Cerebral Cortex October 2013;23:2322–2336
doi:10.1093/cercor/bhs222
Advance Access publication August 13, 2012
Sex Differences in the Human Brain and the Impact of Sex Chromosomes
and Sex Hormones
E. Lentini1, M. Kasahara2, S. Arver3 and I. Savic1
1
Department of Women and Child Health, 2Department of Clinical Neuroscience and 3Department of Medicine, Karolinska
Institute, Stockholm, Sweden
Address correspondence to Ivanka Savic, Department of Clinical Neuroscience, Stockholm Brain Institute, Karolinska Institutet, Retziusväg 8,
171 77 Solna, Sweden. Email: [email protected]
Ivanka Savic designed and wrote the paper; Emilia Lentini analyzed the data and performed research; Stefan Arver performed
research; and Maki Kasahara ran analyses.
While there has been increasing support for the existence of cerebral sex differences, the mechanisms underlying these differences
are unclear. Based on animal data, it has long been believed that
sexual differentiation of the brain is primarily linked to organizational
effects of fetal testosterone. This view is, however, in question as
more recent data show the presence of sex differences before the
onset of testosterone production. The present study focuses on the
impact that sex chromosomes might have on these differences. Utilizing the inherent differences in sex and X-chromosome dosage
among XXY males, XY males, and XX females, comparative voxelbased morphometry was conducted using sex hormones and sex
chromosomes as covariates. Sex differences in the cerebellar and
precentral gray matter volumes (GMV) were found to be related to
X-chromosome dosage, whereas sex differences in the amygdala,
the parahippocamus, and the occipital cortex were linked to testosterone levels. An increased number of sex chromosomes was
associated with reduced GMV in the amygdala, caudate, and the
temporal and insular cortices, with increased parietal GMV and
reduced frontotemporal white matter volume. No selective, testosterone independent, effect of the Y-chromosome was detected. Based
on these observations, it was hypothesized that programming of the
motor cortex and parts of cerebellum is mediated by processes
linked to X-escapee genes, which do not have Y-chromosome homologs, and that programming of certain limbic structures involves
testosterone and X-chromosome escapee genes with Y-homologs.
Keywords: brain, MRI, XXY, sex chromosome, sex dimorphism,
sex hormone
Introduction
Many neuropsychiatric disorders show an uneven sex distribution in regard to their prevalence, and, in some types, their
age at onset and treatment response (Bao and Swaab 2010).
The origins of this are largely unknown. One possible explanation is that sex differences in cerebral anatomy, connectivity
and function render men and women differently susceptible
to certain aspects of psychopathology.
Cerebral sex differences have been shown at the group
level in regard to structural volumes as well as regional
volumes of gray matter (GM) and white matter (WM), even
when correcting for individual differences in total brain
volume. In general, women have been found to have larger
structural volumes in the hippocampus, caudate nucleus
(Filipek et al. 1994; Murphy et al. 1996; Giedd et al. 1997,
2006), and the anterior cingulate gyrus (Paus et al. 1996),
whereas the volume of the amygdala (Giedd et al. 1997, 2006;
© The Author 2012. Published by Oxford University Press. All rights reserved.
For Permissions, please e-mail: [email protected]
Neufang et al. 2009) has been found to be larger in men. With
some exceptions (Goldstein et al. 2001; Carne et al. 2006),
men have been found to have larger GM volumes in the
mesial temporal lobe, the cerebellum, and the lingual gyrus
(Good et al. 2001; Carne et al. 2006). Women, on the other
hand, seem to have larger GM volumes in the precentral
gyrus, the anterior cingulate, the right inferior parietal, and
the orbitofrontal cortex (Nopoulos et al. 2000; Good et al.
2001; Luders et al. 2005; Luders, Gaser et al. 2009; Luders,
Sanchez et al. 2009; Savic and Arvis 2011). In addition, there
are reports of larger GM volumes among women in the dorsolateral prefrontal cortex (Schlaepfer et al. 1995) and planum
temporale (Witelson et al. 1995).
To date, the mechanisms underlying the observed sex differences remain uncertain. Unraveling these mechanisms is of
considerable interest not only for improving our understanding
of some of the basic aspects of human biology, but also for
investigating the pathophysiology of those disorders which
have a skewed sex distribution, including attention deficit hyperactivity disorder (ADHD), autism spectrum disorders,
(male/female prevalence ratio of 6:1–2:1 in various reports),
major depression, and anxiety (male/female ratio of ca. 1:3).
These conditions are associated with changes in the basal
ganglia, the amygdala, the anterior cingulate gyrus, which, according to some studies, are brain regions whose structural
volumes differ between healthy men and women (American
Psychiatric Association 2000; Baron-Cohen et al. 2000; Andreasen 2005; McAlonan et al. 2008; McAlonan et al. 2008; Balint
et al. 2009; Lai et al. 2012; Tomasi and Volkow 2012).
For many years, the dominant hypothesis among biologists
was that sex differences were linked to intrauterine exposure
to testosterone (Phoenix et al. 1959; Ehrhardt and MeyerBahlburg 1979), which after metabolization to estrogen
would induce organizational effects on an initially undifferentiated “female” brain, leading to its masculinization during a
limited period of development. This hypothesis was based on
animal experiments showing that the absence of androgen in
male rats and exposure to androgen in female rats during
sexual differentiation of the brain leads to a complete inversion of sexual behavior (Phoenix et al. 1959; Ehrhardt and
Meyer-Bahlburg 1979). The possible effects of excess intrauterine testosterone may be studied in humans by investigating
females with adrenal cortical hyperplasia (CAH). Contrary to
expectations, no sex atypical features in any of the cerebral
volumes measured were found in CAH women, compared
with female controls (Merke et al. 2003; Giedd et al. 2006;
Ciumas et al. 2009). This absence of masculinization in
women, despite their high level of fetal testosterone, suggests
that factors other than fetal testosterone might be involved in
the development of cerebral sex dimorphism in humans. Such
a scenario is supported by elegant studies by Arnold’s group,
showing that cerebral sex dimorphism exists long before
the fetus starts to produce testosterone (Arnold et al. 2004;
Arnold and Chen 2009). Further support is given through findings using animal data, which have demonstrated that several
genes are expressed in the brain in a sexually dimorphic
manner (Jazin and Cahill 2010). Altogether, these more recent
scientific achievements encourage investigations of genomic
effects on the human brain. In this respect, the possible impact
of X-chromosome gene dosage deserves special attention.
In the somatic cells of females with 46, XX, 1 of the 2 Xchromosomes is randomly inactivated. About 15% of X-linked
genes escape this process, and are denoted as escapee genes
(Carrel and Willard 2005). In 46 XX females, these genes will
be expressed from both X-chromosomes, whereas in 46 XY
males, only from one X-chromosome. Only a few of the
escapee genes have homologs on the Y chromosome. They
are located in the so-called pseudoautosomal region (Xu and
Disteche 2006). Consequently, the remaining X-linked escapee
genes may play a major role in sexual differentiation by virtue
of their higher expression in XX females.
The possible effects of sex chromosome linked gene
expression on cerebral sex dimorphism in humans can by
studied by examining the unique condition of sex chromosome aneuploidy (Good et al. 2003; Shen et al. 2004; Giedd
et al. 2007). To the best of our knowledge, magnetic resonance imaging (MRI) data from subjects with sex chromosome
aneuploidy have not been analyzed regarding sex dimorphism of cerebral GM and WM. Such studies are, however,
warranted, as they may potentially generate important information about the effects of sex chromosome gene dosage.
Optimally, they should be carried out so that the possible
effects of sex hormones can be assessed in parallel (not withstanding that some hormonal effects may be indirect Xchromosome gene effects). This would require measuring the
sex hormone levels and evaluating the possible effects of fetal
testosterone. Fetal testosterone cannot, for natural reasons, be
assessed in adult subjects. Several scientific reports suggest,
however, that the ratio between the length of the second and
fourth digit (2D:4D), may serve as its correlate (Manning et al.
1998, 2004; Williams et al. 2000; Lutchmaya et al. 2004;
Coates et al. 2009; Honekopp and Watson 2010).
In the present MR study, men with Klinefelter’s syndrome
(47, XXY males) were studied together with 46, XX and 46, XY
controls to explore whether sex chromosomes (X-chromosomes
in particular) and sex hormones have an impact on the described sex differences in the GM and WM volumes. An attempt
was also made to find out whether the possible effects of sex
chromosomes and sex hormones could be regionally differentiated. XXY males have a supernumerary X-chromosome, and
although their phenotype may vary, the findings from brain
imaging studies have been rather consistent (Giedd et al. 2007;
Lenroot et al. 2009). The testosterone levels in XXY males have
been found to be normal or subnormal during the prenatal
period and up to puberty (Carson et al. 1982; Ratcliffe et al.
1994), before becoming drastically reduced (Aksglaede et al.
2006) and calling for testosterone supplementation. The gender
roles of XXY males do not differ from other males; their identity
is male, and the majority of XXY men are heterosexual.
Based on the following data, we had 2 main reasons for expecting to find cerebral differences between XXY males and
controls, and assumed that these differences would be primarily related to 2 types of genetic mechanisms. First, in XXY
males there could be an excessive expression of genes that lie
in the pseudoautosomal regions of the X-chromosome. These
X escapee genes have Y-chromosome homologs, and may be
expressed in a higher dose in XXY males than in both XX and
XY controls, leading to cerebral differences in comparison to
“both control groups.” Second, like in XX females there are
also other X escapee genes, although their exact percentage
has not been estimated in XXY males (Vawter et al. 2007),
and these genes do not have Y-chromosome analogues. It is
assumed that they are expressed in excess in XXY males “only
in relation to XY males” but not to typical XX females. Thus,
provided that chromosome-linked processes were involved,
some differences were expected only in relation to XY males,
others in relation to both control populations.
We, therefore, hypothesized: (1) that XXY men would have
different GM and WM volumes from controls and that these
differences are due to sex chromosome aneuploidy, and hypogonadism (in this order), rather than the social environment. (2) That differences in regional GM and WM volume
between XXY and XY men, which are also found between XX
women and XY men (features more closely resembling those
in females), would be primarily due to the genes located on
the extra X-chromosome. (2) Cerebral differences in XXY
males, when compared with both control groups, should, on
the other hand, be due to trisomy.
(3) Because of hypogonadism, the testosterone levels of XXY
males are lower than in XY males (prior to testosterone substitution) but higher than in XX females. Consequently, regional
changes in XXY males could be attributed to their lower testosterone levels, if they resulted in features ”in between” those of
XX women and XY men (for example, a larger regional GM
volume than in XX women, but smaller than in XY men).
To test these hypotheses, 45 XX females, 41 XY males, and
33 XXY males were investigated using MRI and voxel based
morphometry (VBM). The data generated were analyzed consecutively at several levels, and explored with increasing complexity utilizing the options to add nuisance variables and
variables of interest in the explorative multi-regression model
used [statistical parametric mapping 5 (SPM5), Wellcome
Foundation].
Experimental Procedures
Subjects
Thirty-eight males with Klinefelter’s syndrome were recruited
consecutively from the Center of Andrology, Department of
Medicine, Karolinska University Hospital, Sweden, along with
86 controls from the general public. The exclusion criteria included not being 20–55 years, having karyotypic mosaicism; a
heredity for, history of, or current psychosis; a personality disorder; a major or bipolar depression; a neurological disease;
and alcohol or substance abuse problems. Because our
primary purpose was to evaluate the possible effects of Xchromosome dosage on sex differences in the brain, XXY subjects with comorbid autism and ADHD were also excluded.
Mild dyslexia was present in about 60% of the patients. After
exclusion of 5 patients due to depression, substance abuse, or
Cerebral Cortex October 2013, V 23 N 10 2323
hallucinations, the population included in the analysis consisted of 33 XXY males (age 39 ± 10 years, range 21–55, education 13.2 ± 2.7 years), 41 XY males (age 35 ± 7 years, range
27–50, education 16.2 ± 2.4), and 45 XX females (age 35 ± 7
years, range 26–51, education 15.9 ± 2.2).
All subjects underwent a medical examination at the screening visit, including an evaluation of their medical histories
and routine laboratory tests. Comorbid psychiatric disorders
and personality disturbances were assessed according to the
Diagnostic and Statistical Manual of the American Psychiatric
Association, 4th edition (DSM–IV®; 13) by a psychiatrist. This
assessment included a Structured Questionnaire for DSM-IV®
Axis I and II (Structured Clinical Interview for DSM-IV®
(SCID-I and II) (American Psychiatric Publishing Inc., 1997),
as well as scores for depression (Beck Depression Inventory
scores (Beck 1961). The average Beck Depression Inventory
score was 8.6 ± 7.7 for the XXY males, 2.4 ± 2.3 for the male
controls, and 4.7 ± 2.9 for the female controls. Although
patients scored significantly higher than controls (df = 2,
F = 9.345, P = 0.00025), their values were within the normal
range. Thirty patients had been diagnosed in early adolescence and 3 received the diagnosis as adults during the
course of infertility investigations. All but 2 of the patients
(who were deemed to not need testosterone) had received testosterone supplementation when diagnosed and were being
treated at the time of the study. About half of the patient population was receiving daily gels, the other half received injections. None of the patients received medication immediately
before drawing blood samples. The karyotyping test for Klinefelter’s syndrome was performed by assessing the metaphase
chromosomes in cells derived from whole blood samples according to the standard procedure.
All the participants were right-handed [Edinburgh Handedness Inventory (Oldfield 1971)]. All were heterosexual (scored
Kinsey 0) according to Kinsey’s Heterosexual/Homosexual
Table 1
Demographic data
Age
Educationc,d
Right D2:D4
Left D2:D4
Oestradiol (plasma)e,f
Testosterone
(bioactive)c,d,e
Total GMb,e,f
Total WMd,e,f
Total CSFg,f
Tissue volumee,f,h
Total intracranial
volumeb,c,e
Finger Ratios
The 2D:4D ratio of both hands was measured using steel
vernier calipers. An investigator who was blind to the diagnosis measured the fingers directly, on the ventral side of the
hand, between the basal crease and the fingertip (Manning
et al. 1998, 2004; Lutchmaya et al. 2004; Manno 2008; Ciumas
et al. 2009). For 15 subjects, the 2D:4D ratios were also
measured by a second rater, and the inter-rater correlation
was calculated with linear regression (Pearson’s coefficient, P
< 0.05). Nonparametric statistics was employed for group
comparisons because the Shapiro–Wilk test showed that the
normality of data distribution could not be assumed among
the XXY patients.
Venous Blood Samples
Venous blood samples were collected from the controls as
well as from the patients, in the morning between 8 and 10
am. Plasma levels of testosterone (nmol/L), 17β-estradiol
( pmol/L) (radioimmunoassay, Testosterone RIA DSL-4000, Diagnostic Systems Laboratory Inc.), and the sex hormonebinding globulin were analyzed in the Chemical Laboratory
Diagnostics at the Karolinska University Hospital. The levels
of bioavailable testosterone (nmol/L) were calculated using an
equation developed by Sodergard et al. (1982). To avoid bias
due to the menstrual cycle phase, women were measured
during the mid-follicular phase, when levels of 17β-estradiol
are relatively stable and low.
MRI
Unit
a,b
Scale, (Kinsey et al. 2003), used as described previously
(Berglund et al. 2006; Savic and Lindstrom 2008). The major
demographic data are presented in Table 1.
All participants were Caucasian. All provided written informed consent and received reimbursement after participation. The study was approved by the Regional Ethical
Review Board and the Radiation Safety Committee at the
Karolinska Institute.
Year
Year
Ratio
Ratio
pmol/L
nmol/L
cm3
cm3
cm3
cm3
cm3
Female controls
N = 45
Male controls
N = 41
XXY Patients
N = 33
Mean
Mean
Mean
SD
35.1
7.3
15.9
2.2
0.99
0.03
0.99
0.03
164.1 107.5
0.46
0.28
679.1
440.4
312.9
1119.5
1432.3
62.3
44.2
97.6
98.9
136.9
35.0
16.2
0.97
0.97
75.8
6.10
768.5
517.6
376.7
1286.0
1662.7
SD
6.8
2.4
0.02
0.03
28.6
1.47
68.9
51.2
67.7
111.2
135.3
39.1
13.2
0.99
0.99
98.8
11.16
724.6
473.6
372.9
1183.8
1556.8
F
SD
10.5
4.08
2.7 14.08
0.04
0.04
47.3 10.10
6.48 43.12
65.6
44.1
76.2
112.6
139.1
19.97
29.13
7.97
27.75
30.69
Note: F values from group comparisons (one-way ANOVA). Possible differences in digit ratios
were calculated with Kruskal–Wallis test.
a
Difference between XXY and XX, P < 0.05.
b
XXY XY difference P < 0.05.
c
Difference between XXY and XX, P < 0.001.
d
XXY XY difference P < 0.001.
e
Difference between XY and XX, P < 0.001.
f
Difference between XXY and XX, P < 0.01.
g
Difference between XY and XX, P < 0.01.
h
XXY XY difference P < 0.01.
i
Difference between XY and XX, P < 0.05.
2324 The Underpinnings of Sex Dimorphism in Human Brain Anatomy
•
Lentini et al.
MRI Data Acquisition
All MRI data was acquired on a whole-body 1.5-Tesla MRI
medical scanner (General Electric) equipped with an 8-channel
coil. The MRI protocol included the following scans: 1)
3D-weighted T1 spoiled grass (SPGR) images with 1 mm isotropic voxel size according to a previously described protocol
(Ciumas et al. 2010), and 2) 2D T2-weighted fast spin echo
images in the axial plane (effective echo time = 56 ms, repetition time = 2500 ms, field of view = 24 cm, 23 slices of 3 mm
thickness). Only the 3D-weighted T1 SPGR images were used
for VBM.
MRI Data Pre-Processing
Voxel-based morphometry (Ashburner 2007) analysis was
performed using the Gaser Toolbox (http://dbm.neuro.
uni-jena.de/vbm/) with SPM5 (The Wellcome Department of
Imaging Neuroscience, University College London; www.fil.
ion.ucl.ac.uk/spm/) and Matlab 7.3 (Math Works). The VBM
preprocessing included 5 steps: 1) visual inspection for scanner
artifacts and gross anatomical abnormalities for each subject; 2)
setting the image origin at the anterior commissure-posterior
commissure line; 3) using the hidden Markov random field
option in the segmentation of the VBM5 toolbox to minimize
the noise level of the segmentation; and 4) using the Diffeomorphic Anatomical Registration Through Lie Algebra toolbox
(DARTEL, Wellcome Department of Imaging Neuroscience,
University College London; http://www.fil.ion.ucl.ac.uk/spm)
for a high-dimensional normalization protocol. John Ashburner’s chapter in its standard version was followed, including
the Montreal Neurological Institute (MNI) space transformation
(Ashburner 2007). 5) To restore the original volume information within each voxel, voxel values in the segmented
images were modulated (multiplied) by the Jacobian determinants derived from the spatial normalization step. The analyses
of modulated data allowed direct comparisons of regional
differences in the amount of each tissue type. After preprocessing, the segmented images were checked for homogeneity across samples, smoothed with an 8-mm full-width at halfmaximum Gaussian filter, modulated, and normalized, before
being used for the statistical analyses.
Statistical Analyses of Group Differences
Overall Tissue Volumes and Hormone Levels
First, hormone levels and total volumes of GM, WM, and cerebrospinal fluid (CSF) were tested for normal distribution.
Group differences in TV (total volume of GM + WM), total intracranial volume (TIV, calculated as GM + WM + CSF), the
relative volume of GM (total GM volume/TIV), WM (total WM
volume/TIV), and CSF (total CSF volume/TIV) were calculated
using separate one-way analysis of covariances (ANOVAs)
with group as the between factor, followed by Tukey’s post
hoc tests (P < 0.05). The same approach was applied to test
the possible group differences in plasma testosterone and estradiol, age, and education.
VBM Data
The VBM data were analyzed using the VBM toolbox of SPM
5 (SPM 5, Wellcome Foundation).
Group Comparisons. First, group comparisons of regional
GM and WM were carried out to identify the possible sex
differences among the controls, and to investigate how XXY
males differed from each control group. This was done using
a full factorial design, with age and total tissue volume (TV)
as the nuisance variables to adjust for individual differences
in these factors. Significant clusters for this first-level analysis
were assessed at P = 0.001 for the peak voxel value, taking
into consideration false discovery rate (FDR) correction for
multiple comparisons at P < 0.05 from the SPM output tables
(Genovese et al. 2002), and employing cluster extent correction
at P < 0.05. In addition, a correction for nonstationary
smoothness (Hayasaka et al. 2004) was applied using VBM5
toolbox.
Next, we conducted conjunctional analyses to evaluate
whether, and in which regions, XXY males exhibited
“female,” or “male,” features. This analysis (Friston et al.
1999) assumed that both XXY males and XX females differed
in certain aspects from XY males and that these differences
could be overlapping, thus showing common clusters for XX–
XY and XXY–XY contrasts, and vice versa. “Female” features
in XXY males were expected primarily because of their supernumerary X-chromosome and, to some extent, their hypogonadism. It was further assumed that XXY males, because they
have 3 rather than 2 sex chromosomes, would differ from
both control groups at least in some regions. Possible conjunctional clusters were calculated for (A) XX–XY and XXY–
XY, (B) XY–XX and XY–XXY, (C) XX–XXY and XY–XXY, and
(D) XXY–XX and XXY–XY. The conjunctional analyses were
carried out separately for GM and WM with age and TV as the
nuisance variables, using peak voxel threshold at P = 0.001. A
correction for nonstationary smoothness was employed. Clusters in A and B were expected primarily in regard to sexually
dimorphic areas, and the cluster level significance was, therefore, P < 0.05 uncorrected. Here we also employed small
volume correction (10-mm sphere) and FDR correction at P <
0.05. For C and D, P < 0.05 corrected was used, as there was
no a priori hypothesis about cluster location.
The peak voxel threshold was at P = 0.001 for all the
comparisons,.
In the following analyses, we specifically investigated possible effects of sex hormones and sex chromosome genes, and
focused on regions, which showed group differences in the
first-level analysis.
Analyses of the Impact of Sex Hormones on Regional GM and
WM Volumes. The next set of analyses was used to evaluate
the impact of sex hormones on the observed sex differences
among the controls. Whether, and in which regions, the GM
and WM volumes in controls correlated with circulating
estrogen and testosterone levels was investigated. Only the
data from controls were included to avoid possible bias due
to the fact that the plasma levels of these hormones in XXY
males was likely to be affected by testosterone supplementation
and were therefore a less reliable indicator of any potential
long-term activational effects of sex hormones on the brain.
The possible relationship between circulating sex hormones and the GM and WM volumes was tested using multivariate regression analyses in which each pixel value of GM
and WM volume was regressed against 17β-estradiol and testosterone, using age and TV as the nuisance variables, and the
entire brain as the search space (multiple regression analysis,
SPM5). Based on a number of previous reports (Peper et al.
2008, 2009; Neufang et al. 2009; Paus et al. 2010; Witte et al.
2010; Mueller et al. 2011), the assumption was made that circulating sex hormone levels would be significantly correlated
with the GM and WM volumes in regions that corresponded
to the sex dimorphic clusters detected in XX–XY and XY–XX
contrasts. To avoid bias due to the fact that levels of circulating estradiol and testosterone are inherently different in men
and women, and would cluster around 2 ends of the
regression line, the respective values were z-normalized for
each group. The significance level for clusters in regions
showing sex differences in the first-level analysis (the simple
XX vs. XY group comparison, Table 2) was set at P < 0.05 uncorrected ( peak threshold at P = 0.001). For all other regions,
the cluster level was P < 0.05 corrected.
In addition, the possible impact of fetal testosterone on
regional GM and WM volumes was tested. This was done in a
separate analysis, employing right-hand 2D:4D ratios as the
linear regressor of interest, and using age and TV as the nuisance variables. The significance levels were the same as for
the calculations of possible effects of circulating sex hormones. When significant clusters were detected, it was then
tested whether these clusters were still found when the circulating hormone levels were added to the same multiple
Cerebral Cortex October 2013, V 23 N 10 2325
Table 2
Group differences in gray and white matter volumes
Gray matter volumes
Region
Male–female controls
L uncus and amygdala
Cerebellum (the semilunar lobe and vermis)
Lingual gyrus
Superior temporal gyrus, WM
Occipital WM
Female–male controls
Anterior cingulate cortex, subcallosum
R inferior frontal gyrus (+ portion of the orbitofrontal cortex)
Precentral gyrus
XXY–male controls
Frontal pole
Anterior cingulate cortex, subcallosum
R precentral gyrus
Precuneus, (+ precentral gyrus)
Male controls–XXY
L insular cortex and L inferior temporal gyrus
L hippocampus, L amygdala
R hippocampus, R amygdala
R caudate, R putamen,
R insular cortex, R uncus
R inferior temporal gyrus
R superior temporal gyrus
Thalamus, hypothalamus
Cerebellum (semilunar lobe)
XXY–female controls
L postcentral gyrus,
R postcentral gyrus
Cuneus and the lingual gyrus
Lingual gyrus
Precuneus GM and WM
Female controls–XXY
L caudate
L insular cortex and superior and inferior temporal gyrus
L hippocampus, L amygdala
R caudate
R insular cortex and superior temporal gyrus
Thalamus
L cerebellum (semilunar lobe)
L insular WM
R temporal lobe WM
L temporal lobe WM
Z level
White matter volumes
Size, cm3
4.7
6.6
3.8
0.9
4.3
Coordinates
4.0
1.3
3.8
3.5
5.1
6.3
5.0
4.3
1.6
3.1
13.2a
16 44 53
6 3 −27
45 −10 46
−12 −69 58
−58 −7 28
8.0a
−38 −4 −24
−20 −5 −21
29 −2 −6
21 22 −4
54 22 −19
59 −1 −37
31 −17−6
−5 −5 −2
19 −93 −27
−32 −87 −38
22 −63 −19
6
16.0a
0.4
6.4
20.0a
4.4
4.4
4.3
5.4
1.6
5
5.1
4.3
5
16.0a
4.7
4.6
12.0a
4.7
3.7
6.4
2.6
Coordinates
4.9
4.5
5.3
2
45 −44 17
21 −76 33
4.7
4.5
0.9
0.4
−18 −40 60
19 −62 41
5.4
16.0a
−46 −10 −19
5.4
16.0a
37 4 −21
5.1
6.2
57 −14 −2
3.8
2.4
30 −60 −37
5.0
4.9
2.4
0.6
14 −82 −9
8 −62 43
5.0
4.4
4.9
2.0
21.0
6.4
−31 1 32
58 −20 5
−46 −7 −18
−16 21−18
41 38 19
32 −14 51
1–20 67
−39 −16 35
1.6
3.5
1.4
4.2
6.9
4.2
4.2
5.8
4.8
5.8
Size, cm3
−13 −2 −25
55 −67 −18
17 −94 −20
4.8
4.8
4.0
5
Z level
−51−42 58
55 −26 57
1 −89 19
10 −89 −17
12 −80 43
−18 21 −11
−35 14 −6
−20 −13 −17
18 23 −9
50 6 −14
−8 −10 8
−27 −82 −33
Note: Clusters denoting GM and WM volumes calculated with the SPM5 VBM toolbox, using TV and age as covariates of no interest. Significant clusters calculated using voxel threshold at P = 0.001,
FDR corrected at P < 0.05, with cluster significance level at P < 0.05 corrected, minimum cluster size 0.2 cm3. MNI coordinates indicate the peak values and the indicated regions coverage of the
respective cluster. R = right; L = left.
a
Confluent cluster.
regression analyses, in order to determine if an independent
factor had been observed or not.
Analysis of the Dosage Effects of Sex Chromosomes on the
Regional GM and WM Volumes. In order to test for any
possible major effects of sex chromosomes on regional GM
and WM volumes, all 3 groups were included. Two further
explorative multiple regression analyses were used to test
the possible effects of both the number of X- and
Y-chromosomes, which were used as the covariates of interest
(2 separate analyses, voxel threshold at P = 0.001, cluster
extent at P < 0.05 corrected). First, only digit ratios, TV, and
age were included as the nuisance variable. When the sex
hormones showed an effect in the “hormone effect analysis”
(see previous paragraph), which only included controls, the
z-transformed plasma levels of the sex hormones from all 3
populations were then added as the nuisance variables, and
2326 The Underpinnings of Sex Dimorphism in Human Brain Anatomy
•
Lentini et al.
the multiple regression analysis was re-run to control for the
possible effects of sex hormones on the results. The reason
for including sex hormone levels only in the adjunct analysis
was that testosterone and also estrogen levels in XXY males
was likely to be affected by testosterone supplementation
therapy. Adult sex hormone levels could, however,
theoretically, have modulating effects on regional TVs, which
could be identified by adding this factor into the model.
Finally, whether trisomy (the fact that patients had 3 sexchromosomes) explained some of the observed differences
between the XXY patients and both control groups was
tested. Potentially, this analysis could reveal dosage effects
of the X-escapee genes that have homologs on the Ychromosome. Individual, preprocessed WM and GM images
were, therefore, regressed in a separate analysis against the
number of sex chromosomes (2 for XX females and XY males
and 3 for XXY males), using right-hand 2D:4D ratios, TV, and
age as the nuisance variables. Again, the analysis was repeated with the z-normalized sex hormone levels as an
additional nuisance variable to evaluate if this maneuver
modified the results.
Results
Background Data
The demographical data are presented in Table 1. Out of the
initially recruited 124 subjects 5 were excluded due to psychiatric symptoms (see procedures section). No age difference
was found between the male and female controls. The patient
group was significantly older than both control groups, and
there was no age difference between the 2 control groups (P
= 0.019, F = 4.08, df = 2; P = 0.043 for XY vs. XXY, P = 0.035
for the XX vs. XXY, and P = 0.81 for XX vs. XY). Both male
and female controls had significantly higher education than
the patients (P < 0.0001, F = 14.1, df = 2); there was no such
difference between the 2 control groups (XX vs. XY P = 0.91;
XX vs. XXY and XY vs. XXY, P < 0.0001). The 3 study groups
did not differ in regard to handedness or sexual orientation,
and no gross anatomical abnormalities were found, according
to an experienced neuroradiologist.
The TIV and the total TV were normally distributed, as
were the total volumes of GM, WM, and CSF. Group differences were detected in TIV (P < 0.001, F = 30.7, df = 2) as well
as in TV (P < 0.001, F = 27.8, df = 2), They consisted of significantly smaller volumes in females in relation to both XY and
XXY males (P < 0.0001 for TIV for both comparisons and P <
0.001 and P < 0.008, respectively, for the TV comparison)
(Table 1), and significantly smaller TIV and TV in XXY males
compared with XY men (P = 0.017 and P = 0.002, respectively). In addition, there was a significant group difference in
the relative GM volume (GM/TIV) (P = 0.005, F = 5.6, df = 2),
with a greater volume in XX females than in XXY males (P =
0.003), but not XY males (P = 0.15). No difference was detected between XXY and XY males (P = 0.28). The groups differed also in the relative WM volume (WM/TIV) (P = 0.034, F
= 3.47, df = 2), with lower values among XXY males compared
with XY males (P = 0.03), but not XX females (P = 0.11). No
statistically significant difference was found between XX
females and XY males (P = 0.82). Finally, there were group
differences in CSF/TIV (P = 0.0017, F = 6.71, df = 2), with
larger values among XXY males compared with both XX
females (P = 0.002) and XY males (P = 0.02). No significant
group difference was detected in regard to the total GM/WM
ratio (P = 0.06, F = 2.87, df = 2).
Measures of digit ratios, carried out by 2 raters, were highly
correlated (r = 0.92; P < 0.001). The results presented here
were based on measurements from rater 1, as rater 2 performed ratings for only 15 subjects. Even though the mean
values of 2D:4D ratios were higher among XX females and
XXY males compared with XY males (Table 1), a Kruskal–
Wallis test, with the subject group as the independent factor,
showed that there was no significant group difference (P =
0.29). When extending the analysis to all the subjects that
were initially included, a significant group difference was
found for the 2D:4D ratio of the right hand (P = 0.033, χ 2 =
6.8, df = 2) but not the left hand (P = 0.070, χ 2 = 5.2, df = 2).
A post hoc test (Wilcoxon W) showed right-hand digit ratios to
be lower among XY males when compared with XX females
and XXY males (P = 0.019, and P = 0.032, for the respective
comparison), and no difference between XX females and XXY
males (P = 0.680).
VBM Data
Regional Group Differences in GM and WM Volumes
Possible differences between the male and female controls
were first examined to define the sex dimorphic GM and WM
clusters. As previously reported (Savic and Arvis 2011), the
female controls had larger GM volumes than the male controls
bilaterally in the precentral gyrus, the subcallosum, and the
right inferior frontal gyrus (including a portion of the orbitofrontal cortex). Male controls, on the other hand, showed a
larger GM volume in the cerebellum and the left uncus and a
portion of amygdala, and larger occipital GM and WM
volumes (Table 2, Fig. 1). Except for the inferior frontal/orbitofrontal cluster, these clusters also appeared when conjunctional analyses were conducted to detect regional differences,
which 2 groups shared in relation to the third (Table 3). Conjunctional analyses showed that both XX and XXY subjects
had a greater GM volume in the precentral gyrus than XY
Figure 1. Sex dimorphism and its relation to circulating testosterone and X-chromosome. Upper panel: Sexual dimorphism in the GM showing greater (red) and smaller (blue)
volumes in XX compared with XY controls. A significant positive correlation between GM volume and z-normalized active plasma testosterone is indicated (green). Nuisance
variables: age, total TV, right 2D:4D ratio (Clusters calculated using peak threshold at P = 0.001, cluster significance at P < 0.05 uncorrected (for regression with testosterone,
shown in green), P < 0.05 corrected and including FDR correction (see Procedures section) for the clusters illustrating sex differences (shown in red and blue). Lower panel: GM
clusters showing a significant association with the number of X-chromosomes (red, positive association; blue, negative association). Nuisance variables: age, total TV. The
hypothalamic cluster disappeared when also adding z-normalized plasma testosterone as the nuisance variable, while the other clusters remained. Clusters calculated with peak
threshold at P = 0.001, P < 0.05 corrected. All the clusters are superimposed on the GM template from the entire study group. R = right side L = left side.
Cerebral Cortex October 2013, V 23 N 10 2327
Table 3
Conjunctional clusters
Gray matter volumes
Region
XX-XY and XXY-XY
Anterior cingulate cortex, subcallosum
Precentral gyrus (bilateral)
XY-XX and XY-XXY
Cerebellum (vermis)
L Amygdala
The temporo-parietal junction
XXY-XX and XY-XX
Occipital cortex (covering the lingual gyrus)
XX-XXY and XX-XY
L subcallosum + orbitofrontal cortex
XX-XXY and XY-XXY
Bilateral insular cortex, bilateral caudate
L hippocampus, bilateral amygdala
Thalamus
R temporal, insular and frontal WM
L temporal, insular and frontal WM
R fronto-opercular WM
XXY-XX and XXY-XY
L, R parietal lobe (bilateral inferior lobuli and precuneus)
Precuneus
L postcentral gyrus
R postcentral gyrus
Z level
White matter volumes
Size, cm3
Coordinates
3.1
3.8
3.7
1.6
1.5
0.4
−8 40 1
32 −15 51
3 −20 65
5.4
3.0
2.0
0.2
1 −82 −24
−15 −2 −26
5.0
12.0
3 −91 −19
14 −67 −7
3.9
0.9
−16 29 −18
4.9
4.6
4.5
4.3
5.0
7.4
3.5
1.6
−20 20 −11
18 23 −9
−20 −10 −19
−7 −10 0
4.9
4.2
4.1
4.7
Z level
Size, cm3
4.1
0.6
44 −44 17
3.8
0.9
−14 −83 −7
4.4
4.9
3.9
3.4
2.2
2.0
58 − 20 5
−46 −7 −18
29 14 12
Coordinates
−9 −56 53
11 −81 41
−63 −20 46
57 −25 57
16.4
1.6
1.4
4.5
Note: All clusters are calculated using age and TV as nuisance variables. Peak threshold at P = 0.001, cluster level at P < 0.05 uncorrected for XX-XY and XXY-XY, XY-XX, and XY-XXY, and P < 0.05
corrected for the remaining contrasts (see Methods for clarification); italics indicate clusters at P < 0.1 corrected (with peak threshold P = 0.001). MNI coordinates. R = right; L = left.
males, whereas their GM volume in the cerebellum was
smaller. The only difference from XX females that was
common among both male groups was the larger occipital
GM volume ( primarily covering the lingual gyrus), (Tables 2
and 3).
A further finding was that XXY males differed significantly
from both control groups by having a smaller GM volume
bilaterally in the insular cortex, in the superior temporal
gyrus, the caudate and thalamus, in the left hippocampus the
amygdala, and in the cerebellum (the semilunar lobes) in
comparison to XY males as well as XX females. Furthermore,
their GM volume was larger in the inferior parietal lobe,
whereas the WM volumes in the temporal lobe and insula
were significantly smaller in comparison to both the male and
female controls (Table 2). Because XXY males had significantly lower education than controls, the comparisons with
controls were re-run adding education as the nuisance variable. The results did not change.
Sex Hormone Levels, 2D:4D Ratios, and Regional GM and
WM Volumes
A positive correlation was detected between bioactive plasma
testosterone and the GM volume in the parahippocampus (a
cluster also covering the amygdala), the occipital (lingual
gyrus and cuneus), and the insular cortex on both sides (clusters covering parts of putamen). There was also a tendency
for inverse correlation with testosterone in the parietal GM
(Table 4, Fig. 1). No significant correlation with estradiol
was detected. Neither did we detect any significant correlation with the right-hand 2D:4D ratio among controls
(a cluster showing a positive correlation with the GM
volume in the precentral gyrus appeared when also including the XXY patients).
2328 The Underpinnings of Sex Dimorphism in Human Brain Anatomy
•
Lentini et al.
Table 4
Correlation with sex hormone levels
Region
Gray matter volumes
White matter correlations
Z
Size,
level cm3
Z
Size,
level cm3
Testosterone, positive correlation
R Parahippocampal gyrus
4.1
including part of R amygdala
L Parahippocampal gyrus
3.3
covering part of the amygdala
Occipital cortex (the lingual 3.3
gyrus, and cuneus)
L insular cortex
3.5
R insular cortex
3.9
Testosterone, negative correlation
Parietal cortex (superior
4.1
parietal lobule)
Coordinates
2.4
39 −38 −12
2.2
−36 −36 −12
3.2
−7 −91 10
2.4
2.9
−39 6 3
36 20 5
0.9
39 −56 62 4.2
0.8
Coordinates
−43 −16 −49
Note: Values calculated using age and TV as nuisance variable; Height threshold at P = 0.001,
minimum cluster size 0.2 cm3, uncorrected P < 0.05 for the hypothesis-based regions (see
methods), and P < 0.05 corrected for the other areas. Values in italics are calculated at P =
0.001, corrected P < 0.1, and illustrate trend values. MNI coordinates indicate the peak value,
and the indicated regions coverage of the respective cluster. R = right; L = left. No significant
correlations were detected with estrogen levels.
Sex Chromosomes and the Regional GM and WM Volumes
Association
with
the
Number
of
Xand
Y-Chromosomes. X-chromosome dosage was associated with
a greater GM volume bilaterally in the precentral gyrus
(Table 5), and a smaller GM volume in the cerebellar vermis,
the caudate, and the left inferior temporal gyrus. These
associations remained after adding testosterone as a nuisance
variable. The number of X-chromosomes was also associated
with a smaller GM volume in the hypothalamus, a smaller
fronto temporal WM volume, and a greater parietal GM volume.
These latter clusters, however, disappeared after adding
Table 5
Association with sex chromosomes
Gray matter volumes
Z level
Region
X-chromosome, positive association
Precentral gyrus
Parietal cortex (superior parietal lobule, precuneus)
a
X-chromosome, negative association
Hypothalamusa
Cerebellum (vermis)
R caudate
L caudate
L inferior temporal gyrus
L fronto-temporal WMa
R fronto-temporal WMa
Y chromosome, positive association
Occipital cortex (occipital gyri, cuneus, lingual gyrus)a
Y chromosome, negative association
R insular cortex and angular gyrusa
Subcallosuma
Sex chromosome number, positive association
Parietal cortex (including precuneus)
Sex chromosome number, negative association
R caudate + amygdala + insular cortex
L caudate + amygdala + insular cortex
Hypothalamusa
Cerebellar hemispheresa
R amygdale
R superior temporal gyrus
L temporal WM
R temporal WM (includes uncinate, external capsule)
White matter correlations
Size, cm3
Coordinates
3.9
4.1
4.6
2.6
1.1
4.2
47 −7 58
−56 −15 28
−14 −67 56
−19 −50 55
4
5.6
4.5
4.6
4.6
2.4
6.5
1.7
1.7
1.8
−2 −5 −4
9 −42−41
19 22 −5
−18 21 −7
−57 2 −38
3.8
6.2
14 −67 −7
0 −94 24
4.5
3.8
4.1
1.6
42 −28 2
−15 23 −19
4.6
4.4
6.5
1.7
−10 −39 43
15 −75 31
5.1
4.9
4.1
4.9
9.0
6.8
2.0
12.0
4.4
4.7
4.1
2
Z level
Size, cm3
Coordinates
4.0
4.5
2.4
7
−18 3 −2
19 −27 −10
4.4
4.1
3.3
1.3
−48 7 −21
57 −09 −05
28 1 −10
19 22 −6
−20 21 −7
−5 −4 0
50 −75−34
−55 −63 −49
22 −2 −31
50 6 −12
Note: Calculations showing association with X-chromosome, and with the total number of sex chromosomes, taking all 3 groups in the analysis and using right hands D2:D4 ratio, age and TV as
nuisance variable. Clusters calculated at peak threshold P = 0.001, cluster level P < 0.05 corrected (italics denote P < 0.1 corrected). MNI coordinates indicate the peak level, and the regions describe
the coverage of the respective cluster. R = right; L = left.
a
Cluster that disappears when adding testosterone as nuisance variable.
z-testosterone as the nuisance variable, suggesting an interaction
with testosterone (Table 5, Fig. 1), which is congruent with
the previously reported (Neufang et al. 2009) and presently
detected correlations between testosterone and GM in these
regions (Table 4). Including z-estradiol did not further change
the results.
With respect to the Y-chromosome, a positive association
was detected with the GM volume in the occipital lobe, and a
negative association with the GM volume in the right insular
cortex, and the subcallosum. These clusters did not appear
when adding testosterone as a nuisance variable, suggesting
that at least some of the Y-chromosome associations were due
to testosterone (Table 5). Again, including z-estradiol did not
further change the results.
Association with the Total Number of Sex Chromosomes. Broadly
in accordance with conjunctional analysis, which revealed
several singular features in XXY men, it was found that a
higher number of sex chromosomes was associated with
greater GM volume in large portions of the parietal lobe, and
with smaller GM volume bilaterally in the caudate, amygdala,
and the insular cortex, in the right superior temporal gyrus,
and the cerebellar hemispheres (the cerebellar cluster
disappeared when subsequently adding z-testosterone as
nuisance covariate in the regression analysis). The number
of sex chromosomes was also associated with a smaller
frontotemporal WM volume in both hemispheres (a cluster
covering the uncinate fascicle, external capsule, and the right
internal capsule), (Table 5, Fig. 2).
Discussion
The present comparative study between XXY males and XY
and XX controls explores the mechanisms underlying sexual
dimorphism in regional volumes of cerebral GM and WM. In
contrast to previous studies, which were either purely descriptive or limited to investigations of sex hormone effects, the
present study included investigations of multiple etiological
factors. We hypothesized that cerebral differences between
XXY and XY males would mainly be a downstream effect of
the fact that XXY males have 2 instead of 1 X-chromosome,
and 3 instead of 2 sex chromosomes overall. Cerebral differences between XXY males and XX females, on the other
hand, could be attributed to the males’ Y-chromosome, their
trisomy, and their at least slightly higher testosterone levels,
as well as to the higher estrogen levels in XX females either
separately or in combination. In addition, the differences
between XX females and XY males could be ascribed to the
presence or absence of the Y-chromosome, the difference in
the number of X-chromosomes, as well as to the difference in
testosterone and/or estrogen levels (Supplementary Table).
This provided a multifactorial dataset for exploring several
Cerebral Cortex October 2013, V 23 N 10 2329
Figure 2. Association between GM and WM volume and the number of sex chromosomes. Upper panel: GM regions with positive (red/yellow) and negative (blue/green)
associations with the number of sex chromosomes. Nuisance variables: age, TV, right 2D:4D ratio. The cerebellar cluster disappeared when adding z-normalized plasma
testosterone as the nuisance variable. Lower panel: WM regions showing a negative association with the number of sex chromosomes. Nuisance variables: age, TV, right 2D:4D
ratio. Clusters, calculated with peak threshold at P = 0.001, P < 0.05 corrected, are superimposed on the GM template (upper paner) and WM template (lower panel) of the
entire study group. R = right side L = left side.
possible mechanisms that might be underlying the differences
in regional GM and WM volumes among the 3 groups investigated and, ultimately, the sexual dimorphism in the measured
parameters.
Broadly in line with previous studies (Filipek et al. 1994;
Murphy et al. 1996; Paus et al. 1996; Giedd et al. 1997; Goldstein et al. 2001; Good et al. 2001; Luders et al. 2005; Raz
et al. 2005; Carne et al. 2006; Chen et al. 2007; Hsu et al.
2008), relatively larger GM volumes were detected in XX controls bilaterally in the precentral gyrus, the subcallosum, and
the right inferior frontal gyrus, whereas XY controls had
larger GM volumes in the cerebellum, the left uncus and
amygdala, and larger GM and WM volumes in the occipital
lobe (Table 2). XXY males shared some features with XY
males, some with XX females, and also displayed singular features (Tables 2 and 3).
Overall, the group differences found in regard to WM were
paralleled by differences in GM volumes (Table 2). This is not
surprising, considering that GM is composed predominantly
of neurons, while WM is composed mainly of axons connecting these neuronal bodies.
Multiple regression analyses showed that sex differences in
the precentral gyrus (overlapping with the motor cortex) and
cerebellum (mainly overlapping with the vermis) were primarily linked to the number of X-chromosomes (Table 5), whereas
the larger occipital GM volume in men was linked to testosterone levels (Tables 3 and 4). No specific Y-chromosome effects
other than those related to testosterone were detected. Sex
chromosome trisomy was found to interact with several limbic
structures.
In the ensuing discussion, we will address several methodological issues, then discuss the present finding in relation
to relevant literature, and finally speculate about the possible
implications vis-à-vis mechanisms behind sex differences
in cerebral anatomy and the skewed sex distribution in
psychopathology.
Methodological Consideration
One may argue that our midsize sample could hamper detection
of some group differences. Although this is theoretically possible, it should be stressed that our findings have a precedence
2330 The Underpinnings of Sex Dimorphism in Human Brain Anatomy
•
Lentini et al.
in that differences between XY males and XX females were
detected in regions previously reported to be sexually dimorphic, and also that the observed changes in XXY patients
were consistent with previous reports (Raz et al. 1995;
Murphy et al. 1996; Paus et al. 1996; Giedd et al. 1997, 2006,
2007; Nopoulos et al. 2000; Goldstein et al. 2001; Luders et al.
2005). This reproducibility of data provides reliability to the
present multigroup comparisons.
The use of the 2D:4D ratio deserves comment. Since the
influential article of Manning et al. (1998) on this antrophomorphic measure of fetal testosterone exposure, over 300
related works have appeared in the literature (Voracek and
Loibl 2009). The exact mechanisms by which 2D:4D and prenatal androgen exposure are related are, however, not entirely
clear. Knickmayer et al. recently reported in a longitudinal
study of neonates that the interaction of salivary testosterone
and CAG repeats predicted right-hand digit ratio at 12 months
and left hand digit ratio at 12 months and 24 months but not
at 2 weeks of age. Given these observations the authors
suggested that it may be more appropriate to interpret the
2D:4D ratio in adulthood as an index of early testosterone
exposure rather than prenatal exposure per se (Knickmeyer
et al. 2011). While these data need to be replicated, they
motivate a further evaluation of how exactly this anthropomorphic measure reflects testosterone exposure.
Another potential methodological concern regards differences in brain size. Because of their larger brains, the outer
brain limits in men are farther away from the center of the
coil in the MR scanner, and, therefore, could be located in
less homogenous parts of the field. This, in turn, could result
in lower signal intensity in these locations, leading to regionally decreased GM in male subjects. As a consequence,
gender differences in GM volumes in the most extreme parts
of the brains (frontal pole, pre- and postcentral gyrus, and occipital pole) could potentially be due to this confound. While
our findings of lower GM volumes in the precentral gyrus in
XY males could be due to this, such a confound is incompatible with the enlarged GM volume in the occipital GM, and the
fact that other structures located far from the center of the
coil, such as the frontal pole, did not show lower GM volumes
in XY men. Furthermore, the XXY males had larger whole
brain volumes than the XX women and yet several “female”
features were found. Also of note is that larger regional GM
volumes have been found in females even when comparing
their GM with that of males of matched brain size (Luders
et al. 2009).
The groups differed with respect to age, TIV, TV, and their
relative GM and WM volumes, but not in their GM/WM ratios.
Age and TV were, therefore, used as covariates in all the calculations. As pointed out by Leonard et al. (2008), frontal
lobe WM increases with the overall brain size more steeply
than GM. The higher precentral GM volumes in the XX and
XXY groups could, therefore, be an effect of the smaller brain
size in these subjects (relatively less frontal lobe WM). To test
this, the group comparisons of regional TVs were re-run
using total GM volume as the covariate. The differences presented in Table 2 remained, suggesting that they were independent of the type of correction variable used (TV or total
GM volumes).
The GM volume cluster in the occipital lobe and cerebellum
in the XY–XX contrast (Table 1) could, despite the careful
normalization procedure with the DARTEL algorithm, potentially be confounded by the edge effects due to difference in
head size. To test this, in a post hoc analysis, GM and WM
templates were constructed separately for the male and female
controls and superimposed on each other for a visual comparison. No difference was detected in size or shape between the
templates, confirming that the occipital-cerebellar GM cluster
was not an effect of a bias in cranial normalization.
A further issue requiring discussion is the use of sex
hormone levels measured at one time point as a regressor to
reveal potential sex hormone effects on cerebral dimorphism.
As in previous studies, correlation analyses of GM and WM
volumes were based on blood samples collected on one
occasion (Peper et al. 2008; Neufang et al. 2009; Witte et al.
2010). Hormone levels vary, however, with activity, stress,
and sleeping patterns. Although we tried to standardize these
factors, and can claim that the measures of hormone levels
and cerebral GM and WM volumes were temporally related
(blood samples in the majority of subjects were taken on the
same day as the MRI scans), it should be acknowledged that
an approach utilizing multiple measurements of serum
hormone levels over time might have been be more precise
for trying to link circulating hormones to brain morphology.
Such a bias should, however, cause a type II rather than type
I error, meaning that the detected correlations should be
reliable.
The covariation analyses with z-score values of sex hormones, employed in the present study, provide inferences for
how differences in testosterone (or estrogen) affect brain morphology “within-gender.” Although post hoc analysis using
absolute hormone yielded similar clusters, the inherent colinearity with gender is a clear limitation when trying to test
whether testosterone or estrogen may have important effects
on brain regions “across genders.”
It should be mentioned that use of cluster significance in
regard to VBM, (employed in the present study), has been discussed by some researchers, mainly due to the problem of
nonuniformity (Ashburner and Friston 2000). Nonuniformity
correction was, therefore applied in all the analyses. We also
carried out post hoc analysis with the alternative approach
using a voxel height threshold at P = 0.001 with FDR correction at P < 0.05, and small volume correction with 10-mm
sphere or 5 × 10 × 5-mm box) when a specific hypothesis was
tested. Interestingly, with the exception of cluster sizes, and
the subsignificant clusters in Table 3 that were not detected in
the post hoc analysis, the results did not differ for any of the
calculations, suggesting that the reported findings are robust.
Finally, it should also be noted that since regional GM and
WM volumes are composite measures of different micro units,
such as neuronal bodies, dendrites, synapses, axons, myelin,
glial cells, the present findings of enlargements or reductions
of GM and WM volumes cannot be directly translated into
function.
Possible Effects of Sex Hormones on Tissue Volumes and
Sex Dimorphism
Positive correlations were found between bioactive testosterone and the GM volume in the parahippocampus (including a
portion of amygdala), the putamen, and in the insular and the
occipital cortices, as well as a tendency for negative correlation with the GM volume in the parietal lobe. These data
accord with the notion that the density of steroid receptors is
high in these regions (Simerly et al. 1990; Goldstein et al.
2001; Lepore et al. 2008), and with the observation that boys
with familial male precocious puberty and an early excess of
androgen secretion have higher GM volumes in the putamen
and the parahippocampal gyrus (Mueller et al. 2011). They
are also in accordance with Neufang et al.’s (2009) findings of
testosterone having an inverse correlation with parietal lobe
GM volume and a positive correlation with GM volume in the
amygdala. The present study, however, was not able to reproduce findings from investigations of pubertal cohorts which
showed that increasing levels of testosterone in boys contributed to the emerging sex differences in the hippocampus
during adolescence (Neufang et al. 2009), and that increasing
levels of 17β-estradiol in pubertal girls correlated with GM increases in the middle frontal-, inferior temporal-, and middle
occipital gyri (Peper et al. 2009). We also did not find positive
estrogen and negative testosterone correlations in the inferior
frontal gyrus, as previously reported in adult persons (Witte
et al. 2010). This inconsistency might be explained by the age
of the participants investigated. Although there are some
reports that activation effects of sex hormones on brain tissue
may continue into adulthood (Pol et al. 2006), their major
impact is expected earlier. Longitudinal MRI studies show
that hormonal changes during puberty are associated with
dynamic patterns of sex dimorphism in the brain, with enhancement of this dimorphism in some regions and deflation
and even reversal in others (Giedd et al. 1997). Thus, investigations focused on the developmental trajectory may differ
from those carried out in adults in terms of the results they
reach on the hormonal relationships. Another potential explanation relates to the use of correlation analysis. Linear correlation in selected cerebral regions of interest (Witte et al.
2010) may differ from explorative analyses used in the
present study. Witte et al.’s study also showed a slightly different pattern regarding sex differences than some of the other
studies (for example, greater GM volume in men in the cingulate cortex, the precentral gyrus, and the right precuneus)
(Good et al. 2001; Chen et al. 2007; Paus et al. 2010). The presently observed sex differences broadly agree with findings in
the current literature, and the GM volume correlations with
testosterone were mainly detected in the regions showing sex
differences (Tables 2 and 5). Thus, assuming that the
Cerebral Cortex October 2013, V 23 N 10 2331
individual trajectory of sex hormone levels over a lifespan is
stable (the relation of one individual’s hormone level in
relation to another’s is relatively unchanged), as suggested by
Forest et al. (1976), the presently found testosterone correlations should be relevant to sex differences.
When it comes to the 2D:4D ratio, our findings showed
that the right-hand ratios of the XXY males tended to be
higher compared with XY males (although not statistically significant), but were similar to those of the XX females
(Table 1). The right-hand 2D:4D ratio also correlated with the
precentral GM volume (when including XXY males in the
analysis), which could suggest that low action of fetal testosterone is linked to a greater precentral GM volume. The correlation with the precentral GM was, however, detected also
when employing the X-chromosome as the covariate of interest, and persisted when adding the 2D:4D ratio as the
nuisance variable. This implies that the effect of the Xchromosome dominated and that the GM correlation with the
2D:4D ratio could be a downstream effect of the 2 Xchromosomes in XXY males. Such a scenario is congruent
with the fact that the androgen receptor, which mediates the
effects of fetal testosterone, is coded by X-chromosome
genes, and that the receptor-subtype with low affinity to androgens is more common among XXY men (Wikstrom et al.
2006; Knickmeyer et al. 2011). It also fits with the fact that
degeneration of the testosterone producing seminiferous
tubules in XXY males begins at the fetal stage (Aksglaede
et al. 2006), although their fetal testosterone is reported to be
within the lower normal range.
Possible Effects of Sex Chromosomes on Tissue Volumes
and Sex Differences
The results of the regression analyses involving sex chromosomes will be discussed in terms of X-chromosome and sex
chromosome dosage, not withstanding that this is a simplification. We prefer, however, to use the term chromosome
dosage rather than gene dosage because this is what has
been investigated, and to avoid speculations about gene
expression, gene compensation, and gene transcription.
Although the long chain of downstream mediators between
the number of X and sex chromosomes and brain TVs evidently precludes any detailed elaboration of the mechanisms
involved, the present observations nonetheless show a
regional genetic influence on some sex differences in the
human brain.
X-Chromosome Dosage
As in the majority of previous studies, XXY patients showed
GM and WM reductions in a number of cortical and subcortical regions (Patwardhan et al. 2000; Rose et al. 2004; DeLisi
et al. 2005; Giedd et al. 2007; van Rijn et al. 2008; Lenroot
et al. 2009; Rezaie et al. 2009; Steinman et al. 2009), as well as
in the TIV (Tables 1 and 2). They also exhibited an increase in
parietal GM volume, as has also been found in prepubertal
Klinefelter boys by Bryant et al. (2011), whereas Giedd reported increased parietal WM volume, instead (Giedd et al.
2007). This discrepancy may be related to the type of normalization (as in the present study, Bryant et al. used the total TV
as the covariate).
The present study is, to the best of our knowledge, the first
VBM study of XXY males to include female controls as well as
measurements of sex hormones and digit ratios. The observed
2332 The Underpinnings of Sex Dimorphism in Human Brain Anatomy
•
Lentini et al.
similarities between XXY males and XX females, and the
differences that these groups have in common vis à vis XY
males, which persisted after regressing out the hormonal
interaction, suggest that the X-chromosome dosage has an
influence on the GM volumes in the precentral gyrus and the
cerebellum ( primarily the vermis). This is of interest because
of the previously reported sex differences in these regions
(Rhyu et al. 1999; Raz et al. 2001; Luders et al. 2005, 2009;
Savic and Arvis 2011), and the reports that women with X0
monosomy, have smaller precentral GM and larger cerebellar
GM volume than XX women (Brown et al. 2002; Molko et al.
2003; Cutter et al. 2006; Marzelli et al. 2011). This similarity
with XY men and reciprocity to our findings in XXY men
argues for processes related to the dosage of X-chromosome
genes, which do not have Y-chromosome homologs. The
possible effects of X-chromosome dosage on cerebral tissue
may be both direct and indirect, but this does not disqualify
the argument, as such effects, nevertheless, are mediated by
X-chromosome genes. As to the cerebellum, it is also worth
mentioning that regional cerebellar volumes begin to show
sex differences early in childhood, although the developmental trajectories differ between boys and girls (Tiemeier et al.
2010), and that gene expression in the cerebellum has been
reported to be sex differentiated (Vawter et al. 2004). The cerebellum is, however, a large structure and further studies
are needed to understand the programming of its specific
subregions.
Sex Chromosome Dosage
Explorative regression analyses showed that the GM volumes
in the amygdala, the insular cortex, temporal lobe cortex,
caudate, and parietal cortex were linked to the number of sex
chromosomes. The GM volume clusters of the inferior temporal gyrus and caudate were detected also when the number
of X-chromosomes was used as a regressor. They were,
however, primarily constituted by the GM volume reductions
in XXY males, which were found in relation to both control
groups, and thus seem to be related to sex chromosome
trisomy rather than to the X-chromosome dosage.
The presently detected trisomy effects will be discussed in
the context of previous findings in subjects with sex chromosome aneuploidy. Women with X0 monosomy have been
found to exhibit hypertrophy of the structural and GM
volume of the amygdala, as well as in the GM volume in the
superior and, in some studies, the inferior temporal gyrus
(Good et al. 2003; Molko et al. 2003; Kesler et al. 2004; Marzelli et al. 2011). Since the amygdala volume has also been
found to be larger in XY males than in XX females, it has
been suggested to be inversely correlated to X-chromosome
gene dosage, which in turn could have a major influence on
the sexually dimorphic development of this structure (Good
et al. 2003). However, considering that amygdala volume has
been found to be only slightly reduced in XXX females but
markedly reduced in XXY males (Patwardhan et al. 2002),
and that circulating testosterone has been correlated with increasing amygdala GM volume (see Table 4) (Neufang et al.
2009), it might be more plausible that amygdala volume is influenced by “sex chromosome” gene dosage (rather than
“X-chromosome” gene dosage) and testosterone. This also
may apply to the association between trisomy and atrophy of
the superior temporal gyrus, as the posterior portion of this
region mediates language functions, and it has been reported
that language dysfunction in XXY males correlated with an increased expression of a PAR gene [GTPBP6 (Vawter et al.
2007)]. A similar scenario may also account for the present
observations regarding the parietal lobe. In addition to the
finding of trisomy-related GM hypertrophy, which accords
with previously reported parietal atrophy in X0 females
(Brown et al. 2002), we found an inverse correlation between
plasma testosterone and parietal GM volume, suggesting that
impaired testosterone-mediated pruning (Chen et al. 2007;
Koscik et al. 2009) could have added to the increase of parietal GM volume in XXY males, when compared with XY males.
The detected negative association between the number of
sex chromosomes and the GM volume in the caudate (Molko
et al. 2003) is more difficult to explain. The caudate nucleus is
usually reported to be larger in females (Giedd et al. 1997)
(Sowell et al. 2002; Bramen et al. 2011), occurring before they
reach puberty, and is not particularly affected by hormonal
perturbations, which argues for genetic effects. In the
majority of MR studies, however, caudate volume has been
found to be smaller in X0 females just as in XXY males
(Cutter et al. 2006; Marzelli et al. 2011), although there are
some reports of increases (Molko et al. 2003, 2004). It is difficult to attribute these results to the detected sex chromosome
dosage effect, and it would require parallel studies of XX and
XY controls together with both X0 and XXY subjects in order
to gain more insight.
In summary, it is proposed that there are X-dosage effects
in the precentral gyrus and parts of cerebellum. These are of
interest since many forms of mental retardation are linked to
the X-chromosome, as are several hereditary disorders of the
motor system (Fragile X, familiar X-linked dystonias, X-linked
cerebellar hypoplasia, adrenoleukodystrophy, and spinocerebellar degeneration with X-linked inheritance). The study also
indicates that processes related to testosterone and sex
chromosome genes have a combined effect on GM development in the superior temporal gyrus, the amygdala, the
insular cortex, and the parietal cortex. The observed impact
of sex chromosomes is in line with reports of sexdifferentiated gene expression in the brain (Dewing et al.
2003; Vawter et al. 2007), as well as with the notion that cerebral sex differences appear prior to the production of sex
hormones. We hypothesize that the presently detected effects
of trisomy (expressed in differences in XXY men in relation to
both male and female controls) are primarily due to
X-chromosome escapee genes with Y chromosome homologs,
regardless of the fact that homologous genes may be differentially regulated in males and females (Xu et al. 2002). By
identifying brain areas that exhibit the effects of sex chromosomes, the present results add to the animal data about genes
located on X and Y chromosomes that could contribute to sex
differences in GM and WM volumes. This is of interest since
many common psychiatric conditions, such as autism, ADHD,
schizophrenia, and depression, are characterized by sex
differences in regard to their rates, age at onset, course, and
symptoms. Moreover, these conditions often involve abnormalities in the caudate and amygdala (Grossman et al. 2008;
Waddell and McCarthy 2012) and tend to be more prevalent
among those with various forms of sex-chromosome aneuploidy (DeLisi et al. 2005; Marco and Skuse 2006). Studies of
sex chromosome aneuploidies thereby provide a unique tool
to help understand some of the mechanisms behind the early
development of sex differences and their clinical implications.
Supplementary Material
Supplementary material can be found at: http://www.cercor.oxfordjournals.org/.
Notes
The authors thank the Insurance Council for Working Life and Social
Science (FAS), the Swedish Research Council, and VINNOVA for their
financial support. The authors are very grateful to Professor Hans
Forsberg for the fruitful discussions. Dr Savic had full access to all of
the data in the study and takes responsibility for the integrity of the
data and the accuracy of the data analysis. Conflict of Interest: None
declared.
References
Aksglaede L, Wikstrom AM, Rajpert-De Meyts E, Dunkel L, Skakkebaek
NE, Juul A. 2006. Natural history of seminiferous tubule degeneration in Klinefelter syndrome. Hum Reprod Update. 12:39–48.
American Psychiatric Association 2000. Diagnostic and statistical
manual of mental disorders. Washington (DC): American Psychiatric Association.
Andreasen NC. 2005. Vulnerability to mental illnesses: gender makes
a difference, and so does providing good psychiatric care. Am J
Psychiatry. 162:211–213.
Arnold AP, Chen X. 2009. What does the “four core genotypes”
mouse model tell us about sex differences in the brain and other
tissues? Front Neuroendocrinol. 30:1–9.
Arnold AP, Xu J, Grisham W, Chen X, Kim YH, Itoh Y. 2004. Minireview: Sex chromosomes and brain sexual differentiation. Endocrinology. 145:1057–1062.
Ashburner J. 2007. A fast diffeomorphic image registration algorithm.
NeuroImage. 38:95–113.
Ashburner J, Friston KJ. 2000. Voxel-based morphometry—the
methods. NeuroImage. 11:805–821.
Balint S, Czobor P, Komlosi S, Meszaros A, Simon V, Bitter I. 2009.
Attention deficit hyperactivity disorder (ADHD): gender- and
age-related differences in neurocognition. Psychol Med.
39:1337–1345.
Bao AM, Swaab DF. 2010. Sex differences in the brain, behavior, and
neuropsychiatric disorders. Neuroscientist. 16:550–565.
Baron-Cohen S, Ring HA, Bullmore ET, Wheelwright S, Ashwin C,
Williams SC. 2000. The amygdala theory of autism. Neurosci Biobehav Rev. 24:355–364.
Beck AT. 1961. A systematic investigation of depression. Compr Psychiatry. 2:163–170.
Berglund H, Lindstrom P, Savic I. 2006. Brain response to putative
pheromones in lesbian women. Proc Natl Acad Sci USA.
103:8269–8274.
Bramen JE, Hranilovich JA, Dahl RE, Forbes EE, Chen J, Toga AW,
Dinov ID, Worthman CM, Sowell ER. 2011. Puberty influences
medial temporal lobe and cortical gray matter maturation differently in boys than girls matched for sexual maturity. Cereb Cortex.
21:636–646.
Brown WE, Kesler SR, Eliez S, Warsofsky IS, Haberecht M, Patwardhan A, Ross JL, Neely EK, Zeng SM, Yankowitz J et al. 2002. Brain
development in Turner syndrome: a magnetic resonance imaging
study. Psychiatry Res. 116:187–196.
Bryant DM, Hoeft F, Lai S, Lackey J, Roeltgen D, Ross J, Reiss AL.
2011. Neuroanatomical phenotype of Klinefelter syndrome in childhood: a voxel-based morphometry study. J Neurosci. 31:6654–6660.
Carne RP, Vogrin S, Litewka L, Cook MJ. 2006. Cerebral cortex: an
MRI-based study of volume and variance with age and sex. J Clin
Neurosci. 13:60–72.
Carrel L, Willard HF. 2005. X-inactivation profile reveals extensive
variability in X-linked gene expression in females. Nature.
434:400–404.
Carson DJ, Okuno A, Lee PA, Stetten G, Didolkar SM, Migeon CJ.
1982. Amniotic fluid steroid levels. Fetuses with adrenal
Cerebral Cortex October 2013, V 23 N 10 2333
hyperplasia, 46,XXY fetuses, and normal fetuses. Am J Dis Child.
136:218–222.
Chen X, Sachdev PS, Wen W, Anstey KJ. 2007. Sex differences in
regional gray matter in healthy individuals aged 44–48 years: a
voxel-based morphometric study. NeuroImage. 36:691–699.
Ciumas C, Linden Hirschberg A, Savic I. 2009. High fetal testosterone
and sexually dimorphic cerebral networks in females. Cereb
Cortex. 19:1167–1174.
Ciumas C, Wahlin TB, Espino C, Savic I. 2010. The dopamine system
in idiopathic generalized epilepsies: identification of syndromerelated changes. Neuroimage. 51:606–615.
Coates JM, Gurnell M, Rustichini A. 2009. Second-to-fourth digit ratio
predicts success among high-frequency financial traders. Proc Natl
Acad Sci USA. 106:623–628.
Cutter WJ, Daly EM, Robertson DM, Chitnis XA, van Amelsvoort TA,
Simmons A, Ng VW, Williams BS, Shaw P, Conway GS et al. 2006.
Influence of X chromosome and hormones on human brain development: a magnetic resonance imaging and proton magnetic resonance spectroscopy study of Turner syndrome. Biol Psychiatry.
59:273–283.
DeLisi LE, Maurizio AM, Svetina C, Ardekani B, Szulc K, Nierenberg J,
Leonard J, Harvey PD. 2005. Klinefelter’s syndrome (XXY) as a
genetic model for psychotic disorders. Am J Med Genet B Neuropsychiatr Genet. 135B:15–23.
Dewing P, Shi T, Horvath S, Vilain E. 2003. Sexually dimorphic gene
expression in mouse brain precedes gonadal differentiation. Brain
Res Mol Brain Res. 118:82–90.
Ehrhardt AA, Meyer-Bahlburg HF. 1979. Prenatal sex hormones and
the developing brain: effects on psychosexual differentiation and
cognitive function. Annu Rev Med. 30:417–430.
Filipek PA, Richelme C, Kennedy DN, Caviness VS, Jr. 1994. The
young adult human brain: an MRI-based morphometric analysis.
Cereb Cortex. 4:344–360.
Forest MG, De Peretti E, Bertrand J. 1976. Hypothalamicpituitary-gonadal relationships in man from birth to puberty. Clin
Endocrinol (Oxf). 5:551–569.
Friston KJ, Holmes AP, Price CJ, Buchel C, Worsley KJ. 1999. Multisubject fMRI studies and conjunction analyses. NeuroImage.
10:385–396.
Genovese CR, Lazar NA, Nichols T. 2002. Thresholding of statistical
maps in functional neuroimaging using the false discovery rate.
NeuroImage. 15:870–878.
Giedd JN, Castellanos FX, Rajapakse JC, Vaituzis AC, Rapoport JL.
1997. Sexual dimorphism of the developing human brain. Prog
Neuropsychopharmacol Biol Psychiatry. 21:1185–1201.
Giedd JN, Clasen LS, Lenroot R, Greenstein D, Wallace GL, Ordaz S,
Molloy EA, Blumenthal JD, Tossell JW, Stayer C et al. 2006.
Puberty-related influences on brain development. Mol Cell Endocrinol. 254–255:154–162.
Giedd JN, Clasen LS, Wallace GL, Lenroot RK, Lerch JP, Wells EM,
Blumenthal JD, Nelson JE, Tossell JW, Stayer C et al. 2007.
XXY (Klinefelter syndrome): a pediatric quantitative brain magnetic resonance imaging case-control study. Pediatrics. 119:
e232–e240.
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.
Good CD, Johnsrude I, Ashburner J, Henson RN, Friston KJ, Frackowiak RS. 2001. Cerebral asymmetry and the effects of sex and
handedness on brain structure: a voxel-based morphometric analysis of 465 normal adult human brains. NeuroImage. 14:685–700.
Good CD, Lawrence K, Thomas NS, Price CJ, Ashburner J, Friston KJ,
Frackowiak RS, Oreland L, Skuse DH. 2003. Dosage-sensitive
X-linked locus influences the development of amygdala and orbitofrontal cortex, and fear recognition in humans. Brain.
126:2431–2446.
Grossman LS, Harrow M, Rosen C, Faull R, Strauss GP. 2008. Sex
differences in schizophrenia and other psychotic disorders: a
20-year longitudinal study of psychosis and recovery. Compr Psychiatry. 49:523–529.
2334 The Underpinnings of Sex Dimorphism in Human Brain Anatomy
•
Lentini et al.
Hayasaka S, Phan KL, Liberzon I, Worsley KJ, Nichols TE. 2004. Nonstationary cluster-size inference with random field and permutation methods. 22:676–687.
Honekopp J, Watson S. 2010. Meta-analysis of digit ratio 2D:4D
shows greater sex difference in the right hand. Am J Hum Biol.
22:619–630.
Hsu JL, Leemans A, Bai CH, Lee CH, Tsai YF, Chiu HC, Chen WH.
2008. Gender differences and age-related white matter changes of
the human brain: a diffusion tensor imaging study. NeuroImage.
39:566–577.
Jazin E, Cahill L. 2010. Sex differences in molecular neuroscience:
from fruit flies to humans. Nat Rev. 11:9–17.
Kesler SR, Garrett A, Bender B, Yankowitz J, Zeng SM, Reiss AL.
2004. Amygdala and hippocampal volumes in Turner syndrome: a
high-resolution MRI study of X-monosomy. Neuropsychologia.
42:1971–1978.
Kinsey AC, Pomeroy WR, Martin CE. 2003. Sexual behavior in the
human male. 1948. Am J Public Health. 93:894–898.
Knickmeyer RC, Woolson S, Hamer RM, Konneker T, Gilmore JH.
2011. 2D:4D ratios in the first 2 years of life: Stability and relation
to testosterone exposure and sensitivity. Horm Behav. 60(3):
256–263.
Koscik T, O’Leary D, Moser DJ, Andreasen NC, Nopoulos P. 2009. Sex
differences in parietal lobe morphology: relationship to mental
rotation performance. Brain Cogn. 69:451–459.
Lai DC, Tseng YC, Hou YM, Guo HR. 2012. Gender and geographic
differences in the prevalence of autism spectrum disorders in children: analysis of data from the national disability registry of
Taiwan. Res Dev Disabil. 33:909–915.
Lenroot RK, Lee NR, Giedd JN. 2009. Effects of sex chromosome aneuploidies on brain development: evidence from neuroimaging
studies. Dev Disabil Res Rev. 15:318–327.
Leonard CM, Towler S, Welcome S, Halderman LK, Otto R, Eckert
MA, Chiarello C. 2008. Size matters: cerebral volume influences
sex differences in neuroanatomy. Cereb Cortex. 18:2920–2931.
Lepore G, Gadau S, Mura A, Arru B, Mulliri G, Zedda M, Farina V.
2008. Androgen receptor immunoreactivity in rat occipital cortex
after callosotomy. Eur J Histochem. 52:163–168.
Luders E, Gaser C, Narr KL, Toga AW. 2009. Why sex matters: brain
size independent differences in gray matter distributions between
men and women. J Neurosci. 29:14265–14270.
Luders E, Narr KL, Thompson PM, Woods RP, Rex DE, Jancke L, Steinmetz H, Toga AW. 2005. Mapping cortical gray matter in the
young adult brain: effects of gender. NeuroImage. 26:493–501.
Luders E, Sanchez FJ, Gaser C, Toga AW, Narr KL, Hamilton LS, Vilain
E. 2009. Regional gray matter variation in male-to-female transsexualism. NeuroImage. 46:904–907.
Lutchmaya S, Baron-Cohen S, Raggatt P, Knickmeyer R, Manning JT.
2004. 2nd to 4th digit ratios, fetal testosterone and estradiol. Early
Hum Dev. 77:23–28.
Manning JT, Scutt D, Wilson J, Lewis-Jones DI. 1998. The ratio of
2nd to 4th digit length: a predictor of sperm numbers and concentrations of testosterone, luteinizing hormone and oestrogen.
Hum Reprod. 13:3000–3004.
Manning JT, Stewart A, Bundred PE, Trivers RL. 2004. Sex and ethnic
differences in 2nd to 4th digit ratio of children. Early Hum Dev.
80:161–168.
Manno FA, 3rd. 2008. Measurement of the digit lengths and the anogenital distance in mice. Physiol Behav. 93:364–368.
Marco EJ, Skuse DH. 2006. Autism-lessons from the X chromosome.
Soc Cogn Affect Neurosci. 1:183–193.
Marzelli MJ, Hoeft F, Hong DS, Reiss AL. 2011. Neuroanatomical
spatial patterns in Turner syndrome. NeuroImage. 55:439–447.
McAlonan GM, Suckling J, Wong N, Cheung V, Lienenkaemper N,
Cheung C, Chua SE. 2008. Distinct patterns of grey matter abnormality in high-functioning autism and Asperger’s syndrome. J Child
Psychol Psychiatry. 49:1287–1295.
Merke DP, Fields JD, Keil MF, Vaituzis AC, Chrousos GP, Giedd JN.
2003. Children with classic congenital adrenal hyperplasia have
decreased amygdala volume: potential prenatal and postnatal hormonal effects. J Clin Endocrinol Metab. 88:1760–1765.
Molko N, Cachia A, Riviere D, Mangin JF, Bruandet M, Le Bihan D,
Cohen L, Dehaene S. 2004. Brain anatomy in Turner syndrome:
evidence for impaired social and spatial-numerical networks.
Cereb Cortex. 14:840–850.
Molko N, Cachia A, Riviere D, Mangin JF, Bruandet M, Le Bihan D,
Cohen L, Dehaene S. 2003. Functional and structural alterations of
the intraparietal sulcus in a developmental dyscalculia of genetic
origin. Neuron. 40:847–858.
Mueller SC, Merke DP, Leschek EW, Fromm S, Vanryzin C, Ernst M.
2011. Increased medial temporal lobe and striatal grey-matter
volume in a rare disorder of androgen excess: a voxel-based morphometry (VBM) study. Int J Neuropsychopharmacol. 14(4):
445–457.
Murphy DG, DeCarli C, McIntosh AR, Daly E, Mentis MJ, Pietrini P,
Szczepanik J, Schapiro MB, Grady CL, Horwitz B et al. 1996. Sex
differences in human brain morphometry and metabolism: an in
vivo quantitative magnetic resonance imaging and positron emission tomography study on the effect of aging. Arch Gen Psychiatry. 53:585–594.
Neufang S, Specht K, Hausmann M, Gunturkun O, HerpertzDahlmann B, Fink GR, Konrad K. 2009. Sex differences and the
impact of steroid hormones on the developing human brain.
Cereb Cortex. 19:464–473.
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. 98:1–13.
Oldfield RC. 1971. The assessment and analysis of handedness: the
Edinburgh inventory. Neuropsychologia. 9:97–113.
Patwardhan AJ, Brown WE, Bender BG, Linden MG, Eliez S, Reiss AL.
2002. Reduced size of the amygdala in individuals with 47,XXY
and 47,XXX karyotypes. Am J Med Genet. 114:93–98.
Patwardhan AJ, Eliez S, Bender B, Linden MG, Reiss AL. 2000. Brain
morphology in Klinefelter syndrome: extra X chromosome and
testosterone supplementation. Neurology. 54:2218–2223.
Paus T, Nawaz-Khan I, Leonard G, Perron M, Pike GB, Pitiot A, Richer L,
Susman E, Veillette S, Pausova Z. 2010. Sexual dimorphism in the
adolescent brain: Role of testosterone and androgen receptor in
global and local volumes of grey and white matter. Horm Behav.
57(1):63–75.
Paus T, Tomaiuolo F, Otaky N, MacDonald D, Petrides M, Atlas J,
Morris R, Evans AC. 1996. Human cingulate and paracingulate
sulci: pattern, variability, asymmetry, and probabilistic map. Cereb
Cortex. 6:207–214.
Peper JS, Brouwer RM, Schnack HG, van Baal GC, van Leeuwen M,
van den Berg SM, Delemarre-Van de Waal HA, Boomsma DI,
Kahn RS, Hulshoff Pol HE. 2009. Sex steroids and brain structure in
pubertal boys and girls. Psychoneuroendocrinology. 34:332–342.
Peper JS, Brouwer RM, Schnack HG, van Baal GC, van Leeuwen M,
van den Berg SM, Delemarre-Van de Waal HA, Janke AL, Collins
DL, Evans AC et al. 2008. Cerebral white matter in early puberty is
associated with luteinizing hormone concentrations. Psychoneuroendocrinology. 33:909–915.
Phoenix CH, Goy RW, Gerall AA, Young WC. 1959. Organizing action
of prenatally administered testosterone propionate on the tissues
mediating mating behavior in the female guinea pig. Endocrinology. 65:369–382.
Pol HEH, Cohen-Kettenis PT, Van Haren NEM, Peper JS, Brans RGH,
Cahn W, Schnack HG, Gooren LJG, Kahn RS. 2006. Changing
your sex changes your brain: influences of testosterone and estrogen
on adult human brain structure. Eur J Endocrinol. 155:S107–S114.
Ratcliffe SG, Read G, Pan H, Fear C, Lindenbaum R, Crossley J. 1994.
Prenatal testosterone levels in XXY and XYY males. Horm Res.
42:106–109.
Raz N, Gunning-Dixon F, Head D, Williamson A, Acker JD. 2001. Age
and sex differences in the cerebellum and the ventral pons: a prospective MR study of healthy adults. AJNR. 22:1161–1167.
Raz N, Lindenberger U, Rodrigue KM, Kennedy KM, Head D, Williamson A, Dahle C, Gerstorf D, Acker JD. 2005. Regional brain
changes in aging healthy adults: general trends, individual differences and modifiers. Cereb Cortex. 15:1676–1689.
Raz N, Torres IJ, Acker JD. 1995. Age, gender, and hemispheric
differences in human striatum: a quantitative review and new
data from in vivo MRI morphometry. Neurobiol Learn Mem. 63:
133–142.
Rezaie R, Daly EM, Cutter WJ, Murphy DG, Robertson DM, DeLisi LE,
Mackay CE, Barrick TR, Crow TJ, Roberts N. 2009. The influence
of sex chromosome aneuploidy on brain asymmetry. Am J Med
Genet B Neuropsychiatr Genet. 150B:74–85.
Rhyu IJ, Cho TH, Lee NJ, Uhm CS, Kim H, Suh YS. 1999. Magnetic
resonance image-based cerebellar volumetry in healthy Korean
adults. Neurosci Lett. 270:149–152.
Rose AB, Merke DP, Clasen LS, Rosenthal MA, Wallace GL, Vaituzis
AC, Fields JD, Giedd JN. 2004. Effects of hormones and sex
chromosomes on stress-influenced regions of the developing pediatric brain. Ann N Y Acad Sci. 1032:231–233.
Savic I, Arvis S. 2011. Sex dimorphism of the brain in male-to-female
transsexuals. Cereb Cortex. 21(11):2525–2533.
Savic I, Lindstrom P. 2008. PET and MRI show differences in cerebral
asymmetry and functional connectivity between homo- and heterosexual subjects. Proc Natl Acad Sci USA. 105:9403–9408.
Schlaepfer TE, Harris GJ, Tien AY, Peng L, Lee S, Pearlson GD. 1995.
Structural differences in the cerebral cortex of healthy female and
male subjects: a magnetic resonance imaging study. Psychiatry
Res. 61:129–135.
Shen D, Liu D, Liu H, Clasen L, Giedd J, Davatzikos C. 2004. Automated morphometric study of brain variation in XXY males. NeuroImage. 23:648–653.
Simerly RB, Chang C, Muramatsu M, Swanson LW. 1990. Distribution
of androgen and estrogen receptor mRNA-containing cells in the
rat brain: an in situ hybridization study. J Comp Neurol.
294:76–95.
Sodergard R, Backstrom T, Shanbhag V, Carstensen H. 1982. Calculation of free and bound fractions of testosterone and estradiol-17
beta to human plasma proteins at body temperature. J Steroid
Biochem. 16:801–810.
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 Neurol. 44:4–16.
Steinman K, Ross J, Lai S, Reiss A, Hoeft F. 2009. Structural and functional neuroimaging in Klinefelter (47,XXY) syndrome: a review of
the literature and preliminary results from a functional magnetic
resonance imaging study of language. Dev Disabil Res Rev.
15:295–308.
Tiemeier H, Lenroot RK, Greenstein DK, Tran L, Pierson R, Giedd JN.
2010. Cerebellum development during childhood and adolescence: a longitudinal morphometric MRI study. NeuroImage.
49:63–70.
Tomasi D, Volkow ND. 2012. Gender differences in brain functional
connectivity density. Hum Brain Mapp. 33:849–860.
van Rijn S, Aleman A, Swaab H, Vink M, Sommer I, Kahn RS. 2008.
Effects of an extra X chromosome on language lateralization: an
fMRI study with Klinefelter men (47,XXY). Schizophr Res.
101:17–25.
Vawter MP, Evans S, Choudary P, Tomita H, Meador-Woodruff J,
Molnar M, Li J, Lopez JF, Myers R, Cox D et al. 2004. Genderspecific gene expression in post-mortem human brain: localization
to sex chromosomes. Neuropsychopharmacology. 29:373–384.
Vawter MP, Harvey PD, DeLisi LE. 2007. Dysregulation of X-linked
gene expression in Klinefelter’s syndrome and association with
verbal cognition. Am J Med Genet B Neuropsychiatr Genet.
144B:728–734.
Voracek M, Loibl LM. 2009. Scientometric analysis and bibliography
of digit ratio (2D:4D) research, 1998–2008. Psychol Rep. 104:
922–956.
Waddell J, McCarthy MM. 2012. Sexual Differentiation of the Brain
and ADHD: What Is a Sex Difference in Prevalence Telling Us?
Curr Top Behav Neurosci. 9:341–360.
Wikstrom AM, Painter JN, Raivio T, Aittomaki K, Dunkel L. 2006.
Genetic features of the X chromosome affect pubertal development and testicular degeneration in adolescent boys with Klinefelter syndrome. Clin Endocrinol (Oxf). 65:92–97.
Cerebral Cortex October 2013, V 23 N 10 2335
Williams TJ, Pepitone ME, Christensen SE, Cooke BM, Huberman AD,
Breedlove NJ, Breedlove TJ, Jordan CL, Breedlove SM. 2000.
Finger-length ratios and sexual orientation. Nature. 404:455–456.
Witelson SF, Glezer II, Kigar DL. 1995. Women have greater density of
neurons in posterior temporal cortex. J Neurosci. 15:3418–3428.
Witte AV, Savli M, Holik A, Kasper S, Lanzenberger R. 2010. Regional
sex differences in grey matter volume are associated with sex
2336 The Underpinnings of Sex Dimorphism in Human Brain Anatomy
•
Lentini et al.
hormones in the young adult human brain. NeuroImage.
49:1205–1212.
Xu J, Burgoyne PS, Arnold AP. 2002. Sex differences in sex chromosome gene expression in mouse brain. Hum Mol Genet.
11:1409–1419.
Xu J, Disteche CM. 2006. Sex differences in brain expression of Xand Y-linked genes. Brain Res. 1126:50–55.