Abstract

To clarify the developmental brain changes during adolescence, brain morphology was compared between healthy younger adolescent and elder adolescent subjects using both voxel-based morphometry (VBM) and volumetric region-of-interest (ROI) analysis of magnetic resonance imaging (MRI). High-resolution three-dimensional MRI scans were acquired in 23 (10 males and 13 females) younger adolescent subjects (13–14 years) and 30 (15 males and 15 females) elder adolescent subjects (19–21 years). Whole-brain analysis by VBM revealed that the elder adolescent subjects had significantly more gray matter in the left medial temporal regions than the younger adolescent subjects and significantly less gray matter in the left medial frontal region (Brodmann area 6). In the volumetric analysis, significantly less cerebral gray matter volume and significantly greater cerebral white matter volume were found in elder adolescents compared with younger adolescents. The volume of the hippocampus was significantly larger in male elder adolescents than in male younger adolescents. The volume of the parahippocampal gyrus did not differ between younger and elder adolescent subjects. These results suggest a robust maturational process ongoing in the human hippocampus during adolescence, especially in males. The possible relevance of these findings to progress in myelination and implications in psychiatric disorders were discussed.

Introduction

Adolescence is a transitional developmental period in which major physical, psychological, cognitive and social transformations occur. Behavioral changes associated with adolescence include changes in social behaviors (changing relations with parents and developing satisfying relationships outside the family) and in sexual and aggressive drives, preponderance in risk-taking behaviors, and achievement of identity, as well as dramatic quantitative and qualitative growth in cognitive abilities (Spear, 2000; King, 2002). The human brain development is considered to be closely associated with behavioral achievements. To better understand the maturational process in behavior, it is essential to uncover brain development during adolescence. This may also provide a clue for understanding the mechanisms underlying neuropsychiatric disorders such as schizophrenia, which typically develops in late adolescence or early adulthood, demonstrates morphologic abnormalities in fronto-temporo-limbic structures (Shenton et al., 2001; Suzuki et al., 2002) and has been implicated in abnormal brain development (Feinberg, 1983; Weinberger, 1995).

As evidenced by post-mortem and animal studies, one of the most prominent changes in the adolescent brain is a massive elimination or ‘pruning’ of cortical synapses (Huttenlocher, 1994; Rakic et al., 1994). Post-mortem studies have also revealed a protracted progression of myelination during adolescence and adulthood (Yakovlev and Lecours, 1967; Benes, 1989; Benes et al., 1994). In vivo examinations of normal brain development by magnetic resonance imaging (MRI) have demonstrated that gray matter decreases and white matter increases from childhood through adulthood (Jernigan et al., 1991; Pfefferbaum et al., 1994; Caviness et al., 1996; Reiss et al., 1996; Sowell et al., 1999a, 2002, 2003; De Bellis et al., 2001). The changes in gray and white matter volumes are considered to reflect the dendritic pruning process and myelination/axonal growth, respectively. These maturational changes have been reported to show regionally variable patterns; a reduction in cortical gray matter occurs primarily in the dorsal parietal and frontal regions (Giedd et al., 1999; Sowell et al., 1999b, 2001), along with a decrease in subcortical gray matter (Giedd et al., 1996a; Thompson et al., 2000) and an increase in white matter in the internal capsule and arcuate fasciculus (Paus et al., 1999). Some reports have highlighted gender-specific maturational changes of the developing brain (Giedd et al., 1996a,b, 1999; De Bellis et al., 2001). However, the full spatial and temporal distribution and significance of structural changes in the brain during adolescence are not yet established. Notably, only a few MRI studies have focused on the developmental changes of the hippocampus during adolescence (Giedd et al., 1996b; Sowell and Jernigan, 1998). Further, to our knowledge, few studies have focused on the age range of adolescence by comparing the brain morphology between the beginning and the end of adolescent period.

We performed cross-sectional comparisons of high-resolution MRI between younger adolescent and elder adolescent subjects using both a whole-brain analysis by voxel-based morphometry (VBM) and a volumetric region-of-interest (ROI) analysis of the hippocampus and parahippocampal gyrus.

Materials and Methods

Subjects

Twenty-three healthy younger adolescent subjects (10 males and 13 females; age range = 13–14 years) and 30 healthy elder adolescent subjects (15 males and 15 females; 18–21 years) were included in this study. Demographic data of the subjects are presented in Table 1. All subjects were right-handed, and were screened by interviews using questionnaires for perinatal, early developmental, educational, medical, neurological or psychiatric abnormalities. Parents were also interviewed for the younger adolescent subjects. Digit Span, Vocabulary, and Block Design subtests of the Wechsler Intelligence Scale for Children-Third Edition (WISC-III) were administered to all younger adolescent subjects. Nineteen of the 30 elder adolescent subjects completed the homologous three subtests of the Wechsler Adult Intelligence Scale — Revised (WAIS-R). The Vocabulary and Block Design subtests provided an estimate of full-scale IQ (Silverstein, 1982). The Minnesota Multphasic Personality Inventory (MMPI) was administered to all late adolescent subjects, and subjects were excluded if any T-score exceeded 70. The subjects were recruited from among the community by an advertisement except 11 elder adolescent subjects who were recruited from among medical and pharmaceutical students in the early phase of this study and did not undergo the WAIS-R subtests. There was no significant difference between younger adolescent subjects and elder adolescent subjects in gender, handedness, parental education or estimated IQ. Written informed consent was obtained from all subjects and also from parents of the younger adolescent subjects. This study was approved by the Committee on Medical Ethics of Toyama Medical and Pharmaceutical University.

Table 1

Demographic characteristics of younger and elder adolescent subjects

 Younger adolescent subjects
 
   Elder adolescent subjects
 
   

 
Male (n = 10)
 
 Female (n = 13)
 
 Male (n = 15)
 
 Female (n = 15)
 
 
Handedness 10 right-handed  13 right-handed  15 right-handed  15 right-handed  
Age (years) 13.2 (0.4) 13.5 (0.5) 19.3 (1.0) 20.0 (0.9) 
Height (cm)a 158.1 (9.3) 155.9 (5.6) 170.2 (4.3) 158.0 (4.5) 
Weight (kg)b 46.7 (7.2) 49.4 (12.4) 65.3 (12.5) 50.0 (4.7) 
Education (years) 8.0 (0.0) 7.9 (0.3) 13.8 (0.8) 14.1 (0.8) 
Parental education (years) 14.5 (1.5) 13.4 (1.5) 13.0 (1.8) 13.0 (1.6) 
WISC-III/WAIS-Rc         
    Digit span 10.0 (2.5) 8.6 (1.6) 8.1 (2.4) 9.0 (3.2) 
    Block design 11.2 (3.6) 10.6 (2.2) 9.3 (3.8) 11.6 (3.4) 
    Vocabulary 10.5 (2.5) 11.1 (3.4) 6.8 (2.2) 8.5 (3.4) 
    Estimated IQ
 
105.1
 
(14.7)
 
105.1
 
(13.2)
 
91.4
 
(11.7)
 
103.5
 
(16.9)
 
 Younger adolescent subjects
 
   Elder adolescent subjects
 
   

 
Male (n = 10)
 
 Female (n = 13)
 
 Male (n = 15)
 
 Female (n = 15)
 
 
Handedness 10 right-handed  13 right-handed  15 right-handed  15 right-handed  
Age (years) 13.2 (0.4) 13.5 (0.5) 19.3 (1.0) 20.0 (0.9) 
Height (cm)a 158.1 (9.3) 155.9 (5.6) 170.2 (4.3) 158.0 (4.5) 
Weight (kg)b 46.7 (7.2) 49.4 (12.4) 65.3 (12.5) 50.0 (4.7) 
Education (years) 8.0 (0.0) 7.9 (0.3) 13.8 (0.8) 14.1 (0.8) 
Parental education (years) 14.5 (1.5) 13.4 (1.5) 13.0 (1.8) 13.0 (1.6) 
WISC-III/WAIS-Rc         
    Digit span 10.0 (2.5) 8.6 (1.6) 8.1 (2.4) 9.0 (3.2) 
    Block design 11.2 (3.6) 10.6 (2.2) 9.3 (3.8) 11.6 (3.4) 
    Vocabulary 10.5 (2.5) 11.1 (3.4) 6.8 (2.2) 8.5 (3.4) 
    Estimated IQ
 
105.1
 
(14.7)
 
105.1
 
(13.2)
 
91.4
 
(11.7)
 
103.5
 
(16.9)
 

Values are mean (SD). WISC-III, Wechsler Intelligence Scale for Children — Third Edition; WAIS-R, Wechsler Adult Intelligence Scale — Revised.

a

Height in male elder adolescent subjects is taller than that in each of the other three subject groups (P < 0.01, analysis of variance [ANOVA] followed by post hoc Tukey's test).

b

Weight in male elder adolescent subjects is larger than that in each of the other three subject groups (P < 0.01, ANOVA followed by post hoc Tukey's test).

c

WISC-III subtests was given to all younger adolescent subjects, while 19 of 30 elder adolescent subjects completed WAIS-R subtests.

MRI Acquisition

MRI scans were acquired with a 1.5 T scanner (Vision, Siemens Medical System, Inc., Erlangen, Germany). A three-dimensional (3-D) T1-weighted gradient-echo sequence FLASH (Fast Low-Angle Shots) with 1 × 1 × 1 mm voxels was used. Imaging parameters were: TE = 5 ms; TR = 24 ms; flip angle = 40°; field of view = 256 mm; matrix size = 256 × 256.

Voxel-based Comparison of Whole Brain Gray Matter

Voxel-based morphometry was performed according to the methods described by Ashburner and Friston (2000). After transformation of 3-D magnetic resonance images to the ANALYZE format, they were processed using the Statistical Parametric Mapping (SPM) 99 software (Wellcome Department of Cognitive Neurology, Institute of Neurology, London, UK) running in MATLAB 5.3 (Mathworks Inc., Sherborn, MA) on a Windows-platform computer. The images of each subject were spatially normalized by transforming all subjects' data to the same standard stereotactic space (Talairach and Tournoux, 1988). The normalization procedure includes estimation of the optimum 12-parameter affine transformation (linear normalization) and 7 × 8 × 7 basis functions (nonlinear normalization). The spatially normalized images were then segmented into gray matter, white matter, and cerebrospinal fluid (CSF) (Ashburner and Friston, 1997) with a correction for non-uniformity of image intensity. The segmentation procedure in SPM99 employs a modified mixture model cluster analysis to identify voxel intensities matching particular tissue types (gray matter, white matter and CSF) combined with an a priori knowledge of the spatial distribution of these tissues in normal subjects, derived from probability maps. The segmented images were processed to automatically remove any remaining non-brain matter. The gray matter segments were smoothed with a 12 mm full width at half maximum isotropic Gaussian kernel to reduce confounds caused by individual differences in gyral anatomy. The intensity in each voxel of the smoothed data is a locally weighted average of gray matter concentration from a region of surrounding voxels, the size of the region being defined by the size of the smoothing kernel (Ashburner and Friston, 2000).

Comparison between the younger and elder adolescent groups was performed by an analysis of covariance (AnCova) model for global normalization with overall grand mean scaling (Friston et al., 1990). This statistical option normalizes the segmented brain images to the same total amount of gray matter while preserving regional differences in gray matter concentration. Gender was also treated as a nuisance covariate. To test the statistical significance of regionally specific group effects, the SPM{T} statistic was used to evaluate two linear contrasts (more or less gray matter in younger adolescents than in elder adolescents). Statistical significance was defined as P < 0.05 corrected for all voxels.

Volumetric Analysis of Regions of Interest

Image processing for volumetric ROI analysis has previously been described in detail (Takahashi et al., 2002; Zhou et al., 2003). Briefly, on a Unix workstation (Silicon Graphics, Inc., Mountain View, CA), the image data were processed with the software package Dr. View 5.0 (Asahi Kasei Joho System Co., Ltd., Tokyo, Japan). Brain images were realigned in three dimensions and reconstructed into entire contiguous coronal slices of 1 mm thickness, perpendicular to the anterior commissure–posterior commissure line. The whole cerebrum was separated from the brainstem and cerebellum. According to the Alpert algorithm (Alpert et al., 1996), the signal-intensity histogram distributions across the whole cerebrum were used to segment the voxels semi-automatically into gray matter, white matter and CSF. Using the thresholds between the tissue compartments, volumes of whole hemispheric gray matter and white matter were calculated. Intracranial volume (ICV) was also measured as described previously (Zhou et al., 2003). Cerebrospinal fluid volume was calculated by subtracting the whole cerebral volume from the volume of the supratentorial part of the intracranial cavity.

The hippocampus and the parahippocampal gyrus were manually outlined on consecutive 1 mm coronal slices, from anterior to posterior, with the corresponding sagittal and axial planes simultaneously presented for reference. The demarcations of the hippocampus and parahippocampal gyrus in a representative subject are presented in Figure 1. The anterior boundary of the hippocampus was determined by reference to the sagittal plane since the boundary between the hippocampus and the amygdala is more readily identified on the sagittal plane (Convit et al., 1999). The alveus was used to differentiate the hippocampal head from the amygdala and served as the superior boundary of the whole hippocampus. The inferior boundary was the white matter of the parahippocampal gyrus. The lateral and medial boundaries were the inferior horn of the lateral ventricle and the mesial edge of the temporal lobe, respectively. For the parahippocampal gyrus, the most anterior slice was defined by the appearance of the white matter tract (temporal stem) linking the temporal lobe with the rest of the brain. The parahippocampal gyrus was separated laterally by using a line drawn from the most lateral border of the hippocampal flexure to the collateral sulcus, and superiorly by the inferior gray border of the hippocampal formation. The most posterior slice for both the hippocampus and the parahippocampal gyrus was at the level of the last appearance of the fiber of the fornix. Volumes of gray and white matter of the hippocampus or of the parahippocampal gyrus were measured together.

Figure 1.

Presentations of the hippocampus (red) and the parahippocampal gyrus (blue) measured with region-of-interest analysis. The illustrations were taken from mutually orthogonal transaxial (A), sagittal (B) and coronal (C) planes. A three-dimensional reconstructed image of the two regions is also shown (D). Note: The amygdala is excluded from the measurements. a, anterior; i, inferior; l, left; p, posterior; r, right; s, superior.

Figure 1.

Presentations of the hippocampus (red) and the parahippocampal gyrus (blue) measured with region-of-interest analysis. The illustrations were taken from mutually orthogonal transaxial (A), sagittal (B) and coronal (C) planes. A three-dimensional reconstructed image of the two regions is also shown (D). Note: The amygdala is excluded from the measurements. a, anterior; i, inferior; l, left; p, posterior; r, right; s, superior.

One trained rater (H.H.), who was blinded to the subjects' identity, measured the volumes of the regions of interest presented in this study. The intrarater intraclass correlation coefficients (ICCs) in five randomly sampled brains were 0.96 for the hippocampus and 0.97 for the parahippocampal gyrus. The inter-rater ICCs were 0.93 for the hippocampus and 0.94 for the parahippocampal gyrus.

Statistical analyses were performed using repeated measures multivariate analysis of variance with ICV as a covariate (MANCOVA) for each region, with group (younger adolescents, elder adolescents) and gender (male, female) as between-subject factors and hemisphere (right, left) as a within-subject factor. Post hoc Tukey's honestly significant difference tests modified for unequal sample sizes were employed to follow up the significant main effects or interactions yielded by MANCOVA. Statistical significance was defined as P < 0.05 (two-tailed).

Results

Voxel-based Morphometric Analysis

SPM maps showing regional gray matter differences between the groups are illustrated in Figure 2 in which the presetting of height threshold, P = 0.001 (uncorrected) was adopted, for illustrative purposes only, to demonstrate also tendencies toward more or less gray matter. In this presetting, the elder adolescent subjects had more gray matter than the younger adolescents in the bilateral medial temporal regions and the hypothalamus (Fig. 2A). On the other hand, the elder adolescents had less gray matter than the younger adolescents in the dorsolateral frontal and parietal regions, and in the right cerebellum (Fig. 2B).

Figure 2.

Regional gray matter differences between younger adolescent subjects and elder adolescent subjects. Statistical parametric maps in three orthogonal projections show voxels where regional gray matter was larger in elder adolescents than in younger adolescents (A) and smaller in elder adolescents than in younger adolescents (B), respectively. Threshold was set at P < 0.001 uncorrected for illustrative purposes only to show also tendencies toward more or less gray matter. l, left; r, right.

Figure 2.

Regional gray matter differences between younger adolescent subjects and elder adolescent subjects. Statistical parametric maps in three orthogonal projections show voxels where regional gray matter was larger in elder adolescents than in younger adolescents (A) and smaller in elder adolescents than in younger adolescents (B), respectively. Threshold was set at P < 0.001 uncorrected for illustrative purposes only to show also tendencies toward more or less gray matter. l, left; r, right.

According to the criterion of statistical significance of P < 0.05 corrected for all voxels, extracted regions with significant peak coordinates are shown in Table 2. The elder adolescent subjects had significantly more gray matter than the younger adolescents in the left medial temporal regions. The significant peak coordinates were located in the left hippocampus and the left parahippocampal gyrus. On the other hand, the elder adolescents had significantly less gray matter than the younger adolescents in the left medial frontal region. The significant peak coordinate of this region was located in the medial part of the superior frontal gyrus corresponding to Brodmann area 6. Since separate comparisons in each gender did not show significant difference in regional gray matter between the younger adolescents and the elder adolescents, probably due to the small sample size, the male and female subjects were combined for analysis with gender as a nuisance covariate.

Table 2

Stereotaxic coordinates of the maxima demonstrated significant gray matter difference between younger adolescent subjects and elder adolescent subjects and corresponding p and T values

Coordinates
 
  Pa T Location Direction 
x
 
y
 
z
 

 

 

 

 
−24 −4 −28 0.035 5.38 Left medial temporal Early adolescent < late adolescent 
−28 −24 −12 0.044 5.31 Left medial temporal Early adolescent < late adolescent 
−10
 
4
 
54
 
0.034
 
5.40
 
Left medial frontal (BA 6)
 
Early adolescent > late adolescent
 
Coordinates
 
  Pa T Location Direction 
x
 
y
 
z
 

 

 

 

 
−24 −4 −28 0.035 5.38 Left medial temporal Early adolescent < late adolescent 
−28 −24 −12 0.044 5.31 Left medial temporal Early adolescent < late adolescent 
−10
 
4
 
54
 
0.034
 
5.40
 
Left medial frontal (BA 6)
 
Early adolescent > late adolescent
 

BA, Brodmann's area.

a

P values corrected for total volume.

Volumetric Analysis of Regions of Interest

Volumes of the intracranial cavity, CSF, whole hemispheric gray matter and white matter, hippocampus, and parahippocampal gyrus are presented in Table 3. MANCOVA for the whole gray matter revealed significant main effects of group (F = 49.52, df = 1, 48, P < 0.001) and hemisphere (F = 19.40, df = 1, 49, P < 0.001); the elder adolescent subjects had significantly smaller gray matter volumes than the younger adolescent subjects (post hoc test, P < 0.001), and the gray matter volume was larger in the left hemisphere than in the right hemisphere (post hoc test, P < 0.001). In MANCOVA for the whole white matter, there were also significant main effects of group (F = 13.27, df = 1, 48, P < 0.001) and hemisphere (F = 127.67, df = 1, 49, P < 0.001); in contrast to gray matter, the elder adolescent subjects had significantly larger white matter volumes than the younger adolescent subjects (post hoc test, P = 0.005), and the white matter volume was larger in the right hemisphere than in the left hemisphere (post hoc test, P < 0.001). There was no other main effect or interaction in MANCOVAs for the whole gray or white matter volume. Further, the CSF volume was significantly larger in elder adolescent subjects than in the younger adolescent subjects (main effect of group in ANCOVA; F = 22.09, df = 1, 48, P < 0.001 and post hoc test; P < 0.001).

Table 3

Volumes of brain structures measured with region-of-interest analysis in younger and elder adolescent subjects

Brain region Younger adolescent subjects
 
   Elder adolescent subjects
 
   

 
Male (n = 10)
 
 Female (n = 13)
 
 Male (n = 15)
 
 Female (n = 15)
 
 
Intracranial volume 1540 (196.5) 1400 (135.5) 1541 (107.5) 1355 (103.5) 
Cerebrospinal fluida 93.2 (27.6) 96.2 (27.7) 139.3 (47.7) 127.7 (36.4) 
Cerebral gray matterb         
    Left 429 (51.4) 390 (37.5) 383 (40.2) 324 (19.0) 
    Right 417 (51.2) 379 (38.6) 375 (40.1) 317 (18.9) 
Cerebral white matterc         
    Left 176 (31.4) 154 (23.2) 187 (28.3) 179 (27.2) 
    Right 193 (33.0) 172 (25.9) 207 (33.5) 191 (27.9) 
Hippocampusd         
    Left 2.91 (0.34) 2.85 (0.30) 3.29 (0.34) 2.80 (0.21) 
    Right 3.01 (0.30) 2.95 (0.37) 3.49 (0.32) 3.02 (0.31) 
Parahippocampal gyruse         
    Left 7.37 (0.76) 6.67 (0.79) 7.63 (0.98) 6.58 (0.87) 
    Right
 
7.85
 
(0.64)
 
7.09
 
(0.78)
 
7.66
 
(0.98)
 
6.79
 
(0.65)
 
Brain region Younger adolescent subjects
 
   Elder adolescent subjects
 
   

 
Male (n = 10)
 
 Female (n = 13)
 
 Male (n = 15)
 
 Female (n = 15)
 
 
Intracranial volume 1540 (196.5) 1400 (135.5) 1541 (107.5) 1355 (103.5) 
Cerebrospinal fluida 93.2 (27.6) 96.2 (27.7) 139.3 (47.7) 127.7 (36.4) 
Cerebral gray matterb         
    Left 429 (51.4) 390 (37.5) 383 (40.2) 324 (19.0) 
    Right 417 (51.2) 379 (38.6) 375 (40.1) 317 (18.9) 
Cerebral white matterc         
    Left 176 (31.4) 154 (23.2) 187 (28.3) 179 (27.2) 
    Right 193 (33.0) 172 (25.9) 207 (33.5) 191 (27.9) 
Hippocampusd         
    Left 2.91 (0.34) 2.85 (0.30) 3.29 (0.34) 2.80 (0.21) 
    Right 3.01 (0.30) 2.95 (0.37) 3.49 (0.32) 3.02 (0.31) 
Parahippocampal gyruse         
    Left 7.37 (0.76) 6.67 (0.79) 7.63 (0.98) 6.58 (0.87) 
    Right
 
7.85
 
(0.64)
 
7.09
 
(0.78)
 
7.66
 
(0.98)
 
6.79
 
(0.65)
 

Values are mean (SD) of measured volume (cm3). For MANCOVA results, see text. Post hoc comparisons followed MANCOVA with intracranial volume as a covariate revealed:

a

Cerebrospinal fluid volume is larger in elder adolescent subjects than in younger adolescent subjects (P < 0.001).

b

Cerebral gray matter volume is larger in younger adolescent subjects than in elder adolescent subjects (P < 0.001) and on left hemisphere than on right hemisphere (P < 0.001).

c

Cerebral white matter volume is larger in elder adolescent subjects than in younger adolescent subjects (P = 0.005) and on right hemisphere than on left hemisphere (P < 0.001).

d

Hippocampal volume is larger in male elder adolescent subjects than in male younger adolescent subjects (P = 0.002) and on the right hemisphere than on left hemisphere (P < 0.001).

e

Parahippopcampal gyrus volume is larger on the right hemisphere than on left hemisphere (P = 0.004).

MANCOVA for the hippocampus showed significant main effects of group (F = 11.73, df = 1, 48, P = 0.001) and hemisphere (F = 18.98, df = 1, 49, P < 0.001), and a significant group-by-gender interaction (F = 7.01, df = 1, 48, P = 0.011); post hoc tests revealed that the hippocampal volume was larger in the male elder adolescent subjects than in the male younger adolescent subjects (P = 0.002), although there was no significant difference in the female subjects (P = 0.999). The right hippocampus was significantly larger than the left (post hoc test, P < 0.001). MANCOVA for the parahippocampal gyrus revealed only a significant main effect of hemisphere (F = 8.77, df = 1, 49, P = 0.005); the parahippocampal gyrus volume was larger on the right than on the left (post hoc test, P = 0.004).

Discussion

This structural MRI study demonstrated that healthy male subjects in late adolescence had a significantly larger volume of the hippocampus than those in early adolescence, suggesting developmental changes ongoing in the human hippocampal formation during adolescence, especially in males. The fact that both volumetric ROI and VBM analyses led to a similar conclusion seems compelling, although the male-specific changes were revealed only in the ROI analysis. This study also showed significantly smaller total gray matter volume and significantly greater total white matter volume in elder adolescents compared with younger adolescents. These findings are consistent with a large number of previous MRI studies which reported age-related decreases in cerebral gray matter volumes and increases in cerebral white matter volumes in childhood and adolescence (Jernigan et al., 1991; Pfefferbaum et al., 1994; Caviness et al., 1996; Reiss et al., 1996; Sowell et al., 1999a, 2002, 2003; De Bellis et al., 2001). Larger CSF volumes in the elder adolescents than those in the younger adolescents are also congruent with previous studies (Jernigan et al., 1991; Pfefferbaum et al., 1994; Reiss et al., 1996; Sowell et al., 2002, 2003), and suggest that, as a whole, the regressive changes are more predominant than the progressive changes in adolescence.

The present VBM analysis could compare relative amount of gray matter between the groups, since the images of each individual were normalized to the same total amount of gray matter (see Materials and Methods). This means that an increase or decrease in gray matter in circumscribed areas might lead not only to the appearance of change in these areas, but also to an apparent change in the opposite direction in unchanged areas. In the present case, however, such false-positive results seem unlikely to have occurred in the VBM results, because similar findings were obtained from the volumetric ROI comparisons controlled for ICV, which was almost perfectly matched between the younger and elder groups.

To our knowledge, there have been two MRI studies which examined the maturational changes in the hippocampus during the age range covering adolescence (Giedd et al., 1996b; Sowell and Jernigan, 1998). The volume of the mesial temporal lobe, including the hippocampus, amygdala, uncal cortex and parahippocampal gyrus, was reported to show age-related increase in healthy subjects aged 8–38 years (Sowell and Jernigan, 1998). Another study also demonstrated an age-related increase in volume of the hippocampus in female subjects aged 4–18 years (Giedd et al., 1996b). The result of the present study is consistent with those of the previous studies in suggesting that volume expansion of the hippocampus occurs, but indicates volume changes more specifically during adolescence. However, it should be noted that the present study is limited by the fact that the trajectories of hippocampal development can not be examined by regression analyses because of a gap of five years between the younger and elder adolescent groups.

In the hippocampal formation, postnatal neuronal proliferation occurs only in the dentate gyrus, in both rodents and primates. In rats, the numbers of dentate granule cells have been reported to increase during both the juvenile and adult periods (Bayer et al., 1982: Kuhn et al., 1996). There has also been evidence for continuous generation of neurons in the hippocampal dentate gyrus of adult monkeys (Kornak and Rakic, 1999). These reports suggest the occurrence of neurogenesis in the dentate gyrus during adolescence and adulthood in humans, but direct data have not been available.

In contrast, myelination, a broadly accepted marker for the functional maturation of the central nervous system, appears to continue long after birth in the human hippocampus. Postnatal increases of myelination in the superior medullary lamina, which links the hippocampal formation with the entorhinal and cingulate cortices, have been reported to occur during childhood, adolescence, and even adulthood (Benes, 1989; Benes et al., 1994). It has also been demonstrated that myelination in the perforant path lasts in childhood until adolescence, after which, however, the pattern remains unchanged (Arnold and Trojanowski, 1996). Considering the robust increases in the extent of myelination reported in post-mortem studies (Benes, 1989; Benes et al., 1994), it is likely that the volume of the hippocampus is expanded by an increase in myelination in adolescence.

In our volumetric ROI analysis, larger hippocampal volume observed in the elder adolescents in comparison with the younger adolescents may be explained by such volume expansion resulting from the increase in myelination during adolescence. In the VBM analysis, however, it is not clear how increased myelination in the hippocampus would result in an increase in tissue volume quantified as gray matter on MRI. In a small structure with complex gray and white matter composition, such as the human hippocampus, it can be speculated that an increase in myelination increases the volume of the structure but does not sufficiently alter tissue signal characteristics to change the classification of the automatically determined gray matter voxels, as discussed in the previous study (Sowell and Jernigan, 1998).

It is well known that normal gender differences exist in human brain anatomy (Nopoulos et al., 2000; Goldstein et al., 2001; Gur et al., 2002), although the timing and course of such differential development are not fully understood. With respect to the postnatal brain maturation, recent MRI studies demonstrated that males had more prominent age-related gray matter decreases (De Bellis et al., 2001) and white matter increases (Giedd et al., 1999; De Bellis et al., 2001) compared with females during childhood and adolescence. The post-mortem study revealed that myelination in the hippocampus occurred earlier in females than in males during childhood and adolescence (Benes et al., 1994). The MRI study also demonstrated hippocampal volume increases in female subjects aged 4–18 years, which were younger than our subjects (Giedd et al., 1996b). Animal studies have suggested that estrogen has stimulating effects on neuron proliferation (Tanapat et al., 1999), dendritic spine increases (Gould et al., 1990) and synaptogenesis (Woolley et al., 1996) in the hippocampus. Estrogen has also been reported to induce myelination in the rat brain (Prayer et al., 1997). Considering these findings together, it may be suggested that, in our female subjects, maturational processes in the hippocampus had proceeded earlier than in males and had become blunted to be detected as a volume difference between the younger and elder subjects. However, it is necessary to include subjects with the age range missing in the present study or to longitudinally follow up the same subjects from childhood through late adolescence to specifically address this issue.

Links between the limbic structures and the neocortex are thought to be responsible for the integration of emotion with cognition (Benes, 1994). The hippocampus, which has reciprocal connections with the cingulate and entorhinal cortices, is one of the important components in the corticolimbic circuitry of the human brain. The volume expansion of the hippocampus in the advanced phase of human brain development during adolescence may involve a more effective interplay between cognitive processing and emotional reactivity mediated by such circuitry. The finding of male-specific changes in the hippocampal volume in the present study may have some relevance to the development of gender differences in cognition and emotion.

Another possible implication of the present findings may stem from recent neuroanatomic findings in schizophrenia, in which volume reduction in the hippocampal region has repeatedly been reported even in its first episode and predominantly in male patients (Lawrie and Abukmeil, 1998; Shenton et al., 2001). Schizophrenia is associated with deterioration in emotional experience and cognition, and has a typical age at onset during late adolescence and early adulthood. A possibility is suggested that some abnormality in morphological maturation of the hippocampus resulting in disruption of normal volume expansion during adolescence may be involved in the development of psychotic symptoms in schizophrenia, or may increase the susceptibility to develop schizophrenia as has been hypothesized (Seidman et al., 2002; Kurachi, 2003). Given that the maturational process in the hippocampus during adolescence is more active in males than in females, an insult to it would be able to cause structural abnormalities in the hippocampus predominantly in male subjects. Indeed, there is evidence that the severity of psychotic symptoms is significantly correlated with the hippocampal volume reduction in male patients with schizophrenia (Bogerts et al., 1993). Decreases in myelination-related gene expression observed in post-mortem brains of schizophrenia patients (Hakak et al., 2001: Davis et al., 2003) also lend support to the likelihood that disturbed myelination in the hippocampus during adolescence plays some role in the development of schizophrenia.

In conclusion, while the present data are from a cross-sectional sample and need to be replicated in a longitudinal study, the findings suggest that a robust maturational process is ongoing during adolescence in the human hippocampus. These findings may have some implications not only in normal development in cognition and emotion during adolescence, but possibly in the mechanisms underlying neuropsychiatric disorders such as schizophrenia.

This research was supported by the Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (12470193) and the Japanese Ministry of Education, Culture, Sports, Science and Technology (12210009). The authors would like to thank Dr Lisha Niu, Ms Ikiko Yamashita and Mr Kouichi Mori for their support with this research.

References

Alpert NM, Berdichevsky D, Levin Z, Morris ED, Fischman AJ (
1996
) Improved methods for image registration.
Neuroimage
 
3
:
10
–18.
Arnold SE, Trojanowski JQ (
1996
) Human fetal hippocampal development. I. Cytoarchitecture, myeloarchitecture, and neuronal morphologic features.
J Comp Neurol
 
367
:
274
–292.
Ashburner J, Friston KJ (
1997
) Multimodal image coregistration and partitioning: a unified framework.
Neuroimage
 
6
:
209
–217.
Ashburner J, Friston KJ (
2000
) Voxel-based morphometry-the methods.
Neuroimage
 
11
:
805
–821.
Bayer SA, Yackel JW, Puri PS (
1982
) Neurons in the rat dentate gyrus granular layer substantially increase during juvenile and adult life.
Science
 
216
:
890
–892.
Benes FM (
1989
) Myelination of cortical-hippocampal relays during late adolescence.
Schizophr Bull
 
15
:
585
–593.
Benes FM (
1994
) Development of the corticolimbic system. In: Human behavior and the developing brain (Dawson G, Fischer KW, eds), pp. 176–206. New York: Guilford Press.
Benes FM, Turtle M, Khan Y, Farol P (
1994
) Myelination of a key relay zone in the hippocampal formation occurs in the human brain during childhood, adolescence, and adulthood.
Arch Gen Psychiatry
 
51
:
477
–484.
Bogerts B, Lieberman JA, Ashtari M, Bilder RM, Degreef G, Lerner G, Johns C, Masiar S (
1993
) Hippocampus–amygdala volumes and psychopathology in chronic schizophrenia.
Biol Psychiatry
 
33
:
236
–246.
Caviness VS, Kennedy DN, Richelme C, Rademacher J, Filipek PA (
1996
) The human brain age 7–11 years: a volumetric analysis based on magnetic resonance images.
Cereb Cortex
 
6
:
726
–736.
Convit A, McHugh P, Wolf OT, de Leon MJ, Bobinski M, De Santi S, Roche A, Tsui W (
1999
) MRI volume of the amygdala: a reliable method allowing separation from the hippocampal formation.
Psychiatry Res Neuroimag
 
90
:
113
–23.
Davis KL, Stewart DG, Friedman JI, Buchsbaum M, Harvey PD, Hof PR, Buxbaum J, Haroutunian V (
2003
) White matter changes in schizophrenia: evidence for myelin-related dysfunction.
Arch Gen Psychiatry
 
60
:
443
–456.
De Bellis MD, Keshavan MS, Beers SR, Hall J, Frustaci K, Masalehdan A, Noll J, Boring AN (
2001
) Sex differences in brain maturation during childhood and adolescence.
Cereb Cortex
 
11
:
552
–557.
Feinberg I (
1983
) Schizophrenia: caused by a fault in programmed synaptic elimination during adolescence.
J Psychiatry Res
 
17
:
319
–334.
Friston KJ, Frith CD, Liddle PF, Dolan RJ, Lammertsma AA, Frackowiak RSJ (
1990
) The relationship between global and local changes in PET scans.
J Cereb Blood Flow Metab
 
10
:
458
–466.
Giedd JN, Snell JW, Lange N, Rajapakse JC, Casey BJ, Kozuch PL, Vaituzis AC, Vauss YC, Hamburger SD, Kaysen D, Rapoport JL (
1996
a) Quantitative magnetic resonance imaging of human brain development: ages 4–18.
Cereb Cortex
 
6
:
551
–560.
Giedd JN, Vaituzis C, Hamburger SD, Lange N, Rajapakse JC, Kaysen D, Vauss YC, Rapoport JL (
1996
b) Quantitative MRI of the temporal lobe, amygdala, and hippocampus in normal human development: ages 4–18 years.
J Comp Neurol
 
366
:
223
–230.
Giedd JN, Blumenthal J, Jeffries NO, Castellanos FX, Liu H, Zijdenbos A, Paus T, Evans AC, Rapoport JL (
1999
) Brain development during childhood and adolescence: a longitudinal MRI study.
Nature Neurosci
 
2
:
861
–863.
Goldstein JM, Seidman LJ, Horton NJ, Makris N, Kennedy DN, Caviness VS Jr, Faraone SV, Tsuang MT (
2001
) Normal sexual dimorphism of the adult human brain assessed by in vivo magnetic resonance imaging.
Cereb Cortex
 
11
:
490
–497.
Gould E, Woolley CS, Frankfurt M, McEwen BS (
1990
) Gonadal steroids regulate dendritic spine density in hippocampal pyramidal cells in adulthood.
J Neurosci
 
10
:
1286
–1291.
Gur RC, Gunning-Dixon F, Bilker WB, Gur RE (
2002
) Sex differences in temporo-limbic and frontal brain volumes of healthy adults.
Cereb Cortex
 
12
:
998
–1003.
Hakak Y, Walker JR, Li C, Wong WH, Davis KL, Buxbaum JD, Haroutunian V, Fienberg AA (
2001
) Genome-wide expression analysis reveals dysregulation of myelination-related genes in chronic schizophrenia.
Proc Natl Acad Sci USA
 
98
:
4746
–4751.
Huttenlocher PR (
1994
) Synaptogenesis in human cerebral cortex. In: Human behavior and the developing brain (Dawson G, Fischer KW, eds), pp. 137–152. New York: Guilford Press.
Jernigan TL, Trauner DA, Hesselink JR, Tallal PA (
1991
) Maturation of human cereberum observed in vivo during adolescence.
Brain
 
114
:
2037
–2049.
King RA (
2002
) Adolescence. In: Child and adolescent psychiatry: a comprehensive textbook (Lewis M, ed.), pp. 332–342. Philadelphia, PA: Lippincott.
Kornack DR, Rakic P (
1999
) Continuation of neurogenesis in the hippocampus of the adult macaque monkey.
Proc Natl Acad Sci USA
 
96
:
5768
–5773.
Kuhn HG, Dickinson-Anson H, Gage FH (
1996
) Neurogenesis in the dentate gyrus of the adult rat: age-related decrease of neuronal progenitor proliferation.
J Neurosci
 
16
:
2027
–2033.
Kurachi M (
2003
) Pathogenesis of schizophrenia. Part II. Temporo-frontal two-step hypothesis.
Psychiatry Clin Neurosci
 
57
:
9
–15.
Lawrie SM, Abukmeil SS (
1998
) Brain abnormality in schizophrenia: a systematic and quantitative review of volumetric magnetic resonance imaging studies.
Br J Psychiatry
 
172
:
110
–120.
Nopoulos P, Flaum M, O'Leary D, Andreasen NC (
2000
) Sexual dimorphism in the human brain: evaluation of tissue volume, tissue composition and surface anatomy using magnetic resonance imaging.
Psychiatry Res Neuroimag
 
98
:
1
–13.
Paus T, Zijdenbos A, Worsley K, Collins DL, Blumenthal J, Giedd JN, Rapoport JL, Evans AC (
1999
) Structural maturation of neural pathways in children and adolescents: in vivo study.
Science
 
283
:
1908
–1911.
Pfefferbaum A, Mathalon DH, Sullivan EV, Rawles JM, Zipursky RB, Lim KO (
1994
) A quantitative magnetic resonance imaging study of changes in brain morphology from infancy to late adulthood.
Arch Neurol
 
51
:
874
–887.
Prayer D, Roberts T, Barkovich AJ, Prayer L, Kucharczyk J, Moseley M, Arieff A (
1997
) Diffusion-weighted MRI of myelination in the rat brain following treatment with gonadal hormones.
Neuroradiology
 
39
:
320
–325.
Rakic P, Bourgeois J-P, Goldman-Rakic PS (
1994
) Synaptic development of the cerebral cortex: implications for learning, memory, and mental illness.
Prog Brain Res
 
102
:
227
–243.
Reiss AL, Abrams MT, Singer HS, Ross JL, Denckla Mbyte (
1996
) Brain development, gender and IQ in children: a volumetric imaging study.
Brain
 
119
:
1763
–1774.
Seidman LJ, Faraone SV, Goldstein JM, Kremen WS, Horton NJ, Makris N, Toomey R, Kennedy D, Caviness VS, Tsuang MT (
2002
) Left hippocampal volume as a vulnerability indicator for schizophrenia: a magnetic resonance imaging morphometric study of nonpsychotic first-degree relatives.
Arch Gen Psychiatry
 
59
:
839
–849
Shenton ME, Dickey CC, Frumin M, McCarley RW (
2001
) A review of MRI findings in schizophrenia.
Schizophr Res
 
49
:
1
–52.
Silverstein AB (
1982
) Two- and four-subtest short forms of the Wechsler Adult Intelligence Scale — Revised.
J Consult Clin Psycholgy
 
50
:
415
–418.
Sowell ER, Jernigan TL (
1998
) Further MRI evidence of late brain maturation: limbic volume increase and changing asymmetries during childhood and adolescence.
Dev Neuropsychol
 
14
:
599
–617.
Sowell ER, Thompson PM, Holms CJ, Batth R, Jernigan TL, Toga AW (
1999
a) Localizing age-related changes in brain structure between childhood and adolescence using statistical parametric mapping.
Neuroimage
 
9
:
587
–597.
Sowell ER, Thompson PM, Holms CJ, Jernigan TL, Toga AW (
1999
b) In vivo evidence for post-adolescent brain maturation in frontal and striatal regions.
Nat Neurosci
 
2
:
859
–861.
Sowell ER, Thompson PM, Tessner KD, Toga AW (
2001
) Mapping continued brain growth and gray matter density reduction in dorsal frontal cortex: inverse relationships during postadolescent brain maturation.
J Neurosci
 
21
:
8819
–8829.
Sowell ER, Trauner DA, Gamst A, Jernigan TL (
2002
) Development of cortical and subcortical brain structures in childhood and adolescence: a structural MRI study.
Dev Med Child Psychiatry
 
44
:
4
–16.
Sowell ER, Peterson BS, Thompson PM, Welcome SE, Henkenius AL, Toga AW (
2003
) Mapping cortical change across the human life span.
Nat Neurosci
 
6
:
309
–315.
Spear LP (
2000
) The adolescent brain and age-related behavioral manifestations.
Neurosci Biobehav Rev
 
24
:
417
–463.
Suzuki M, Nohara S, Hagino H, Kurokawa K, Yotsutsuji T, Kawasaki Y, Takahashi T, Matsui M, Watanabe N, Seto H, Kurachi M (
2002
) Regional changes in brain gray and white matter in patients with schizophrenia demonstrated with voxel-based analysis of MRI.
Schizophr Res
 
55
:
41
–54.
Takahashi T, Kawasaki K, Kurokawa K, Hagino H, Nohara S, Yamashita I, Nakamura K, Murata M, Matsui M, Suzuki M, Seto H, Kurachi M (
2002
) Lack of normal structural asymmetry of the anterior cingulate gyrus in female patients with schizophrenia: a volumetric magnetic resonance imaging study.
Schizophr Res
 
55
:
69
–81.
Talairach J, Tournoux P (
1988
) Co-planar stereotaxic atlas of the human brain. Stuttgart: Georg Thieme.
Tanapat P, Hastings NB, Reeves AJ, Gould E (
1999
) Estrogen stimulates a transient increase in the number of new neurons in the dentate gyrus of the adult female rat.
J Neurosci
 
19
:
5792
–5801.
Thompson PM, Giedd JN, Woods RP, MacDonald D, Evans AC, Toga AW (
2000
) Growth patterns in the developing brain detected by using continuum mechanical tensor maps.
Nature
 
404
:
190
–193.
Weinberger DR (
1995
) Schizophrenia as a neurodevelopmental disorder. In: Schizophrenia (Hirsch SR, Weinberger DR, eds), pp. 293–323. Oxford: Blackwell.
Woolley CS, Wenzel HJ, Schwartzkroin PA (
1996
) Estradiol increases the frequency of multiple synapse boutons in the hippocampal CA1 region of the adult female rat.
J Comp Neurol
 
373
:
108
–117.
Yakovlev PI, Lecours AR (
1967
) The myelogenetic cycles of regional maturation of the brain. In: Regional development of the brain in early life (Minkowski A, ed.), pp. 3–70. Oxford: Blackwell.
Zhou S-Y, Suzuki M, Hagino H, Takahashi T, Kawasaki Y, Nohara S, Yamashita I, Seto H, Kurachi M (
2003
) Decreased volume and increased asymmetry of the anterior limb of the internal capsule in patients with schizophrenia.
Biol Psychiatry
 
54
:
427
–436.

Author notes

1Department of Neuropsychiatry, Toyama Medical and Pharmaceutical University, Toyama, Japan, 2Department of Psychology, Toyama Medical and Pharmaceutical University, Toyama, Japan, 3Department of Radiology, Toyama Medical and Pharmaceutical University, Toyama, Japan and 4Department of Physiology, Toyama Medical and Pharmaceutical University, Toyama, Japan