Molecular Psychiatry (2009) 14, 988–991
& 2009 Nature Publishing Group All rights reserved 1359-4184/09 $32.00
www.nature.com/mp
LETTERS TO THE EDITOR
Prenatal sex differences in the human brain
10
8
6
4
male
female (ind.1,2,3)
SR
Y
2
R
P
PC S4Y
D 1
H
1
D 1Y
D
X3
U Y
S
N P9
LG Y
N
EI 4Y
F1
AY
U
TY
TM ZFY
S
C B4
Yo Y
rf1
5
PR B
KY
The presence of genetic sex differences in the adult
human brain is now recognized.1 We hypothesized
that the basis of this sex bias is already established in
the brain before birth. Here, we show that several
genes encoded in the Y-chromosome are expressed in
many regions of the male prenatal brain, likely having
functional consequences for sex bias during human
brain development.
The marked sex differences in age at onset,
prevalence and symptoms for numerous neuropsychiatric disorders2 indicate the importance to study
the emergence of a sex bias during human brain
development. A recent article includes for the first
time comprehensive data on the human brain transcriptome before birth.3 This elegant work reveals a
large number of gene and alternative splicing differences specific to certain regions of the brain during
development. A total of 12 regions in the midgestation human brain were analyzed in both male and
female human fetal brain specimens. This data set
provides for the first time the opportunity to evaluate
the existence of embryonic sex bias in the human
brain. To do this, we re-analyzed human fetal brain
expression data from the Gene expression Omnibus
(GEO accession GSE 13344, Human Exon 1.0 ST
Array) according to sex. The database included 95
array hybridizations, of which 72 corresponded with
different brain regions from three midgestation female
fetuses and 23 were from equivalent brain regions
from a male. The largest sex differences observed
were in genes encoded on the Y-chromosome, showing the existence of prenatal gender bias in brain
expression. Figure 1a shows the expression levels of
11 Y-chromosome-encoded genes, including RPS4Y1,
PCDH11Y, DDX3Y, USP9Y, NLGN4Y, EIF1AY, UTY,
ZFY, TMSB4Y, CYorf15B and PRKY. For 10 genes out
of these 11, expression was detected in all the 12
brain regions analyzed (Supplementary Figure 1),
suggesting that they may be present throughout the
whole brain during development. PRKY on the other
hand was expressed in a more restricted manner in
the cortex samples, with low expression in cerebellum and basal ganglia. Some evidence of specific
splice variant expression was found. For example,
only one out of three ZFY transcripts produced
positive signals in cortex, striatum and thalamus
(Supplementary Figure 1).
The expression of more than one-third of the genes
encoded on the Y-chromosome in prenatal human
brain indicates their importance for sex-biased brain
development. These genes are not only expressed in
the brain before birth but some of them are also
known to have sex differences in adult brain,1,4
whereas others are expressed during infancy, but
reduced later on during their lifetime.5
Intriguingly, SRY, a well-known determinant of
testicle development during midgestation,6 showed
no evidence of expression in any of the brain regions
analyzed (Figure 1b, and Supplementary Figure 1),
suggesting that the main somatic sex determinants
may be different for the brain and gonads during
human gestation.
In humans, all 11 genes described here are encoded
in the male-specific region of the Y-chromosome,7
with RPS4Y1 and ZFY located in the p-arm very close
to SRY and most of the remaining genes located in the
q-arm. Interestingly, the expression of Y-linked genes
log2 expression
Molecular Psychiatry (2009) 14, 988–989. doi:10.1038/
mp.2009.79
mean bg
Figure 1 Expression of Y-linked genes in male prenatal
brain. (a) Identification of 11 Y-linked genes that are
expressed in the male prenatal brain. The genes were
identified by comparing microarray measurements of
mRNA levels in 12 different brain regions in prenatal male
brain (yellow, narrays = 23) and in prenatal female brain (blue,
narrays = 72). As female samples do not contain Y-linked
genes, the signals obtained in females were used as
measurements of the local probe background level for each
individual Y-linked gene, and the criterion for expression
was a mean fold-change of at least two as compared with the
female signals. (b) SRY did not show evidence of expression
in any of the male prenatal brain samples. The band on the
box represents the median, and the lower and upper hinge
of the box represent the first and third quartile. Whiskers
designate the most extreme data points which are no more
than 1.5 times the interquartile range from the box. Round
circles show outliers. The horizontal dashed line represents
the overall mean signal level of all Y-linked genes in
females, shown as an indicator of the global background
signal levels in the arrays.
Letters to the Editor
in prenatal brain is only partially conserved between
rodents and humans. USP9Y, DDX3Y and UTY have
known orthologous genes in the mouse Y-chromosome, and their expression is conserved in terms of
sex bias prenatally in both groups.8 Other genes with
prenatal sex bias in humans, such as RPS4Y1 and
EIF1AY, have rodent homologs encoded in somatic
chromosomes and are not known to be sexually
dimorphic. PCDH11Y and CYorf15B only have known
mouse homologs in the X-chromosome and are not
known to have sex differences in the brain. TMSB4Y
and NLGN4Y lack known homologous in rodents.
Finally, ZFY, which encodes for a transcription factor
in the Y-chromosome in both groups, is expressed in
human brain before birth, but absent in prenatal
mouse brain.8 The differences in prenatal expression
of Y-linked genes mentioned above suggest that parts
of the programming of gender biases in the brain are
human-, or at least, primate-specific.
Although the importance of X-chromosome-encoded
genes for mental function has been well established,9 the
relevance of Y-chromosome genes on brain function is
less known. The results presented here, together with
the well-known rapid evolution of Y-chromosomes,10
clearly point to the importance of future investigation
on Y-linked gene function in the developing human
brain.
Conflict of interest
The authors declare no conflict of interest.
B Reinius and E Jazin
Department of Development & Genetics, Evolutionary
Biology Centre, Uppsala University, Uppsala, Sweden
E-mail: [email protected] or
[email protected]
References
1 Reinius B, Saetre P, Leonard JA, Blekhman R, Merino-Martinez R,
Gilad Y et al. PLoS Genet 2008; 4: e1000100.
2 Paus T, Keshavan M, Giedd JN. Nat Rev Neurosci 2008; 9:
947–957.
3 Johnson MB, Kawasawa YI, Mason CE, Krsnik Z, Coppola G,
Bogdanovic D et al. Neuron 2009; 62: 494–509.
4 Galfalvy HC, Erraji-Benchekroun L, Smyrniotopoulos P, Pavlidis P,
Ellis SP, Mann JJ et al. BMC Bioinformatics 2003; 4: 37.
5 Weickert CS, Elashoff M, Richards AB, Sinclair D, Bahn S, Paabo S
et al. Mol Psychiatry 2009; 14: 558–561.
6 Maatouk DM, Capel B. Curr Top Dev Biol 2008; 83: 151–183.
7 Skaletsky H, Kuroda-Kawaguchi T, Minx PJ, Cordum HS, Hillier L,
Brown LG et al. Nature 2003; 423: 825–837.
8 Xu J, Burgoyne PS, Arnold AP. Hum Mol Genet 2002; 11:
1409–1419.
9 Skuse DH. Hum Mol Genet 2005; 14(Spec No 1): R27–R32.
10 Goto H, Peng L, Makova KD. J Mol Evol 2009; 68: 134–144.
Supplementary Information accompanies the paper
on the Molecular Psychiatry website (http://www.
nature.com/mp)
Stem cell signaling in
newly diagnosed,
antipsychotic-naive
subjects with nonaffective
psychosis
989
Molecular Psychiatry (2009) 14, 989–991. doi:10.1038/
mp.2009.45
Widespread metabolic abnormalities have been described in newly diagnosed, antipsychotic-naive
patients with nonaffective psychosis, including a
shorter telomere,1 abnormal glucose tolerance,2 an
increase in the pro-inflammatory molecule interleukin-62 and an increased pulse pressure.1 Stromalderived factor 1-alpha (SDF-1a), the major chemokine
for adult stem cells (SCs), is involved in glucose
regulation3 and is abnormal in diabetes.4 SDF1 is
critical in homing, maintenance and mobilization of
adult SCs.5 We tested the hypothesis that newly
diagnosed, antipsychotic-naive patients with nonaffective psychosis would have a decreased concentration of circulating SDF-1a compared with control
subjects.
The psychosis (N = 24) and control (N = 24)
subjects were matched for race (Caucasian), gender,
age, body mass index, smoking habit, cortisol blood
levels, socioeconomic status and catchment area
(Table 1). The two groups were matched before
assaying SDF-1a concentrations. All subjects were
interviewed using the Structured Clinical Interview
for DSM-IV Axis I Disorders. Psychopathology was
rated using the Positive and Negative Syndromes
Scale (PANSS). Subjects in the psychosis group had a
maximum lifetime antipsychotic exposure of 1 week
and no antipsychotic use in the 30 days before the
study, and had a diagnosis of nonaffective psychosis.
Exclusion criteria for the control subjects included a
history of psychosis or major depressive disorder.
Additional general inclusion criteria were age from
18 to 64 years, no history of serious medical or
neurological condition and not using any medication
that impacts glucose tolerance. All subjects gave
informed consent for participation in the study,
which was conducted under the supervision of the
local institutional review board.
Blood was drawn between 0800 and 0900 hours
after an overnight fast. Plasma was collected on ice
using EDTA and centrifuged for 15 min at 1000 g
within 30 min of collection. An additional centrifugation step of the separated plasma at 10 000 g for
10 min at 2–8 1C was performed for complete
platelet removal. Samples were aliquoted and stored
at
80 1C. Quantitative determination of human
CXCL12/SDF-1a was determined with Quantikine
Molecular Psychiatry