0022-5347/02/1685-2142/0 Vol.

168, 2142–2148, November 2002

THE JOURNAL OF UROLOGY® Printed in U.S.A.
Copyright © 2002 by AMERICAN UROLOGICAL ASSOCIATION, INC.® DOI: 10.1097/01.ju.0000034424.55595.33

Review Article

ANDROGEN IMPRINTING OF THE BRAIN IN ANIMAL MODELS AND

HUMANS WITH INTERSEX DISORDERS: REVIEW AND
RECOMMENDATIONS
ZOLTAN HRABOVSZKY AND JOHN M. HUTSON
From the Surgical Department, Royal Children’s Hospital, Melbourne, Victoria, Australia

ABSTRACT

Purpose: Psychosexual development, gender assignment and surgical treatment in patients

with intersex are controversial issues in the medical literature. Some groups are of the opinion
that gender identity and sexual orientation are determined prenatally secondary to the fetal
hormonal environment causing irreversible development of the nervous system. We reviewed the
evidence in animal and human studies to determine the possible role of early postnatal androgen
production in gender development.
Materials and Methods: An extensive literature review was performed of data from animal
experiments and human studies.
Results: Many animal studies show that adding or removing hormonal stimulus in early
postnatal life can profoundly alter gender behavior of the adult animal. Human case studies show
that late intervention is unable to reverse gender orientation from male to female. Most studies
have not permitted testing of whether early gender assignment and treatment as female with
suppression/ablation of postnatal androgen production leads to improved concordance of the
gender identity and sex of rearing.
Conclusions: Animal studies support a role for postnatal androgens in brain/behavior devel-
opment with human studies neither completely supportive nor antagonistic. Therefore, gender
assignment in infants with intersex should be made with the possibility in mind that postnatal
testicular hormones at ages 1 to 6 months may affect gender identity. A case-control study is
required to test the hypothesis that postnatal androgen exposure may convert ambisexual brain
functions to committed male behavior patterns.
KEY WORDS: androgens, gender identity, testis, brain, sex differentiation disorders

A newborn with ambiguous genitalia causes severe dis-
tress to the family and immediately raises several questions
about gender, the type and timing of treatment for the anom-
aly and the possible outcome. The physician is expected to
answer these questions but in many cases it is not easy. In
several aspects intersex disorders are not yet completely
understood and the medical literature is also divided about
issues of gender assignment and surgical treatment.
We provide an up-to-date summary of the most impor-
tant data gained from animal experiments and human
studies on androgenic control of brain development. We
searched for evidence that may relate to whether postnatal
hormone levels have any role in masculinizing the devel-
oping nervous system, which could be relevant for the
timing of surgery, or whether these changes occur exclu-
sively prenatally. In the light of our current knowledge of
this field we establish some guidelines for the clinician on
gender assignment and the subsequent treatment of chil-
dren with ambiguous genitalia.
ANIMAL EXPERIMENTS
Rodents. In 1959 Phoenix et al reported that adult female
guinea pigs exposed to androgens during gestation showed
less feminine copulatory behavior (lordosis) and more mas-
2142
culine behavior (mounting) than control females.1 According
to their hypothesis testicular steroids could alter the devel-
oping nervous system permanently, masculinizing and de-
feminizing sexual behavior of the animal as an adult. This
long-lasting “organizational” effect contrasted with the “acti-
vational” effects of steroids on behavior when steroid treat-
ment in adult animals caused only a temporary behavioral
change. The organizational influences of hormones typically
occur during early critical periods of development, when mas-
culine neural circuits develop or are maintained and femi-
nine circuits are prevented from developing or are allowed to
atrophy. The mechanisms involved in these hormonal influ-
ences on neural development have been described most
clearly in studies of rodents.
Hormonal Influences on Reproductive Behavior: Genetically
female rats treated with testosterone during the first few days
of life do not show the typical female reproductive behavior of
lordosis in adulthood even if given activating injections of fe-
male steroid hormones, estradiol and progesterone in adult-
hood.2 In contrast, their behavior is more like that of genetic
males. Similarly male rats castrated at birth show lordosis in
response to activating doses of estradiol and progesterone, and
do not show male sexual behavior after testosterone treatment
in adulthood.3 Prenatal antiandrogen treatment also prevents
masculinization of mating behavior in males.4 These early hor-
 ANDROGEN IMPRINTING OF BRAIN AND INTERSEX DISORDERS
mone manipulations can permanently alter the neural poten-
tial for the animal to show sex typical reproductive behavior.
Ovarian hormones do not appear to be essential for the
development of female reproductive behavior. Female rats
ovariectomized or treated with anti-estrogens after birth
show sexual behavior similar to that of intact females.5 The
absence of androgens or their metabolites appears to be suf-
ficient for feminine behavioral development.
Paradoxically estrogen treatment was even more effective
than androgen treatment for masculinizing reproductive be-
havior in neonatal female rodents.6 Estradiol can be formed
from testosterone through the action of the enzyme aro-
matase. This enzyme has been localized to several brain
regions that also are known to contain estrogen receptors
and be involved in the control of reproductive behavior.7
According to the estrogen hypothesis, during normal male
development testosterone enters the cells in these brain re-
gions, where it is converted to estradiol by aromatase. Estra-
diol then interacts with estrogen receptors to promote neural
differentiation in the masculine direction. While testosterone
can enter the brain easily in males, developing female rats
are protected from maternal estrogens, which are bound to
!-fetoprotein and unable to penetrate the blood-brain barri-
er.8 This hypothesis has been supported by evidence that
treating male rats with anti-estrogens or aromatase inhibi-
tors prevents the normal masculine development of repro-
ductive behavior despite the high level of endogenous testos-
terone.7 These drugs also block behavioral masculinization in
testosterone treated newborn females.9
Sexual Differentiation of Brain Structures: Levels of go-
nadal hormones during early development also influence sex-
ual differentiation of the brain structure. Several areas with
cytoarchitectural sexual dimorphism have been described in
the brain and spinal cord.10 The most extensively studied
structure is the preoptic area, which is involved in mating
behavior and the control of gonadotrophin secretion.11 The
volume of a cell dense subregion of the preoptic area called
the sexually dimorphic nucleus of the preoptic area is several
times larger in adult male rats than in females.11 Neurons of
the preoptic area in male and female rats contain more
estrogen receptors than any other brain region.12 While the
sexually dimorphic nucleus of the preoptic area preferen-
tially binds estrogen in the 2 sexes, it also accumulates
testosterone and dihydrotestosterone.13 Adult hormone ma-
nipulations do not alter the volume of the sexually dimorphic
nucleus in either sex, although exposure of developing female
rats to testosterone or estrogens leads to a larger, more
masculine sexually dimorphic nucleus of the preoptic area in
adulthood and castration of newborn males results in a more
feminine one.14 The masculinizing effect of early estrogen
treatment is stronger than that of early androgen treatment,
suggesting that estrogens have a direct effect on stimulating
the development of the sexually dimorphic nucleus of the
preoptic area in male rats.14, 15
It appears that testosterone has 2 major effects on the fetal
rodent brain. Testosterone up-regulates the production of
androgen receptors in sexually dimorphic brain areas.16 In
addition, androgen receptor activation by testosterone stim-
ulates aromatase synthesis, which results in increased estra-
diol production within the sexually dimorphic nuclei.17
Increased levels of estradiol are then responsible for devel-
oping the anatomical differences of the brain in the male
direction.
Hormonal Influences on Gonadotropin Regulation: Male
and female mammals differ in their patterns of gonadotropin
hormone release.18 In adult females gonadotropin hormone
release is cyclic, producing fluctuations in the ovarian hor-
mones associated with estrus or menstrual cycles. In con-
trast, in males gonadotropin hormone secretion is tonic,
maintaining high levels of testicular hormones in adulthood.
Gonadotropin hormone release from the pituitary gland is
2143
under the control of hypothalamic nuclei.19 Differentiation of
the cyclic or tonic pattern depends on the influence of gonadal
hormones early in life. Prenatal and postnatal antiandrogen
treatment in male rats produces the cyclic pattern of gonad-
otropin hormone release.20
Early postnatal gonadectomy in male or female rats also
results in the cyclic pattern of gonadotropin hormone secre-
tion.18, 21 Early postnatal treatment with aromatizable an-
drogens or high doses of estrogens in female rats produces a
male (tonic) pattern of gonadotropin hormone release.18, 22, 23
These observations suggest that disintegration of the cyclic
pattern and differentiation of the tonic pattern of gonadotro-
pin hormone release are directly under estrogenic rather
than androgenic control.
Birds. Sexual Dimorphism of the Song System: Another
useful model in which to study the early differentiating ef-
fects of gonadal steroids on the developing brain is the neural
circuit for song in songbirds. Courtship song and the brain
regions that control it show striking sexual dimorphism in
zebra finches.24 Only males sing and the telencephalic nuclei
in the motor pathway that generate song are larger in vol-
ume in males and contain more and larger cells than in
females.24 Males sing more often when testosterone levels
are high in adulthood but females do not sing at all, even if
treated with testosterone as adults.25 Estrogens and andro-
gens given in gonadally intact, hatchling female zebra
finches permanently masculinize the song system, increasing
the size of vocal control regions.26 These neonatally mascu-
linized females also sing in adulthood.27 While the most
potent virilizing agent is estradiol, nonaromatizable andro-
gens have relatively little masculinizing effect in hatchling
females.28
However, experiments in zebra finches have yielded some
contradictory results. Steroid treatment in females can effec-
tively masculinize neural centers as well as behavior but
castration,29 antiandrogen,30 antiestrogen31 or aromatase in-
hibitor treatment32 failed to prevent male zebra finches from
having a masculine brain and the singing that accompanies
it. Studies measuring plasma sex steroids in the weeks after
hatching did not show higher androgen levels in males than
in females.33 These results suggest that testicular secretions
are not exclusively responsible for masculine development of
the neural song system in this species. The brain of male
songbirds may have an alternative, steroid independent
means of developing in the masculine direction.
HUMAN STUDIES
While data from animal studies should be applied to hu-
mans with extreme caution, the similarity of findings in
diverse numbers of species suggests that processes that have
evolved in lower animals may be preserved in humans.
The best studied sexually dimorphic brain structure in
humans is the sexually dimorphic nucleus of the preoptic
area. It has been shown that males have a larger nucleus
with more neurons than females but this sex difference in
neuronal number is not detectable in children younger than
6 to 10 years.34
Another strategy for asking whether fetal steroids affect
the human brain is to study sex differences in human behav-
ior and determine whether inadvertent exposure to excess
amount of hormones alters sexually dimorphic behavior. Sev-
eral aspects of human behavior show reliable sex differences,
including gender identity, sexual orientation, rough play,
juvenile toy preferences, physical aggression, and certain
aspects of verbal, visual-spatial and mathematical abilities.
The most characteristic differences are those in gender iden-
tity and sexual orientation.35
Strong evidence that hormones contribute to human be-
havioral development has come from studies of girls with
congenital adrenal hyperplasia. This disorder is caused by an
2144 ANDROGEN IMPRINTING OF BRAIN AND INTERSEX DISORDERS
enzymatic defect resulting in overproduction of adrenal ste-
roids, including androgens. Females with congenital adrenal
hyperplasia are exposed to androgens during gestation and
are typically born with some degree of genital virilization.
Postnatal exposure to androgens would also occur until ad-
renal hormone replacement therapy is begun.
These individuals are usually assigned the female sex and
raised as girls. Many have a tomboyish personality, play with
toys that are more often favored by boys and participate in
more rough play than other girls.36, 37 Behavioral masculin-
ization in congenital adrenal hyperplasia cases does not seem
to be related to the degree of genital virilization.38 As women,
they are usually sexually attracted to men but are more
likely to have bisexual or homosexual interests than women
without the condition.37, 39 Although the majority of patients
assigned and reared as female have a female gender identi-
ty,40 some become gender dysphoric in adolescence or adult-
hood and an even smaller number choose to live as males.41
It has been suggested that masculinized behavior in females
with congenital adrenal hyperplasia could result from paren-
tal doubt and self-doubt regarding sexual identity after early
genital virilization.38 However, an androgen effect on the
developing brain, similar to that in other species, seems to be
an equally plausible explanation for the behavioral differ-
ences.
Information on patients with 5!-reductase deficiency and
17"-hydroxysteroid dehydrogenase deficiency can also be in-
terpreted as supporting a hormonal contribution to human
psychosexual development. Individuals with these rare en-
zymatic defects are unable to produce certain androgens
needed for normal prenatal masculinization of the external
genitalia. These newborns are often assigned the female gen-
der because minimal virilization is apparent at birth and
raised as girls. However, at puberty when testicular testos-
terone secretion increases, the external genitalia virilize and
in many patients male gender identity and sexual behavior
develop.42, 43 An explanation of such results is that fetal
and/or postnatal androgens had masculinized the brain, so
that social rearing notwithstanding the individuals were des-
tined to think and, therefore, behave as men. However, it is
difficult to rule out that these individuals were treated dif-
ferently than normal girls.
In patients with the complete androgen insensitivity syn-
drome there is a severe defect in androgen receptors, so that
developing XY male fetuses ignore the masculinizing effect of
androgens and show an unambiguous female exterior.44 They
are accepted as female at birth and raised as girls but are
identified when menses fail to begin. If the testes are not
removed, normal female breasts develop from the increased
conversion of testosterone to estrogen at puberty. Such indi-
viduals are feminine in gender identity and sexual orienta-
tion.44 Their cognitive abilities are also similar to that of
normal females.45 The development of feminine characteris-
tics in these patients is supported by the insensitivity of the
brain to androgens and their unambiguous upbringing as
girls.
There is a remarkable interspecies difference in the man-
ifestation of androgen insensitivity in humans and lower
mammals. Androgen insensitive male rats have a feminine
phenotype but a masculine nervous system and behav-
ior,46, 47 which underline the importance of estrogen receptors
for the sexual differentiation of behavior in this species.
Conversely the feminine behavior of androgen insensitive
humans indicates that sexual steroids act exclusively
through androgen receptors to masculinize the nervous sys-
tem or androgen receptors must be functional for estrogen
receptor activation to be effective or occur at all.10
Studies of women who had been exposed to diethylstilbes-
trol during gestation suggest that fetal estrogens are also able
to masculinize human sexual behavior. Diethylstilbestrol
was extensively used in the United States between 1947 and
1971 as hormonal support for at-risk pregnancy until the
relationship of prenatal diethylstilbestrol exposure to adeno-
carcinoma of the vagina and cervix was identified.48
Diethylstilbestrol exposed women showed bisexual or homo-
sexual interest more frequently than siblings and matched
controls.49 They showed greater cognitive lateralization (sim-
ilar to men) than their nondiethylstilbestrol treated sisters,50
which suggests that diethylstilbestrol slightly masculinized
them. Diethylstilbestrol exposed females appear to be unaf-
fected in regard to gender identity, childhood play and inter-
ests, and cognitive abilities.51 Findings in these women have
special significance because they are unlikely to reflect
psychosocial factors. In contrast to women with congenital
adrenal hyperplasia, these patients are born with typical
female genitalia, and so the possibility that genital abnor-
malities discouraged heterosexual interest can be excluded.
Further evidence supporting a contribution of the early
hormone environment to gender identity and behavioral de-
velopment comes from a pair of male monozygotic twins, in 1
of whom the penis was accidentally ablated during circumci-
sion at age 7 months.52 A decision was made to reassign the
infant to the female sex and rear him as a girl. Surgical
castration with initial genital reconstruction was done at age
21 months. Subsequent followup showed that by early ado-
lescence the patient had rejected the female identity and he
began to live as a male at age 14 years.52 Later the patient
recalled that he had never been comfortable as a girl and he
had an exclusive sexual attraction to females.
Although human sexual development is rather complicated
because the social environment and upbringing can also sig-
nificantly contribute to the ultimate expression of human
sexual behavior and gender identity, available data from
human studies appear to confirm a role for early hormone
environment in shaping the developing human nervous sys-
tem.
THE CRITICAL PERIOD FOR SEXUAL DIFFERENTIATION
There seems to be a developmental period during which
the brain is more sensitive to the organizational effects of
gonadal hormones than at any other time. This interval is
called the critical period for sexual differentiation and it is
species specific.53 The critical period is an empirical concept
and does not represent a clearly defined stage of develop-
ment. Available data suggest that the beginning of the crit-
ical period may follow differentiation of testicular Leydig
cells and the onset of testosterone secretion.54 In rats typical
Leydig cells first appear between days 16 and 18 after con-
ception.55 Testosterone is significantly higher in male than in
females rat fetuses from day 18 after coitus through day 5
post partum.54 The greatest differences in absolute values
were observed on days 18 and 19, when a distinct surge
occurred in males. Resko et al reported that the concentra-
tion of plasma testosterone in males decreased gradually
after birth, reaching a baseline value between ages 10 and 30
days.56 By administering testosterone to female and neona-
tally castrated male rats they noted that the period during
which the volume of the sexually dimorphic nucleus of the
preoptic area is sensitive to exogenous androgen begins
strictly on day 18 of gestation and extends through postnatal
day 5.57 Interestingly it was recently noted that castration
can decrease the volume of the sexually dimorphic nucleus of
the preoptic area in male rats even as late as postnatal day
29.58 A possible explanation for this paradox could be the
dependence of sexually dimorphic nucleus of the preoptic
area cells on gonadal hormones for survival during develop-
ment. Administering exogenous testosterone may be effective
for a shorter period because cells of the sexually dimorphic
nucleus of the preoptic area may only survive for a limited
time after neonatal castration. However, in intact males
 ANDROGEN IMPRINTING OF BRAIN AND INTERSEX DISORDERS
these cells remain alive, but dependent on exposure to tes-
tosterone for survival.58
It has also been observed that there are different times
when specific neural and behavioral characteristics are max-
imally sensitive to steroid hormone influences. For example,
the effect of castration has different time courses on the
luteinizing hormone surge and lordosis behavior in rats.58
Only females and males castrated on postnatal day 1 and not
later showed luteinizing hormone surges after estrogen and
progesterone priming in adulthood. There was a stepwise
decrease in adult lordosis behavior in male rats castrated
postnatally and the response decreased with increasing age
at castration until day 7. These data suggest the existence of
several specific critical periods of different lengths that are
involved in discrete developmental processes of sexual differ-
entiation of the central nervous system, in contrast to a
single critical period for all developmental mechanisms.58
Another important aspect is that hormonal influences on
sexual differentiation of the brain and behavior are seem-
ingly not all or nothing. Each masculine and feminine char-
acteristic exists on a continuum and the amount, duration or
timing of hormone exposure determines its position on the
continuum.35
In humans to our knowledge the exact timing of the critical
period for sexual differentiation of the central nervous sys-
tem is unknown. However, we can assume that during a
critical period testosterone levels in the circulation are sig-
nificantly higher in males than in females, similar to the
results of animal experiments. Fetal testis incubation in vitro
and perfusion studies in vivo have shown that testicular
testosterone production occurs as early as 7 to 8 weeks of
gestation.59 By 14 to 18 weeks Leydig cells occupy half of the
volume of the fetal testes, after which they gradually involute
until term.60 Serum testosterone in the male fetus attains a
peak at 14 to 16 weeks.61 At this time values are in the adult
male range, which is essential for male genital differentia-
tion.61 After 24 weeks there is no significant sex difference in
umbilical arterial serum testosterone,62 although the obser-
vation that amniotic fluid levels even in late pregnancy are
higher in males than in females61 suggests continued testic-
ular secretion.
While some investigators noted higher testosterone at
term in the umbilical cord and peripheral venous blood sam-
ples in males than in females,63 others failed to detect a
significant sex difference.64 In female infants testosterone
concentration remains constantly low during year 1 of life.65
In males the level increases from birth to a peak at 1 to 3
months of life and thereafter it decreases to prepubertal
levels by ages 4 to 6 months (see figure). If we accept that as
Mean serum testosterone in male fetus and infant61
2145
in lower animals, testosterone in humans also masculinizes
the central nervous system, we must suppose that these 2
peaks of serum testosterone in males have special signifi-
cance for sexual development in the male direction.
HYPOTHESIS
Gender identity depends on stepwise exposure to androgen
secretion prenatally, postnatally and at puberty. Although
the exact role of postnatal androgen production is unknown,
we propose that it is a crucial adjunct for masculine devel-
opment of the brain in animals and humans. Furthermore,
when treating intersex conditions in infants, we should as-
sume that postnatal androgens may masculinize behavior at
least until definitive case-control studies prove otherwise. If
postnatal androgens have a key role as we suspect, infants
with early female gender assignment and suppression or
ablation of androgen production should have better long-
term outcome for psychosexual development. Moreover, if
postnatal sexual steroid imprinting occurs, reassigning in-
fants with intersex to the female gender should be avoided
after ages 4 to 6 months.
GENDER ASSIGNMENT
Gender assignment in an individual with intersex is a
serious and sometimes difficult decision with life-long impli-
cations. Not only do patients need a gender that they identify
with psychosexually, but also it is equally important to pro-
duce functional genitalia consistent with the gender as-
signed. Money et al first emphasized the role of upbringing in
gender identity development.66 Their studies in the 1950s of
patients with hermaphroditism led to the belief that individ-
uals are psychosexually neutral at birth and become differ-
entiated as male or female during the course of various
growth experiences. They believed that unambiguously rais-
ing a child with a physical intersex condition as a member of
the assigned sex would be more important for gender identity
development than chromosomal, gonadal or hormonal sex.67
Money et al recommended early gender assignment with
early corrective surgery in an attempt to identify unambig-
uously the child with the chosen sex of rearing. The most
important factor for making the decision was “the morphol-
ogy of the external genitals and the ease with which these
organs can be surgically reconstructed to be consistent with
the assigned sex.”67 The assumption that individuals are
psychosexually undifferentiated at birth was gradually repu-
diated as further research demonstrated that prenatal hor-
mones can influence certain aspects of human sex dimorphic
behavior in the same direction as in other mammals.
2146 ANDROGEN IMPRINTING OF BRAIN AND INTERSEX DISORDERS
Currently there are 2 opposing viewpoints in the medical
literature regarding gender assignment and the treatment of
individuals with ambiguous genitalia. Traditionally gender
assignment and genital reconstruction have been quite uni-
formly done shortly after birth. This practice helps the indi-
vidual to avoid internal conflicts and stigmatization by the
family or peer environment and facilitates parental accep-
tance of the child as male or female. However, in recent years
some intersex support groups, such as the Intersex Society of
North America, comprising many adults born with ambigu-
ous genitalia who are now dissatisfied with the outcome,
have raised their voices against early genital reconstruc-
tion.68 Diamond and Sigmundson have the opinion that the
internal and external genitals should be left alone until after
puberty, when the patient, who is the ultimate decision
maker, is able to express individual needs and wants as a
sexual being and provide informed consent to corrective sur-
gery.69 They argue that gender identity, sexual orientation
and sexual preferences are personal choices, and neither the
family nor physician can predict how an individual with
intersex would choose to live or with what type of genitals.
According to their principles the patient should undergo gen-
der assignment at birth depending on the nature of the
diagnosis, although no corrective surgery on the external or
internal genitalia should be performed.69 The parents must
be consistent in rearing the child as a boy or girl. However,
after puberty it is the patient who eventually decides
whether to live as male, female or even intersexual depend-
ing on how the central nervous system dictates sexual iden-
tity in the individual. Opponents of this view argue that the
ambiguous appearance of the genitalia may hinder gender
identity development and unambiguous rearing by the par-
ents is virtually impossible when the genitals are not consis-
tent with the sex chosen. While corrective surgery with cur-
rent techniques can offer good results in cases of severe
hypospadias, not performing the operation imposes obvious
practical difficulties in genotypically male patients. In cer-
tain intersex disorders retaining dysgenetic or streak gonads
with a potential for tumor development can be a hazardous
practice. Furthermore, we do not know what role the physi-
ologically increased level of postnatal testosterone in males
has on later psychosexual development. If it has a role, it
would seem wise in any patient with intersex assigned the
female gender to remove the testes in the first few weeks of
life well before the serum testosterone peak.
RECOMMENDATIONS
Rapid detailed investigations must be performed in new-
borns with genital ambiguity to evaluate the chromosomal
gonadal sex, hormonal environment, and anatomy of the
external and internal genital organs to clarify the type of
anomaly. The patient needs gender assignment as early as
possible, preferentially within the first few weeks of life.
Fortunately gender assignment is relatively straightforward
in the majority of individuals with intersex.
In newborns with ambiguous genitalia secondary to con-
genital adrenal hyperplasia or intrauterine exposure to high
levels of androgens regardless of the degree of virilization
female gender assignment is warranted for several reasons.
Because the internal organs in these patients are essentially
intact, they have the potential for normal sexual function and
fertility. Although there is more rough play than may be
expected of girls, and a higher incidence of bisexuality and
homosexuality than in the general population, these individ-
uals almost always identify as female if gender assignment is
made early in life.40 Most often individuals with congenital
adrenal hyperplasia who want the gender changed to male
have been discovered in late infancy or childhood, or have not
been receiving adrenal suppressive medication consistently
since birth, allowing androgen levels to remain high to mas-
culinize further the nervous system.41 We perform feminiz-
ing genitoplasty within the first 2 months of life to ensure
that the appearance of the external genitalia is consistent
with the female sex of rearing and relieve parental anxiety
about the child regarding the concerns of relatives and
friends. When the patient presents after 4 to 6 months and
has been erroneously raised as a male because of almost
complete virilization of the external genitalia, we recommend
the maintenance of the originally assigned gender and per-
form phallic surgery to enable the patient to function as a
male.
Individuals with complete androgen resistance have an
unambiguous female phenotype, gender identity and sexual
orientation. Rearing the child as female is the right decision
in gender assignment. The testes can be removed at diagno-
sis or preserved until puberty to induce spontaneous femini-
zation. After puberty they should be removed because of the
increased risk of malignant change.70
The most difficult decision is assigning the appropriate
gender in under virilized genetic males with ambiguous gen-
italia. A wide spectrum of anomalies can produce under vir-
ilization in males, such as defects in testosterone synthesis or
metabolism and poor tissue response to androgens. Special
attention should be given to the appearance of the external
genitalia and their ability to respond to androgen or human
chorionic gonadotropin stimulation. Strikingly masculinized
patients should be assigned the male gender, particularly
when phallic enlargement is noted in response to exogenous
androgens. On the other hand, when the phallus is inade-
quate and does not increase in size with stimulation, female
gender assignment is most prudent when the patient pre-
sents in the neonatal period. To make the decision we must
also consider the expected course and outcome of that partic-
ular anomaly. Patients with 5!-reductase deficiency and
17"-hydroxysteroid dehydrogenase deficiency require spe-
cialized laboratory evaluation to make the diagnosis but they
are good candidates for male gender assignment because
marked spontaneous virilization can be expected at puber-
ty.42, 43 Should the female gender be elected the patient needs
early orchiectomy and estrogen replacement therapy at pu-
berty.
In true hermaphrodites when male gender assignment is
being considered, serum testosterone levels before and after
human chorionic gonadotropin stimulation are said to have a
predictive value. If levels are consistently low despite stim-
ulation, poor masculinization at puberty can be expected.71
While in individuals raised as males spermatogenesis and
testicular hormone production are characteristically defi-
cient,71 others raised as females may ovulate, become preg-
nant and give birth to normal offspring.72 After gender as-
signment gonads and internal organs discordant for the sex
of rearing should be removed. Males need regular evaluation
because tumor may form in testicular tissue.73
In 46 XY pure and in mixed gonadal dysgenesis the gonads
have a propensity toward malignant tumors.74 All dysgenetic
testes and streak gonads should be removed regardless of the
gender assigned. The preferred sex of rearing is female.
However, markedly virilized individuals can be raised as
male and in selected situations the testis can be retained if it
is in the scrotal position.75
CONCLUSIONS
Animal experiments in songbirds and mammals have pro-
vided evidence that prenatal and early postnatal sexual hor-
mone secretions have an organizational effect on the devel-
oping nervous system, which results in sexual dimorphism of
brain structure and behavior. Studies in individuals with
intersex and traumatized genitalia who underwent sex reas-
signment suggest that a mechanism similar to that in ani-
mals may also exist in our species. However, sexual develop-
 ANDROGEN IMPRINTING OF BRAIN AND INTERSEX DISORDERS
ment of the human brain is much more complicated since
social factors also contribute to the ultimate expression of
human sexual behavior and gender identity.
The 2 periods during early human development when an-
drogens are elevated in males compared with females are the
prenatal period from about weeks 8 to 24 of gestation and the
postnatal period from about months 1 to 6 of life (see figure).
We propose that hormone induced masculinization on the
developing brain occurs during these 2 periods of hormone
elevation. Moreover, when decisions about gender assign-
ment and treatment are made early, it should allow for
possible postnatal virilization.
Early gender assignment followed by early removal of the
source of excess androgens when the female gender is chosen
may prevent the brain from further masculinization after
birth. This result requires prompt administration of hydro-
cortisone in patients with congenital adrenal hyperplasia
and removal of all testicular tissue in genotypically male
individuals before postnatal androgen production causes fur-
ther masculine development of the nervous system. The best
outcome may be expected when early gender assignment and
genitoplasty are combined with unambiguous upbringing in
the sex chosen for the patient with the intersex anomaly.
Infants diagnosed late and treated beyond ages 4 to 6 months
should not undergo sex reversal to the female gender since
prenatal and postnatal imprinting can be expected.
REFERENCES
1. Phoenix, C. H., Goy, R. W., Gerall, A. A. and Young, W. C.:
Organizing action of prenatally administered testosterone pro-
pionate on the tissues mediating mating behavior in the fe-
male guinea pig. Endocrinology, 65: 369, 1959
2. Ward, I. L.: Differential effect of pre- and postnatal androgen on
the sexual behavior of intact and spayed female rats. Horm
Behav, 1: 25, 1969
3. Gerall, A. A., Hendricks, S. E., Johnson, L. L. and Bounds, T. W.:
Effects of early castration in male rats adult sexual behavior.
J Comp Physiol Psychol, 64: 206, 1967
4. Clemens, L. G., Gladue, B. A. and Coniglio, L. P.: Prenatal
endogenous androgenic influences on masculine sexual behav-
ior and genital morphology in male and female rats. Horm
Behav, 10: 40, 1978
5. Young, W. C., Goy, R. W. and Phoenix, C. H.: Hormones and
sexual behavior. Science, 143: 212, 1964
6. Whalen, R. E. and Nadler, R. D.: Suppression of the development
of female mating behavior by estrogen administered in in-
fancy. Science, 141: 273, 1963
7. McEwen, B. S., Lieberburg, I., Chaptal, C. and Krey, L. C.:
Aromatization: important for sexual differentiation of the neo-
natal rat brain. Horm Behav, 9: 249, 1977
8. Vermeulen, A. and Verdonck, L.: Studies on the binding of tes-
tosterone to human plasma. Steroids, 11: 609, 1968
9. Fadem, B. H. and Barfield, R. J.: Neonatal hormonal influences
on the development of proceptive and receptive feminine sex-
ual behavior in rats. Horm Behav, 15: 282, 1981
10. Cooke, B., Hegstrom, C. D., Villeneuve, L. S. and Breedlove,
S. M.: Sexual differentiation of the vertebrate brain: principles
and mechanisms. Front Neuroendocrinol, 19: 323, 1998
11. Gorski, R. A., Gordon, J. H., Shryne, J. E. and Southam, A. M.:
Evidence for a morphological sex difference within the medial
preoptic area of the rat brain. Brain Res, 148: 333, 1978
12. Herbison, A. E. and Theodosis, D. T.: Immunocytochemical iden-
tification of oestrogen receptors in preoptic neurones contain-
ing calcitonin gene-related peptide in the male and female rat.
Neuroendocrinology, 56: 761, 1992
13. Jacobson, C. D., Arnold, A. P. and Gorski, R. A.: Steroid autora-
diography of the sexually dimorphic nucleus of the preoptic
area. Brain Res, 414: 349, 1987
14. Dohler, K. D., Coquelin, A., Davis, F., Hines, M., Shryne, J. E.
and Gorski, R. A.: Pre- and postnatal influence of testosterone
propionate and diethylstilbestrol on differentiation of the sex-
ually dimorphic nucleus of the preoptic area in male and
female rats. Brain Res, 302: 291, 1984
15. Dohler, K. D., Coquelin, A., Davis, F., Hines, M., Shryne, J. E.,
Sickmoller, P. M. et al: Pre- and postnatal influence of an
2147
estrogen antagonist and an androgen antagonist on differen-
tiation of the sexually dimorphic nucleus of the preoptic area
in male and female rats. Neuroendocrinology, 42: 443, 1986
16. Lu, S. F., McKenna, S. E., Cologer-Clifford, A., Nau, E. A. and
Simon, N. G.: Androgen receptor in mouse brain: sex differ-
ences and similarities in autoregulation. Endocrinology, 139:
1594, 1998
17. McCarthy, M. M.: Molecular aspects of sexual differentiation of
the rodent brain. Psychoneuroendocrinology, 19: 415, 1994
18. Pfeiffer, C. A.: Sexual differences of the hypophyses and their
determination by the gonads. Am J Anat, 58: 195, 1936
19. Gorski, R. A.: Localization and sexual differentiation of the ner-
vous structures which regulate ovulation. J Reprod Fertil, 1:
67, 1966
20. Neumann, F. and Elger, W.: Permanent changes in gonadal
function and sexual behavior as a result of early feminization
of male rats by treatment with an antiandrogenic steroid.
Endokrinologie, 50: 209, 1966
21. Takewaki, K.: Some experiments on the control of hypophyseal-
gonadal system in the rat. Gen Comp Endocrinol, suppl., 1:
309, 1962
22. Gorski, R. A. and Barraclough, C. A.: Effects of low dosages of
androgen on the differentiation of hypothalamic regulatory
control of ovulation in the rat. Endocrinology, 73: 210, 1963
23. Greene, R. R., Burrill, M. W. and Ivy, A. C.: Experimental inter-
sexuality: the paradoxical effects of estrogens on the sexual
development of the female rat. Anat Rec, 74: 429, 1939
24. Nottebohm, F. and Arnold, A. P.: Sexual dimorphism in vocal
control areas of the songbird brain. Science, 194: 211, 1976
25. Arnold, A. P.: Behavioral effects of androgen in zebra finches
(Poephila guttata) and a search for its action. Ph. D. Thesis,
Rockefeller University, New York, 1974
26. Gurney, M. E.: Hormonal control of cell form and number in the
zebra finch song system. J Neurosci, 1: 658, 1981
27. Pohl-Apel, G.: The correlation between the degree of brain mas-
culinization and song quality in estradiol treated female zebra
finches. Brain Res, 336: 381, 1985
28. Grisham, W. and Arnold, A. P.: A direct comparison of the
masculinizing effects of testosterone, androstenedione, estro-
gen, and progesterone on the development of the zebra finch
song system. J Neurobiol, 26: 163, 1995
29. Arnold, A. P.: The effects of castration and androgen replace-
ment on song, courtship, and aggression in zebra finches
(Poephila guttata). J Exp Zool, 191: 309, 1975
30. Schlinger, B. A. and Arnold, A. P.: Androgen effects on the
development of the zebra finch song system. Brain Res, 561:
99, 1991
31. Mathews, G. A., Brenowitz, E. A. and Arnold, A. P.: Paradoxical
hypermasculinization of the zebra finch song system by an
antiestrogen. Horm Behav, 22: 540, 1988
32. Wade, J., Schlinger, B., Hodges, L. and Arnold, A. P.: Fadrozole,
a potent and specific inhibitor of aromatase in the zebra finch
brain. Gen Comp Endocrinol, 94: 53, 1994
33. Hutchison, J. B., Wingfield, J. C. and Hutchison, R. E.: Sex
differences in plasma concentrations of steroids during the
sensitive period for brain differentiation in the zebra finch.
J Endocrinol, 103: 363, 1984
34. Hofman, M. A. and Swaab, D. F.: The sexually dimorphic nu-
cleus of the preoptic area in the human brain: a comparative
morphometric study. J Anat, 164: 55, 1989
35. Hines, M.: Abnormal sexual development and psychosexual is-
sues. Baillieres Clin Endocrinol Metab, 12: 173, 1998
36. Berenbaum, S. A. and Hines, M.: Early androgens are related to
childhood sex-typed toy preferences. Psychol Sci, 3: 203, 1992
37. Ehrhardt, A. A. and Meyer-Bahlburg, H. F.: Effects of prenatal
sex hormones on gender-related behavior. Science, 211: 1312,
1981
38. Slijper, F. M. E.: Androgens and gender role behaviour in girls
with congenital adrenal hyperplasia (CAH). Prog Brain Res,
61: 417, 1984
39. Money, J., Schwartz, M. and Lewis, V. G.: Adult erotosexual
status and fetal hormonal masculinization and demasculiniza-
tion: 46,XX congenital virilizing adrenal hyperplasia and
46,XY androgen-insensitivity syndrome compared.
Psychoneuroendocrinology, 9: 405, 1984
40. Money, J. and Ehrhardt, A. A.: Gender dimorphic behavior and
fetal sex hormones. Rec Prog Horm Res, 28: 735, 1972
41. Meyer-Bahlburg, H. F. L., Gruen, R. S., New, M. I., Bell, J. J.,
2148 ANDROGEN IMPRINTING OF BRAIN AND INTERSEX DISORDERS
Morishima, A., Shimshi, M. et al: Gender change from female
to male in classical adrenal hyperplasia. Horm Behav, 30: 319,
1996
42. Imperato-McGinley, J., Guerrero, L., Gautier, T. and Peterson,
R. E.: Steroid 5alpha-reductase deficiency in man: imprinting
and the establishment of gender role. AMA Arch Neurol Psych,
77: 333, 1957
43. Farkas, A. and Rosler, A.: Ten years experience with masculin-
izing genitoplasty in male pseudohermaphroditism due to 17
beta-hydroxysteroid dehydrogenase deficiency. Eur J Pediatr,
152: S88, 1993
44. Wilson, J. D., George, F. W. and Griffin, J. E.: The hormonal
control of sexual development. Science, 211: 1278, 1981
45. Masica, D. N., Money, J., Ehrhardt, A. A. and Lewis, V. G.: IQ,
fetal sex hormones and cognitive patterns: studies in the tes-
ticular feminizing syndrome of androgen insensitivity. Johns
Hopkins Med J, 124: 34, 1969
46. Jacobson, C. D.: The characterization, ontogeny and influence of
androgen on the sexually dimorphic nucleus of the preoptic
area. Ph. D. Thesis, University of California, Los Angeles,
California, 1980
47. Olsen, K. L.: Androgen-insensitive rats are defeminised by their
testes. Nature, 279: 238, 1979
48. Herbst, A. L., Ulfelder, H. and Poskanzer, D. C.: Adenocarcinoma
of the vagina. Association of maternal stilbestrol therapy with
tumor appearance in young women. N Engl J Med, 284: 878,
1971
49. Meyer-Bahlburg, H. F. L., Ehrhardt, A. A., Rosen, L. R., Gruen,
R. S., Veridiano, N. P., Vann, F. H. et al: Prenatal estrogens
and the development of homosexual orientation. Dev Psychol,
31: 12, 1995
50. Hines, M. and Shipley, C.: Prenatal exposure to diethylstilbes-
trol (DES) and the development of sexually dimorphic cogni-
tive abilities and cerebral lateralization. Dev Psychol, 20: 81,
1984
51. Lish, J. D., Meyer-Bahlburg, H. F., Ehrhardt, A. A., Travis, B. G.
and Veridiano, N. P.: Prenatal exposure to diethylstilbestrol
(DES): childhood play behavior and adult gender-role behavior
in women. Arch Sex Behav, 21: 423, 1992
52. Bradley, S. J., Oliver, G. D., Chernick, A. B. and Zucker, K. J.:
Experiment of nurture: ablatio penis at 2 months, sex reas-
signment at 7 months, and a psychosexual follow-up in young
adulthood. Pediatrics, 102: e9, 1998
53. MacLusky, N. J. and Naftolin, F.: Sexual differentiation of the
central nervous system. Science, 211: 1294, 1981
54. Weisz, J. and Ward, I. L.: Plasma testosterone and progesterone
titers of pregnant rats, their male and female fetuses and
neonatal offspring. Endocrinology, 106: 306, 1980
55. Merchant-Larios, H.: The onset of testicular differentiation in
the rat: an ultrastructural study. Am J Anat, 145: 319, 1976
56. Resko, J. A., Feder, H. H. and Goy, R. W.: Androgen concentra-
tions in plasma and testis of developing rats. J Endocrinol, 40:
485, 1968
57. Rhees, R. W., Shryne, J. E. and Gorski, R. A.: Termination of the
hormone-sensitive period for differentiation of the sexually
dimorphic nucleus of the preoptic area in male and female
rats. Brain Res Dev Brain Res, 52: 17, 1990
58. Davis, E. C., Shryne, J. E. and Gorski, R. A.: A revised critical
period for the sexual differentiation of the sexually dimorphic
nucleus of the preoptic area in the rat. Neuroendocrinology,
62: 579, 1995
59. Siiteri, P. K. and Wilson, J. D.: Testosterone formation and
metabolism during male sexual differentiation in the human
embryo. J Clin Endocrinol Metab, 38: 113, 1974
60. Mancini, R. E., Vilar, O. and Lavieri, J. C.: Development of
Leydig cells in the normal human testis. Am J Anat, 112: 203,
1963
61. Smail, P. J., Reyes, F. I., Winter, J. S. D. and Faiman, C.: The
fetal hormonal environment and its effect on the morphogen-
esis of the genital system. In: Pediatric Andrology. Edited by
S. J. Kogan and E. S. E. Hafez. The Hague: Martinus Nijhoff,
1981
62. Takagi, S., Yoshida, T., Tsubata, K., Ozaki, H., Fujii, T. K.,
Nomura, Y. et al: Sex differences in fetal gonadotropins and
androgens. J Steroid Biochem, 8: 609, 1977
63. Herruzo, A. J., Mozas, J., Alarcon, J. L., Lopez, J. M., Molina, R.,
Molto, L. et al: Sex differences in serum hormone levels in
umbilical vein blood. Int J Gynaecol Obstet, 41: 37, 1993
64. Rivarola, M. A., Forest, M. G. and Migeon, C. J.: Testosterone,
androstendione and dehydroepiandrosterone in plasma during
pregnancy and at delivery: concentration and protein binding.
J Clin Endocrinol Metab, 28: 34, 1968
65. Forest, M. G., Sizonenko, P. C., Cathiard, A. M. and Bertrand, J.:
Hypophyso-gonadal function in humans during the first year
of life. I. Evidence for testicular activity in early infancy. J Clin
Invest, 53: 819, 1974
66. Money, J., Hampson, J. G. and Hampson, J. L.: An examination
of some basic sexual concepts: the evidence of human her-
maphroditism. Bull Johns Hopkins Hospital, 97: 301, 1955
67. Money, J., Hampson, J. G. and Hampson, J. L.: Imprinting and
the establishment of gender role. AMA Arch Neurol Psych, 77:
333, 1957
68. Recommendations for Treatment: Intersex Infants and Children.
San Francisco: Intersex Society of North America, 1995
69. Diamond, M. and Sigmundson, H. K.: Management of intersex-
uality: guidelines for dealing with persons with ambiguous
genitalia. Arch Pediatr Adolesc Med, 151: 1046, 1997
70. Snyder, H. McC., III: Management of ambiguous genitalia in the
neonate. In: Urologic Surgery in Neonates and Young Infants.
Edited by L. R. King. Philadelphia: W. B. Saunders Co., pp.
346 –385, 1988
71. Aaronson, I. A.: True hermaphroditism. A review of 41 cases
with observations on testicular histology and function. Br J
Urol, 57: 775, 1985
72. Kim, M. H., Gumpel, J. A. and Graff, P.: Pregnancy in a true
hermaphrodite. Obstet Gynecol, 53: 40S, 1979
73. Guaschino, S., Stola, E., Spinillo, A., Marchetti, M. and Aguzzi,
A.: True hermaphroditism: diagnosis and surgical treatment.
Clin Exp Obstet Gynecol, 15: 74, 1988
74. Amice, V., Amice, J., Bercovici, J. P., Riviere, D. and Corolleur,
M. J.: Gonadal tumor and H-Y antigen in 46,XY pure gonadal
dysgenesis. Cancer, 57: 1313, 1986
75. Giwercman, A., von der Maase, H. and Skakkebaek, N. E.:
Epidemiological and clinical aspects of carcinoma in situ of the
testis. Eur Urol, 23: 104, 1993