[PDF] Expression and Distribution of Androgen-Binding Protein/Sex

View - Download Expression and Distribution of Androgen-Binding Protein/Sex

Previous PDF

Next PDF

BIOLOGY OF REPRODUCTION 56, 14-20 (1997)

Expression and Distribution of Androgen-Binding Protein/Sex Hormone-Binding Globulin in the Female Rodent Reproductive System'

David R. Joseph,

2 4

Stephen G.A. Power,

34
and Peter Petrusz s

Departments of Pediatrics

4 and Anatomy and Cell Biology,

North Carolina, Chapel Hill, North Carolina 27599

ABSTRACT

Androgen-binding protein (ABP)/sex hormone-binding glob- ulin gene expression has been described in the rat testicular Sertoli cell and brain. The extracellular protein is thought to regulate the bioavailability of sex steroids, but may have a more complex function as a hormone or growth factor. Transgenic mice were developed with a 5.5-kilobase (kb) rat DNA fragment containing the ABP gene with all 8 exon sequences and 1.5 kb upstream of the transcription start site. Expression of the gene was observed in the testis and brain, but not in other examined tissues of the transgenic mice. In this paper we describe ABP gene expression in ovaries of transgenic mice that contain the rat gene; a lower level of ABP mRNA was also detected in the transgenic uterus. Northern blot analysis also detected ABP mRNA in rat ovary. The hybridizing species in the rat and trans- genic mouse ovaries and uteri were the size of testicular ABP mRNA (1.7 kb). Except in the transgenic mouse brain, there was no detectable hybridizing RNA in the other transgenic tissues examined. The plasma, ovary, and uterus of the transgenic mice all contained elevated ABP (dihydrotestosterone [DHT]-binding)

activities as compared to those of wild-type littermates; otherwild-type and transgenic tissues were negative for DHT binding.

Immunohistochemistry revealed increased immunoreactivity in the transgenic oviduct and uterus, but not the ovary. In the ovi- duct, the intense immunoreactivity was associated with the ep- ithelium, whereas in the uterus it was primarily associated with the luminal epithelium and glands. Phenotypic abnormalities of the homozygous transgenic mice included reduced fecundity re- sulting in small litters. We conclude that ABP may function in the female reproductive system to increase the local concentra- tions of sex steroids or to sequester them in key target organs. Studies in the female will aid in elucidating the functions of ABP in male and female reproduction.

INTRODUCTION

Androgen-binding protein (ABP) and sex hormone-bind- ing globulin (SHBG) are extracellular proteins that bind dihydrotestosterone (DHT), testosterone, and estradiol with high affinity [1-4]. They are encoded by a single gene and share the same amino acid sequence, but they differ in their sites of synthesis [5, 6]. ABP is produced by the Sertoli cells of the testicular seminiferous tubules and is secreted into the luminal fluid [7, 8]. After transport to the epidid- ymis, it is internalized by the caput epithelium [9, 10] by what is thought to be a receptor-mediated process [11-13].

Accepted August 14, 1996.

Received March 7, 1996.

'This work was supported by PHS grants RO1-HD21744 (PI, David Joseph), 5-P30-HD-18968 (PI, Frank S. French, The Laboratories for Re- productive Biology), NIH contract 263-MD-314558 (PI, Frank S. French and Peter Petrusz), and a grant from the Andrew Mellon Foundation. 'Correspondence and current address: David

Joseph, Applied Genetics

Laboratories, BDI, The University of Florida, 12085 Research Drive, Ala- chua, FL 32615. FAX: (904) 462-0875. 'Current address: Department of Obstetrics and Gynecology, The Uni- versity of Western Ontario, London, ON, Canada. The Laboratories for Reproductive Biology, University of Although the best-characterized testicular binding protein is rat ABP, the male mouse reproductive system also has an ABP, but at a level 2-4% of that in the rat [14]. The liver of most adult animals, including humans, secretes SHBG into the blood, where SHBG circulates as the major sex steroid-binding protein [2, 3]. In the mouse and rat, the adult liver does not produce SHBG, but the fetal rat liver does synthesize and secrete SHBG [15, 16]. Also, in the rat, the ABP/SHBG gene is expressed in the adult brain [17] and embryo [18]. The functions of the ABP/SHBG gene products remain poorly understood [1-4, 19]. SHBG is thought to regulate the bioavailability of circulating sex steroids; one theory proposes that only free steroids are available for uptake by cells. In the male reproductive tract, ABP is thought to regulate spermato- genesis and sperm maturation [20, 21], but its actual function is unknown. The identification in several tissues of ABP/ SHBG membrane-binding proteins that have the binding properties of a receptor suggests that these proteins have a much broader function [11, 22-24], possibly as hormones or growth factors [3]. Supporting this idea, two laboratories have demonstrated increases in intracellular cAMP levels in re- sponse to SHBG administration [25-27].

To study the rat ABP/SHBG

gene, transgenic mice that express the rat gene were developed [28]. A 5.5-kilobase (kb) genomic DNA fragment, containing 1.5 kb upstream of the transcription start site and all 8 coding exons, was found to be capable of directing the tissue-specific gene expression in the testis. There were increased levels of ABP in the testis, epididymis, and blood of the transgenic mice- levels associated with a loss of male fertility [29, 30]. We demonstrate here that the transgenic female mouse ovary and uterus also express the ABP gene. Moreover, the ex- periments suggest that an excess of ABP in the female re- productive system leads to reduced fecundity. We will refer to the female transgenic ABP/SHBG gene product as ABP

MATERIALS AND METHODS

Animal Housing and Breeding

Mice were housed in the UNC

Division of

Laboratory

Animal Medicine under controlled conditions and given food and water ad libitum. Mice were killed with carbon dioxide gas before collection of tissues. All protocols were approved by the Animal Care and Use Committee and con- form to the PHS Animal Welfare Assurance Policy. Initially the line 24 [28] transgenic mice were propagated by mat- ings of the hemizygotes with C57BL6/6J x DBA2/2J Fl mice (Jackson Labs., Bar Harbor, ME). Subsequently, the ABP transgenic line was propagated by crossing hemizy- gotes or mating young homozygous males with hemizygous females. Transgenic mice were identified by hybridization blot analysis of tail DNA or by plasma DHT-binding as- says. Homozygous transgenic mice were identified by quantitative hybridization as described below. Their ho-

14Downloaded from https://academic.oup.com/biolreprod/article/56/1/14/2760648 by guest on 21 September 2023

ABP IN FEMALE REPRODUCTION

mozygosity was confirmed by breeding experiments, which yielded the predicted genotypes and phenotypes. There was a perfect correlation of the homozygous genotype with a motor disorder that is described below. Nontransgenic mice are referred to as wild type. Wild-type mice used for com- parisons in these experiments were littermates of transgenic mice derived from hemizygous crosses.

Hybridization Blot Analyses

DNA hybridization was used to identify the transgenic mice. DNA was isolated from mouse tails as previously described [28], sheared by forcing the solution through a

22-gauge needle, and purified by column chromatography

(Elutip; Schleicher and Schuell Co., Keene, NH). DNA samples were alkali denatured, neutralized, and transferred to a nitrocellulose membrane using a blot filtration mani- fold. After hybridization [31] with ABP cDNA [32], the signals were visualized by autoradiography. Visual inspec- tion or densitometric analysis of the autoradiogram deter- mined which mice were wild type, hemizygotes, and ho- mozygotes. The density of the autoradiographic DNA sig- nal from the homozygote DNA was approximately twice that of the hemizygote DNA signal. Wild-type mouse ABP

DNA was not detectable under these conditions.

Northern blot analyses were done as previously de- scribed [33]. Briefly, total tissue RNA was isolated by gua- nidine isothiocyanate-CsC1 ultracentrifugation. After quan- titation by UV absorbance measurements, aliquots of de- natured RNA were fractionated by agarose gel electropho- resis and stained with ethidium bromide. Equal intensity of fluorescence of each sample ensured that equal amounts of RNA from each tissue would be compared in the Northern blot [34]. The total RNA or poly(A) RNA was denatured with glyoxal and fractionated by agarose gel electrophoresis [35]. After transfer to a nylon membrane, ABP mRNA was visualized by hybridization with ABP cDNA and autora- diography [33].

Immunohistochemistry

Tissues were removed, fixed with Bouin's solution, em- bedded in paraffin, and sectioned (10 pLm) for histochemistry. Sections were stained with hematoxylin and eosin to evaluate the morphology. For immunohistochemistry, the paraffin-em- bedded tissues were sectioned and deparaffinized. In some experiments, sections were treated with trypsin to help expose the antigen (0.25 mg/ml trypsin for 10 min at room temper- ature). After incubation with the primary rabbit antiserum and the secondary antiserum (affinity-purified sheep anti-rabbit IgG), immunoreactive ABP was visualized by the double per- oxidase antiperoxidase complex method [36]. The peroxidase reaction was done with 0.05 M Tris buffer, pH 7.6, containing

0.002% hydrogen peroxide and saturated diaminobenzidine.

The sections were counterstained with toluidine blue, dehy- drated with ethanol and xylene, coverslipped, and analyzed by microscopy. Sections stained without primary antiserum were used as negative controls. One primary rabbit antiserum (#244) was raised against the purified rat ABP and has been demonstrated by a number of criteria to be specific [9]. The other primary antiserum was raised against an ABP- derived peptide (#654; amino acid residues 88-105,

GDTNTEDDWFMLGLRDGQ) and functioned specifically

for immunohistochemistry [16]. In these experiments, speci- ficity of the immunoreactivity was tested by preadsorption of the 654 antiserum with the peptide immunogen (10 tpM); blocking of the signal demonstrated specificity. Both antisera FIG. 1. Northern blot analysis of female transgenic tissues. Denatured total RNA (10 pig/lane) was fractionated by agarose gel electrophoresis, transferred to a nylon membrane, and hybridized with ABP cDNA. Lane

1, wild-type brain; lane 2, wild-type liver; lane 3, wild-type lung; lane 4,

wild-type ovary; lane 5, wild-type uterus; lane 6, wild-type kidney; lane

7, transgenic brain; lane 8, transgenic liver; lane 9, transgenic lung; lane

10, transgenic ovary; lane 11, transgenic uterus; lane 12, transgenic kid-

ney; lane 13, no RNA; lane 14, transgenic testis. reacted with rat and mouse ABP, which share 89% of their amino acid residues; but antiserum 244 produced a much stronger signal-to-noise ratio than peptide 654 antiserum. The hCG antiserum was raised in this laboratory and was previ- ously characterized for immunohistochemistry [37].

DHT-Binding Assays

The preparation of tissue extracts and the DHT-binding assay for ABP have been previously described [16, 38]. Briefly, tissue extracts or plasma samples were incubated with 6 nM [ 3

H]DHT or [

3

H]DHT and a 100-fold excess of

unlabeled DHT for 1 h at 0°C. After charcoal treatment for

2 min to remove unbound DHT, the radioactivity in the

supernatant fluid was determined. Under these conditions, rat ABP is 90% saturated. Specific binding was calculated as cpm (labeled DHT) -cpm (labeled DHT plus unlabeled DHT). Mean values and standard deviations () were cal- culated by standard methods. There were 6-8 animals in each group. Student's t-test was used to determine whether differences in the mean values ( standard deviation) were significant. DHT-binding values, expressed as ng ABP, were based on the assumption of one DHT-binding site per mo- lecular dimer (Mr 90 000).

RESULTS

Tissue-Specific ABP/SHBG Gene Expression in Female

Transgenic Mice

A previous study demonstrated that the transgenic male mice contained detectable rat ABP mRNA in the testis, but not the kidney, spleen, or liver [28]. In a later study, North- ern blot hybridization with ABP cDNA as probe revealed that the line 24 transgenic male brain also contained hybrid- izing ABP-specific mRNA, which was longer and more het- erogeneous than the normal 1.7-kb mRNA (unpublished re- sults). Similarly, expression of the rat ABP gene was ob- served in the brain of transgenic female mice. Figure 1 (lane

7) demonstrates the presence of ABP mRNA in total RNA

15

Downloaded from https://academic.oup.com/biolreprod/article/56/1/14/2760648 by guest on 21 September 2023

JOSEPH ET AL.

FIG. 2. Northern blot analysis of ABP mRNA in rat ovaries. Rat poly(A) RNA (3 Rpg/lane) was analyzed with ABP cDNA as described for Figure

1. Lane 1, 14-day-old Sprague-Dawley rat; lane 2, 24-day-old rat; lane 3,

30-day-old rat; lane 4, blank; lane 5, adult rat; lane 6, adult rat; lane 7,

no RNA; lane 8, 26-day-old rat testis. of female transgenic mouse brain, but not in transgenic liver, lung, kidney (Fig. 1, lanes 8, 9, and 12), or spleen RNA (not shown). There was also no detectable signal in wild-type brain, liver, ovary, uterus, lung, or kidney RNA (Fig. 1, lanes

1-6). Unlike transgenic testicular ABP mRNA (Fig. 1, lane

14), brain RNA contained no detectable 1.7-kb mRNA

(which encodes ABP). ABP cDNA hybridized with a het- erogeneous brain RNA population greater than 1.7 kb, a pop- ulation not known to be capable of encoding active ABP (Fig. 1, lane 7). Hybridizing transgenic testicular mRNA contains a similar heterogenous ABP mRNA population and the 1.7-kb mRNA (Fig. 1, lane 14). Interestingly, ABP mRNA was also detected in total RNA from the transgenic ovary, but not wild-type ovary (Fig. 1, lanes 10 and 4). In contrast to the hybridizing testicular and brain RNA, most of the detectable ABP RNA is processed to the active 1.7-kb form. A faint band of reactivity was also observed in trans- genic uterine RNA (Fig. 1, lane 11). The experiments described above demonstrated that the rat ABP gene was expressed in the female transgenic mice. To determine whether the endogenous ABP gene was ex- pressed in female rats, ovaries were examined for the pres- ence of ABP mRNA. Hybridization of Northern blots with ABP cDNA revealed the presence of ABP mRNA in ovar- ian poly(A) RNA from rats of various ages (Fig. 2). The level of hybridizing RNA gradually increased from 14 days of age to adulthood. Similar results were obtained with an analysis of total rat ovarian RNA from various ages (data not shown). As seen with the poly(A) RNAs, the ABP mRNA level in total RNAs also increased until adulthood.

Morphology of Transgenic Tissues

Tissues representing wild-type and transgenic (hemizy- gous and homozygous) female mice of ages 3 mo to 2 yr were examined by light microscopy after staining with he- matoxylin and eosin. Tissues included ovary, oviduct, uter- us, brain, spinal cord, lung, liver, pituitary, adrenal, kidney, lymph node, thymus, spleen, pancreas, salivary gland, skin, bone, cartilage, striated muscle, thyroid, stomach, and small intestine. The structures of all the transgenic female tissues examined appeared to be the same as in the wild-type lit- termates (data not shown). The pituitary sections from fe- male transgenic mice of various ages (3 mo to 1.0 yr) were immunostained with hCG antisera. Antibody treatment spe- cifically stained the gonadotrophs, which appeared to have normal distribution, structure, and gonadotropin content (data not shown). There was no obvious enlargement and vacuolation, which would have indicated hyperactivity. Immunocytochemical Localization of ABP in Transgenic

Female Tissues

To determine the cellular location of immunoreactive ABP, the transgenic tissues were examined by immunohis- tochemistry. Two antisera were used to identify the ABP- containing structures. Antiserum 244 was raised against pu- rified rat epididymal ABP [9], and antiserum 654 was raised against a rat ABP-derived peptide [16]. Intense immunoreac- tivity in a transgenic tissue that was not present in the wild- type mouse tissue indicated transgenic ABP Furthermore, specificity was assessed through demonstration of a lack of reactivity with normal rabbit serum and with antiserum 654 preadsorbed with the peptide immunogen. The oviductal and uterine ABP immunoreactivities presented below were dem- onstrated to be specific by these criteria. No specific immunoreactivity was detected in the wild- type female reproductive tissues. In the ovary (data not shown), oviduct (Fig. 3A), and uterus (Fig. 3B), only a very low level of apparently nonspecific staining was observed. In the wild-type and transgenic ovaries there was no de- tectable difference between the ABP immunoreactivities. Figure 3C illustrates the lack of immunoreactivity in the transgenic ovary. Similarly, the rat ovary contained no de- tectable specific immunoreactivity in any region (data not shown). It should be pointed out that the stroma and muscle in many sections of rat and mouse tissues yielded light staining with both ABP antisera and normal rabbit serum; this reactivity with antiserum 654 was not blocked with the FIG. 3. Immunohistochemical analysis of female transgenic reproduc- tive tissues. Transgenic mouse tissue sections were reacted with ABP an- tiserum 244 (without trypsin treatment), and the immunoreactivity was visualized by the double peroxidase antiperoxidase complex method with diaminobenzidine as substrate (brown reaction product). The counterstain was with toluidine blue. A) Wild-type oviduct. x50. Note the lack of staining in the epithelium. B) Wild-type uterus. x50. There is little im- munoreactivity in any region, including the small glands (arrowheads) and lumen (asterisk). C) Transgenic ovary. x32. Note the lack of immuno- reactivity in the follicles (arrowheads), but, adjacent to the ovary, the in- tense immunoreactivity in the degenerating mesonephric ducts (arrows). D) Proximal transgenic oviduct. x125. There are large amounts of im- munoreactivity in some epithelial cells in folds of the mucosa (arrows). The arrowheads point to vacuoles in the epithelium. E) Transgenic ovi- duct. x100. The glandular-like structures in the epithelium contain in- tense immunoreactivity (arrowheads). F) Transgenic oviduct. x250. Note the intense immunoreactivity in the secretory cells (arrows). G) Transgenic oviduct and proximal uterus. x50. There is immunoreactivity in some epithelial cells along the length of the oviduct in the epithelium and lu- men of the distal oviduct as it enters the uterus (arrows). The arrowheads point to immunoreactivity in the proximal oviduct epithelium. H) High- power magnification (125) of transgenic distal oviduct in G. Note the intense immunoreactivity in the lumen. I) Transgenic uterus. x125. Note staining in the lumen of the small glands (arrow). J) Transgenic uterus. x80. There is intense immunoreactivity in the epithelium and lumen of the uterine glands (arrowheads). K) Transgenic uterine endometrium. x 125. Note immunoreactivity in the supranuclear region of the glandular (arrows) and luminal epithelium. Asterisk, uterine lumen. L) Transgenic intestinal lymphoid tumor. x312. Approximately 1-2% of the cells con- tain immunoreactivity in the cytoplasm. 16

Downloaded from https://academic.oup.com/biolreprod/article/56/1/14/2760648 by guest on 21 September 2023

ABP IN FEMALE REPRODUCTION 17

Downloaded from https://academic.oup.com/biolreprod/article/56/1/14/2760648 by guest on 21 September 2023

JOSEPH ET AL.

peptide. Also, Figure 3C demonstrates intense ABP im- munoreactivity in the epithelial cells of the degenerating mesonephric ducts in a transgenic mouse. This duct system was not observed in the wild-type mouse sections, appar- ently because of random sampling (i.e., no effort was made to locate the wild-type ducts). In contrast to observations in the wild-type mouse ovi- duct and uterus, intense specific ABP immunoreactivity was found in the oviduct and uterus of transgenic mice. Figure 3, D-F, demonstrates this reactivity in various regions of the oviduct. In each region, ABP staining was present in the cytoplasm of some epithelial cells and absent in other cells. Immunoreactivity was particularly intense in small gland-like structures in folds of the mucosa (Fig. 3E, arrows). In oviducts of some animals, only the secretory epithelial cells (peg cells) and not the ciliated cells con- tained immunoreactive ABP (Fig. 3F). As seen in Figure

3D, many of the epithelial cells in normal and transgenic

oviducts contained large vacuoles. These vacuoles were as- sociated with immunoreactive cells as well as nonreactive cells and were also present in the wild-type oviduct, es- pecially in mice older than 6 mo. The transgenic uterus also contained significant immu- noteactivity. Interestingly, the epithelium in the segment of the oviduct entering the uterus contained intense ABP stain- ing (Fig. 3G). Moreover, at higher magnification it can be seen that the lumen in this region contains large amounts of immunoreactivity (Fig. 3H). Also, in other regions of the uterus, the epithelium of glands stained intensely in the ep- ithelium and the lumen (Fig. 3, I and J). Figure 3K dem- onstrates that some regions of the transgenic uterine epithe- lium contained ABP immunoreactivity, which appeared to be primarily supranuclear. In these regions of the endome- trium there was no visible staining in the lumen. The two patterns of staining (i.e., small glands or supranuclear) were not observed in the same uterus (i.e., either one or the other was present in a given animal), suggesting that the physio- logical state may affect the ABP location. The effects of changing hormone levels during the estrous cycle on the location of ABP immunoreactivity were not investigated. Al- though the wild-type rat uterus did not yield ABP immu- noreactivity with the standard methods, trypsin treatment of the tissue slices did reveal ABP immunoreactivity in the glands that was similar to immunoreactivity in the transgenic mice (data not shown). There was no visible immunoreac- tivity in the rat oviduct with or without trypsin treatment. Several other transgenic and wild-type tissues were also examined for cellular immunoreactivity with antiserum 244 (data not shown). No nonreproductive tissues were observed to contain dramatic increases in specific immunoreactivity in the transgenic mice. However, the liver of the normal and transgenic animals had reactivity in the cytoplasm of the hepatocytes. In the liver, immunoreactivity was apparently more intense in the transgenic than in the wild-type tissues, but the difference was minor compared to differences in the oviduct and uterus. Other examined tissues, except for those of the central nervous system, were negative. These tissues included the adrenal gland, kidney, lymph node, thymus, spleen, salivary gland, smooth and striated muscle, cardiac muscle, pancreas, stomach, salivary gland, anterior and pos- terior pituitary, and cartilage. The central nervous system immunoreactivity in the transgenic mice (unpublished re- sults) will be described in a separate publication.

Plasma and Tissue DHT-Binding Activity

To determine the tissue location of rat ABP, plasma and extracts of various tissues from wild-type and hemizygous transgenic mice were assayed for DHT-binding activity. In the female transgenic mice, ABP DHT-binding activity was ele- vated in the ovary (373 + 39 ng ABP/g tissue) and uterus (290

54 ng ABP/g tissue), but not in any other tissue ex-

amined, including liver, brain, kidney, and muscle ( 30 ng ABP/g). Small amounts of activity were present in the wild- type ovary (68 15 ng ABP/g) and uterus (95 27 ng ABP/g). Likewise, the plasma levels of DHT-binding activity were elevated in the female hemizygous transgenic mice (228 + 28 ng ABP/ml) as compared to wild-type female mice (<

30 ng ABP/ml). Homozygous female mouse plasma con-

tained approximately twice the level of DHT binding (424 +

55 ng ABP/ml) as the hemizygous mouse plasma.

Phenotypic Characteristics

The hemizygous female transgenic mice had no obvious phenotypic differences from the wild-type mice. The exter- nal appearances, walking characteristics, weight, reproduc- tive potential, and life expectancies of the hemizygous fe- male mice appeared normal. Although the longevity of the homozygous female mice appeared normal (greater than 2 yr), they differed from the wild-type and hemizygous mice in several other phenotypic characteristics. Differences were observed in the weights of the mice. From 50 to 150 days of age, the weights of the homozygous female trans- genic mice (21 + 2.6 g) were not significantly different from those of wild-type mice (24 ± 2.1 g, p = 0.07); how- ever, after this age, the differences increased with increas- ing age. Over 200 days, wild-type (30 5.9 g) and hem- izygous mice (30 + 5.5 g) had the same weight, whereas the homozygous females were clearly smaller than wild- type or hemizygous mice (25 1.9 g, p = 0.027 and p =

2.7 x 10

-5 , respectively). The most obvious phenotypic difference between the ho- mozygous transgenic mice and the wild-type mice was in their walking characteristics. The male and female homo- zygous mice exhibited a serious gait disorder. Both sexes walked with high stepping and shaking motions in their hind limbs. They were able to move forward with difficulty on sawdust in the cage, but had extreme difficulty moving forward on a hard surface. This defect became noticeable at about 20 days of age and reached full expression at 50-

60 days.

quotesdbs_dbs24.pdfusesText_30

[PDF] androgène

[PDF] inhibine

[PDF] fsh

[PDF] sertoli

[PDF] testosterone

[PDF] cellules de sertoli

[PDF] discours abraham lincoln 1860

[PDF] le pouvoir des mots citations

[PDF] le pouvoir des mots livre

[PDF] pyramide de maslow motivation

[PDF] maslow pdf

[PDF] pyramide de maslow enfance

[PDF] pyramide des besoins de maslow marketing

[PDF] pyramide de maslow critique

[PDF] pyramide de maslow exemple concret