000). Setting the stage for cognition: genesis of the primate cerebral cortex. In:

Gazzaniga, M. S. (Ed.), The New Cognitive Neurosciences (2nd ed.) (pp. 7-22). Cambridge,

Massachusetts: Massachusetts Institute of Technology Press.

Rakic, P. (2001). Professor and chairman of neurobiology: research interests [online].

Available: http://info.med.yale.edu/neurobio/rakic/rakic.html (January 3, 2003).

Volpe, J.J. (2001). Neurology of the Newborn (4th ed.). Philadelphia: W.B. Saunders Company.

Wallace, M.N., Roeda, D., & Harper, M.S. (1997). Deoxyglucose uptake in the ferret auditory

cortex. Experimental Brain Research 117, 488-500.