Lung (2010) 188:229–233 DOI 10.1007/s00408-010-9231-x
Levels of Dihydrotestosterone, Testosterone, Androstenedione, and Estradiol in Canalicular, Saccular, and Alveolar Mouse Lungs Eric Boucher • Pierre R. Provost • Audrey Devillers Yves Tremblay
•
Received: 8 December 2009 / Accepted: 26 January 2010 / Published online: 17 February 2010 Ó Springer Science+Business Media, LLC 2010
Abstract Androgens and estrogens are known regulators of fetal and postnatal lung development, but their levels in the developing lung have never been determined. We present here, for the first time, a gas chromatography-mass spectrometry (GC/MS) quantification of dihydrotestosterone, testosterone, androstenedione, and estradiol in canalicular, saccular, and alveolar stage lungs of both sexes. Testosterone, androstenedione, and estradiol were observed in all the analyzed lung samples from gestation day (GD) 16.5 to postnatal day (PN) 30, totalizing 383 individual mice. Levels of these three steroids decreased between birth and PN 5. In contrast, dihydrotestosterone was detected only in male samples on GD 19.5, PN 0, and PN 30. A significant sex difference was observed for testosterone and androstenedione but not for estradiol. Steroid levels were also determined in skinned hind legs for comparison. Three-way analysis of variance revealed that tissue (lung or leg) had a significant effect on testosterone levels for both sexes, but not on androstenedione and estradiol levels. Low but significant testosterone and androstenedione levels were observed in all the females and in prepubertal male samples. These levels must be E. Boucher P. R. Provost A. Devillers Y. Tremblay (&) Reproduction Axis, Perinatal and Child Health, CHUQ Research Center, 2705 Laurier Boulevard, Rm T-1-49, Quebec City, QC, Canada e-mail:
[email protected] URL: http://www.crchuq.ca E. Boucher P. R. Provost A. Devillers Y. Tremblay Centre de Recherche en Biologie de la Reproduction (CRBR), Laval University, Quebec City, QC, Canada P. R. Provost Y. Tremblay Department of Obstetrics and Gynecology, Faculty of Medicine, Laval University, Quebec City, QC, Canada
sufficient to induce androgen receptor activation, as suggested by our recent report showing the presence of androgen receptor in the nucleus of several lung cells in corresponding developmental ages and sexes. Keywords Lung development Androgens Estrogens Fetus Neonates
Introduction Sex steroids such as androgens and estrogens are wellknown regulators of lung development (reviewed in [1]). Indeed, androgens are known to delay lung maturation and the surge of surfactant production in males compared to females [2, 3]. Androgens have other potential roles in lung development, as demonstrated by their proposed role in branching morphogenesis [4, 5]. In contrast, estrogens accelerate surfactant production [6, 7] and are involved in alveolar formation [8, 9] and morphometry [10]. During gestation the fetal lung is exposed to circulating testosterone and androstenedione originating from fetal and neonatal testes [11], placenta [12], and maternal [13] and fetal [14] ovaries. Circulating androgens could also arise, at least partially, from fetal adrenal precursor, as suggested by Cyp17 expression in the mouse fetal adrenal [15]. Mice are born during the first days of the saccular stage, before the beginning of alveolarization [16]. This differs from human lung development, in which saccularization and a large part of alveolarization take place in utero [16]. Consequently, neonatal mouse lungs cease to be exposed to maternal- and placental-derived androgen precursors after the beginning of the saccular stage. Indeed, circulating androstenedione levels in neonatal mice have been shown to be very low [17], as for testosterone titers in females and prepubertal males.
123
230
Lung (2010) 188:229–233
So far, available data mainly concern circulating and intratesticular concentrations of steroids. They were mostly obtained by radioimmunoassay, a technique that does not present the same level of specificity and sensitivity than more recent approaches. Besides, genes involved in androgen and estrogen synthesis and inactivation were shown to be expressed throughout lung development. Therefore, pulmonary estrogen, androgen, and androgen precursor concentrations are needed for the study of the modulation of androgen action in the developing lung. For these reasons, we present in this study quantification of dihydrotestosterone, testosterone, androstenedione, and estradiol levels by gas chromatography and mass spectrometry (GC/MS) in canalicular, saccular, and alveolar mouse lungs. For comparison purposes, levels in skinned hind legs were also determined.
Determination of Dihydrotestosterone, Testosterone, Androstenedione, and Estradiol Titers in Murine Lungs and Hind Legs Steroid hormones were extracted as previously described [19]. Dihydrotestosterone, testosterone, androstenedione, and estradiol were quantified by gas chromatography and
Method Animals and Tissue Preparation Protocol and procedures were approved by the animal care and use committee and the institutional review board of the Centre de Recherche du Centre Hospitalier Universitaire de Que´bec (CPA-CHUQ; protocols 2005-156 and 2008-071). Animal care, mating, and sacrifice were done as previously described [18]. Fetal/neonatal lungs and skinned hind legs were collected and flash frozen. For each litter, the lungs and skinned hind legs of BALB/c mice were pooled by sex. Up to three litters were pooled to create each biological replicate (Table 1). Except when indicated, a total of three biological replicates for each sex were analyzed for each time point.
Table 1 Number of fetuses/pups used in this study according to development time and sex Sex Replicatea
Male
Female
1
2
3
1
2
3
GD 16.5
18
16
19
GD 17.5
10
9
9
16
15
20
8
16
GD 19.5
5
6
10
9
8
7
PN 0
8
8
8
7
8
11
15
PN 2
5
7
8
7
11
9
PN 5
5
5
4
6
5
4
PN 10
2
b
4
b
3
5
2
b
b
PN 30
2
2b
3b
3
7b
a
2
6b
Male and female fetuses/pups from the same litters were separately pooled to prepare the corresponding replicates b
Hind legs were not collected for those litters
123
Fig. 1 Testosterone (a), androstenedione (b), and estradiol (c) levels in mouse lungs and skinned hind legs from GD 16.5 to PN 30. Levels are reported in nanograms (ng) (testosterone, androstenedione) or picograms (pg) (estradiol) of steroid per gram of wet weight. Values are mean ± SEM of three biological replicates, except for hind leg tissues on PN 10 and PN 30 for which a single pooled sample was used to measure testosterone, androstenedione, and estradiol. In some samples, one or more steroid(s) were not detected
Lung (2010) 188:229–233
231
negative chemical ionization mass spectrometry (GC/MS) as previously reported [20]. Statistics Statistics were performed by three-way analysis of variance (ANOVA) using Systat 12 software (Systat Software, Inc., Chicago, IL, USA). Steroid level was defined as the dependent variable. Developmental age (GD 16.5 through PN 30), tissue measured (lung or hind leg), and sex were defined as potential sources of variance (categorical variables). All possible interactions were tested. Differences were considered significant when P \ 0.05.
Results and Discussion Dihydrotestosterone (data not shown), testosterone (Fig. 1a), androstenedione (Fig. 1b), and estradiol (Fig. 1c) levels were measured by GC/MS in 48 pools of lungs and 40 pools of hind legs, prepared using 383 individual fetuses/pups taken between GD 16.5 and PN 30 (Table 1). Hind legs were chosen as reference tissue. Androgens have been shown to play a role in fetal skin maturation [21, 22], and genes involved in androgen metabolism have been detected in fetal skin [23]. Therefore, skin was removed from the hind legs before freezing. Developmental age was found to be a significant source of variance for testosterone, androstenedione, and estradiol (Fig. 1, Table 2). As such, testosterone and androstenedione levels were found to vary according to a temporal pattern similar to that seen for testosterone production in the developing mouse testis [24, 25]. In fact, testosterone and androstenedione levels in male lungs and hind legs remain elevated until PN 2, after which they decrease to reach those of the female tissues until puberty (PN 30; Fig. 1a, b). Levels in female tissues follow a similar pattern from GD 16.5 to PN 2, although with much less amplitude
and a faster decrease after birth compared to males (Fig. 1a, b). This might be partly attributed to parturition and the subsequent absence of placental- and maternalderived androgens. Contrary to the situation for testosterone, the measured tissue (lung or hind leg) was not found to be a significant source of variance for androstenedione and estradiol levels (Table 2). Indeed, both tissues exhibit a similar temporal pattern for androstenedione and estradiol (Fig. 1b, c). Estradiol levels were similar and followed the same temporal pattern (Fig. 1c, Table 2) as the published fetal and maternal circulating levels for the rat [25] and mouse [12]. Thus, both androstenedione and estradiol levels measured in lungs and hind legs probably reflect circulating levels of both steroids. In contrast, lung testosterone levels consistently showed more variation over developmental age than their hind leg counterparts (Fig. 1a). This observation is supported by statistical analysis, which also suggests that this tissue-specific variation is influenced by sex (Table 2). This agrees with the existence of a lung-specific testosterone metabolism. Indeed, we have previously shown that the developing lung expresses the molecular machinery necessary for the control of androgen synthesis and inactivation throughout the canalicular, saccular, and alveolar stages of lung development [18, 26–30]. Local modulation of androgen action could take place at the cellular level, as evidenced by colocalization of 17b-HSD 2 and AR [18, 26, 30]. Furthermore, AR shows nuclear localization in several lung cells for both sexes [18, 26, 30], even though lung testosterone levels show a major sex difference. Antenatal and postnatal mouse lungs are also known to express 5a-reductase, which converts testosterone into dihydrotestosterone, the most potent androgen, and 3a-hydroxysteroid dehydrogenase, which inactivates dihydrotestosterone into 5a-androstane-3a,17b-diol [18, 31]. Significant levels of dihydrotestosterone were detected only in some samples, the lungs and hind legs of males collected on GD 19.5, PN 0, and PN 30. Levels of
Table 2 Analysis of variance in steroid levels according to the developmental age, sex, and tissue variables Sources of variance
Testosterone a
Age Tissue Sex
Androstenedione
F ratio
P value
6.835 5.711
