Abstract

While brain gray matter is primarily associated with sensorimotor processing and cognition, white matter modulates the distribution of action potentials, coordinates communication between different brain regions, and acts as a relay for input/output signals. Previous studies have described morphological changes in gray and white matter during childhood and adolescence, which are consistent with cellular genesis and maturation, but corresponding events in infants are poorly documented. In the present study, we estimated the total number of cells (neurons, oligodendrocytes, astrocytes, and microglia) in the cerebral white matter of 9 infants aged 0–33 months, using design-based stereological methods to obtain quantitative data about brain development. There were linear increases with age in the numbers of oligodendrocytes (7–28 billion) and astrocytes (1.5–6.7 billion) during the first 3 years of life, thus attaining two-thirds of the corresponding numbers in adults. The numbers of neurons (0.7 billion) and microglia (0.2 billion) in the white matter did not increase during the first 3 years of life, but showed large biological variation.

Introduction

Quantitative stereological studies of the normal human brain have hitherto focused largely on cell population in gray matter of the adult (Pakkenberg and Gundersen 1997; Pelvig et al. 2008), and on development of the fetal brain (Samuelsen et al. 2003). Less attention has been given proliferation of cell populations during the early postnatal period, and development of the subcortical white matter is in particular need of documentation. The human brain is only 30% of its adult size at birth versus approximately 65% in the macaque monkey and approximately 45% in chimpanzee, and attains approximately 80% of its adult weight during the first 2–3 postnatal years (Prechtl 1986; Jensen 1996). Thus, man is an extreme case for perinatal brain development, with the period of maximal brain growth extending from the last half of gestation into the first year of life (Dobbing and Sands 1973). Approximately half of the mass of the neonate human brain is white matter, in which the predominant cell type consists of glia, that is oligodendroglia and astrocytes, which together provide the myelin for neuronal fibers and maintain a cellular and extracellular environment necessary for neuronal functioning. There is a growing recognition that glia, possibly through a glial network, may undertake intracellular communication, complementing that of the neurons themselves (Baumann and Pham-Dinh 2001). Although oligodendrocytes are well known for their production of myelin, the full range of their functions remains a puzzle; satellite oligodendrocytes without any connection of the myelin sheath have been described (Baumann and Pham-Dinh 2001). The development of the non-neuronal elements of the human brain has been a neglected topic.

The human forebrain develops from the cranial part of the neural tube (Stiles and Jernigan 2010), and the future brain cells originate from proliferating progenitor cells situated in the areas surrounding the telencephalic ventricles, that is the ventricular zone, the subventricular zone, and the ganglionic eminence. From these regions, the newly generated and differentiated neurons migrate radially along a scaffold provided by radial glial cells, or tangentially toward the cortical plate (Rakic 1988; Kornack 2000; Parnavelas et al. 2000). Some neurons stop their migration in the subplate, a transient zone identified in the fetus, which gradually dissolves during early infancy, and merges with the intermediate zone to become the white matter of the adult brain (Kostovic and Rakic 1980, 1990; Judas et al. 2010). The subventricular zone, which is present in the late gestational and early postnatal mammalian brain, is the major source of astrocytes and oligodendrocytes; maturation of these cells occurs in early postnatal life, where there is a large expansion of the fiber network. In the late prenatal and early postnatal period, when the majority of subplate neurons die, the remaining cells come to comprise the interstitial neurons of the adult white matter (Kostovic and Rakic 1980, 1990; Chun and Shatz 1989; Ghosh and Shatz 1992). Although oligodendrocyte proliferation occurs early, myelination takes place late in development; in general, the pattern of myelination proceeds from proximal to distal structures in the long tracts, from the phylogenetically older to newer structures, that is in a caudo-cephalic direction in the cerebral while matter. First myelination begins in the spinal cord at as early as 22 weeks of gestation, but commences gradually in the hemispheric white matter during the first 2 postnatal years, and continues to develop until late adolescence (Friede 1989). Thus, even at term, the hemispheres are almost totally devoid of myelin, and the high water content of the white matter probably contributes for the extreme fragility of the immature brain (Larroche 1998). The primary sensory pathways are myelinated first, while the associative areas myelinate later, and last of all the anterior frontal and anterior temporal cortices. Myelin plays an inhibitory role to growing axons, and contributes to the anatomic stability of the mature, intact nervous system, by preventing random regrowth of axon sprouts once pathways have been established and myelinated. High-speed salutatory conduction, fidelity of signaling for long distances, and space economy are the 3 major advantages conferred to the vertebrate nervous system by the myelin sheath. The importance of myelin in maintaining normal human brain function is highlighted by its involvement in an array of different neurological diseases, such as leukodystrophies and multiple sclerosis in the central nervous system and peripheral neuropathies in the peripheral nervous system (Baumann and Pham-Dinh 2001).

Structural MRI and molecular biological studies have shown morphological and quantitative changes in gray and white matter during human childhood and adolescence, consistent with ongoing cellular maturation and the formation of new cells (Winick 1968; Dobbing and Sands 1973; Toft et al. 1995; Giedd et al. 2010; Sanchez et al. 2012). Glial progenitors in the cortex and white matter continue to proliferate in infancy, throughout childhood and into adulthood (Miller 2002), in concert with forebrain myelination (Aubert-Broche et al. 2008). Increasing cell size, glial proliferation, and marked myelination all contribute to the substantial brain growth in the first 3 years of life (Voigt and Pakkenberg 1983; Jernigan et al. 2011). Despite this knowledge about the developing brain, there is no quantitative data on the proliferation of cell populations in white matter during the postnatal period. In this study, we aimed to quantify the total number of neurons and glial cells in the subcortical white matter of 9 infant brains, ranging in age from newborn to 3 years of age. To this end, we used modern stereological cell counting methods for unbiased estimation of cell populations. The project took advantage of a unique collection of postmortem brain specimens from infants who died without primary neurological deficits that were collected and preserved at the brain bank at Bispebjerg Hospital.

Materials and Methods

Subject Characteristics

The study was based on hemispheres from the brains of 9 infants (age range 0–33 months) obtained postmortem from 1983 to 1991, all in accordance with Danish law regarding autopsied human tissue. The brains were selected from a larger sample of infant brains, but limited to those individuals without disease of the central nervous system, with the single except of the 23-month-old infant who died from diatesis hemorrhagica and with meningococcal meningitis. Table 1 details the clinical information on cause of death, brain weight, and storage time in fixative. None of the brains had tumors or hemorrhages, and none showed signs of cortical spongiosis, vasculitis, or encephalitis. The brains had been stored in different fixatives for various periods of time, so the brain weights are not directly comparable; despite this caveat brain weights were all within the normal range (Voigt and Pakkenberg 1983), except for a 19-month-old infant's brain, that had been fixed in picric acid for 28 years. The material was limited to 9 brains as a consequence of our strict inclusion criteria.

Table 1

Subject characteristics

Age of subjects Sex Cause of death Brain weight (g) Time in fixative (years) 
Newborn (6 min) Asphyxia 420 11 
14 days Asphyxia, pneumonia 318 20 
4 months Complications during surgery for truncus arteriosus persistens 548 25 
8 months Mb. Cordis congenitalis 615 26 
9 months Mb. Cordis congenitalis 647 26 
15 months Complications during diaphragmatic hernia surgery 1034 20 
19 months Mb. Cordis congenitalis 696 28 
23 months Diatesis hemorrhagica caused by severe tracheobronchitis mucopurulenta 1274 26 
33 months Sequelae to multiple dog bites 933 25 
Age of subjects Sex Cause of death Brain weight (g) Time in fixative (years) 
Newborn (6 min) Asphyxia 420 11 
14 days Asphyxia, pneumonia 318 20 
4 months Complications during surgery for truncus arteriosus persistens 548 25 
8 months Mb. Cordis congenitalis 615 26 
9 months Mb. Cordis congenitalis 647 26 
15 months Complications during diaphragmatic hernia surgery 1034 20 
19 months Mb. Cordis congenitalis 696 28 
23 months Diatesis hemorrhagica caused by severe tracheobronchitis mucopurulenta 1274 26 
33 months Sequelae to multiple dog bites 933 25 

Tissue Processing

All brains had been fixed for at least 10 years in 0.1 M sodium phosphate-buffered 4% formaldehyde (pH 7.2) at room temperature. Brains that were very soft were transferred to a solution of 50% formaldehyde (Merck 37%, Germany, 1.04003), 25% saturated picrid acid (Prolabo), and 25% demineralized water. Thus, differing fixation histories influenced their volume shrinkage. After fixation, the cerebrum was detached from the brainstem at the level of the third cranial nerve, and the meninges were removed. One hemisphere from each brain was used for stereology (5 left and 4 right hemispheres). The right or the left hemisphere was selected following the rules of systematic uniform random sampling (SURS; Gundersen et al. 1999). The entire hemisphere was coronally cut into 2.5-cm slabs, dehydrated, and embedded in paraffin in a LeicaASP300 tissue processor, followed by coronal sectioning on a Leica SM2400 sliding microtome set at 40 µm thickness. Exhaustive sectioning resulted in a total of 2100–3800 sections from each hemisphere, which were sampled with a known and predetermined interval. A wetted filter paper was placed on the paraffin block to allow collection of the sampled section without the use of water baths. When the section adhered to the filter paper, it was mounted on a double-silane-coated microscope slide by placing the paper with the section facing down on the slide and pressing gently with a printing roller. This procedure resulted in an almost complete absence of tissue deformation artifacts, such as usually arise from the use of water baths. Specifically, there was no shrinkage in the z-axis, and the final tissue section height after processing remained about 40 μm as in Eriksen and Pakkenberg (2007).

Upon slide mounting, the sampled sections were dried at 40 °C for 24 h, followed by heating to 60 °C for 60 min. They were then dewaxed in xylene for 2 × 30 min, followed by 2 × 20 min in 99% ethanol, 2 × 10 min in 96% ethanol, 5 min in 70% ethanol, 5 min in 35% ethanol, 5 min in 17% ethanol, and 5–10 min in distilled water. The rehydrated sections were stained for 45 min using a modified Giemsa stain containing 200 mL of Giemsa stock solution (Merck, Germany, product 1.09204) and 800 mL of potassium hydrogen phosphate (67 mmol/L, pH 4.4), which was filtered before use. Finally, the sections were differentiated with 0.5% acetic acid and dehydrated by passage through 96% ethanol, 99% ethanol for 2 × 2.5 min, and xylene for 2 × 15 min.

Criteria for Visually Determining Cell Type

As previously described (Pelvig et al. 2003, 2008; Fabricius et al. 2013), determination of cell type was performed based on morphology. The cells were identified as neurons if they had a single large nucleolus free of any surrounding heterochromatin, a typical pale chromatin pattern in a triangularly rounded nucleus, and visible cytoplasm surrounding the nucleus (Duffy 1983; Berry and Butt 1997; Fig. 1). Oligodendrocytes are often found in clusters and are characterized by a small rounded or oval nucleus with dense chromatin structure and a perinuclear halo (Fuller and Burger 1997; Baumann and Pham-Dinh 2001). Astrocytes are defined as cells with a round and pale nucleus with heterochromatin concentrated in granules in a rim below the nuclear membrane and a relatively translucent cytoplasm. A small nucleolus is not always identified, but when present, it often shows eccentric localization. Astrocytes are often found in isolation, and their nuclear membrane has a sharp profile. Microglia cells are defined by a small elongated or comma-shaped nucleus with dense peripheral chromatin (Pelvig et al. 2003). Endothelial cells were not included in the study.

Figure 1.

Representative images of (a) a neuron, (b) an oligodendrocytes, (c) an astrocyte, and (d) a section of a brain. Neurons have a combination of a single large nucleolus, typically a triangularly rounded nucleus, which are surrounded by a visible cytoplasm; oligodendrocytes are characterized by a small rounded or oval nucleus with a dense chromatin structure and a perinuclear halo; astrocytes have a round or triangular and pale nucleus with the heterochromatin concentrated in small granules and sometimes small nucleoli, often situated eccentrically; microglia (not shown) have small elongated or comma-shaped nucleus with a dense peripheral chromatin. The bar in a, b, and c = 10 μm; the bar in d = 1 cm.

Figure 1.

Representative images of (a) a neuron, (b) an oligodendrocytes, (c) an astrocyte, and (d) a section of a brain. Neurons have a combination of a single large nucleolus, typically a triangularly rounded nucleus, which are surrounded by a visible cytoplasm; oligodendrocytes are characterized by a small rounded or oval nucleus with a dense chromatin structure and a perinuclear halo; astrocytes have a round or triangular and pale nucleus with the heterochromatin concentrated in small granules and sometimes small nucleoli, often situated eccentrically; microglia (not shown) have small elongated or comma-shaped nucleus with a dense peripheral chromatin. The bar in a, b, and c = 10 μm; the bar in d = 1 cm.

Stereological Design

For white matter volume estimations, the section sampling fraction was 1/100 for all brains, providing 19–35 sections per brain. For the estimation of cell densities in the subcortical white matter, the sampling fraction was 1/200–1/400, yielding 9–11 sections per hemisphere. For precision of estimates see, for example, Pakkenberg and Gundersen (1988) and Slomianka and West (2005). The sections were selected following the rules of SURS (Gundersen et al. 1999) to ensure that all parts of the subcortical white matter had an equal probability of being sampled. Estimations of white matter regions included the subcortical white matter, corpus callosum, internal and external capsules, and anterior and posterior commissures.

Volume Estimation

Estimates of subcortical white matter volume were obtained by employing Cavalieri's principle using point probes that were systematically and randomly applied to each cross-sectional area. The Cavalieri volume was estimated by multiplying the average distance between the sections, T = t × k, where t is the average section thickness (∼40 µm) and k is the inverse sampling fraction of the sections (100). T was multiplied by the total cross-sectional area of the white matter, which was obtained by multiplying the total number of counted points in the region of interest by the area per test point (Gundersen, Bendtsen, et al. 1988; Gundersen et al. 1999).

When tissue is embedded in paraffin, substantial tissue shrinkage is inevitable. Normally, this can be accommodated by multiplying the volume by a shrinkage correction factor. However, this was not possible in this study due to the differences in shrinkage artifacts caused by the variety of fixative compositions and durations for the 9 samples. Accordingly, estimated cell densities cannot be compared directly, as they are influenced to differing extents by shrinkage. Nonetheless, the product of the density and the volume after final processing gives an unbiased estimate of the total cell numbers, and as such, can be compared among the brains, irrespective of shrinkage.

Cell Density and Number Estimation

The numerical cell densities of neurons and glial cells were estimated by applying the optical disector (Gundersen, Bagger, et al. 1988). The sections were analyzed on a BX-50 Olympus microscope equipped with a Heidenhain electronic microcator with a digital readout for measurements in the z-direction (with a precision of 0.5 µm), and an XY motorized stage (M2000, Prior). Live images from a high-resolution color camera (JAI CV-3200, Japan) mounted on a ×2.5 photo eyepiece (Olympus, DK) were forwarded to a computer running the CAST Stereology software package (v. 2.1.5.8, Visiopharm, Hørsholm, Denmark). Sampling in the z-axis was performed with a disector height of 20 µm and upper and lower guard zones of at least 5 µm. Area frame was 360 µm2 for oligodendrocytes and 3250 µm2 for the rest of the cells. The cells were counted using a ×60 oil immersion objective (Olympus, PlanApo N, NA 1.42), giving a final onscreen magnification of ×2700. Neurons and glial cells (subdivided into astrocytes, oligodendrocytes, and microglia) were counted only if they came into focus with visible nucleoli inside the counting frame. When the number of cells in all disectors, ΣQ, had been counted for 1 brain, the numerical density, NV, was estimated as described previously (Pelvig et al. 2008). The total number of particles was estimated as the product of the reference volume and the numerical density in the samples, multiplied by 2, to provide bilateral data. Cells that could not be clearly identified were separately classified as unidentified cells. They only constituted a negligible fraction of the total cell counts. Epithelial cells were not included in the counts.

Stereological Analysis

The precision of the estimates is expressed as the coefficient of error (CE) and was calculated as reported previously (Gundersen et al. 1999). As a measure of the variability of the estimates within a group, the observed interindividual coefficient of variation (CV) (SD/mean) is reported in parentheses after group means in the Results section. In this study, a mean of 231 points (range 121−458) was counted for each brain, corresponding to total white matter volumes, with an average CE(P) of <4%. For the density estimations, a mean of 170 (range 110−345) neurons was counted in 397 disectors, and a mean of 588 (range 376−918) glial cells was counted in 181 disectors per infant brain. The glial cells were distributed by type as follows: mean counts of 223 (range 115−331) oligodendrocytes, 343 (range 224−562) astrocytes, and 24 (range 10−45) microglia. This resulted in CE(Q) estimates of 8.4% for neurons, 7.1% for oligodendrocytes, 5.9% for astrocytes, and 22.8% for microglia. A mean of 8.8 cells per brain could not be identified, which represented <1.2% of the total cell number. The sum of counted cells is acceptable (CE(Q) <10%), with the exception of the microglia, which was estimated with less precision due to their rarity.

Statistical Analysis

The total cell numbers for oligodendrocytes and astrocytes appeared to increase linearly with age, so we applied a linear regression analysis to describe the total number of cells, using age in months as the independent regression variable.

The coefficient of determination, R2, is the proportion of the variation explained by the regression model, which is an index of how well the model fits the dependent variable.

Results

Number of Cells in the White Matter

The mean number of white matter cells was 23.6 × 109, ranging from 9.6 to 35.9 × 109.

Neuron number: the 9 infants were estimated to have a mean of 0.68 × 109 (CV = SD/mean = 0.28) white matter neurons (range 0.49–0.94 × 109) (Fig. 2a).

Figure 2.

Total bilateral number of cells in the subcortical white matter by age (n = 9). (a) Neurons, (b) microglia, (c) oligodendrocytes, and (d) astrocytes. The gray symbol in b indicates the male subject who died from meningitis. Open symbols: female and solid symbols: males.

Figure 2.

Total bilateral number of cells in the subcortical white matter by age (n = 9). (a) Neurons, (b) microglia, (c) oligodendrocytes, and (d) astrocytes. The gray symbol in b indicates the male subject who died from meningitis. Open symbols: female and solid symbols: males.

Glial cell number: the total number of glia cells was 22.8 × 109 (range 8.71–35.6 × 109). The glial cells were subdivided into oligodendrocytes, astrocytes, and microglia (Fig. 1). There was a mean of 19.2 × 109 oligodendrocytes (range 7.06–28.9 × 109) (Fig. 2c), 3.4 × 109 astrocytes (range 1.46–6.7 × 109) (Fig. 2d), and 0.20 × 109 (0.32) (range 0.14–0.33 × 109) microglia (Fig. 2b).

Linear Increase in Oligodendrocytes and Astrocytes During Childhood

We observed a linear increase in oligodendrocyte and astrocyte counts during the first 3 years of life. For oligodendrocytes, the regression gave y = 0.57 × 109x + 12.1 × 109, where y is the total number of oligodendrocytes, and x is the number of months after birth. The goodness of the fit was R2 = 0.84, and the slope significantly different from zero (P = 0.0005), with 95% confidence intervals (95% CIs) of 0.35–0.79 × 109 for the slope, and 8.52–15.7 × 109 for the ordinate intercept (Fig. 2c). The number of astrocytes increased with age with linear regression as y = 0.17 × 109x + 1.39 × 109. The goodness of the fit was R2 = 0.89, and the slope is significantly different from zero (P = 0.0001), with 95% CIs of 0.11–0.22 × 109 for the slope, and 0.55–2.22 × 109 for the ordinate intercept (Fig. 2d).

We applied linear regression analysis to describe the total number of neurons and microglia as well, but the slopes of the linear regression analyses were not significantly different from zero, indicating an absence of proliferation during the 3 year postnatal period.

White Matter Cell Distribution

Neurons accounted for a mean of 3.4% (range 1.4−8.0%), whereas oligodendrocytes constituted 81.3% (range 74.0−87.7%), astrocytes 13.9% cells (range 8.8−18.8%), and microglia 1.1% (range 0.6−1.9%) of all cells counted. Figure 3 shows an example of the cell distribution in the white matter of brain hemispheres from 4 cases: newborn, 8 months, 19 months, and 33 months of age.

Figure 3.

Cell distribution in 4 infants: newborn, and aged 8, 19, and 33 months. Oligo, oligodendrocytes; astro, astrocytes; micro, microglia; UnID, unidentified cells.

Figure 3.

Cell distribution in 4 infants: newborn, and aged 8, 19, and 33 months. Oligo, oligodendrocytes; astro, astrocytes; micro, microglia; UnID, unidentified cells.

Discussion

The major finding of this study is that the total number of oligodendrocytes and astrocytes in the subcortical white matter of human infants increases as a linear function of age during the first 3 years of life. The numerical increase in oligodendrocyte numbers was from approximately 7 billion to 28 billion, indicating a mean increase of 600 million oligodendrocytes per month during the first 3 years of life. Considering the large degree of variance in the total number of brain cells between individuals, it is quite extraordinary that the postnatal increase in oligodendrocyte numbers can be expressed so well by a linear model. In our material, the proliferation of subcortical white matter astrocytes is also directly related to age, increasing from 1.5 to 6.7 billion during the first 3 years of life. The linear regression slope indicated a mean birth rate of 170 million astrocytes per month. In contrast to the linear proliferation of oligodendrocytes and astrocytes, we found a constant total number of neurons and microglia in the white matter during the first 3 years of postnatal life. In total, approximately 750 million surviving new cells are added to the white matter each month, during the first 3 years of life, corresponding to a birth rate of 17 000 new cells per minute or nearly 300 per second.

There is as yet no quantitative data on the total number of cells in the white matter of adult humans; for the purpose of comparison, we have estimates of the total number of neurons, oligodendroglia, astroglia, and microglia in the entire brain white matter from 3 normal adults in their 70s (unpublished data). For this group, the total number of oligodendroglia was 37 billion, suggesting an increase of at least 10 billion after 3 years of age. The astroglia count was 9 billion in the adult white matter, compared with the approximately 6 billion at 3 years of age. It seems likely that the proliferation of oligodendrocytes and astroglia is 65–75% complete at 3 years of age, and apt to continue in the following years, although the linear growth phase must be approaching completion. The distribution of the different cell types in the infants was approximately constant with age (Fig. 3).

The mean total neuron number in white matter was the same in the infants (0.68 × 109) as in adult brains (0.69 × 109) (unpublished data). In general, neurons resident in the white matter are considered remnants from the fetal subplate, and are comprised of a heterogeneous mixture of excitatory and inhibitory neurons (Judas et al. 2010; Connor et al. 2011). These neurons are believed to influence cortical and thalamic circuitry by local effects, and to exhert a regulation of microvasculature blood flow in the brain. The sparse population of microglia in the subcortical white matter (0.20 × 109) did not increase with age and was not different in the adult brain (0.28 × 109) (unpublished data).

Distinguishing between the largest glia cells and the smaller neurons in Giemsa stained tissue can be a problem. The use of strict morphological criteria, in the absence of staining with immunohistochemistry, may not be sufficient for correct assignment of cell type in all cases. A more reliable separation of neurons and glia might have been obtained through the use of sensitive and robust antibodies for a specific glial marker such as glial fibrillary acidic protein. This was not possible in the present study due to the long-term, formalin fixation of the brain tissue. However, our morphological criteria for cell differentiation have proved to be consistent in a number of stereological studies (Pelvig et al. 2003, 2008; Fabricius et al. 2013), including immunohistochemistry (Hou et al. 2012).

Another limitation of this study is the low sample size, which was necessary due to our strict inclusion criteria. Some of the infants had died from a congenital heart condition, and it cannot be excluded that heart disease, with impaired cerebral perfusion, might have influenced the total number of brain cells. However, except for one infant, the brain weights were within normal limits, and the total number of cells showed no correlation to the pathological diagnoses. Since the statistical power of the study is low due to inclusion of only one brain per time point, these data must be considered as preliminary; more studies with a larger number of brains per time point would be necessary to substantiate our findings. Finally, the results for microglia are less precise than for other cell types, due to the naturally low abundance of these cells in human white matter under normal conditions. However, it is in line with our expectations that the number of microglia cells did not correlated to age.

Improved knowledge of the development of white matter in the human brain may have important implications for neuropsychiatric disease. Indeed, increased densities and altered distributions of white matter neurons in the frontal, temporal, and parietal lobe white matter have been implicated as a causative factor in complex central nervous system disorders, such as schizophrenia (Eastwood and Harrison 2003; Connor et al. 2011). Because of methodological differences in previous studies, which place main focus on different white matter zones (i.e., “superficial” and “deep” white matter), there is little consensus regarding which regions are affected in schizophrenia (Connor et al. 2011). However, there is some link between the disorder and changes in white matter composition during neurodevelopment, which is further implied by the occurrence of abnormalities in oligodendrocyte number and morphology, as reported in postmortem studies (Roy et al. 2007). Diffusion tensor imaging studies have demonstrated altered white matter microstructure in autism and bipolar disorder, which may have arisen from alterations in axonal organization and white matter tracts during development (Adler et al. 2006; Travers et al. 2012).

This study provides data on the cell populations in subcortical white matter of infants aged 0−33 months, being the first study to quantify proliferation of oligodendrocytes and astrocytes in the white matter during early childhood in a unique collection of brains from young children. This knowledge provides valuable information about the maturation of the brain white matter in the early phase of childhood, and is intended to serve as a reference for future attempts to understand the natural history of diseases that are associated with changes in the white matter, such as, for example, multiple sclerosis and neuropsychiatric disorders (Roy et al. 2007; Fields 2008; Nakahara et al. 2012).

Author's Contribution

B.P. conceived and supervised the project and assisted in writing the manuscript. R.K.S. performed the experiments, collected, and analyzed the data, and wrote the manuscript. M.K. assisted in data collection.

Funding

This work was supported by the Danish Council for Independent Research, Medical Sciences (FSS: 12-124750).

Notes

Dr Bjarne Beck is acknowledged for collecting some of the brain specimens. The authors thank Ms Susanne Sørensen for her skillful technical assistance. The authors acknowledge professional editing of the manuscript by Inglewood Biomedical Editing. Conflict of Interest: The authors declare no competing financial interests.

References

Adler
CM
Adams
J
DelBello
MP
Holland
SK
Schmithorst
V
Levine
A
Jarvis
K
Strakowski
SM
.
2006
.
Evidence of white matter pathology in bipolar disorder adolescents experiencing their first episode of mania: a diffusion tensor imaging study
.
Am J Psychiatry
 .
163
:
322
324
.
Aubert-Broche
B
Fonov
V
Leppert
I
Pike
GB
Collins
DL
.
2008
.
Human brain myelination from birth to 4.5 years
.
Med Image Comput Comput Assist Interv
 .
11
:
180
187
.
Baumann
N
Pham-Dinh
D
.
2001
.
Biology of oligodendrocyte and myelin in the mammalian central nervous system
.
Physiol Rev
 .
81
:
871
927
.
Berry
M
Butt
AM
.
1997
.
Structure and function of glia in the central nervous system
. In:
Graham
DI
Lantos
PL
, editors.
Greenfield's neuropathology
 .
London
:
Arnold
. p.
63
83
.
Chun
JJ
Shatz
CJ
.
1989
.
Interstitial cells of the adult neocortical white matter are the remnant of the early generated subplate neuron population
.
J Comp Neurol
 .
282
:
555
569
.
Connor
CM
Crawford
BC
Akbarian
S
.
2011
.
White matter neuron alterations in schizophrenia and related disorders
.
Int J Dev Neurosci
 .
29
:
325
334
.
Dobbing
J
Sands
J
.
1973
.
Quantitative growth and development of human brain
.
Arch Dis Child
 .
48
:
757
767
.
Duffy
PE
.
1983
.
Astrocytes: normal, reactive and neoplastic
 .
New York
:
Raven Press
.
Eastwood
SL
Harrison
PJ
.
2003
.
Interstitial white matter neurons express less reelin and are abnormally distributed in schizophrenia: towards an integration of molecular and morphologic aspects of the neurodevelopmental hypothesis
.
Mol Psychiatry
 .
8
:
769
.
821–831
.
Eriksen
N
Pakkenberg
B
.
2007
.
Total neocortical cell number in the mysticete brain
.
Anat Rec (Hoboken)
 .
290
:
83
95
.
Fabricius
K
Jacobsen
JS
Pakkenberg
B
.
2013
.
Effect of age on neocortical brain cells in 90+ year old human females—a cell counting study
.
Neurobiol Aging
 .
34
:
91
99
.
Fields
RD
.
2008
.
White matter in learning, cognition and psychiatric disorders
.
Trends Neurosci
 .
31
:
361
370
.
Friede
R
.
1989
.
Gross and microscopic development of the central nervous system. developmental neuropathology
 .
Berlin Heidelberg
:
Springer
. p.
2
20
.
Fuller
GN
Burger
PC
.
1997
.
Central nervous system
. In:
Sternberg
SS
, editor.
Histology for pathologists
 .
Philadelphia
:
Lippincott Williams & Wilkins
. p.
243
282
.
Ghosh
A
Shatz
CJ
.
1992
.
Involvement of subplate neurons in the formation of ocular dominance columns
.
Science
 .
255
:
1441
1443
.
Giedd
JN
Stockman
M
Weddle
C
Liverpool
M
Alexander-Bloch
A
Wallace
GL
Lee
NR
Lalonde
F
Lenroot
RK
.
2010
.
Anatomic magnetic resonance imaging of the developing child and adolescent brain and effects of genetic variation
.
Neuropsychol Rev
 .
20
:
349
361
.
Gundersen
HJ
Bagger
P
Bendtsen
TF
Evans
SM
Korbo
L
Marcussen
N
Moller
A
Nielsen
K
Nyengaard
JR
Pakkenberg
B
et al
1988
.
The new stereological tools: disector, fractionator, nucleator and point sampled intercepts and their use in pathological research and diagnosis
.
APMIS
 .
96
:
857
881
.
Gundersen
HJ
Bendtsen
TF
Korbo
L
Marcussen
N
Moller
A
Nielsen
K
Nyengaard
JR
Pakkenberg
B
Sorensen
FB
Vesterby
A
et al
1988
.
Some new, simple and efficient stereological methods and their use in pathological research and diagnosis
.
APMIS
 .
96
:
379
394
.
Gundersen
HJ
Jensen
EB
Kieu
K
Nielsen
J
.
1999
.
The efficiency of systematic sampling in stereology—reconsidered
.
J Microsc
 .
193
:
199
211
.
Hou
J
Riise
J
Pakkenberg
B
.
2012
.
Application of immunohistochemistry in stereology for quantitative assessment of neural cell populations illustrated in the Gottingen minipig
.
PLoS ONE
 .
7
:
e43556
.
Jensen
PKA
.
1996
.
Menneskets oprindelse og udvikling
 .
Copenhagen, Denmark
:
G.E.C. Gad
.
Jernigan
TL
Baare
WF
Stiles
J
Madsen
KS
.
2011
.
Postnatal brain development: structural imaging of dynamic neurodevelopmental processes
.
Prog Brain Res
 .
189
:
77
92
.
Judas
M
Sedmak
G
Pletikos
M
Jovanov-Milosevic
N
.
2010
.
Populations of subplate and interstitial neurons in fetal and adult human telencephalon
.
J Anat
 .
217
:
381
399
.
Kornack
DR
.
2000
.
Neurogenesis and the evolution of cortical diversity: mode, tempo, and partitioning during development and persistence in adulthood
.
Brain Behav Evol
 .
55
:
336
344
.
Kostovic
I
Rakic
P
.
1980
.
Cytology and time of origin of interstitial neurons in the white matter in infant and adult human and monkey telencephalon
.
J Neurocytol
 .
9
:
219
242
.
Kostovic
I
Rakic
P
.
1990
.
Developmental history of the transient subplate zone in the visual and somatosensory cortex of the macaque monkey and human brain
.
J Comp Neurol
 .
297
:
441
470
.
Larroche
JC
.
1998
.
The central nervous system
. In:
Wigglesworth
JS
Singer
DB
, editors.
Textbook of fetal and perinatal pathology
 .
Chicago
:
Blackwell Science
. p.
672
731
.
Miller
RH
.
2002
.
Regulation of oligodendrocyte development in the vertebrate CNS
.
Prog Neurobiol
 .
67
:
451
467
.
Nakahara
J
Maeda
M
Aiso
S
Suzuki
N
.
2012
.
Current concepts in multiple sclerosis: autoimmunity versus oligodendrogliopathy
.
Clin Rev Allergy Immunol
 .
42
:
26
34
.
Pakkenberg
B
Gundersen
HJ
.
1997
.
Neocortical neuron number in humans: effect of sex and age
.
J Comp Neurol
 .
384
:
312
320
.
Pakkenberg
B
Gundersen
HJ
.
1988
.
Total number of neurons and glial cells in human brain nuclei estimated by the disector and the fractionator
.
J Microsc
 .
150
:
1
20
.
Parnavelas
JG
Anderson
SA
Lavdas
AA
Grigoriou
M
Pachnis
V
Rubenstein
JL
.
2000
.
The contribution of the ganglionic eminence to the neuronal cell types of the cerebral cortex
.
Novartis Found Symp
 .
228
:
129
139
.
Pelvig
DP
Pakkenberg
H
Regeur
L
Oster
S
Pakkenberg
B
.
2003
.
Neocortical glial cell numbers in Alzheimer's disease. A stereological study
.
Dement Geriatr Cogn Disord
 .
16
:
212
219
.
Pelvig
DP
Pakkenberg
H
Stark
AK
Pakkenberg
B
.
2008
.
Neocortical glial cell numbers in human brains
.
Neurobiol Aging
 .
29
:
1754
1762
.
Prechtl
HF
.
1986
.
New perspectives in early human development
.
Eur J Obstet Gynecol Reprod Biol
 .
21
:
347
355
.
Rakic
P
.
1988
.
Specification of cerebral cortical areas
.
Science
 .
241
:
170
176
.
Roy
K
Murtie
JC
El-Khodor
BF
Edgar
N
Sardi
SP
Hooks
BM
Benoit-Marand
M
Chen
C
Moore
H
O'Donnell
P
et al
2007
.
Loss of erbB signaling in oligodendrocytes alters myelin and dopaminergic function, a potential mechanism for neuropsychiatric disorders
.
Proc Natl Acad Sci USA
 .
104
:
8131
8136
.
Samuelsen
GB
Larsen
KB
Bogdanovic
N
Laursen
H
Graem
N
Larsen
JF
Pakkenberg
B
.
2003
.
The changing number of cells in the human fetal forebrain and its subdivisions: a stereological analysis
.
Cereb Cortex
 .
13
:
115
122
.
Sanchez
CE
Richards
JE
Almli
CR
.
2012
.
Neurodevelopmental MRI brain templates for children from 2 weeks to 4 years of age
.
Dev Psychobiol
 .
54
:
77
91
.
Slomianka
L
West
MJ
.
2005
.
Estimators of the precision of stereological estimates: an example based on the CA1 pyramidal cell layer of rats
.
Neuroscience
 .
136
:
757
767
.
Stiles
J
Jernigan
TL
.
2010
.
The basics of brain development
.
Neuropsychol Rev
 .
20
:
327
348
.
Toft
PB
Leth
H
Ring
PB
Peitersen
B
Lou
HC
Henriksen
O
.
1995
.
Volumetric analysis of the normal infant brain and in intrauterine growth retardation
.
Early Hum Dev
 .
43
:
15
29
.
Travers
BG
Adluru
N
Ennis
C
Tromp do
PM
Destiche
D
Doran
S
Bigler
ED
Lange
N
Lainhart
JE
Alexander
AL
.
2012
.
Diffusion tensor imaging in autism spectrum disorder: a review
.
Autism Res
 .
5
:
289
313
.
Voigt
J
Pakkenberg
H
.
1983
.
Brain weight of Danish children. A forensic material
.
Acta Anat (Basel)
 .
116
:
290
301
.
Winick
M
.
1968
.
Changes in nucleic acid and protein content of the human brain during growth
.
Pediatr Res
 .
2
:
352
355
.