CHARACTERIZATION OF LEYDIG CELL
DEVELOPMENT IN THERAT TESTIS
by
Juan Zhai
Submitted to the Faculty of the School of Graduate Studies
of the Medical College of Georgia in Partial Fulfillment
of the Requirements for the Degree of
Doctor of Philosophy
MAY, 1996 CHARACTERIZATION OF LEYDIG CELL DEVELOPMENT
IN THERAT TESTIS
This dissertation is submitted by Juan Zhai and has been examined and approved by an appointed committee of the faculty of the School of Graduate
Studies of the Medical College of Georgia.
The signatures which appear below verify the fact that all required changes have been incorporated and that the dissertation has received final approval with reference to content, fonn and accuracy of presentation.
This dissertation is therefore accepted in partial fulfillment of the requirements for the degree ofDoctor ofPhilosophy.
Date
' ACKNOWLEDGMENTS
I would like to acknowledge and express my appreciation to my major advisor Dr. Tom Abney and my advisory committee, Drs. Virendra Mahesh, Tom Mills, William Hill and Ken Lanclos, for their support and advice.
The author is appreciative of the advice and help from Drs. Darrell Brann, Roni Bollag and Lawrence Hendry.
A special thanks is extended to my frie~ds Chris Reilly and Edward Pan for their support and being there to listen to all my complaints.
Thanks are extended to Angela Foreman and Gene Canady for their friendship and help.
lll DEDICATION
This dissertation is dedicated to my father, Shouyi Zhai, whose love and wisdom had made a great contribution to my life and career.
IV TABLE OF CONTENTS
Page Introduction and specific aims 1 Background 5 1 Steroidogenic pathway in Leydig cells 6 Ontogenic development of Leydig cells 9 Regulation ofLeydig cell function 17 Ethylene dimethanesulphonate (EDS)-treated rats An animal model to study Leydig cell development and function 33 Materials and Methods 38 Animals and treatment 3 8 Hormones and chemicals 38 Buffers and solutions 42 Serum testosterone determination 44 Cell isolation 45 Histochemical study (313-HSD staining) 46 In vitro incubation and testosterone production 46 RNA extraction 47 Preparation of eDNA probes 48 Northern blot analysis 50 Estrogen receptor binding assay 51 Protein determination 52 Immunohistochemical study on the testicular tissue 52 Immunohistochemical study on the isolated precursor cells and Leydig cells 54 Reverse transcription-polymerase chain reaction (RT -PCR) 55 Gel electrophoresis ofRT-PCR products 57 Southern blotting and quantitative analysis 57 Flow cytometer analysis 58 Comparative study of estrogen receptor levels in the precursor cells and Leydig cells by immunofluorescent analysis 59 Statistical analysis 60 Results 62 Characterization of Leydig cell degeneration and regeneration after EDS treatment 62 Histochemical analysis of the precursor cell and Leydig cell fraction by 313-hydroxysteroid dehydrogenase (3 13-HSD) staining 64 Analysis of the precursor cell and Leydig cell fractions from the control and EDS-treated rats by flow cytometry 67 Characterization of in vitro testosterone production by Leydig cells and precursor cells during differentiation 74
LH receptor and steroidogenic enzyme (P-450.,. and P-45017a)
v mRNA changes during the differentiation process 80 Estrogen receptor binding in the testis of the control and EDS-treated rats 90 Immunohistochemical studies of estrogen receptor in the testis of the control and EDS treated rats 95 Immunohistochemical detection of estrogen receptor in the isolated Leydig cells and precursor cells 95 Detection of estrogen receptor mRNA in Leydig cells and precursor cells by reverse transcription-polymerase chain reaction 99 Comparative study of estrogen receptor levels in the precursor cells and Leydig cells by immunofluorescent analysis 109 Regulation of precursor Leydig cell differentiation by LH/hCG and other possible factors Ill - Discussion 118 Characterization ofLeydig cell degeneration and regeneration after EDS treatment 118 Histochemical studies of the precursor cell and Leydig cell fraction by 3 ~-hydroxysteroid dehydrogenase (3 ~-HSD) staining 124 Analysis of the precursor cell and Leydig cell fractions from the control and EDS-treated rats by flow cytometry 127 Characterization of in vitro testosterone production by Leydig cells and precursor cells during differentiation 131
LH receptor and steroidogenic enzyme (P-450"' and P-45017a) message RNA changes during the differentiation process 135 Estrogen receptor and its mRNA levels in the testis and in the isolated precursor cells and Leydig cells 140 Regulation of precursor Leydig cell differentiation by LH!hCG and other possible factors 143 Summary 146 Literature cited 148
VI LIST OF FIGURES
Page Figure I. Steroidogenic pathway for the synthesis of testosterone 7
Figure2. Amino acid sequence, orientation and proposed topology of the rat LH/hCG receptor in the plasma membrane 20
Figure 3. Genomic organization of the rat LH/hCG receptor 22
Figure 4. Gel electrophoresis of tlie LH receptor plasmid digestion 49
Figure 5. Genomic structure of the rat estrogen receptor 56
Figure 6. Temporal changes in serum testosterone levels after a single intraperitoneal injection ofEDS 63
Figure 7. The influence of a single injection ofEDS on testicular weight 65
Figure 8. The effect of a single intraperitoneal injection ofEDS on the interstitial cell number 66
Figure 9. Histochemical study of the Leydig cell and precursor cell fractions by 3f3-HSD staining 68
Figure 10. Cell size of the Leydig cell and precursor cell fractions in the control and day 20 post-EDS rats analyzed by flow cytometry 69
Figure 11. Granularity of the Leydig cell and precursor cell fractions by flow cytometry 71
Figure 12. Dio fluorescence, indicative of mitochondrial activity/number, in the Leydig cell and precursor cell populations by flow cytometry 72
Figure 13. Nile red fluorescence, indicative of lipid droplets content, in the Leydig cell and precursor cell fractions by flow cytometry 73
Figure 14. In vitro testosterone production by Leydig cells
vii in the control and EDS-treated rats 75
Figure 15. In vitro testosterone production by precursor cells in the control and EDS-treated rats 76
Figure 16. Changes of the cell number in the Leydig cell fraction after EDS treatment 78
Figure 17. Changes of the cell number in the precursor cell fraction after EDS treatment 79
Figure 18. Gel electrophoresis of the total RNA extracted from Leydig cells and precursor cells 81
Figure 19. Northern blot analysis of the LH receptor in the control Leydig cells and precursor cells and in the precursor cells after EDS treatment Figure A. 83 Figure B. 84 Figure C. 85 Figure D. 86
Figure 20. Northern blot analysis ofP-450css in the control Leydig cells and precursor cells and in the precursor cells after EDS treatment 88
Figure 21. Northern blot analysis ofP-45017., in the control Leydig cells and precursor cells and in the precursor cells after EDS treatment 89
Figure22. Northern blot analysis ofLH receptor, P-450css and P-45017amRNA in the Leydig cells of the control and day 36 and 60 post-EDS rats Figure A. 91 Figure B. 92
Figure 23. Changes in the specific estrogen binding capacity of rat testicular tissue after a single intraperitoneal injection ofEDS 93
Figure 24. Representative Scatchard plot analysis of 3H-estradiol binding by control and day 16 post-EDS treated rat testicular cytosols 94
viii Figure 25. Localization of the testicular estrogen receptor by immunohistochemical analysis in the controls and day 10 EDS-treated rats Figures A and B. 96 Figures C and D. 97 Figures E and F. 98
Figure26. Detection of the estrogen receptor by immunohistochemical analysis in the Leydig cell and precursor cell fractions from the controls and day 10 EDS-treated rats 100
Figure 27. RT-PCR detection of estrogen receptor mRNA in Leydig cells and precursor cells from controls and precursor cells from EDS-treated animals 101
Figure 28. Specificity of the RT-PCR products amplified by the primer sets 103
Figure 29. Gel electrophoresis of the coamplification ofrat estrogen receptor and rabbit P -globin cDNAs 104
Figure 30. Quality control of the RT-PCR Figures A and B 106 Figure C 107
Figure 31. Southern blot and quantitative analysis of the RT-PCR products generated from precursor cells of the control and EDS-treated rats 108
Figure 32. Southern blot and quantitative analysis of the RT-PCR products generated from Leydig cells of the control and EDS-treated rats 110
Figure 33. Comparative study of the estrogen receptor levels in Leydig cells and precursor cells of the control and day 10 post-EDS rats 112
Figure 34. Average gray value, indicative of immunofluorescence of estrogen receptor, in Leydig cell and precursor cell populations 114
Figure 35. Stimulation of testosterone production by hCG in the precursor cells of the immature and day 20 post-EDS rats 116
ix Figure 36. Effects ofhCG or/and IGF-I on the in vitro testosterone production by precursor cells at day 20 post-EDS treatment 117
X LIST OF TABLES
Page Table I. Comparison of estrogen receptor levels by immunofluorescent study in Leydig cells and precursor cells of the control and day 10 post-EDS rats 113
Table 2. Summary of precursor cell differentiation and Leydig cell regeneration after EDS treatment 145
XI 2 factor in Leydig cell development and function, evidence has accumulated that local factors play an important role, either in conjunction with or dependent upon LH.
In recent years, the ethylene dimethanesulphonate (EDS)-treated rat has become a useful animal model for studying Leydig cell development and function. EDS, an alkylating agent, rapidly and selectively destroys mature Leydig cells in the rat testis. A single intraperitoneal injection ofEDS results in a complete destruction of Leydig cells within two days. However, Leydig cell regeneration occurs within two to three weeks after EDS treatment. The regeneration of Leydig cells in the testis is complete by five to seven weeks after EDS treatment. The regeneration of Leydig cells after EDS treatment is primarily dependent upon high levels of serum LH. In addition, previous studies have shown that daily treatment of estradiol suppressed the regeneration of Leydig cells in the EDS-treated rats.
It is widely accepted now that the regenerated Leydig cells after EDS treatment are derived from the differentiation of a precursor cell population in the interstitial tissue.
However, very little is known about these Leydig precursor cells. Several candidates for ' precursor cells have been proposed by different investigators, of which mesenchymal cells appear to be the leading candidate. There is still a lack of understanding as to how the relationship between precursor cells and mature Leydig cells in the testicular interstitium is maintained in adult rats and what triggers the precursor cells to differentiate after EDS treatment. Knowledge concerning the morphological and biochemical characteristics of the precursor cells is very limited. The differentiation process of precursor cells is not well understood at this time and warrants further investigation. Based upon data from previous studies, we proposed the following hypothesis. 3
Hypothesis: Both Leydig cell and precursor cell populations are present iri the interstitial tissue of the mature rat testis. Leydig cells possess LH receptor and steroidogenic function, while precursor cells manifest different biochemical features and remain undifferentiated. The equilibrium between these two cell populations is maintained by LH and local factors. After Leydig cells are destroyed by EDS treatment, the equilibrium is altered and precursor cells begin to differentiate into Leydig cells. During the differentiation process, precursor cells acquire the morphological and functional characteristics of Leydig cells. A new equilibrium is established when Leydig cell regeneration is accomplished.
To test this hypothesis, five specific aims were designed and investigated in the present study.
Specific aims:
1. Examine the development of testosterone production and hCG responsiveness
in the Leydig cell and precursor cell populations after EDS treatment.
2. Characterize the expression oflH receptor gene in Leydig cells and precursor
cells and the changes of LH receptor mRNA during the precursor cell
differentiation process.
3. Characterize gene expression of steroidogenic enzymes, e.g., cholesterol
side chain cleavage enzyme (P-450=) and 17a.-hydroxylase (P-45Q711 ), in
Leydig cells and precursor cells during the differentiation process.
4. Investigate the localization ofthe estrogen receptor in the testis of the control
and EDS-treated rats to test the hypothesis that estrogen receptor is present
in both Leydig cells and precursor cells. 4
5. Investigate estrogen receptor gene expression in Leydig cells and precursor
cells and the pattern of estrogen receptor gene expression in these cells during
the differentiation process.
The information provided by these studies will extend our understanding of Leydig precursor cells and their differentiation process. The results obtained from these studies will help us to characterize the precursor cells at the biochemical and molecular level and allow further investigation into the endocrine/paracrine regulation of the precursor cell development. BACKGROUND
Franz Leydig (1850), in the middle of nineteenth century, first described the interstitial cells of the manunalian testis. Since that time, there has been continuous progress in obtaining information on the function and regulation of this testicular compartment. At the beginning ofthis century, a possible endocrine role for the Leydig cells was proposed, and by the 1930s it was evident that the male hormone was testosterone. However, it was not until the early
1960s, when experiments were performed in isolated tissue compartments (seminiferous tubule and interstitial tissue), that it was conclusively demonstrated that the Leydig cells were the main source of testicular androgens. By the late 1960s and early 1970s, the main hormonal interactions arhong the hypothalamus, pituitary and the testis as well as the principal biosynthetic and metabolic pathways for the steroids in Leydig cells had been delineated (Eik nes, 1975). More recently, the development of techniques for the isolation of Leydig cells
(Conn eta!, 1977; Payne et al1980; Aquilano and Dufau, 1984), the establishment ofLeydig cell lines (Ascoli, 1981; Rebios, 1982), the cloning ofthe LH receptor (Loosfelt et al, 1989;
Mcarland et al, I 989) and the steroidogenic enzymes involved in testosterone synthesis
(Hanudoglu, 1992; Labrie eta!, 1992 a; Simpson and Waterman, 1988) have given new insight into the regulation of Leydig cell function. One of the most exciting new areas that has evolved in the last few years is the paracrine and autocrine regulatory interactions through
5 6 which the fine tuning of Leydig cell function is achieved. This shift from endocrine to paracrine research has become critical in the understanding of gonadal differentiation and function (Sharp, 1990; Saez et al, 1989; Tahja, 1989; Skinner, 1991).
Steroidogenic pathway in Leydig cells
The mammalian testis is divided into two compartments: 1) the vascularized interstitial. tissue formed mainly by Leydig cells, macrophages, lymphatic vessels and connective tissue, and 2) the avascular seminiferous tubules containing Sertoli cells and germ cells. Hormone secretion and spermatogenesis are the two main physiological functions of the testis. The principle steroids secreted by the testis are androgens, among which testosterone is the most important product. Almost all testicular androgens are formed and secreted by Leydig cells since at least two of the key enzymes, cholesterol side chain cleavage (P-450,.,) and 3 ~ hydroxysteroid dehydrogenase (3 ~-HSD) are not expressed in the seminiferous tubules
(Liebrman and Prasad, 1990; Hall, 1988). The main, but probably not the only substrate
(Liebrman and Prasad, 1990), for the synthesis of steroid hormones in Leydig cells, as in all steroidogenic tissues, is cholesterol. This substrate can be either synthesized de novo from acetate or delivered into the cells from the plasma lipoproteins (Hall, 1988; Chen et a!, 1980).
The general pathway of steroidogenesis in the testis resembles that of the other steroidogenic tissues. As shown in Fig. 1, the conversion of cholesterol to testosterone involves three hydroxy lations, two cleavages, two dehydrogenations (3 ~ and 17~), and one isomerization
A +s. The first reaction, conversion of cholesterol to pregnenolone catalyzed by P-450,..,, is the rate-limiting step. It has been postulated that under physiological conditions the transfer of Figure 1. Steroidogenic pathway for the synthesis of testosterone.
P-450""': cholesterol side chain cleavage enzyme
P-45017a: !?a.-hydroxylase
1713-HSD: 17j3-hydroxysteroid dehydrogenase
4 5 3j3-HSD: 3j3-hydroxysteroid dehydrogenase (ll. " isomerase) 7
Cholesterol
P-450css l 3P-HSD
Pregnenolone ------~ ·Progesterone
P-:"1-50 17" l 3P-BSD l P-45017" 17a.:.hydroXypregnenolone ...... ;_~__;___-~ 17a-liydroXyprogesterone -: . :- _·.-· ... _- :-·;: ..· .. - -· .;: .-_, ,--_. ... -. -_;. ·-,- . ·... ---. ·• . . . · ..
' .. 17:.20 lyase : J ' ·. ·· · .·- ·. -. ·3p-BSD · .1 '1 i.:io ly~e . . :' Dehydroepiandrosterone ------'-----'---'-~ Androstenedione
17P-HSD l 3P-HSD l 17P-HSD
.Androstenediol------~ Testosterone 8 cholesterol to the inner mitochondrial membrane is the most important step in steroid synthesis (Privalle et al, 1983) and involves several factors (Hall, 1985; Van.Noort et al,
1988). The main acute effect of the pituitary hormones, such as LH and ACTH, is to stimulate the transfer of cholesterol, both through increasing the availability of cytsolic free cholesterol and through enhancing cholesterol transfer between the outer and inner mitochondrial membranes (Yamamoto et al, 1991; Hallet al, 1979). It has been reported that stimulation of cholesterol transfer in Leydig cells by LH is blocked very rapidly by protein synthesis inhibitors. This suggests that LH-stimulated cholesterol transfer involves protein synthesis
(Hallet al, 1979). The product of this cleavage reaction is pregnenolone which is rapidly released from the mitochondria into cytosol by unknown mechanisms. The biosynthesis of androgens from pregnenolone, depending on species and stages of development, takes place
5 4 through either the "' - and/or ~; -pathways. In the fetal and adult human, as well as in the rabbit and pig testis, the ~; 5-pathway appears to be more important (Yannaihara and Troen,
1972; Ewing et al, 1980). However, the rat preferentially utilizes the"'4 -pathway (Weuestem et al, 1987):The conversion of 3~-hydroxy-5-ene steroids into 3-keto-4-ene steroids is catalyzed by3~-HSD. Three types of3~-HSD (Type I-ill) have been cloned in both rat and mouse (Labrie et al, 1992 a; Labrie et al, 1992 b; Bain et al, 1991). Both type I and type II isoforms are expressed in several steroidogenic tissues, including the adrenal; the ovary and the testis, but type I isoform has higher enzyme activity. Type ill 3 ~-HSD is mainly expressed in male liver and its biological function is still unknown. The subsequent conversion involves
two steps, 17a.-hydroxylation and cleavage of the C17•20 bond, which is catalyzed by a single
enzyme, cytochrome P-450 17a.-hydroxylase (P-45017.). This single protein has been purified 9 from the pig adrenal and testis and demonstrated to possess both enzymatic activities (Nakajin
et al, 1981; Nakajin et al, 1984). eDNA cloning ofP-4501701 and expression studies of the
eDNA clones isolated from the testis and adrenal of several species including human
(Bradshaw et al, 1987) and rat (Namiki et al, 1988; Fevold et al, 1989) also have
demonstrated t~at this single enzyme exhibits both 17a-hydroxylase and 17-20 lyase
activities. Different activities of this enzyme in glucocorticoid and androgen pathways lie in
the rnicrosomes. Despite the fact that the pure enzyme isolated from the adrenal has the same
lyase activity as that in the testis, very little lyase activity is observed in adrenal microsomes,
(Hal~ 1991). The product ofP-45017 .., androstenedione, is converted to testosterone by the
microsomal enzyme 1713-hydroxysteroid dehydrogenase (1713-HSD). The enzyme has been
purified and the eDNA has been cloned from human placenta (Lin et al, 1992; Peltoketo et
~ 1988). In both human and rat, 1713-HSD mRNA and enzymatic activities are present not
only in steroidogenic tissues, but in many other tissues as well (Martel et al, 1992).
In addition, androgens can be converted to estrogens by cytochrome aromatase which
is also located in the rnicrosomes. Estrogens are produced in the testis of several species
(Kelch et al, 1972; Raeside et al, 1988; Henri_cks et al, 1988). Estrogen production in most
species is low except in the stallion (Dintinger et al, 1989) and in the boar (Claus et al, 1985)
where the secretion of estrogens is even higher than that of testosterone.
Ontogenic development of Leydig cells
The ontogenesis ofLeydig cells involves two distinct generations in mammals, fetal
type and adult-type Leydig cells. The fetal-type Leydig cell generation is responsible for the
masculinization during fetal and neonatal life through the secretion of testosterone. These 10 fetal Leydig cells regress thereafter and the adult-type Leydig cell population emerge and contribute to pubertal sexual differentiation.
Early in fetal life, the genital primordium is histologically identical in both sexes.
Primordial germ cells, which have migrated from the endoderm of the yolk sac, and somatic cells, which are derived from the mesonephros are contained in the genital ridge. The first clearly recognizable sign of male gonadal differentiation is the formation of the seminiferous cords. Relatively late in the course of testis formation, the fetal Leydig cells appear after the formation of the seminiferous cords (Gondos, 1980; Jost eta!, 1973; Grorge and Wilson,
1988; Wartenberg, 1989). In the rat, seminiferous cords begin to form at 13 days of gestation ·
(Jost et a!, 1981). Sertoli cell precursors, appearing as large "clear" cells, aggregate and surround the germ cells during the formation of seminiferous cords (Jost eta!, 1981; Magre and Jost, 1980). In the mean time, a basal membrane gradually forms around the seminiferous cords. Histologically recognizable Leydig cells appear in the interstitial region on day 15 to day 16 of gestation in the rat (Magre and Jost, 1980; Narbaitz and Adler, 1967; Roosen
Runge and Anderson, 1959; Merchant-Larios, 1975). These fetal Leydig cells demonstrate the ultrastructural characteristics of the steroidogenic cells, containing large mitochondria with tubular cristae, abundant lipid droplets and smooth endoplasmic reticulum (Magre and
Jost, 1980; Narbaitz and Adler, I 967). The fetal testis begins to produce testosterone, both in vivo and in vitro on day I5.5 (Narbaitz and Adler, 1967). It has been reported that 313-
HSD activity is present in the fetal testis by the age of day 15 (Niemi and Ikonen, I 96 I) and the fetal testis is able to convert progesterone into testosterone in vitro as early as day 13 of gestation (Noumura eta!, I966; Gangnerau and Picon, I987). These studies suggest that the 11 functional maturation probably precedes the structural differentiation during Leydig cell development in fetal life. In the differentiation of fetal Leydig cell function in the rat, the rate limiting step is the conversion of cholesterol into pregnenolone, catalyzed by P-450scc enzyme.
The ability ofthe fetal rat testis to specifically bind LH and respond to this pituitary hormone develops between days 14 and 15 of gestation (Gangnerau et al, 1982; Warren et al, 1984), coinciding with the secretion of testosterone from the fetal testis. The development of coupling between cAMP and the testosterone synthesis may precede that between LH binding and cAMP response by several hours (Gangnerau and Picon, 1987). It is concluded from these studies that the differentiation ofLH responsiveness and steroidogenesis is determined independently in the fetal rat testis.
Signals triggering the initial differentiation of the fetal Leydig cells are still unknown.
It has been clearly demonstrated that the initial differentiation of Leydig cells in the rat fetal testis is not dependent upon gonadotropins. The testis determining factor is currently thought to regulate only Sertoli cell differentiation and the organization of seminiferous cords
(McLaren, 1990; McLaren, 1991). This-sexual-genetic-control-independent theory was also demonstrated in XX+-> XY chimeric mouse testes in which both XX and XY contributed to the formation of Leydig cells and only XY contributed to the Sertoli cells (Burgoyne et al,
1988). Gonadal anlage, removed from the fetus between days 12 to 13 and incubated in a synthetic medium without any hormones or growth factors, complete the morphological and functional differentiation of Leydig cells within 3 days (Gangnerau et al, 1982; Agelopoulou et al, 1984; Magre and Jost, 1984). In vivo evidence also indicates that the initial Leydig cell differentiation takes place in a gonadotropin free environment since the fetal rat pituitary 12 gonadotropes begin to differentiate on day 16 (Jennes, 1990) and LH cannot be detected until day 16.5 or 17 (Aubert et al, 1985), while the fetal Leydig cells are first observed on day 15.5
(Magre and Jost, 1980; Matbaitz and Alder, 1967; Roosen-Runge and Anderson, 1959;
Merchant-Larios, 1975). It has been reported that there is no chorionic gonadotropin-like
activity in the rat placenta since attempts to detect LHj3 mRNA have been unsuccessful
(Wurzel et al, 1983; Habert and Picon, 1990), thus further supporting the conclusion that the
initial fetal Leydig cell development is gonadotropin-independent.
Since the differentiation of fetal Leydig cells takes place after that of the seminiferous
cords in all species studied, it has been suggested that Leydig cells probably differentiate
under the paracrine regulation of Sertoli cells. This theory is supported by experiments in
which disruption of the seminiferous cords was correlated to the impairment of Leydig cell
function in fetal mice (Byskov et al, 1983). Furthermore, histologically recognizable clusters
ofLeydig cells are observed only in gonads containing the best developed seminiferous cords
(Jost et al, 1973). It has been recently reported that recombinant FSH stimulates testosterone
production by fetal rat testis explanted on day 14.5 (Lecerf et al, 1993). Since it is widely
accepted that FSH receptors are exclusively localized in Sertoli cells (Bardin et al, 1988),
these data suggest a paracrine effect of Sertoli cells on Leydig cell differentiation.
Nevertheless, Patsavoudi et al (1984) reported that fetal calf serum prevented the formation
of testicular cords, but not the structural and functional differentiation of Leydig cells in organ
culture, indicating that Leydig cell differentiation does not depend on normal histological
organization of the Sertoli cells. Thus, it still remains to be proved whether the initial phase
of Leydig cell development is influenced by Sertoli cell-secreted factors, such as Mullerian 13 inhibiting factor (MIF) which is able to increase testosterone secretion in cultured fetal ovaries by inhibiting aromatase activity (Vigier et al, 1990).
The fetal Leydig ceils in the rat do not form large confluent regions like those in the human. They remain as individual ceils for a longer duration and form irregularly outlined groups toward the end of the pregnancy (Kuopio et al, 1989). The differentiating Leydig ceils are covered by smail patches of ultrastructuraiiy and immunocytochemicaiiy identifiable basement membrane which increases in area with advancing ceil differentiation (Kuopio et al,
1989). The addition ofL-azetidine 2-carboxylic acid, a proline competitor which prevents the secretion offibronectin and laminin, into the medium of organ culture of differentiating testis, inhibits the differentiation and the steroidogenic function of Leydig ceils (Jost et al, 1988).
The inhibition can be prevented by the addition of an excess of proline into the medium. These findings indicate that the extraceilular matrix may play a role in Leydig ceil differentiation, probably in their grouping into epithelial-like glandular islets. This also implies that laminin, coilagen type IV, and other basement membrane components may be products of fetal Leydig ceils. In addition, there is evidence that adult mouse Leydig ceils specificaily bind lymphocytes and macrophages through a nonimmunological mechanism (Rivenson et al, 1981). The significance of these findings still remains obscure, but it is known that macrophages appear in the rat testicular interstitium at day 19 during early fetal Leydig ceil differentiation and produce several regulatory factors which may exert paracrine effects on Leydig ceils (Hutson,
1990).
In the rat, the number ofLeydig ceils greatly increases between days 16.5 to 18.5 of gestation and reaches a maximum on day 20.5 (Tapanainen et al, 1984; Kerr and Kneil, 1988). 14
Testicular steroidogenesis increases considerably during the· same time as the number of
Leydig cells increases. The regulation of the further differentiation offetal Leydig cells and steroid biosynthesis is apparently under the control of pituitary gonadotropins. In one nonphysiological experiment, caffeine administration to the mother decreased the number of rat fetal Leydig cells and their testosterone biosynthesis (Pollard et al, 1990). The biological
significance of these results still remain unexplained, but it may offer a new insight into the
mechanisms regulating the proliferation of the fetal Leydig cells.
Fetal Leydig cells have been suggested to be derived from the mesenchymal precursor
cells in the interstitial region of the testis in many species, including human, rat and pig
(Pelliniemi and Niemi, 1969; Wartenberg, 1989; Magre and Jost, 1980; Moon and Hardy,
1973). Moon and Hardy (1973) studied the fetal pig testis and showed that a relative decrease
in the number of mesenchymal cells corresponds in time and size to an increase in maturing
Leydig cells. Because only a few mitotic figures are seen in fetal Leydig cells, it has been
concluded that these fetal Leydig cells are exciusively derived from the differentiation of
preexisting mesenchymal cells in the interstitium (Moon and Hardy; 1973). This is in
agreement with Orth's study in which rat fetal Leydig cells were shown to have no mitotic
activity and remain unlabeled after injection of3H thymidine (Orth, 1982).
The relative and cumulative volume of fetal Leydig cells clearly decrease after birth,
during the first and the second postnatal weeks in many studies (Roosen-Rounge and
Anderson, 1959; Lording and De Krester, 1972; Kerr and Knell, 1988). This is the
consequence of a decrease in Leydig cell number during this period. Some studies, however,
h3:ve shown that there is a second peak of total fetal Leydig cell number on day 14 of 15 postnatal life (Kuopio et al, 1989 b; Mendis-Handagama et al, 1987), and that the reduction in the cumulative Leydig cell volume results from the decrease in size of these cells. This second peak of fetal Leydig cell number is obscured because of the concomitant increase in the number of adult-type Leydig cells. The fate of fetal-type Leydig cells after the second week of postnatal life is not clear. Although it has been suggested in some studies that they persist as a stable population throughout life (Kerr and Knell, 1988) or they are transformed into adult-type Leydig cells (Mendis-Handagama et al, 1987), it is widely accepted that these fetal-type Leydig cells degenerate after the second postnatal week (Roosen-Runge and
Anderson, 1959; Kuopio et al, 1989b; Wartenberg, 1981). In addition to morphological regression, these cells clearly undergo functional regression. Immunohistochemical studies demonstrate an age-related decrease in 313-HSD activity (Orth, 1980), and in vitro studies show an age-related decrease in basal and ill-stimulated testosterone production in fetal-type
Leydig cells (Habert and Brignaschi, 1991 ). What causes this morphological and functional regression of fetal Leydig cells is unknown. It can not be explained by the lack of gonadotropin stimulation since the plasma LH like activity increases during this period
(Habert and Picon, 1982). Again, paracrine and/or autocrine mechanisms might be involved in the fetal Leydig cell regression.
Fetal-type Leydig cells possess several distinct morphological and functional characteristics. They have a more extensive cytoplasm, abundant lipid droplets and a less distinct peripheral layer ofheterochromatin in the nucleus when compared to the adult Leydig cells (Huhtaniemi and Pelliniemi, 1992). The mitochondria and smooth endoplasmic reticulum are shaped differently from those of the adult Leydig cells. Functionally, fetal rat Leydig cells 16 produce high levels of androgen and have low levels of Sa-reductase and aromatase activity
(Habert and Picon, 1984). The most striking functional difference between fetal and adult
Leydig cells is the resistance of the former cell population to LH-hCG induced desensitization
(Huhtaniemi and Pelliniemi, 1992; Warren et al, 1982). Instead, LH/hCG treatment produces an up-regulation of LH/hCG receptor mRNA and binding sites and as a result, increased steroidogenic activity in fetal Leydig cells (Huhtaniemi and Pelliniemi, 1992).
The adult-type Leydig cells emerge in the testis around day I 0 after birth in the rat and increase in number considerably after day 15, reaching a maximum of25xl06 per testis by the end of puberty (Mendis-Handagama et al, I 987; Tapanainen et al, 1984). These adult Leydig cells are thought to be derived from the differentiation of precursor mesenchymal cells in the interstitium (Prince, 1984; Tapanainen et al, 1984). The undifferentiated mesenchymal cells, with elongated nuclei and sparse cytoplasm, are dispersed in a loose connective tissue matrix
(Prince, 1984; Chemes et al, 1985). Unlike fetal Leydig cells, the differentiation of adult
Leydig cells is dependent upon LH. At the beginning of puberty, the mesenchymal cells begin to differentiate into adult Leydig cells, coinciding with the rise in plasma LH level (Lee et al,
1975; Ketelslegers et al, 1978). The ultrastructural changes during this process involve the appearance and marked increase of the smooth endoplasmic reticulum, mitochondria and lipid droplets, as well as the typical changes in nuclear morphology (Prince, 1984). The undifferentiated mesenchymal cells have been isolated from prepubertal human testis and ' shown to undergo differentiation in vitro in the presence of hCG (Chemes H et al, 1992;
Hardy et al, 1990).
Although there are repeated observations that few if any mitotic figures are seen in 17
Leydig cells during postnatal development, Johnson and Neaves (1981) have argued that the expected frequency of mitotic figures in Leydig cells is extremely low (about 1 in 20,000) because of a length in the population doubling time. The results of the kinetic studies on the development of the adult Leydig cell population in the testes of the pubertal rats show that mesenchymal cells comprise 44% and Leydig cells 16% of the total interstitial cell population at day 19 of gestation. By day 56 postpartum the relationship is reversed, with mesenchymal cells comprising 3% and Leydig cells 49% of the total interstitial population (Hardy et al,
1989). These results suggest a precursor-product relationship between mesenchymal and
Leydig cells because no such reciprocal relationship is observed between Leydig cells and macrophages, pericytes, endothelial, or myoid cells. Autoradiographic analysis of 3H thymidine incorporation into mesenchymal and Leydig cells demonstrate that the falling percentage of the labeled mesenchymal cells between 14-28 days postpartum coincides with an equivalent rise in the percentage of labeled Leydig cells (Hardy et al, 1989). The newly formed Leydig cells divide one or two times between days 28-70 postpartum to complete the growth of Leydig cell number.
Regulation of Leydig cell function
Many studies have shown that Leydig cell structure and differential function is dependent upon LH (Russell eta!, 1992; Swerdloffand Heber, 1981; Teerds et al, 1989).
Hypophysectomy (Swerdloff and Heber, 1981), suppression of gonadotropins by steroid administration (Russell eta!, 1992), or neutralization ofLH by specific antibodies (Dym et a!, 1977) results in Leydig cell atrophy and cellular volume loss. Cytoplasmic smooth endoplasmic reticulum, LH!hCG receptor, steroidogenic enzyme (especially P-45017.. and P- 18
450scd activities and testosterone secretion are greatly diminished by LH deprivation. The number ofLeydig cells, however, is not significantly influenced (Swerdloffand Heber, 1981;
Keeney et al, 1988). Treatment ofthese LH deprived rats with LH/hCG restores the structure and function of Leydig· cells (Ressell et al, 1992; Teerds et al, 1989; Dym et al, 1977). The effects ofLH/hCG treatment on steroidogenic enzyme activities and testosterone production are more quickly observed than those on the morphological changes. The time of complete recovery is correlated to the duration ofLH deprivation by either hypophysectomy (Stocco et al, 1990) or by steroid implantation (Russell et a!, 1992). It also has been observed that the
decline of the activities of the steroidogenic enzymes, P-45017., P-4508cc and 1713-HSD, is much faster than that of3f3-HSD (Keeney eta!, 1988; Purvis et al, 1973). It is concluded from all the results that LH is either the only or the main factor required to maintain Leydig
cell structure and function, at least in mature rats.
LH/hCG ca~ses two types of responses in Leydig cells. The first response involves the induction of a sharp increase in cAMP and steroid synthesis within minutes. This acute effect does not involve RNA synthesis and is mainly mediated by cAMP (Bartke et a!, 1978).
The second type of response involves the long term trophic effect of the hormone on cell structure and function which require RNA and protein synthesis (Payne, 1992). The recent
availability of specific antisera to the P450 enzymes and the isolation of eDNA clones specific for P-450scc and for P-45Cha has made it possible to initiate studies on the regulation of synthesis and steady state levels of these two most important steroidogenic enzymes involved
in testosterone production in Leydig cells. P-45017a and P-450scc are differentially regulated by LH acting via its second messenger cAMP. P-450scc is constitutively expressed in normal 19 mouse Leydig cells and MA-l 0 tumor Leydig cells (Payne, 1990). Chronic cAMP stimulation increases the steady state levels ofP-450scc mRNA and de novo P-4508cc protein synthesis.
In contrast, cAMP is obligatory for de novo synthesis ofP-45017a in normal mouse Leydig cells. P-45017a synthesis ceases in the absence ofLH or cAMP. Newly synthesized protein (s) is required for cAMP induction ofP-45017a mRNA levels since cycloheximide prevents the cAMP induced P-45017a mRNA synthesis, but not for P-45Qcc mRNA (Payne and Sha,
1991). The stimulatory effects of LH on these steroidogenic enzymes have also been demonstrated in the rat (Anderson and Mendelson, 1985).
The first step in LH action is binding ofthe hormone on target cells. LH/hCG receptor belongs to a G protein-coupled receptor family. This glycoprotein receptor, as well as two other receptors in this family, FSH and TSH receptor, contain seven transmembrane domains
(Dohlman et al, 1991). It has a large extracellular domain (300-400 amino acids in length) which is responsible for the recognition and high affinity binding of the glycoprotein hormone
LH (Fig. 2). This is examined by expressing a eDNA encoding for a truncated form of the rat
LH receptor with 1-338 amino acid residues (Xie et al1990). The detergent extracts of the cells expressing the truncated rat LH/hCG receptor bind hCG with equal or higher affinity than cells expressing the wild type receptor. This truncated receptor containing the extracellular domain and devoid of the transmembrane and cytoplasmic domains, is capable of binding hormone but is trapped in an intracellular compartment. The ability of this extracellular domain to bind LH with high affinity has also been confirmed by several other investigators. In fact, the affinity ofLH/hCG binding to detergent extracts prepared from cells transfected with the eDNA encoding N-terrninal 294 amino acids is the same as that detected Figure2. Amino acid sequence, orientation and proposed topology of the rat
llJJhCG receptor in the plasma membrane .. The figure is cited from Segal off
and Ascoli (1993).
Branch-like structures:· potential sites for N-Iinked
glycosylation
Underlined sequences with dashes: . potential tryptic cleavage
sites
Asterisks and dark dots: potential phosphorylation
sites
Rectangles: . weak consensus sequences for
~AMP-dependent · protein
kinase-catalyzed
phosphorylation
Ovals: consensus sequences for
protein · kinase C-catalyzed
phosphorylation 21 in cells transfected with the full length eDNA (Ji and Ji, 1991 ). Further truncations, however, result in a slight reduction in binding affinity and a substantial reduction in the levels of receptor expressed (Braun et al, 1991).
The structural configuration of the coding region of the LH receptor gene isolated from rat consists of 11 exons separated by 10 introns (Tsai-Morris et al, 1991 ). All 10 introns are located within the putative extracellular domain prior to the first transmembrane region.
(Fig. 3). The 5'-non-coding region contains several potential TATA boxes and SP1, AP2 promoter binding sites (Nelson et al, 1994). The presence of multiple transcriptional initiation sites is indicated by primer extension studies. Exons 2-8 approximate the regions of the 14 consecutive 20 amino acid repeated motifs which is also present in FSH and TSH genes, with the exception of a 6-8 amino acid shift in each motif Exons I, 9, 10 and 11 contain important structural elements including the conserved cysteines, the soybean lectin motif, the seven transmembrane domain with cytoplasmic G protein coupling elements, and three putative N linked glycosylation sites. Exon 10 and the beginning of exon 11 along with the lectin motif are unique to the LH receptor and may be involved in specific association with the hormonal ligand. The homologous regions with other members of the glycoprotein receptor family encoded by exons 2-8, and the common amino acid motif that contains 3 .conserved cysteines immediately prior to the first transmembrane region may be involved in common hormonal interactions and in coupling functions, respectively.
The mRNA for LH receptor is expressed mainly in testis and ovary, but also in other tissues where the majority of transcripts remain incompletely spliced (Lei et al, 1993). In both the testis and ovary, there are several transcripts varying from 1.2-7.6 Kb in length (Saez, Figure 3. Genomic organization ofthe rat LH/hCG receptor. Exons are represented by
the open rectangles and introns by triangles. The size of the different introns
is indicated on the top of each triangle. The exons are drawn to scale indicated
under exon 1. The first 10 exons encode for the majority of the extracellular
domain. Exon 11 encodes for a small portion of the extracellular domain, all
the transmembrane regions and the entire C-terminal cytoplasmic tail. 22
14.7 2.2 25 2.3 O.o9 7.2 2.7 4.2 2.0 13.6 kb. kb- kb kb -kb kb kb kb kb kb·
·:Exti:acellular domain ·.Transm!:mbrane and intra«ellular domains 23
1994). The size and the abundance of these transcripts vary from one species to another and from tissue to tissue. This phenomenon may result from variable 3'-noncoding sequence or I from alternative splicing. Since the eDNA encoding the full length receptor contains at least
2100 base pairs, it is obvious that transcripts shorter than 2.1 Kb can not be translated into the whole receptor. Transcripts larger than 2.1 Kb may result from the use of remote transcription start sites or from different sites of addition and/or different lengths of the poly
A tail (Segaloffand Ascoli, 1993). Alternative or incorrect splicing of the LH receptor gene may also account for the multiplicity of the transcripts because of the complex structure of the gene and the presence of several eDNA variants detected during the cloning of the receptor (Segaloff et al, 1990). Recent studies .have shown that transcripts both larger and
smaller than the full length eDNA are amplified from rat ovary using reverse transcription- polymerase chain reaction, clearly implicating that alternative splicing of the gene as a major component contributing to the multiple forms ofLH receptor transcripts (Sokka et al, 1992).
Northern blot analysis with different LH receptor eDNA probes has brought about the
detection of different transcripts. The 1.2 Kb transcript detected inMA-10 cells hybridizes with a eDNA probe corresponding to the excellular domain of the rat LH receptor, but not to a probe corresponding to the transmembrane domain (Wang et al, 1991). Similar results have been reported for the 1.8 Kb transcript detected in the rat testis (LaPolt et al, 1991),
suggesting that these transcripts encode for a truncated form of the LH/hCG receptor corresponding to the extracellular domain which remain trapped within the cells rather than being secreted.
It is now well demonstrated that LH/hCG receptor is a single 90 Kd glycoprotein. 24
Using affinity cross-linking techniques, several research groups have reported findings that support the existence of only one hormone binding unit of molecular size 90 Kd (Neuman et ai, 1991). This is also supported by a recent study in which 5 different monoclonal antibodies that inhibit hCG binding to the membrane receptor recognize only a 90 Kd polypeptide in a detergent extract of ovarian cell membranes (Neuman et al, 1991). It has been well established that cAMP serves as the second messenger for LH action. LH increases adenylate cyclase activity in isolated membranes and cAMP production in intact cells (Ascoli and Segaloff,
1989; Rommerts and Cooke, 1988). Moreover, cAMP derivatives or compounds that raise endogenous cAMP levels mimic the effects of LH/hCG effect on steroid biosynthesis
(Rommerts and Cooke, 1988). Although most studies have demonstrated that cAMP is the principle mediator of the actions ofLH/hCG in gonadal cells, there have been frequent reports about the possibility that other second messengers are involved in mediating the actions of this hormone. With regards to the possible activation of several signaling systems by LH/hCG, the stimulation of the phospholipase C pathway has been considered as a candidate in spite of variable data. Davis et a! (1987) reported that LH stimulates the turnover of phosphoinositides and the formation of inositol phosphates in gonadal cells and as a
2 consequence increases the intracellular Ca + levels. Since the effects of LH on inositol
2 phosphate accumulation and intracellular Ca + are small and require higher concentrations of
LH, the question remains whether phospholipase C activation is a primary response to receptor occupancy or secondary to an increase in cAMP. Recent studies, however, have clearly demonstrated that the LH/hCG receptor can be coupled to at least two different transduction systems. Gudermann et al (1992) reported that addition ofLH to mouse L cells
\, 25 transfected with a mouse LH receptor results in increased levels of cAMP, inositol
2 phosphates, and intracellular Ca •. Moreover, addition of LH to Xenopus oocytes injected with a mouse LH receptor cRNA has been reported to result in the activation of adenylyl
cyclase and mobilization of intracellular Ca2+ stores (Gudermann et al, 1992 b). The effects
2 on inositol phosphates and Ca + mobilization are not secondary to the activation of adenylyl
cyclase by the LH receptor because activation of adenylyl cyclase by other agents do not have
2 a similar effect on inositol phosphates or Ca + mobilization. These results suggest that mouse
LH receptor is capable of activating two distinct G proteins, G, which activates adenylyl
cyclase, and Gq which activates phospholipase C.
As a primary factor to maintain Leydig cell structure and differentiated function, LH
is also required for the maintenance of LH binding sites on Leydig cells at physiological
concentrations. High levels of LHihCG, however, induce receptor loss and steroidogenic
refractoriness in a process called desensitization. LH induced receptor loss is comprised of two categories, uncoupling and down-regulation. Uncoupling is an agonist-induced change
in the functional properties of the receptor without a change in the numbers of receptors. This
is a relatively fast phenomenon that occurs within minutes of addition of the agonist and is
likely to result from posttranslational modifications of the receptor that attenuates its ability to activate the effector system. Studies show that pretreatment with LH/hCG causes a 2-fold reduction in LH/hCG-stimulated adenylyl cyclase activity in gonadal cells (Sanchez-Yagiie
et a!, 1991 ). This reduction in hormonal responsiveness results from the hormone-ligand uncoupling without a reduction in the number of LH/hCG receptor. This is also due to
changes in the functional properties of G, or the catalytic subunit of adenylyl cyclase. The 26 second phase ofLH/hCG-induced receptor loss, down-regulation ofLH/hCG receptor, has been mostly studied in MA-l 0 cells. Results show that exposure of these Leydig tumor cells to LH/hCG results in a time-dependent decrease in the number ofLH!hCG receptor without a change in the affinity of the receptors (Rebois and Fishman, 1984). This LH!hCG-induced down-regulation of LH!hCG receptor consists of two distinct phases. The first phase is characterized by 80% reduction in the levels ofLH!hCG receptors with little or no change in
LH!hCG receptor mRNA. Quantitatively this is the most important phase and it involves an increase in the rate of degradation of the receptor. The second phase is characterized by a further reduction in the levels of LH/hCG receptors that is accompanied by a 40-60% reduction in the levels ofLH!hCG receptor mRNA (Segaloffand Ascoli, 1993). It is therefore
concluded from these results that the LH/hCG-induced LH!hCG receptor down-regulation involves two processes. One is an increase in receptor degradation caused by LH/hCG and the other is a decrease in receptor synthesis that is secondary to a decease in receptor mRNA.
The molecular basis of the transcriptional regulation of LH!hCG receptor has been
studied using luciferase fusion constructs transfected into Leydig cells (Nelson et al, 1994).
The results demonstrate that the proximal186 basepairs (relative to the translation start site)
of the 5'-flanking region of the rat LH/hCG receptor gene represent a basal promotor that
accounts for the Leydig cell specific expression of this receptor. A region that confers negative transcriptional regulation by cAMP maps to nucleotides -40 to -70 of this basal
promotor. Leydig cell specific protein(s) bind to this basal LH!hCG promotor. The binding
of the Leydig cell specific proteins to the promotor region involves an AP-2 consensus
sequence beginning at nucleotide-59. 27
Although LH is absolutely required for Leydig cells, evidence has accumulated that
other hormones and factors also are involved in regulating Leydig cell development and function. The anatomical arrangement of the testis with two compartments, the seminiferous
tubules and the interstitial tissue, separated by the blood-testis barrier, points to an active
interaction among different testicular cells. Indeed, data over the past few years clearly indicate that Leydig cell function is locally regulated, in conjunction with or dependent upon the control of gonadotropins. The first evidence oflocal communication between Leydig cells
and seminiferous tubules was provided by Johnson and Ewing (1971 ), who reported that FSH
enhanced testosterone production by perfused rabbit testis exposed to maximal concentrations
ofLH. It has also been reported that FSH is involved in the development of adult Leydig cell
population. A close correlation between the rise in serum FSH levels and the formation of
Leydig cells around puberty has been demonstrated from several studies (Swerdloff et a!,
1971; Lee eta!, 1975). Furthermore, results from the immature hypophysectomized rats show
that FSH enhances the capacity of the testis to bind LH and secrete testosterone (Odell and
Swerdloff, 1975; Chen et a!, 1977). In Kerr and Sharpe's study (1985), FSH treatment
resulted in a significant increase in Leydig cell number and caused Leydig cell hypertrophy
in immature hypophysectomized rats. Recent studies using highly purified pituitary FSH
(Teers et al, 1989 b) and recombinant human FSH preparations (Vihko KK et al, 1991)
indicate that FSH administration to immature hypophysectomized rats may increase the LH
receptor number and LH receptor mRNA levels. Since FSH receptors are only localized in
Sertoli cells, the effects ofFSH on Leydig cell morphology and function may be mediated
by the secretory factors from the seminiferous tubules. 28
Other evidence to support the regulatory role of seminiferous tubules on Leydig cells is derived from the observation that Leydig cells lying close to stage VII-VIII seminiferous tubules are larger than those lying adjacent to tubules at other spermatogenic stages in the rat
(Bergh, 1982). Coculture of rat Leydig cells with isolated segments of seminiferous tubules at different stages of the spermatogenic cycle also suggests that stage VII tubules positively regulate testosterone production while later stages may exert inhibitory effects (Parvinen et al, 1984). The most convincing evidence comes from the studies of Sertoli-Leydig cell coculture. Coculture of Sertoli cells with Leydig cells isolated from immature pig testis enhances hCG-stimulated testosterone production by Leydig cells. Pretreatment of Sertoli
Leydig cell coculture with FSH further increases the steroidogenic capacity ofLeydig cells and induces a significant increase in LH/CG receptor number (Benahmed et al, 1985).
Similarly, coculture of rat immature Sertoli cells and Leydig cells enhances basal and LH or cAMP stimulated steroid production, and these effects are significantly enhanced by pretreatment of the coculture with FSH (Verhoeven and Caileau, 1990). Furthermore, it has been recently reported that co culture of adult human Leydig cells with human Sertoli cells not only prevents the decline in the steroidogenic capacity observed when Leydig cells are cultured alone, but greatly increases testosterone production by Leydig cells (Lejeune et al,
1993). More evidence of the paracrine r~gulation of Sertoli cells on Leydig cell function is provided by studies using Sertoli cell conditioned medium (SCCM). SCCM stimulates Leydig cell steroidogenesis in different species, including rat (Verhoeven and Caileau J, 1985) and human (Verhoeven and Caileau, 1987). Pretreatment of Sertoli cells with FSH enhances the stimulatory effects of SCCM on Leydig cell steroidogenesis (Verhoeven and Caileau, 1985). 29
In addition to these acute stimulatory effects, SCCM also has long term influences on Leydig
cells. The nature and the strength of these long term effects depend upon the conditions in
which Sertoli cells are cultured and the maturity status of Sertoli and Leydig cells
(Papadopoulos et al, 1987). Sertoli cells secrete more than 80 proteins and FSH regulates the
secretion of some of them, but very few have been identified because these paracrine factors
are probably secreted in very small amounts and have short half lives. One Sertoli cell
secreted factor responsible for the acute steroidogenic effects on Leydig cells has been
purified and has a molecular weight of 80 KD (Cheng et al, 1993).
Growth factors, an important candidate responsible for the local regulating effects on
Leydig cell function, have been studied in many laboratories for the last twenty years. It has
been demonstrated that insulin-like growth factor I (IGF-I) and its mRNA are present in the
testis (Handelsman et al, 1985; Lin et al, 1990). In the rat, IGF-I is secreted from both Sertoli
cells (Smith et al, 1987) and Leydig cells (Caileau eta!, 1990). Testicular IGF-I synthesis is
regulated by gonadotropins. LH and FSH stimulate the secretion ofiGF-I by Leydig cells and
Sertoli cells, respectively (Caileau et al, 1990). IGF-I receptor has been reported to be present
in Leydig cells (Lin et a!, 1986) and hCG treatment up-regulates IGF-I receptor gene
expression in Leydig cells (Nagpal eta!, 1991). Although it has a small mitogenic effect on
Leydig cells isolated from the immature rats (Khan eta!, 1992), the main effects ofiGF-I are
regulating Leydig cell differentiated functions. IGF-I enhances hCG-stimulated cAMP and
testosterone production by Leydig cells from both the pubertal and adult rats (Gelber et al,
1992). The number of LH/hCG receptor in Leydig cells is also increased by IGF-I
administration (Lin et al, 1986). The stimulatory effects ofiGF-I on Leydig cell differentiated 30
functions are further supported in the studies using antiserum to IGF-1. The addition ofiGF-I
antiserum decreases Leydig cell steroidogenesis in both Leydig cell culture and Leydig/Sertoli
cell coculture (Verhoeven and Cailleau, 1990). The above results indicate that IGF-I regulates
Leydig cell function through a paracrine/autocrine mechanism.
Another growth factor involved in regulating Leydig cell function is transforming
growth factor~ (TGF-~). Several studies have demonstrated that rat testis possesses TGF-~
like irrununoreactive material (Teerds and Darrington, 1993; Avallet et al, 1994) and TGF-~
mRNA (Avallet et al, 1994). In the immature animals, both Leydig cells and Sertoli cells
express TGF-~ mRNA and immunoreactive TGF-~ (Avallet eta!, 1994), while Sertoli cells
are found to contain TGF-~ mRNA throughout testicular development (Mullaney and
Skinner, 1993). It is still a controversial as to whether adult Leydig cells express TGF-~.
Specific receptors for TGF-~ with high affinity have been reported to be present in Leydig
cells in the rat testis (Morera et a!, 1988). Compared to IGF-I, TGF-~ has been shown to
have a potent inhibitory effect on Leydig cell differentiated function. Treatment of porcine
Leydig cells with TGF-~1 results in a progressive decrease in LH/hCG receptor number without affecting the binding affinity (Avallet et al, 1987). TGF-~ also causes a decrease in
hCG-induced cAMP and testosterone production in Leydig cells (Avallet et a!, 1987).
Although the in vitro inhibitory effects ofTGF-~ on Leydig cell differentiated function is well
documented, its potential paracrine/autocrine role in vivo is still unknown.
While IGF-1 and TGF-~ are the two most important and well-studied growth factors
involved in regulating Leydig cell function, more and more studies have suggested that other 31 growth factors, including transforming growth factor a. (TGF-a.), epidermal growth factor
(EGF), fibroblast growth factor (FGF), also exert paracrine/autocrine effects on Leydig cell maturation and function. Inhibin, GnRH and different types of cytokines have also been studied and demonstrated to influence Leydig cell function in vitro.
The role of estrogens in testicular function has been a controversial issue for years.
Interest has been growing since the discovery that the interstitium of the rat testis contains a specific, high affinity receptor for estrogen (Brinkmann et at, 1972; Abney, 1976). Early in fetal life, exposure of the male fetus to elevated circulating levels of estrogen increases the incidence of cryptorchidism (Juenemann eta!, 1986). Damber and Bergh (1980) reported an increase in estradiol concentration in the testes of cryptorchid rats. Most of the studies concerning the effect of estrogen on Leydig cells involve desensitization, a process in which a high level hCG administration to animals induces receptor loss, a refractoriness of adenylate cyclase, and a reduction in cAMP and testosterone production. Several groups have demonstrated that hCG induces testicular aromatase, which stimulates estrogen synthesis in the testis (Valladares and Payne, 1979; Canick et at, 1979). Cigorraga et at (1980) reported that the estrogen antagonist tamoxifen prevented the down-regulation of steroidogenesis induced by hCG. Evidence of estrogen inhibitory effect on Leydig cell function has been provided by both in vivo and in vitro studies. Estrogen treatment in vivo causes a decrease in testosterone production without a concomitant decrease in serum LH level (Tsai-Morris et at, 1987), suggesting that the inhibitory actions of estrogen occur directly in the testis and need not involve alterations in pituitary function. Administration of estrogen in vitro results in a decline in Leydig cell testosterone production (Moger, 1980), further confirming the 32 direct effects of estrogen.
In the rat and human, the testis is both an estrogen producing and estrogen target organ. It has been reported that estrogen concentration in spermatic venous blood is 10-50 fold higher than that in peripheral blood in several species (Kelch et al, 1972), suggesting that a physiological concentration of estrogen within the testis may be sufficient to exert a local action. Sertoli cells are the major source of testicular estrogen before puberty, whereas Leydig cells become the major or only site of aromatization after puberty development (Payne et al,
1987; Papadopoulos et al, 1986). Leydig cell aromatase activity and estrogen synthesis is stimulated by hCG administration (Payne et al, 1987). It is generally accepted that the estrogen receptor is localized in Leydig cells in the interstitial compartment of the testis (Lin et al, 1982). Miyashita et al have identified an unoccupied but transformed nuclear estradiol receptor in cultured mouse Leydig cells even in the absence of estrogen (Miyashita et al,
1990). Multiple estrogen binding sites also are reported to be present in mouse Leydig tumor cells (Sato et al, 1985).
The sites of estrogen's inhibitory effect on Leydig cell steroidogenesis involve the specific steroidogenic enzymes. Several groups have shown that estrogen decreases P-450 17~ and 17f3-hydroxysteroid dehydrogenase activity in both adult and immature rats (Brinkmann et al, 1980; Tsai-Morris et al, 1987). Another effect of estrogen on Leydig cells has been suggested to regulate Leydig cell development. Dhar and Setty in their studies demonstrated an absence of mature Leydig cells in the testes of adult rats that had received a single injection of estradiol at 5 days of age ( Dhar and Setty, 1976), indicating Leydig cell development in immature rats is inhibited by estradiol. In vivo administration of estradiol to mature rats for 33 a period of2 days resulted in a 50% decrease in the 3H-thymidine incorporation into isolated interstitial cells (Saez et al, 1978). Later, work by Abney and Carswell (1986) further demonstrated that estradiol treatment in vivo inhibited 3H-thymidine incorporation in Leydig cells isolated from mature rat testis, in the presence or absence of simultaneous hCG administration. All these studies suggest a role of estrogen in Leydig cell proliferation and differentiation.
Ethylene dimethanesulphonate (EDS)-treated rats:
An animal model to study Leydig cell development and function
In recent years, a new animal model has been utilized to study Leydig cell development and function in the rat testis. Ethylene dimethanesulfonate (EDS), an alkylating agent, was recognized as an antitumor chemical in mice and rats over 30 years ago (Sudiura and Stock, 1957). In 1966, Jackson demonstrated its antifertility effect in the adult male rats for the first time. For many years thereafter the precise action of EDS upon the testis remained poorly understood. It was recognized that the antispermatogenic effect ofEDS on the rat testis is not permanent since full spermatogenesis and fertility is restored about 8-10 weeks after EDS injection (Jackson and Jackson, 1984). Later studies using histological techniques and electron microscopy showed that a single intraperitoneal injection ofEDS results in a complete destruction ofLeydig cells in the rat testis (Kerr et al, 1985; Kerr et al,
1987; Molenaar et al, 1986; Jackson et al, 1986). One day after EDS injection the interstitial tissue exhibits degenerating Leydig cells, abundant pyknotic interstitial cells, deposition of cellular debris and extensive networks of fibrillar material. Testicular macrophages progressively remove Leydig cells from interstitial tissue by phagocytosis. This results in the 34 total absence ofLeydig cells 3-14 days post-EDS and the absence of any detectable specific
125I-hCG binding to testis homogenates. Serum testosterone decreases to very low levels while LH and FSH in serum rise dramatically. Macrophage numbers increase by three-fold by day 3 after EDS injection and decrease thereafter but remain the main cell type in the
interstitium until 14 days post-EDS. At 2-3 weeks post-EDS injection, single or paired Leydig
cells appear again in the interstitium in proximity to perivascular and peritubular tissues.
These newly regenerated Leydig cells share similar morphological and steroidogenic features
with fetal type Leydig cells, demonstrated by large amount of lipid droplets in the cytoplasm
(Jackson et a!, 1986) and high levels of Sa-reductase activity (Myers and Abney, 1990).
Between 4-6 weeks post-EDS, the newly formed fetal-like Leydig cells are transformed
moi'J!hologically and functionally into adult Leydig cells. The regeneration ofLeydig cells is
reflected by gradually elevated serum testosterone levels which returns to normal range
between 6-8 weeks after EDS treatment.
Many studies have demonstrated that the regenerated Leydig cells after EDS injection
are derived from precursor cells in the interstitium of the testis (Kerr eta!, 1985; Jackson et
a!, 1986; Kerr eta!, 1987; Teerds eta!, 1989; Teerds eta!, 1990). A total destruction and loss
of Leydig cells occurs by day 2 after EDS treatment and the return of serum testosterone level
is associated with the reappearance ofLeydig cells in the interstitial tissue. This indicates that
there is no possibility that the new generation of Leydig cells arise by recovery and cell
division of residual Leydig cells. The observations that the cells in the interstitium during
Leydig cell regeneration exhibit cytological characteristics intermediate to Leydig cells and
elongated connective tissue cells strongly support the view that these intermediate cells are 35 derived from connective tissue cells (Jackson et al, 1986). This is further supported by studies using 3H thymidine incorporation. Following administration ofEDS, proliferating interstitial cells were labelled in a pulse-chase experiment by way of three 3H thymidine injections on days 2, 3 and 4 post-EDS. It was found that some of the newly formed Leydig cells at day
14 post-EDS were labelled with 'H thymidine, indicating that these Leydig cells were derived from precursor cells.that had incorporated 3H thymidine at days 2, 3 and 4 post-EDS (Teerds et al, 1990). The studies further demonstrated that the proliferative activity of the mesenchymal cells in the interstitium which are presumed to be the precursors ofLeydi~ cells increased considerably at day 2 after EDS administration and returned to control levels at day
7. The labelling index ofLeydig cells was approximately 100 times higher 21 days after EDS administration than that of the untreated controls, suggesting that some Leydig cells were formed from the division ofthe newly formed Leydig cells. It has been concluded from these studies that the regeneration of Leydig cells after EDS administration is the result of two successive waves of proliferation, namely of the precursor cell (mesenchymal cells) and of the newly formed Leydig cells. However, Myers and Abney (1991) reported that injection of rats at days 2 and 4 post-EDS with 3H thymidine did not result in labeled Leydig cells at 30 days post-EDS. This suggests that the initial peak of 3H thymidine incorporation is not associated with Leydig cell development. On the other hand, it is possible that EDS directly or indirectly destroys interstitial cells other than Leydig cells and this increase in 3H incorporation following EDS treatment may represent proliferation of connective tissue cells.
The other observation in the above studies was 3H thymidine injection at 20 days post-EDS resulted in a high percentage ofLeydig cells that were labeled when rats were ~lied 6 hours \
36 or 4 days (24 days post-EDS treatment) after 3H injection. The high number oflabeled non
Leydig cells present during this period suggests that precursor cell proliferation and differentiation may be involved in this process.
The regeneration of Leydig cells after EDS treatment is dependent upon LHihCG. It is clearly demonstrated that almost no Leydig cells were formed in the rats bearing a testosterone implant duriilg the first 4 weeks after EDS administration (Teerds et al, 1989 c).
Serum LH levels in these rats were suppressed to less than 2 ng/ml by testosterone implant and a normal FSH level was maintained (Molenaar et al, 1986 b). Daily injection of hCG beginning directly after or 3 days after EDS administration stimulated the repopulation of
Leydig cells with apparently normal morphology and steroidogenic activities on day 35 post
EDS. Similarly, when hypophysectomized rats were treated with EDS to destroy their Leydig cells, administration of hCG was able to induce Leydig cell regeneration (Molenaar et a!,
1986). These results indicate that precursor cells depend on LH/hCG for their proliferation and differentiation into Leydig cells.
Although LHthCG is exclusively required for Leydig cell regeneration after EDS treatment, it is not the only factor that governs this process. If some form of seminiferous tubule damage/dysfunction (e.g., cryptorchidism, x-irradiation, treatment with busulfun) is introduced to rats at the time ofEDS treatment, it exerts an influence on the pattern of Leydig cell regeneration. Kerr and Donachie (1986) reported that if rats were made unilaterally cryptorchid at the time ofEDS injection, Leydig cells regenerated faster in the cryptorchid testis than in the contralateral scrotal testis. Moreover, Leydig cell regeneration was most profound around the most damaged tubules. These findings suggest that multiple factors from 37 the adjacent seminiferous tubules are involved in Leydig cell regeneration presumably through a paracrine mechanism.
The EDS-treated rat model also provides a possibility to study the effects of estrogens on Leydig cell development. Histological and functional studies have demonstrated that daily administration of estradiol to EDS-treated rats, with or without simultaneous hCG administration, inhibited the regeneration of Leydig cells (Abney and Myers, 1991).
Furthermore, estradiol was most effective in blocking Leydig cell regeneration when administered between days 5 and 16 post-EDS treatment, a time when histologically identifiable Leydig cells were absent. This suggests that the Leydig precursor cells might be the target site of estradiol inhibitory action. MATERIALS AND METHODS
Animals and treatment
Mature (60-65 days of age) and immature (21 days of age) male Sprague-Dawley rats were purchased from Harlan Industries (Indianapolis, IN) and housed in a 12 hour light-dark environment with water and rat chow provided ad libitum. Mature rats received a single intraperitoneal injection ofEDS, at a concentration of 100 mg/Kg BW. These EDS- treated rats were sacrificed by decapitation or C02 asphyxiation on days 2, 4, 10, 16, 20, 24, 30, 36,
45 and 60 post treatment. Immature rats were sacrificed at the age of21 days for the in vitro study of precursor cell differentiation.
Hormones and chemicals
All chemicals for the synthesis ofEDS were distilled prior to use. Ethylene glycol and pyridine were purchased from Fisher Scientific (Pittsburgh, PA) and methane sulfonyl chloride was obtained from Sigma Chemical Co. (St. Louis, MO). The methods used to synthesize
EDS was previously described by Jackson and Jackson (1984) and modified by Myers and
Abney (1990). Briefly, 9.4 g ethylene glycol was mixed, with 45 ml pyridine which was cooled down to 5-1 0°C in a salt/ice water mixture. Thirty four grams of methane sulfonyl chloride was added dropwise (approximately 750 ullminute) to the mixture while it was kept at 10°C.
The resulting mixture, containing EDS, was allowed to warm up to 25°C. Thereafter, it was
38 39 added to an acid-ice bath composed of 45 ml concentrated sulphuric acid poured over approximately 800 ml crushed ice with constant stirring. An oily white precipitate was formed
and collected by filtration on a Whatman 40 filter paper (Whatman International Ltd,
Maidstone, England). ·The precipitate was washed with cold water and methanol and air dried overnight. This white powder was dissolved in cold chloroform (Fisher Scientific). Ice cold methanol was added slowly until crystals were formed. The synthesized EDS was twice
crystallized from cold chloroform and collected by filtration. A final yield of approximately
10 g with a melting point of 45°C was obtained. EDS was dissolved in 1:1 (v/v) dimethyl
sulfoxide (Sigma Chemical Co.) and water for injection.
[1,2,6,7-3H testosterone (94 Ci/mmol), [1,3,5(10)- H] 17P-estradiol (85-115
Ci/mmol) were purchased from New England Nuclear (Boston, MA). Unlabeled testosterone
and estradiol were provided by Steraloids (Wilton, NH). Etiocholan-3 P-ol-17-one was
obtained from Sigma Chemical Co. hCG used for stimulation of testosterone production in vitro was a product of Series Laboratory, Inc. (Phoenix, AZ). The antisera to testosterone (x-
181 and R-15P) were purchased from Radioassay Systems Laboratories Inc. (Carson, CA)
and ICN Pharmaceutical Inc. (Costa Mesa, CA), respectively.
Acetone, chloroform, EDTA (disodium salt), ethidium bromide, formaldehyde, gelatin, hematoxylin stain (Gill's formulation #3), magnesium chloride, potassium chloride, potassium phosphate monobasic, propane, SDS (sodium dodecyl sulfate), sodium chloride,
sodium citrate, sodium hydroxide, sodium phosphate monobasic, sodium phosphate dibasic,
sodium phosphate dibasic heptahydrate, trichloritic acid, and xylene were obtained from
Fisher Scientific. 40
Ampicillin, ammonium sulfate, boric acid, chloramphenicol, collagenase (type L), dextran sulfate (500,00 MW), diethyl pyrocarbonate (DEPC), 13-D(+) glucose, glucose oxidase (type VII-S), HEPES, Mops (3-[N-morpholino ]propane-sulfonic acid), 13- nicotinamide adenine dinucleotide (!3-NAD), nitro blue tetrazolium (NBT), Percoll, polyvinylpyrrolidone (PVP), silane (y-Methacryloxypropyltrimethoxysilane), sodium acetate, sodium azide, soybean trypsin inhibitor, tetracycline, thimerosal, Triton X-100, trizma base
(Tris[hydroxymethyl]-aminomethane) and trizma hydrochloride were purchased from Sigma
Chemical Co.
AMV reverse transcriptase, bulk deoxynucleotide triphosphate ( dNTPs ), Klenow fragment, Lambda Hind III DNA marker, PCR marker, Oligo (dT)15 primer, restriction enzymes EcoR I, Pst I and Xbal I, RNasin ribonuclease inhibitor, Taq DNA polymerase,
Wizard DNA purification system (maxi, mini, PCR) were obtained from Promega
Corporation (Madison, MI).
Escoscint scintillation fluid was obtained from National Diagnostics (Atlanta, GA).
RNAzol was purchased from Biotex Laboratories (Houston, TX). Gene clean kit was obtained from Bio 101 (La Jolla, CA). Agarose (ultra pure DNA grade), low melt agarose, xylene cyanole FF and Zeta probe blotting membrane were obtained from Bio-rad (Hercules,
CA). DMEM, Ham's F12, Penicillin Streptomycin, rabbit 13-globin mRNA, 18S/28S RNA standard and SS RNA standard were purchased from Gibco BRL (Grand Island, NY). Yeast extract and tryptone were provided by Difco (Detroit, MI). Bromo phenol blue was a product ofHarlco (Philadelphia, PA). Dextran and a.l2P-dCTP were obtained by ICN Pharmaceutical
Inc. Genescreen plus hybridization transfer membranes were products of New England 41
Nuclear (Boston, MA). Density marker beads, Nick 1M column Sephadex G-50 and
Oligolabelling kit were purchased from Pharmacia Inc. (Piscataway, NJ).
The rabbit polyclonal antibody (ER 715) against rat estradiol receptor was provided by the National Institutes of Health (Bethesda, MD). A synthetic estrogen receptor peptide was also included for specificity studies. ABC kit, avidin-biotin blocking kit, biotinylated goat anti-rabbit IgG, normal goat serum, the diaminobenzidine (DAB) substrate kit and vectashied mounting solution were purchased from Vector (Burlingame, CA). Tissue-Tek O.C.T. compound was obtained from Baxter Scientific (Atlanta, GA). Bodipy goat anti-rabbit fluorescent antibody, Dio (3,3'-dipentyloxacarbocyanine) and Nile red (9-diethylamino-5H benzo[u]phenoxazine-5-one) were products of Molecular Probes, Inc. (Eugene, OR). Linker antibody (goat anti-rabbit IgG) and ClonoP AP were obtained from Stemburger Monoclonals
Inc. (Baltimore, MD). Goat anti-HRPIFITC was purchased from Jackson ImmuunoResearch
Laboratories, Inc. (West Grove, PA).
DH5.. Escherichia coli and Escherichia coli containing pHcGAPD were provided by
American Type Culture Collection (ATCC, Rockville, MD). Plasmid with rat P-4508cc eDNA insert was a gift from Dr. JoAnne S. Richards at Baylor College of Medicine (Houston, TX).
Rat LH receptor eDNA Escherichia coli was a gift from Dr. Aaron J.W. Hsueh at University of California, San Diego. Escherichia coli containing rat estrogen receptor eDNA was a gift from Dr. Vrrendra B. Mahesh and originally provided by Dr. Muramatsu (Koike et al, 1987).
P45017 .. eDNA was provided by Dr. Richard H. Fevold at University of Montana and generously given by Dr. Virendra B. Mahesh. 42
Buffers and solutions
LB (Luria-Bertani) medium: 10 g tryptone, 5 g yeast extract, 5 g NaCI/liter
DMEMIF12 buffer: 1:1 mixture of DMEM and Fl2, 1.2 giL
sodium bicarbonate, 15 mM HEPES, 0.1%
BSA, 25 mg!L soybean trypsin inhibitor,
pH7.4
lOXHBSS: 4 g KCl, 0.6 g KH2P04, 80 g NaCl, 0.9 g
(Hanks' balanced salt solution) N~HP04 ·7H20, 10 g D-glucose/Iiter
Percell buffer: 10 ml lOX HBSS, 35 mg NaHC03, 420 mg
HEPES, 25 mg BSA, 90 ml distille~ H20
Isotonic Percell: 100 mi Percell, 9 ml lOX HBSS
PGB (phosphate gelatin buffer): 0.154 M NaCl, 0.039 M NaH2P04, 0.016
M N~HPO ,4 0.01% merthiolate, 0.1%
gelatin
0.1 M phosphate buffer 115.5 mi 0.2 M sodium phosphate monobasic,
384.5 ml 0.2 M sodium phosphate dibasic,
distilled H20 500 ml, pH 7.3
10 mM PBS (phosphate buffered saline) : 11.6 mi 0.2 M sodium phosphatemonobasic,
38.4 ml 0.2 M sodium phosphate dibasic, 8.5
g sodium chloride/liter
50XTAE: 242 g Tris base, 57.1 ml glacial acetic acid,
100 ml 0.5 M EDTA (pH 8.0)/liter 43
5XTBE: 54 g Tris base, 27.5 g Boric acid, 20 rn1 0.5
M EDTA (pH 8.0)/liter
6X gel loading buffer: 0.25% bromophenol blue, 0.25% xylene
cyanol FF, 30 % glycerol in H20
1OOX Denhardt's solution 1 g polyvinylpyrrolidone (MW 40,000), 1
g bovine serum albumin, 1 g Ficoll 400/50
rn1
10 X MOPS buffer: 0.2 M morpholinopropanesulfonic acid, 0.1
M sodium acetate, 10 mM EDTA, pH 7.2
2ox sse: 3 M NaCI, 0.3 M sodium citrate
20XSSPE: 3 M NaCl, 0.2 M NaH2P04, 0.02 M EDTA,
pH7.4
Prehybridization!hybridization solution: 50% formamide, 5X SSPE, 5X Denhardt's
(Northern blot) solution, 1% SDS, 10% dextran sulphate
RNA denaturing buffer: 360 ul formamide, 80 ul lOX MOPS, 130
ul 3 7% formaldehyde, 50 ul 80% glycerol,
40 ul bromophenol blue (saturated solution),
90 ulH20
1 M Na2PO 4 ·7H2 0, 4 ml 85% H3 P04 , pH
7.2
Prehybridization!hybridization solution: 50% formamide, 0.12 M N~HP04 , 0.25 M
(Southern blot) NaCl, 7% SDS, 1 mM EDTA 44
TE buffer: 10 mM Tris-HCl, 1 mM EDTA, pH 7.6
DPBS: 0.2 g KCl, 0.2 g KH2P04, 0.047 g MgG.J ,
(Delbecco's phosphate buffered saline) 8 g NaCl, 1.15 g N~HPO/liter
*·. All solutions and reagents used for RNA analysis were made in DEPC-treated H20 .
The concentration ofDEPC was 200 ul!L. The others were made in distilled H20.
Serum testosterone determination
EDS-treated rats at different times post-EDS and age-matched untreated controls were sacrificed by decapitation. Blood was collected and centrifuged at 3000 rpm for 10 min for the collection ofthe serum. Steroids were extracted from serum using ether (Meiner and
Abney, 1980). Five ml redistilled ether was added to 1 ml serum and mixed by vortexing for
3 0 sec. The tubes were immediately inserted into a dry ice/methanol bath for the aqueous phase to freeze. The upper organic phase was decanted into a clean tube and kept in a water bath at 37"C under a stream ofN2 gas until the ether was completely evaporated. The residue was dissolved in 1 ml PGB for testosterone radioimmunoassay.
Testosterone levels were determined using the antiserum to testosterone (X-181 or
R-15P). These antibodies exhibit high affinity for testosterone and low cross reactivity for other steroids. The cross reactivity for dihydrotestosterone (DHT) and 17f3-estradiol is less than 0.1%. The antibody was used at a final dilution of1: 140,000 for X-181 and 1:11,200 for
R-15P at which they bound 50% of the 3H testosterone (10, 000 cpm/tube). The unlabeled testosterone amounts of the standard curve were 5, 10, 25, 50, 100, 150, 250 and 500 pglper assay tube. The assays were conducted at 4°C in phosphate-buffered saline containing 0.1%
100 Bloom gelatin (PGB). A 0.2 ml suspension of0.625% Norit A charcoal and 0.0625% 45 dextran prepared in PGB was used to separate the bound and free steroids in the reaction.
(Myers and Abney, 1990). Five ml of Scintillation fluid was added to .the supernatant and the samples were counted in a Beckman scintillation counter. The data were analyzed on the
Beckman EIAIRIA Immunofit program.
Cell isolation
Testes were removed and placed on ice in DMEMIF 12 buffer. The procedures utilized to isolate Leydig cells and precursor cells were carried out essentially as described by
Klinefelter (1987) and Hardy (1990). Testes were decapsulated and the digestion of testes
was performed in DMEMIF12 buffer with 0.625 mg collagenase/mi. Digestion was carried out at 37°C for 25 min with 100 rpm shaking. The dispersed interstitial cells were separated from
the tubular compartment by filtration through surgical gauze and washed twice in DMEM/F12 by centrifugation at 2000 rpm, 5 min each to sediment the cells and remove the collagenase.
The number of the interstitial cells was counted on a hemacytometer. Two hundred and fifty to 400 million interstitial cells were resuspened in Percoll buffer and isotonic Percoll, 2.5 ml each. The 5 ml interstitial cell suspension was added to a tube containing 12 ml Percoll buffer and 18 ml isotonicPercoll, giving the final Percoll concentration of 55%. APercoll gradient tube containing density marker beads of 1.062 g/ml and 1.076 g/ml was included each time.
The tubes were mixed and Percoll gradients were generated by centrifugation at 20,000x g for 60 min at 4°C. A fraction containing precursor cells was removed from the Percoll gradient between a density of 1.064-1.070 g/ml and the Leydig cell enriched fraction was
collected from a density heavier than 1.070 g/ml. Each fraction was diluted with DMEM/F12 and centrifuged to remove the Percoll while pelleting the cells. 46
Histochemical study (313-HSD staining)
Histochemical studies were performed on cells isolated from the testes of the control and EDS day 20 rats. The procedures were carried out as described by Klinefelter et a)
(1987). Fifteen ul of isolated Leydig cell or precursor cell fraction containing 100,000 cells were spread on silane coated slides and the cells were air dried for at least 1 hr. Reagent A containing 1 mg nitro blue tetrazolium dissolved in 0.6 m1 substrate solution (1mg Etiocholan-
3j3-ol-17-one/ml DMSO) was mixed with reagent B containing 10 mg NAD dissolved in 9.5 m1 warm DPBS. The mixture was added to the dried cells. A negative control for each cell preparation was included, without the addition of the substrate in reagent A The reaction was carried out in a humid box for 90 min. Slides were washed with distilled water and fixed for
10 min, using 10% formaldehyde in DPBS with 5% sucrose. Finally, slides were washed with distilled water and coverslips were mounted with 1:1 glycerol!DPBS. Slides were sealed with nail polish and observed under light microscope at IOOX magnification. Two hundred cells in a minimum of three fields of each preparation were counted and the cells containing purple blue staining were recognized as 3j3-HSD positive. The experiments were repeated three times.
In vitro incubation and testosterone production
6 Aliquots of0.25X10 cells from each fraction were incubated in DMEMIF12 media at 37°C in an atmosphere of 5% C02, 5% 0 2 and 90% N2• The media were removed after 3 hr incubation when the viable cells were attached to the plates. Fresh media were added and the cells were incubated in the presence or absence of 50 miD hCG/well for 24 hr, as described earlier (Myers and Abney, 1988). The media were collected and testosterone 47 production was determined by RIA as described above.
RNA extraction
Total RNA was extracted from the precursor cell and Leydig cell fractions isolated from Percoll gradients using RNAzol. Briefly, RNAzol was added to the cell pellets at a concentration of0.2 ml RNAzol/106 cells. RNAzol was pipetted back and forth to mix with the cells. The ceii-RNAzol mixture was placed on ice for 10 min. One tenth volume of chloroform was added and mixed thoroughly by shaking the tubes for 3 0 sec. The mixture was kept on ice for 15 min and then centrifuged at 12,000 xg for 15 min at 4°C. The aqueous phase was collected and an equal volume of cold isopropanol was added. The tubes were mixed and kept at -20°C for 45 min. RNA was then precipitated by centrifugation at 12,000 xg for 15 min, 4°C. RNA pellets were washed twice with 75% ethanol by centrifugation at
10,000 xg, 10 min each. RNA pellets were air dried for 10 min. 1E buffer was added to dissolve the RNA pellets and 100% ethanol was added to make the final ethanol concentration of75%. These RNA samples were stored at -80°C until northern blot analysis and RT-PCR were performed. For the determination of RNA concentration, an aliquot was taken out and Ill 0 volume of3M sodium acetate was added. The tubes were placed at -20°C for 20 min and centrifuged at 12,000 xg for 15 min at 4°C. RNA pellets were air dried and resuspended in 100 ul DEPC H20. The concentration was determined by U.V. absorbance at 260 nm in a UV 160 U UV visible recording spectrophotometer(Shimadzu). The ratio of
260run/280nm was generally greater than 1.6. One absorbance unit at 260 nm is equivalent to 40ugRNA. 48
Preparation of eDNA probes
To prepare the eDNA probes, plasmids with the inserts were transfonned to
Escherichia coli using the CaC12 method (Sanbrook, 1989). LH receptor plasmid (clone P4) contains a fragment corresponding to rat LH receptor eDNA bases 249 to 849 reported by
McFarland et al (1989). This plasmid carries antibiotic resistance to ampicillin and the fragment was inserted at the restriction site EcoR I. To prepare the competent cells for transfonnation, 40 X 106DHsc.Escherichia coli were suspended in 50 mM CaCl/10 mM Tris
HCl (pH 8.0). The tubes were placed on ice for 15 min. Cell suspension was centrifuged at
4000 xg for 5 min and the supernatant was discarded. Cells were resuspended in 50 mM
CaCl/10 mM Tris-HCI. Forty ng of vector DNA was added to the cell suspension and incubated at 42°C for 2 min. One ml ofLB medium was added to the tube and the cells were incubated at 31'C for 30 min. An aliquot of the transfonned cells was spreaded onto a 1.5% agar plate containing 75 ug ampicillin/ml and incubated at 37°C overnight. Single colonies were picked out with sterile rods and suspended in LB media with 75 ug ampicillin/mi. The suspended cells were incubated at 31'C with vigorous shaking for 24 hr. An aliquot of the cell suspension was pelleted and the vector DNA was isolated using the Promega Wizard mini
DNA purification system. One ug of the vector DNA was digested with the restriction enzyme EcoR I at 37°C for 1 hr. The digested mixture was electrophoresed on a 1% agarose gel and the gel was observed under U.V. light. The insert eDNA was identified with the size of 600 bp (Fig. 4) and the size of the vector DNA pGEM-4Z is 2. 7 kb. Once the correct eDNA size was ascertained, the rest of the cells were incubated in 500 mJ LB (75 ug ampicillin/ml) at 31'C with vigorous shaking. Chloramphenicol (2.5 ml, 34 mg!ml in ethanol) Figure 4. Gel electrophoresis of the LH receptor plasmid digestion. Plasmids were
isolated from the Escherichia coli that were transformed with the rat LH
receptor plasmid (clone P4). Digestion of the plasmid was performed
by incubating the plasmid DNA with the restriction enzyme EcoR I. The
digested products of the plasmid were electrophoresed on 1% agarose
gel. The size of the LH receptor eDNA insert and the plasmid is 0.6 Kb
and 2.7 Kb, respectively. 49
+-1en ,_ Q) C) ~,_ "'C co "'C E E <( en z co 0 a..
4.4 Kb- 2.3 Kb- 2.0 Kb_,_
0.5 Kb- 1 2 50. was added after 6-8 hr when the cells were at log phase and incubation was carried on overnight. Cells were harvested and the vector DNA was isolated with Maxi DNA purification system. The vector DNA was digested with EcoR I and the digested mixture was electrophoresed on 1.5% low melting point agarose gel at 4°C. Bands ofLH receptor eDNA
(600 bp) were excised under a U.V.light and the eDNA was purified with gene clean kit.
pGEM vector with P-450""' eDNA insert was transformed to Escherichia coli as described above. This plasmid carries antibiotic resistance to tetracycline. The fragment was inserted at the restriction site EcoR I and the size of the insert is 1.2 Kb. Transformed
Escherichia coli were incubated in LB with 50 ug tetracycline/ml and the eDNA was purified.
Escherichia coli transformed with GAPD eDNA plasmid was purchased from ATCC and the eDNA fragment was inserted at the restriction sites Pst I and Xba I. Rabbit f3-globin eDNA probe was synthesized by PCR. The primer set used in the synthesis ofthe eDNA probe flanks rabbit 13-globin eDNA between bases 97 and 421 (Kafatos eta!, 1977).
Northern blot analysis
Twenty ug total RNA from each cell fraction was utilized for northern blot analysis.
The procedures for northern blot were carried out essentially as described by Shan and Hardy
(1992). Briefly, RNA samples were precipitated as described above and dissolved in 20 ul
RNA denaturing buffer. The denaturation of RNA. was performed at 95°C for 2 min and the tubes were immediately placed on ice. The denatured RNA samples were loaded on 1% agarose gel containing IX MOPS. Electrophoresis was carried out at 80 volts for 4 hr in IX
MOPS and 0.66 M formaldehyde gel running buffer. RNA was transferred onto Genescreen hybridization transfer membranes in lOX SSC for 16-24 hr. Membranes were briefly rinsed 51 in 2X SSPE and fixed by baking at 80°C for 2 hr. Prehybridization was performed at 42°C for
2-4 hr in prehybridizationlhybridization buffer with the addition of 100 ug denatured sheared salmon sperm DNA/mi. Fifty to one hundred ng of eDNA probes were labeled with 50 uCi
32P-dCTP and the separation of the labeled eDNA from the dNTPs was performed on the
Sephadex G-50 columns. Hybridizations were carried out overnight in fresh buffer with 32P labeled eDNA probes as the template. Membranes were rinsed in 2X SSPE, washed in 2X
SSPE/1% SDS and 0.5X SSPE/1% SDS at room temperature, 15 min each. Further washing was performed in O.lX SSPE/1% SDS at 65°C for 15 min. The membranes were wrapped in saran wrap and exposed to Kodak X-ray film at -80°C for an appropriate time between 24 and
48 hr. For the detection of different target mRNA, membranes were stripped by washing twice in O.lX SSC/1% SDS at 95°C, 20 min each. Hybridization with each eDNA probe was performed as described above. Glyceraldehyde 3-phosphate dehydrogenase (GAPD) mRNA was used as an internal control. Quantification for the messenger RNA was performed by scanning the autoradiograms with a Ultrascan XL Laser densitometer and the target mRNA levels were expressed relative to that of the internal control, GAPD. The experiments were conducted two to seven times, using 8-10 rats per group.
Estrogen receptor binding assay
The estradiol binding capacity of whole testicular cytosol in control and EDS-treated rats was determined as described previously (Abney, 1976). Testes were removed, decapsulated and minced on ice. The minces were diluted 1:5 (wt/vol) with TE buffer and homogenized in aPotter-Elvehjem glass-Teflon apparatus by 8-10 strokes. The homogenates were centrifuged at 17,000 xg at 4°C for 10 min. The supernatants were collected and 52 centrifuged at 105,000 xg for 60 min at 4°C in a Beckman L2-65B ultracentrifuge. The supernatants were collected as cytosol. One hundred ul of each cytosol was added to tubes containing increasing concentrations of3H-estradiol in the presence or absence of a 500 fold excess of unlabeled estradiol. The assays were incubated in TE buffer at 4°C for 18 hr.
Separation of bound from free estradiol was performed by the addition of 1.0 m1 of dextran coated charcoal (0.3% Norit A Charcoal and 0.03% dextran in TE buffer) and centrifugation at 2500 rpm for 10 min. Scintillation fluid was added to the supernatant and the samples were counted in a Beckman LS 3801 counter at a efficiency of 50-60% for 3H. The binding data were analyzed by the method of Scatchard (Scatchard, 1949). The experiments were
.conducted three to five times , using 2 to 4 rats per group.
Protein determination
Aliquots of cytosol (200 ul) were precipitated with 0.3 M perchloric acid for 30 min on ice. The precipitates were centrifuged at 1000 xg for 10 min. The supernatants were discarded and the pellets were solubilized in 0.1 M KOH. The cytosolic protein concentrations were then determined by the method of Lowry (1951) using BSA as a standard. Absorbance was determined at 750 nm in a spectrophotometer.
Immunohistochemical study on the testicular tissue
Immunohistochemical studies for identification and localization of the estradiol receptor in the testes of control and EDS-treated rats were carried out as described previously
(Sprangers et a1, 1989; Prins, 1992). Testes were removed from the animal and small sections were covered with tissue-tek O.C.T. compound, and frozen rapidly in liquid propane placed on liquid nitrogen. The frozen testes were stored in plastic tubes in liquid nitrogen. Frozen 53 testes were sectioned at 10 urn on a IEC Minotome cryostatic microtome and mounted on
gelatin coated slides. The slides were placed in acetone with CaCI2 desiccant at -80°C for 48-
72 hr for freeze-substitution. Thereafter, the slides were moved to -20°C and kept for 3 hr.
The mounted sections were fixed with 2% paraformaldehyde in 0.1 M phosphate buffer at 4°C for 20 min. After fixation, slides were placed in 85% ethanol with 1.5% polyvinylpyrrolidone
(PVP) twice for 2 min each. Slides were then washed three times in PBS containing 1.5%
(PVP) at 3 min each. They were then placed in 0.05% sodium Borohydride with 1.5% PVP in PBS twice for 2 min each. Thereafter, slides were washed in PBS containing 1.5% PVP three times for 3 min each and PBS containing 1.5% PVP and 0.1 % gelatin three times for
3 min each. To remove the endogenous peroxidase, sections were incubated with PBS containing 1 mM sodium azide, 10 mM glucose and 1 unit glucose oxidase/ml at 37°C for 30 min and then briefly rinsed in PBS containing 1.5% PVP and 0.1% gelatin. Sections were incubated with 2% normal goat serum (NGS) at 4°C for 20 min and then the extra liquid was discarded. Incubation with the primary antibody (ER 715, 1:1000 in PBS containing 0.1% gelatin) was carried 01,1t at 4°C overnight. ER 715 was raised in rabbit against a peptide representing a portion of the hinge region of the rat estrogen receptor. Preparation, purification and characterization of this antiserum were described previously by Furlow et a!
(1990). For the specificity study, another set of slides were incubated with estrogen receptor antibody in the presence of2 ug synthetic estrogen receptor peptide/ml to block the antibody binding to the tissue. The slides were then washed three times in PBS containing 0.1% gelatin for 3 min each. Avidin-biotin block was performed on the tissue sections. Sections were incubated with 1:1 avidin/2% NGS at 4°C for 15 min. Biotin incubation was carried out at 54 room temperature for 15 min. Liquid was drained off and sections were incubated with biotinylated secondary antibody (goat anti-rabbit IgG) for 30 min at room temperature. Slides were then washed three times in PBS containing 0.1% gelatin at 3 min each. ABC reagent was incubated with the tissue sections at room temperature for 60 min. Slides were washed three times in PBS containing 0.1% gelatin and 50 mM Tris (pH 7.6), respectively at 3 min each. Incubation with diaminobenzidine (DAB) as chromogen was performed for 10 min and
then slides were washed three times in 50 mM Tris and three times in distilled H20 for 3 min each. Sections were counter-stained with 1:4 hematoxylin, washed in distilled H20, dehydrated sequentially in 30%, 50%, 70%, 80%, twice in 95%, and three times in 100% ethanol for 5 min each. The final dehydration was performed in xylene, three times for 5 min each. Coverslips were mounted with Permount. The immunohistochemical studies were performed on 4-6 testes per group and ten fields were observed for each testis.
Immunohistochemical study on the isolated precursor cells and Leydig cells
To further identifY the cellular localization of estrogen receptor, precursor cells and
Leydig cells were isolated from the control and EDS-treated rats and immunohistochemical procedures were performed. Two million cells of each fraction were resuspended and washed in phosphate buffered saline (PBS). Cells were resuspended in 2 m1 PBS and 2 m1 of 4% paraformaldehyde was added to the cell suspension while vortexing. Fixation of the cells with paraformaldehyde was carried out at 4°C for 30 min. Cells were. then pelleted and washed in
PBS. Cell pellets were resuspended in 4 ml PBS and pelleted onto silane-coated slides using a Shandon cytospin centrifuge (Shandon Scientific Ltd), at a concentration of 100,000 cells/200 ul. Slides were washed in PBS at 4°C and incubated with 0.2% Triton-100 X for 5 55 min to permeablize the cells. Immunohistochemical reactions were carried out as described above.
Reverse transcription-polymerase chain reaction (RT-PCR)
An aliquot of I ug total RNA extracted from Leydig cell and precursor cell fractions
~e utilized for RT-PCR reaction. RNA was reverse transcribed by incubating at 37°C for
30 min with I5 units of AMV reverse transcriptase in a 20 ul reaction volume containing 50 mM Tris-HCI (pH 8.3), 50 mM KCI, I 0 mM MgClz, 0.5 mM spermidine, IO mM DTT, 25 unit RNasin ribonuclease inhibitor, 0.5 ug oligo (dT)15 primer and I mM of each dNTP. For quantification. of the ER mRNA by RT-PCR, I pg rabbit globin mRNA was added to each reverse transcription reaction as an internal standard. The reaction was terminated by placing the tubes on ice. Forty ul of IOO% ethanol and 2 ul of3 M sodium acetate was added to the reaction mixture. The tubes were kept at -20°C for at least 20 min and centrifuged at I2,000 xg at 4°C for I5 min. The pellets were air dried and resuspended in IO ul distilled H20.
All oligonucleotide primers used in PCR were synthesized on a 3 80B DNA synthesizer (Applied Biosystems). The estrogen receptor primer set flanks the rat estrogen receptor eDNA sequence from base I283 to 1665 (Fig. 5).The sequence of the estrogen receptor primer set is, 5'-primer: 5' CCG GGG AAG CTC CTG TTT G 3'; 3'-primer: 5' AGA
GAT GCT CCA TGC CTT TGT TAC 3'. The .expected molecular size of the PCR product is 382 bp, which flanks 4 exons and corresponds to a portion of the estrogen binding domain
(Koike et al, I987; Ponglikitmonglol et a!, I988). The primer set for rabbit f3-globin eDNA, as sequenced by Kafatos eta! (1977), flanks the eDNA sequence from base 97 to 421. The oligonucleotide sequence is, 5'-primer: 5' TGT GGG GCA AGG TGA ATC T 3'; 3'-primer: Figure 5. Genomic structure of the rat estrogen receptor. Open triangles refer
to the positions of the splicing sites predicted by comparison to the
human estrogen receptor gene structure.,Domains A-E are cited from Koike
eta!, 1987.· 56
Xbal EcoRI ' splicing sites ' ' " ·u -" : 'u '\7 '\7 '\7 ' ., '1>"1'.,- ' 5' .~ . -- AlB· .. ·.C.· -D - I· I F 1-:'7 3' ' ' .- . . . .. -' . -. ' ' ti:s3 16~5 2i>I 0 _PCR primers: 1283L ~ - _ .... l665U ;-_____;_----il Amplified eDNA: I 382bp _ 57
5' ATT CTT TGC CAA AAT GAT GAG AC 3'. The size of the PCR product for rabbit 13- globin is 324 bp.
An aliquot of2 ul ofthe reverse transcription mixture containing cDNAs was used for
PCR. The PCR reaction was performed oh a Thermal cycler (Perkin Elmer Cetus) in a 50 ul
volume.containing 50 pmol of each 5' and 3' primer, 2.5 mM MgCl2, 50 mM Tris-HCI (pH
9.0), 0.1% Triton X-100, 2 unit~ofTaq DNA polymerase and 0.2m M each dNTP. The mixture was overlaid with mineral oil and denatured at 95°C for 7 min. The PCR reaction was run for 3 5 cycles, each cycle included I min at 95°C, I min at 62°C and I min at 72°C.
Gel electrophoresis ofRT-PCR products
Ten ul of each RT-PCR product was taken from the reaction tubes and 2 ul DNA loading buffer was added. The RT -PeR products were electrophoresed in 2.5% agarose gel containing 0.1 ug ethydium bromide/ml in IX TBE buffer. The gel electrophoresis was carried out at 60 volts. PeR markers were included each time to demonstrate the size of the PeR product. Gels were observed under a U.V. light and pictures were taken using Polaroid film
(Fisher Scientific).
Southern blotting and quantitative analysis
One ul of each RT -PeR product was electrophoresed in 2.5% agarose gel in·IX TBE buffer. After electrophoresis, the gels were denatured in 0.5 N NaOH, I M NaCl for 30 min.
Thereafter, gels were neutralized in 1.5 M Tris-Hel (pH 7.4) and 3M NaCl for 30 min. RT
PeR products in the gels were transferred onto Zeta probe blotting membranes in lOX SSe for 16-24 hr. Membranes were rinsed briefly in 2X SSe and baked at 80°C for 2 hr.
Prehybridization was performed in DNA prehybridization/hybridization buffer for 5 min at 58
42°C. The buffer was replaced and hybridization with 32P labeled rat estrogen receptor eDNA as a template (1 X 106 cpm/ml) was carried out at 42' C overnight. Membranes were then rinsed in 2X SSC, washed in 2X SSC/1% SDS and 0.5X SSC/1% SDS at room temperature, each for 15 min. The last wash was performed in 0.1X SSC/1% SDS at 65°C for 15 min.
Membranes were exposed to Kodak X-ray film for variable times from 15 min to 3 hr.
Hybridization with 32P labeled rabbit 13-globin eDNA was performed after the membranes were stripped twice in 0.1X SSC/1% SDS at 95°C for 20 min each. Quantification ofthe hybridization signals was conducted by scanning the autoradiograms with a densitometer.
Estrogen receptor mRNA levels, reflected by the amount of the RT-PCR product hybridization signals, were standardized by those of the 13-globin in the same reaction tube.
The experiment were conducted four to six times with 5-6 rats each group. RT-PCR and
Southern blot analysis were performed two to three times for each experiment.
Flow cytometer analysis
Precursor cell and Leydig cell fractions were prepared as described above. An aliquot of 250,000 fresh cells was applied to flow cytometer for measurement of the size and granularity of the cells, using axial light loss (ALL) and 90"LS, respectively. For the determination of mitochondria content, the fluorescent probe DiO was prepared as a 1 mM stock solution in dimethylsulfoxide (DMSO) and was stored at 10°C in the dark. At the time of the staining, the stock solution was diluted to 100 uM in PBS and added to an aliquot
(300-500 ul) of the cell suspension such that final concentrations ofDiO and DMSO vehicles were 0.5 uM and 0.05% v/v, respectively (Dive et a!, 1992). DiO was incubated with cell suspensions for 10 min at 37°C. The green DiG-mitochondria fluorescence was examined at 59
SIS-545 nm on flow cytometer. For flow spectrofluoremetric analysis oflipid droplets, nile red was used as a fluorescent probe. Nile red was stored as a stock solution in DMSO (I mg/ml) and was diluted I: I 00 in PBS before staining. The final concentration of nile red was
100 ng/ml. A cell suspension (1 X 106/ml) was incubated with nile red at 37"C for 30 min. The yellow-gold fluorescence for lipid droplets were examined at 488 nm (Dive et al, 1992). ' Comparative study of estrogen receptor levels in the precursor cells and Leydig cells by immunofluorescent analysis
Precursor cells and Leydig cells were prepared and cyto spun onto the slides as described above. Slides were then washed twice in 100 mM Tris (pH 7.6) at 4°C, each for
5 min. Thereafter, slides were kept in acetone/methanol (I: I) and at -20°C for IO min. Slides were then rinsed and washed twice in 100 mM Tris at 4°C, 5 min each. Cells on the slides were incubated with 3% normal goat serum (NGS) for 10 min. Incubation with the primary antibody, rabbit anti-rat· estrogen receptor (I: 1000 in 1% NGS), was carried out at 4°C overnight. For the specificity studies, one more set was incubated with estrogen receptor antibody in the presence of2 ug/ml ofthe synthetic rat estrogen receptor peptide to block the antibody binding. At the end of the prim.ary antibody incubation, slides were rinsed and washed in 100 mM Tris, 5 min each. Slides were then incubated with the linker antibody (goat anti-rabbit Ig G, 1:50 in I% NGS) for'30 min at room temperature. After rinsing and washing in 100 mM Tris, the slides were incubated with rabbit ClonoP AP (I :200 in I% NGS). Then the slides were rinsed and washed again in 100 mM Tris and incubated with 1 :200 goat anti
HRP/FITC (diluted in 1% NGS) for 30 min at room temperature. Slides were rinsed and washed in 100 mM Tris and coverslips were mounted with Vectashield. Cells were observed 60 under a Axiophot fluorescent microscope. The intensity of fluorescence in the cells was analyzed on the Metamorph program.
Statistical analysis
One-way ANOVA: was used in the statistical analysis among the control and EDS
(analysis ofvari~nce) treated rats.
Two-way ANOV A: for the analysis of the flow cytometric data. The differences
between the control and EDS-treated rats and between Leydig
cells and precursor cells were analyzed by two-way ANOVA
Dunnett's t test: for the comparison of multiple groups ofEDS-treated rats to
the controls. Dunnett's t test was used in the analysis of
serum testosterone levels, testicular weight, interstitial cell
number, in vitro testosterone production by Leydig cells and
precursor cells, changes of cell number in the Leydig cell and
precursor cell fractions, LH receptor mRNA (1.8 Kb), Northern
blot results in the Leydig cell fraction (LH receptor, P-450"',
P-450m,), estrogen binding capacity, estrogen receptor mRNA
changes in the precursor cell and Leydig cell fractions.
Fisher's LSD t test: to detect the difference among multiple groups. Fisher's LSD
t test was used in the analysis ofLH receptor mRNA (7.0 Kb),
P-450", mRNA, P-45Q,a mRNA, immunofluorescence of
estrogen receptor in Leydig cells and precursor cells.
Correlation: was used in the quality control study of RT-PCR for the 61
measurement of the correlation coefficient between the amount
of the testicular RNA and the RT-PCR products. p values ofO.OS were assigned for the level of significance for statistical analysis. RESULTS
Characterization of Leydig cell degeneration and regeneration after EDS treatment
These initial experiments were designed to investigate the destruction of Leydig cells by EDS administration and the subsequent Leydig ceil regeneration. Mature male rats ( 60 days of age) were treated with a single dose of EDS (100mg/Kg body weight) by intraperitoneal injection. Animals were sacrificed at different days post-EDS treatment and age matched control rats were included in each experiment.
Trunk blood was collected from the control and EDS-treated rats and steroids were extracted from the serum. Testosterone levels were determined by RIA As shown in Fig. 6, serum testosterone level in control rats was 2.325 ± 0.098 ng/ml. Two days after EDS injection, the serum testosterone level decreased to 0.025 ± 0.003 ng/ml. Serum testosterone level reached the lowest point at day 4 post-EDS treatment (O.D15 ± 0.005 :-tg/ml). A slight, but nonsignificant increase in serum testosterone level was observed at day 10 post-EDS treatment, yet showed no statistical significance. Serum testosterone level increased significantly at day 16 (0.350 ± 0.032 ng/ml) in contrast to that at day 4 post-EDS treatment.
Thereafter, serum testosterone level further increased and was not significantly different from the control values by day 24 after EDS treatment.
Testes were removed from the control and EDS-treated rats and the weight of the
62 Figure 6. Temporal changes in serum testosterone levels after a single intraperitoneal
injection ofEDS (100 mg/kg BW). Age-matched controls (bar) were not
treated. These values represent the mean ± SEM of five to six rats from two
to six separate experiments. One-way ANOVA showed a significant difference
among the control and EDS-treated rats (p<0.005). The results were then
analyzed by Dunnett's t-test. An asterisk denotes a significant difference versus
the controls, p< 0.05. "a" denotes the first significant increase in serum
testosterone level in comparison to the lowest point at day 4 post-EDS
treatment (p<0.05). 63
2.5 ,...... E...... Ol 2 ...... ,c
(]) c 1.5 .....0 ...... (]) .If) .. .· 0 1· ...... If) (]) ...... 0.5 E :J '- (]) (/) 0 0 2 4 10 16 20 24 30 36 45
Days after EDS 64 testes was recorded. As shown in Fig. 7, testicular weight of the control rats was 1.749 ±
0.011 g!testis. Testis weight was not different from that of the control at day 2 after EDS treatment. At day 10 post-EDS treatment, testis weight decreased to 1.012 ± 0.051 g!testis.
Testis weight reached the lowest point at day 16 post-EDS treatment, 0.818 ± 0.038 g!testis.
Thereafter, testis weight increased significantly at day 20 post-EDS treatment (1.047 ±0.04 g!testis) and reached the value of 1.558 ± 0.012 g!testis at day 36 post-EDS treatment, which is not statistically different from that of the control.
Testes were decapsulated and digested with collagenase. The crude interstitial cells were collected. Fig. 8 shows the recoverable interstitial cell number in the control and EDS treated rats. Control rats yielded 89.92 ± 2.95 X 106 interstitial cells/testis. Interstitial cell number decreased to 58.22 ± 1.33 X 106/testis at day 2 post-EDS treatment. The decrease in the interstitial cell number in the testis continued at day 10 after EDS treatment and the lowest value was observed at day 16 post-EDS treatment, yielding 39.48 ± 2.69 X 106/testis.
Thereafter, testicular interstitial cell number increased gradually and a significant increase was observed at day 30 post-EDS treatment (68.33 ± 8.4 X 106/testis). Interstitial cell number returned to control values around 7 weeks post-EDS treatment.
Histochemical analysis of the ·precursor cell and Leydig cell fraction by 3~ hydroxysteroid dehydrogenase (3~-HSD) staining
Leydig cell and precursor cell enriched fractions were isolated from the testes of the control and EDS-treated rats. For the purpose of determining purity of the cell preparations, histochemical studies were conducted using 3 ~-HSD staining. Cells containing 3 ~-HSD were stained as purple-blue. The results of3~-HSD staining for the Leydig cell and precursor cell Figure 7. The influence of a single injection of EDS on testicular weight. The
values represent the mean ± SEM of five to six rats from three to five
experiments. One-way ANOV A showed a significant difference among the
control (bar) and EDS-treated rats (p<0.005). *p controls by Dunnett's t-test. "a" denotes the first significant increase of testicular weight in comparison to the lowest point at day 16 post-EDS treatment (p 2 ·~ .!: . • :. = Ol (J) :5: ' . .!a ~ (f) Q) f- 0 2 10 16 20 24 30 36 45 60 Days after EDS Figure 8. The effect of a single intraperitoneal injection of EDS on the interstitial cell number. The results represent the mean ± SEM of five to six rats from three to five experiment. One-way ANOV A showed a significant difference among the control (bar) and EDS-treated rats (p in comparison to controls by Dunnett's t-test. "a" denotes the first significant increase in the interstitial cell number in comparison.. to the lowest point at day 16 post-EDS treatment (p ,--.., . • (f) ..... 1 (f) Q) ...... lD 1 0 .-- '-' ·* 0 2 10 16 20 24 30 36 45 60 Days after EDS 67 fractions are presented in Fig. 9. Ninety five percent of the cells isolated from the control Leydig cell fraction were deeply stained (Fig. 9C). Fig. 90 demonstrates the negative results of 313-HSD staining for the control Leydig cell fraction in the absence of the substrate, etiocholane. The majority of the precursor cell fraction isolated from the control testis were stained only lightly (Fig. 9A), compared to the negative staining without the substrate (Fig. 9B). Twenty days after EDS administration, a time when Leydig cell regeneration occurs, the overall percentage of stained cells in the precursor cell fraction was 93%, as shown in Fig 9E. About 60% of these cells were lightly stained compared to the negative control without the substrate (Fig. 9F) and 33% of this precursor cell fraction were darkly stained. In the Leydig cell fraction isolated from the testis at day 20 post-EDS treatment, overall 95% of the cells were stained. Sixty five percent of those were intensely stained and 30% stained faintly (Fig. 9G, H), which correlated to the results in the precursor cell fraction at this stage. Analysis of the precursor cell and Leydig cell fractions from the control and EDS treated rats by flow cytometry To characterize the specific physical properties of the precursor cell and Leydig cell fractions, these two cell populations were isolated from the control and EDS-treated rats and subjected to flow cytometry. To compare different cell fractions, two cytometry gates were selected, based upon the distribution of the cells, and applied to each cell fraction. Fig. I 0 summarizes the cell size of the precursor and Leydig cell fractions isolated from the control and day 20 post-EDS treatment rats. The data demonstrate that precursor cells were larger than Leydig cells in both the control and day 20 post-EDS treatment rats. Cell size increased in both precursor cell and Leydig cell fraction after EDS treatment. Two way ANOV A Figure 9. Histochemical study of the Leydig cell and precursor cell fractions by 3 f3-HSD staining. These results are representative of three experiments with two to three rats each. The majority of the precursor cells isolated from the control rats were faintly stained '(A) compare to the precursor cells without the substrate in the reaction (B). More than 90% of the control Leydig cells were intensely stained (C). At day 20 post-EDS treatment, the 3f3-HSD positive cells increased in number in both precursor cell (E)"and Leydig cell (G) fractions. Figures D, F, H represent the results in the absence of the substrate for control Leydig cell, day 20 post-EDS .•precursor cell and Leydig cell fractions, respectively. Magnification: 100 X. 68 A c • Figure 10. Cell size of the Leydig cell and precursor cell fractions in the controls and day 20 post-EDS rats analyzed by flow cytometry. To compare the cell size in different groups, two gates were selected according to the distribution of the cells. The same gates were applied to each cell population and the percentage (expressed as mean ± SEM) of the cells within the gates were compared. The results represent three experiments, each performed using three rats. Two way ANOV A was utilized to analyze the data. a: control rats b: day 20 post-EDS treatment rats. Significant difference between a and b (p i: precursor Leydig cell (PLC) fraction. ii: Leydig cell (LC) fraction. Significant difference between i and ii (p Control PLC 69.6+0.7% 26.2+1.8% ~c ::! 0 a, 1 u •400 600 . 800 1000 SIZE m~------~------~------~,., 84.2±1.0% 12.7+0.9% Control LC .. ~c ::! 0 a, n u .~ .\ ,fS) . • ' 0 ,, :<>ea.... ·400 .., 6.00 .. , .81313 1000 ... SIZE·... · ...... · .... :.. ·_ .. . (S) • ' .-. .- . • .- . ,.,_ EDS PLC .,129.7±2..8% ~c ::! 0 b,i u . 6 . . 41313 6013 81313 1131313 SIZE EDS LC 56.4±3.3% 38.2+4.4% ~c ::! 0 b, ii u 2013 4013 6013 . 8013 113013 SIZE 70 analysis also demonstrates that EDS treatment had an effect on the difference in size between precursor cells and Leydig cells. In other words, the difference between the precursor cells and Leydig cells at day 20 post-EDS treatment was greater than that in the control rats (p The second parameter that was measured was granularity of the cells. The difference of the granularity of precursor cells and Leydig cells in the control and EDS day 20 rats is shown in Fig. II. Flow cytometry data and two way ANOVA analysis revealed a significant difference in granularity between precursor cells and Leydig cells and between the control and EDS-treated rats (p<0.001). At day 20 post-EDS, granularity in the precursor cell fraction increased significailtly. Like cell size, the difference in the granularity between the precursor cell and Leydig cell fractions at day 20 post-EDS treatment was also greater than that of the controls (p<0.001). Fig. 12 and Fig. 13 demonstrate the mitochondria and lipid droplet contents in the precursor cells and Leydig cells, respectively. Mitochondria content was higher in the precursor cells and in EDS-treated rats. Two way ANOVA analysis indicates that this difference is significant between precursor cells and Leydig cells (p Two gates were selected according to the distribution of the cells. The same gates were applied to each cell population and the percentage (expressed as mean± SEM) of the cells within the gates were compared. The results from three experiments, each performed using three rats, were subjected to two way ANOVA. a: control rats b: day 20 post-EDS treatment rats. Significant difference between a and b (p i: precursor Leydig cell (PLC) fraction. ii: Leydig cell (LC) fraction; Significant difference between i and ii (p 100 . 1000 ~~~------~----~------~ Control LC 38.8±1.8% 61.2+1.7% . . ~c 0 a u u" ' A ;EDS PLC ·. -· .. ~.. ' . ·... ·· ...: . ~c "0 b, i u c 100 1000 ~:~------~------~~ 37.0+3.0% 63:2+3.0% EDS LC ~c 0 u" b, ii E 10 _1000 LOG GRAN Figure 12. Dio fluorescence, indicative of mitochondrial activity/number, in the Leydig cell and precursor cell populations by flow cytometry. The data represent three experiments performed using three rats each time. The percentage within the gates are expressed as mean ± SEM. The results were analyzed by two way ANOVA a: control rats b: day 20 post-EDS treatment rats. Significant difference between a and b (p i: precursor Leydig cell (PLC) fraction. ii: Leydig cell (LC) fraction. Significant difference between i and ii (p ~~------, "' 37.5±3.5% Control PLC 60.0±3.9% ~c 0 • u" a, 1 c 1 10" 100. 1000 PMT2 LOG Control LC 53.7+3.8% 37.6+6.1% ~c "0 .. ~ u a, 11 (S) ! ' . ·. f · Hf 100 ·ra00 · ... 1"f'1T? LOG ----· EDSPLC ·15;0+3.2% ~c 0 u" b, i E 10 100 1000 PMT2 LOG EDS LC 34.8±7.8% 60.0+8.3% .... c "0 b, ii. u G 10 100 1000 PMT2·LOG Figure 13. Nile red fluorescence, indicative of lipid droplet content, in the Leydig cell and precursor cell fractions by flow cytometry. The data represent three experiments performed using three rats each time. The percentage within the gates are expressed as mean ± SEM. The results were analyzed by two way ANOVA a: control rats b: day 20 post-EDS treatment rats. Significant difference between a and b (p i: precursor Leydig cell (PLC) fraction. ii: Leydig cell (LC) fraction. .. Significant difference between i and ii (p "'.~------~------~ N Control PLC 38.2±5.5% 61.7±5.3% ..., c ::J o· a, 1 u G> • 1 18. 100 1000 .. PHT4 LOG "'N Control LC 67.6+4.4% 29.7+2.3% ...,. ·c ::J •• 0 a, 11 u G) ._ . . . ~ . -.'l -1' · 10· 100 100·0 ;-· f>MH LOG :-.. - ' .. .· -~ . "'·N· EDS PLC '26.5±6.2% 72.4+7.0% ..., c ::J 0 • u b, 1 E- G> . 1 ·10. 100 1000 PMT4 LOG "'N EDS LC 43.4+8.2% 55.9+8.4% ..., c ::J .. 0 b, 11 u G> • 1 10 100 10013 PMT4 LOG 74 Characterization of in vitro testosterone production by Leydig cells and precursor cells during differentiation To investigate the in vitro testosterone production and hCG responsiveness in Leydig cell and precursor cells during differentiation, aliquots of 0.2Sxi 06 cells isolated from each fraction were incubated in the presence or absence of SO m1U hCG/well for 24 hr at 37°C. Spent media of the cell incubations were collected for the detennination of the testosterone production. Fig. I4 shows the 24 hr testosterone production by the Leydig cell fraction isolated from the control and EDS-treated rats. The basal testosterone production in the control Leydig cells was .20.8 ± 9.8 ng/I06 cells. Addition of hCG resulted in a 7.S fold stimulation in testosterone production equivalent to IS6.9 ± 70 ng/I06 cells. Testosterone production was not detected in this cell fraction between days 2 and I 0 post-EDS treatment because no Leydig cells were present. At day I6 after EDS injection, cells in this Leydig cell fraction again began to produce testosterone and responded to hCG stimulation. Thereafter, testosterone production and hCG responsiveness in the Leydig cell fraction increased gradually and a significant increase was observed at day 24 in contrast to those at day I6 post-EDS treatment. Basal and hCG-stimulated testosterone production by the Leydig cell fraction returned to the control levels by 6-8 weeks post-EDS treatment. Compared to Leydig cells, the precursor cell fraction isolated from the control testes produced very low levels ofbasal testosterone (2 ± I.S ng/106 cells), as shown in Fig IS. This cell fraction also exhibited a low responsiveness to hCG stimulation. The addition of hCG to the culture resulted in only a l.S fold increase in testosterone production (3.I ± 1.8 ng/I06 cells). Testosterone production in the precursor cell fraction at days 2 and IO post-EDS Figure 14. In vitro testosterone production by Leydig cells in the control and EDS treated rats. Testosterone production was expressed as ng/106 cells/24 hr. Data for basal and hCG-stimulated production are expressed as the mean ± SEM from 2-4 experiments, each performed using five to six rats. ND denotes not-detected. Statistical analysis by ANOV A showed a significant difference (p<0.005) among the control and EDS-treated groups. Each group was then compared by Dunnett's t-test to the controls (**p<0.005). "a" denotes the first significant increase in testosterone production in comparison to day 16 post-EDS treatment (p<0.05). ·< 75 250~------, .!!l ~· "' 200 0 ~ 'Ol 5 150 c 0 ··B ·-: :J .. "0e 100 c. (!) c e(!) so .... . :g .... EDS-treated rats. Testosterone production was expressed as ng/1 06 cells/24 hr. Data for basal and hCG-stimulated production are expressed as the mean ± SEM from two to four experiments, each performed using five to six rats. ND denotes not-detected. Statistical analysis by ANOV A showed a significant difference (p groups. Each group was then compared by Dunnett's t-test to the controls (* p production in comparison to day 16 post-EDS treatment (p 60 . ~ .!:!2 Q) ~· u s·o . «> 0 ~ 0) 40 'c '-' c .Q..., u 30 ::J ;,a : ... "0 ,. ,_0 Q. 20 Q) c R ..,Q) Ul 10 ..,0 Ul Q) ND r- 0 0 2 10 16 20 24 30 45 60 Days after EDS 77 treatment was not detectable, comparable to that of the Leydig cell fraction during this time. From day 16 post-EDS treatment, testosterone production in this cell fraction began to increase and a significant increase was observed at day 24 in contrast to that at day 16 post EDS treatment. Testosterone production in the precursor cell fraction reached a maximum between 4-6 weeks post-EDS treatment which was about one third of that in the control Leydig cell fraction. At day 60 after EDS administration, the differentiated precursor cells had formed a new Leydig cell population. Testosterone production in the precursor cell fraction returned to the control values at 1.2 ± 0.5 ng/106 cells for basal testosterone production. The changes in cell number in the Leydig cell and precursor cell fractions during this regeneration process were investigated. As shown in Fig 16, the number of cells isolated from the Percoll gradient in the Leydig cell fraction was 2.31 ± 0.11 X 106/testis in the control rats. This number decreased to 0.41 ± 0.02 X 106/testis at day 2 and 0.27 ± 0.02 X 106/testis at day 10 post-EDS treatment. Thereafter, cell number in Leydig cell fraction increased significantly at day 16 (0.46 ± 0.03 X 106/testis) in contrast to the lowest point at day 10 post-EDS treatment. Cell number in Leydig cell fraction reached the value of 1. 71 ± 0.07 X 106/testis at day 36 post-EDS treatment. At day 45 post-EDS treatment, Leydig cell number had returned to the control levels. Compared to Leydig cells, the number of cells isolated from the Percoll gradient in the precursor cell fraction did not show dramatic changes among the control and EDS-treated groups (Fig. 17). Cell number in the precursor cell fraction of the control rats was 2.09 ± 0.10 X 106/testis. This number increased slightly to 2.41 ± 0.26 X Hf /testis at day 2 post-EDS treatment, yet was not significantly different from that of the control. Precursor cell fraction Figure 16. Changes of the cell number in the Leydig cell fraction after EDS treatment. Animals received a single EDS injection and were sacrificed at different times post-EDS treatment. The Leydig cell fraction was collected on a Percoll gradient with a density heavier than 1.070 g. Statistical analysis by one-way ANOVA showed a significant difference among the control and EDS-treated groups (p by Dunnett's t-test (*p represent three to five experiments, each performed using five to six rats. "a" denotes the first significant increase in the cell number in comparison to the lowest point at day 10 post-EDS treatment (p<0.05). LC: Leydig cell 78 3 ,...... ,(/) (/) 2 ...... Q) ...... ·w .r '.·0 ·,· ... ·' ,-- -- : 1.-5 ...... • .. '- Q) _,_ .0 E ::J a * c * u 0.5 _J 0 2 10 16 20 24 30 36 45 60 Days after EDS Figure 17. Changes of the cell number in the precursor cell fraction after EDS treatment. Animals received a single EDS injection and were sacrificed at different times post-EDS treatment. The precursor cell fraction was collected on a Percoll gradient with a density between 1.064 and 1.070 g. EDS-treated groups were compared to the controls by one-way ANOV A and no significant difference was observed. The results expressed, as mean ± SEM, represent three to five experiments, each performed using five to six rats. PLC: precursor Leydig cell 79 3 ...... (f) ...... (f) 0 2 10 16 20 24 30 36 45 60 Days after EDS 80 displayed a small decrease in cell number between day 10 and day 20 post-EDS treatment (no significant difference). The number of cell in the precursor cell fraction fluctuated slightly at later days, yet showed no significant difference. LH receptor and steroidogenic enzyme (P-450,.. and P-45017..) messenger RNA changes during the differentiation process These studies were designed to characterize the pattern of gene expression of the important functional and biochemical markers of Leydig cells during the differentiation process. Northern blot analysis was utilized to detect the levels of several messenger RNA, including LH receptor, P-450"" and P-45017a. Total RNA was extracted from the precursor and Leydig cell fractions of the control and EDS-treated rat testes. To examine the quality of the RNA samples, 1-2 ug of RNA was electrophoresed on 1% agarose gel and observed under a U.V. light. Intact RNA samples demonstrated clear 28S and 18S bands (Fig. 18). Twenty ug RNA extracted from each cell fraction was electrophoresed on agarose gel and transferred onto Genescreen nylon membranes. Membranes were hybridized with different 32P labeled eDNA probes and exposed to Kodak X -ray films. Quantification of the mRNA levels was performed by scanning the films on a densitometer. Levels of the target mRNA were corrected by that of the "house keeping gene", GAPD. Northern blot results and quantification ofLH receptor mRNA levels in Leydig cell and precursor cell fractions are shown in Fig 19. The Leydig cell fraction from the control rat expressed all four major transcripts ofLH receptor mRNA, 1.8 Kb, 2.5 Kb, 4.2 Kb and 7.0 Kb. The 7.0 Kb transcript is known to be translated into full length functional LH receptor and the 1.8 Kb transcript encodes a truncated form corresponding to the extracellular domain Figure 18. Gel electrophoresis of the total RNA extracted from Leydig ceiis and precursor ceiis. One to two ug of each total RNA was electrophoresed on a 1% agarose gel in 1 X TAE buffer. The gel was viewed under a U.V. light. ' The quality ofthe RNA was demonstrated by the 28S and 18S rRNA bands. LC: Leydig cell PLC: precursor cell •• •• • ! - I- I I U • • I I • • • "' l I • I I I • I I I • • • r 1 I I • I 11 • I I ~ •• I II • I I • -00 82 of the receptor. The ratio of7.0 Kb to 1.8 Kb transcript in the control Leydig cell fraction is 1.02 ± 0.23 (Fig. 19D). Only the truncated 1.8 Kb transcript was expressed in the precursor cell fraction of the control rat. When the amount of the 1.8 Kb transcript mRNA was corrected by GAPD mRNA and was assigned a value of 1 for the control Leydig cells, the level of this truncated transcript was 1.92 ± 0.41 in the control precursor cells (Fig. 19C). Precursor cells at day 2 post-EDS treatment also were found to express only the 1.8Kb transcript and the amount of this transcript was not different from that of the control precursor cells (Fig. 19 C). At day 10 after EDS treatment, all four transcripts were detected in the precursor cell fraction (Fig. 19A). At this time, the level of the 7.0 Kb transcript ofLH receptor mRNA was 0.52 ± 0.11 in the precursor cell fraction relative to the control Leydig cells (Fig. 19B). No significant changes of the 7.0 Kb transcript level in the precursor cell fraction was detected between day 10 and day 20 post-EDS treatment. The amount of the 7.0 Kb transcript for the LH receptor increased significantly at day 24 after EDS injection (1.33 ± 0.09 relative to the control Leydig cells). The peak value of this transcript was 1.62 ± 0.1 at day 36 post-EDS in the precursor cell fraction. Thereafter, the level of the 7.0 Kb transcript in the precursor cell fraction decreased greatly at day 45 post-EDS treatment and no longer was detected in this fraction at 60 days after EDS administration. Compared to the changes of7.0 Kb transcript in the precursor cell fraction, the truncated 1.8Kb transcript decreased significantly at day 10 after EDS treatment and reached a nadir between 16 and 36 days post EDS treatment. Thereafter, coinciding with the decrease in the 7.0 Kb transcript in the precursor cell fraction, this truncated form ofLH receptor mRNA transcript began to increase and returned to the control level at 60 days post-EDS treatment. Figure 19. Northern blot analysis of the LH receptor in the control Leydig cells and precursor cells and in the precursor cells after EDS-treatment. Twenty ug of total RNA extracted from the Leydig cell and precursor cell fractions was electrophoresed on a 0.66 M formaldehyde gel. The RNA in the gel was transferred onto a Genescreen nylol} membrane in 10 X SSC. Membranes were hybridized with the 32P-labeled LH receptor and GAPD eDNA probes. Northern blot results are shown in Fig. A. The amount of the LH receptor mRNA was standardized to that ofthe GAPD and assigned a value of I in the control Leydig cells. Fig. B and C illustrate the levels of the 7.0 Kb and 1.8 Kb transcripts, respectively. The ratio of7.0 Kb and 1.8 Kb transcripts in the control and EDS-treated groups is shown in Fig. D. The results represent the mean± SEM from two to seven experiments, each performed using eight to ten rats. LC: Leydig cell PLC: precursor Leydig cell 83 A 0 CD s ~ CD CD -LO s - CV) - CV) 0 -N ..-- ..-- N N N <.0 -.._ "0 "0 "0 "0 "0 "0 "0""" "0 "0 "0 -0.._ -.._0 """ c:: C/) C/) C/) C/) C/) C/) C/) C/) C/) C/) c:: -0 Cl Cl Cl Cl Cl Cl c:: Cl Cl Cl Cl -0 0 Ll.J Ll.J Ll.J Ll.J Ll.J Ll.J -0 Ll.J Ll.J LJ..J Ll.J 8 8 -0 0 0 0- 0- -0 -0 0 0 0 0 0 _J _J _J _J _J _J _J 0 _J _J _J _J _J 0... 0... 0... 0... 0... 0... 0... _J 0... 0... 0... 0... LH Receptor - 7.0 kb - ~ 4.2 kb- - 2.5 kb- - 1.8 kb - GAPD - 1.8 kb - Figure 19B: Precursor cells ofthe control and EDS-treated rats were compared by Fisher's LSD t-test. Groups a are significantly different from groups b (p B b ...... 1.8 0 u.. 1.6 r:"'__.PLC LC I <( (.9 1.4 cr ':r: 1.2 ,. ..J ··'-""" . l tt (/) ~"+! .. - o·.s; ...!:: -~-c ..t."+!, ~: 0:6. t\~ ...... Q) 0.4 t;f; > •\!...... ro 0.2 ...... :a:; ...... t\"+ 0::: 0 0 0 2 ·1 0 16 20 24 36 45 · GO Days after EDS Figure 19C: The levels of the 1.8 Kb transcripts in the precursor cells were compared by Dunnett's t-test and asterisks indicate significant difference when compared to the control precursor cells. 85 c ,...... , 3 0 0... Days after EQS 86 D 7 [iiiiiiLCl 6 ~ 5 4 ·r.-. . ""' 3· ".• -,· ..0 ~ 1'-- 2 1 0 0 2 10 16 20 24 36 45 60 Days after EDS 87 Fig. 20 A and B show the changes ofP-450,.. mRNA level in the Leydig cell and precursor cell fractions during differentiation. The steroidogenic enzyme P-450,cc mRNA was detected in Leydig cell fraction in the control testis with a size of2.2 Kb. Precursor cells in the control testis did not express the mRNA for this steroidogenic enzyme. P-450,.. mRNA also was not detected in the precursor cell fraction 2 days after EDS injection. Precursor cells began to express P-450"" mRNA at day I 0 post-EDS at a level of0.26 ± 0.08 relative to the control Leydig cells. This low level ofP-450= mRNA was maintained in the precursor cell fraction between day 10 and day 20 post-EDS treatment. P-450"" mRNA in the precursor cell fraction increased significantly to a level of0.75 ± 0.05 at day 24 and reached the maximum (0.99 ± 0.03) at day 36 post-EDS. P-450", mRNA level in the precursor cells decreased significantly at day 45 and was not detected at 60 post-EDS treatment. Gene expression of P-450 17~, another important steroidogenic enzyme, was also investigated. Northern blot analysis (Fig. 21) demonstrated that mRNA of P-450 17~ was expressed in the Leydig cell fraction in the control rat. Similar to P-450"', P-450 17~ mRNA was not detected in the precursor cell fraction in the.control testis. P-450 17~ mRNA showed a similar pattern compared to P-450"' mRNA during the differentiation process of the precursor cells. It was not detected in the precursor cell fraction 2 days after EDS treatment. Precursor cells began to express P-450 17~ mRNA at day 10 post-EDS (0. 74 ± 0.1 relative to the control Leydig cells). P-450 17~ mRNA levels did not change significantly between day 10 and day 20 post-EDS treatment. At day 24 after EDS treatment, P-450 17~ mRNA increased significantly to 1.48 ± 0.14 relative to the control Leydig cells. It reached a peak value at day 36 post-EDS treatment (1.73 ± 0.12) and maintained high level at day 45 post-EDS Figure 20. Northern blot analysis ofP-450"' in the control Leydig cells and precursor cells and in the precursor cells after EDS-treatment. Twenty ug of total RNA extracted from the Leydig cell and precursor cell fractions was electrophoresed on a 0.66 M formaldehyde gel. The RNA in the gel was transferred onto a Genescreen nylon membrane in 10 X SSC. Membranes were hybridized with the 32P-labeled P-450"' and GAPD eDNA probes. Northern blot results are shown in Fig A The amount of the P-450"' mRNA was standardized to that of the GAPD and assigned a value of 1 in the control Leydig cells. Fig. B illustrates the relative units ofP-450.,., mRNA in the control and EDS-treated groups. The results represent the mean± SEM from two to seven experiments, each performed using eight to ten rats. Precursor cells of the control and EDS treated rats were compared by Fisher's LSD t-test. Groups a are significantly different from groups b (p<0.05). LC: Leydig cell PLC: precursor Leydig cell \J I Relative units (CSCC/GAPD) .J:::. G) (.11 ):> 0 \J ("") 0 en 0 ...... I en 9 9 9 ...... > 0 N .+>. 0') CX) N = ·- ·- == LC (Control) o r(fliflfjif/if!!l!!iJii!!i!ii!iiJ!f;!i;ii;!iJ]fj!i;!ft!jJiiJ!I ...... _... . _.. . _... _ • • _...... ~-+. ~·- ...... 6...... ~ -- -- PLC (Control) 0 J, I I. r• PLC (EDS d2) PLC (EDS d1 0) I • PLC (EDS d16) 0 -'0 I ~ 0..) I 1- ru '< PLC (EDS d20) (f) ~ II ru l > w PLC (EDS d24) N I -t\ M- 0 l - PLC (EDS d36) (1) ~ I 1- -; N __._ N m OJ 0 0 CT CT :0') t ~~J ~ I "' "' (/) ~ I ___.e-9 n,) I - 1 • LC (Control) II- PLC (EDS d24) ~ r I I ~ I I IYI IW I PLC (EDS d36) I • PLC (EDS d45) [i!] I I(?) I PLC (EDS d60) 00 00 Figure 21. Northern blot analysis ofP-45017" in the control Leydig cells and precursor cells and in the precursor cells after EDS-treatment. Twenty ug of total RNA extracted from the Leydig cell and precursor cell fractions was electrophoresed on a 0.66 M formaldehyde gel. The RNA in the gel was transferred onto a Genescreen nylon membrane in 10 X SSC. Membr!!nes were hybridized with 32 the P-labeled P-450 17" and GAPD eDNA probes. Northern blot results are shown in Fig. A The amount of the P-45017" mRNA was standardized to that of the GAPD and assigned a value of 1 in the control Leydig cells. Fig. B illustrates the relative units ofP-45017" mRNA in the control and EDS-treated groups. The results represent the mean ± SEM from two to seven experiments, each performed using eight to ten rats. Precursor cells of the control and EDS treated rats were compared by Fisher's LSD t-test. Groups a are significantly different from groups b (p LC: Leydig cell PLC: precursor Leydig cell ""'0 I Relative units (P 4so-17a/GAPD) +::. G) c.n eo· ):> __.0 ""'0 -.....1 0 0 0 0 ...... 0 Q > o N ~ a, Co -'N~01eoN • LC (Control) 0 rfllillf!lf/lfllf!iiliili!l!fllfl!ft I PLC (Control) 0 ~ I • - - PLC (EDS d2) PLC (EDS d1 0) 0 c; l ~ OJ I - PLC (EDS d16) ru '< ~ OJ - PLC (EDS d20) (f) I ~ ~ t PLC (EDS d24) ru N I I -h 0 r-+ - PLC (EDS d36) CD + ~ I \ N .. .j:::. __. N m CXl N 0 I A A (/) ~ t ~ rr = = .j:::. ______, v _ , Vl l I - LC (Control) ~ r- I !WI ..PLC (EDS d24) I PLC (EDS d36) ~ PLC (EDS d45) I PLC (EDS d60) 00 \0 90 treatment. Again, gene expression ofP-450 17~ was no longer detected in the precursor cell . fraction at day 60 post-EDS treatment. To confirm the regeneration of Leydig cells at the mRNA level, total RNA was also extracted from the Leydig cell fraction at day 36 and day 60 post-EDS treatment. RNA was subjected to northern blot analysis and the results are shown in Fig. 22. As demonstrated by serum testosterone level and in vitro testosterone production, Leydig cells had regenerated by this time. LH receptor and steroidogenic enzyme mRNA levels were similar to the control values. Estrogen receptor binding in the testis of the control and EDS treated rats The estrogen binding capacity of whole testicular cytosol was determined and the data were analyzed by the method of Scatchard. As shown in Fig. 23, the specific estrogen binding capacity (B....) in the control testis was 13.57 ± 1.29 frnoVmg cytosolic protein. Four days after EDS treatment, estrogen binding capacity in the testis decreased to 3.44 ± 0.15 frnoVmg cytosolic protein. A significant increase in estrogen binding capacity was observed at day 10 post-EDS treatment (6.28 ± 0.53 frnoVmg cytosolic protein). Thereafter, estrogen binding capacity increased slightly between days 10 and 20 post-EDS treatment, yet remained significantly lower than that of the control rats. By day 30 post-EDS treatment, testicular estrogen binding capacity returned to control values (13.73 ± 1.64 frnoVmg cytosolic protein). Dissociation constants (Kd) were determined from the slopes of the Scatchard plots.· The Kd values obtained at various days after EDS treatment did not differ significantly from the controls. Fig. 24 demonstrates the representative Scatchard analysis for the control and EDS-treated rats. The dissociation constant for estrogen receptor binding in the testis of the Figure 22. Northern blot analysis ofLH receptor, P-450"' and P-45017,. mRNA in the Leydig ceiis of the control and day 36 and 60 post-EDS rats. Twenty ug of total RNA extracted from Leydig ceil was electrophoresed on a 0.66 M formaldehyde gel. The RNA in the gel was transferred onto a Genescreen nylon membrane in 10 X SSC Membranes were hybridized with the 32P labeled eDNA probes. Northern blot results are shown in Fig. A. The amount of the target mRNA was standardized to that of the GAPD and assigned a value of 1 in the control Leydig ceiis. Fig. B iiiustrates the relative units of each mRNA in the control and EDS-treated groups. The results represent mean± SEM from two experiments, each performed using ten rats. Statistical analysis by ANOVA showed no significant difference among the control and EDS-treated groups. LHR: LH receptor 91 A 0 c.o M 0 C.O ...... "0 "0 c:: (/) (/) 0 Cl Cl (.) UJ UJ LH Receptor -7.0 kb -4.2 kb -2.5 kb -1.8 kb P-450css -2.0 kb P-45017a -2.2 kb GAPD -1.8 kb 92 B 0 36 60 Days after EDS Figure 23. Changes in the specific estrogen binding capacity of rat testicular tissue after a single intraperitoneal injection of EDS (100 mg/kg BW). The specific binding is expressed as fino! estradiol bound/mg cytosol protein. Control animals were not treated. Each value represents the mean ± SEM obtained from three to five cytosol preparations, using two to four rats each experiment. Statistical analysis by ANOV A showed a significant difference among the control and EDS-treated groups (p versus the control values (Dunnett's !-test) is denoted by an asterisk. "a" denotes the first significant increase in estrogen binding capacity in comparison to the lowest point at day 4 post-EDS treatment. 93 ,.....,. c 16 .Q) ...... 0 14 I... Q. 12 Ol E 1~0 ~ ...... Q) ~a a· E . .- -·- :_' ~. * ~ . U") 6 ·,_,. ·-·. I .* 0 4 ...... ,r- 2 X Cll E 0 CQ 0 4 10 . 16 20· 30 Days after EDS Figure 24. Representative Scatchard plot analysis of3H-estradiol binding by control and day 16 post-EDS treated rat testicular cytosols. The experimental procedure is outlined in the Materials and Methods section. The data have been corrected for the contribution by nonspecific binding. These data are the means of three experiments. 94 0.12 --.------,---;---, • EDS (Day 16) 0.10 • • Control 0.08 Q) .....Q) • LL -.;. . "'0 . 0.06 .. -C:: :::J cc0 0.04 0.02 • 0 2 4 6 8 10 12 14 Bound (10-15 Moles) 95 control rats was 3.1 ± 0.8 x 10-Hl~1. At day 16 after EDS treatment, the Kd value was not significantly different from that of the control testis. Immunohistochemical studies of estrogen receptor in the testis of the control and EDS treated rats These studies were designed to determine the presence and localization of estrogen receptor in the testis. Using a polyclonal antibody against the rat estrogen receptor, immunohistochemical studies were carried out on frozen sections of the testis. In the control testis, estrogen receptor was detected exclusively in the interstitium (Fig. 25A, B). No staining of estrogen receptor was observed in the seminiferous tubules. At day 10 post-EDS treatment (Fig. 25C, D), testicular interstitial cell number decreased dramatically as expected due to the loss of Leydig cells (Myers and Abney, 1990). Immunohistochemical data, nevertheless, demonstrated that estrogen receptor was present in the testis of the EDS-treated rats when Leydig cells were absent. As in the control testis, estrogen receptor immunostaining was confined to the interstitium and no staining was observed in the seminiferous tubules in these EDS-treated rats. Addition of the synthetic estrogen receptor peptide to the immunohistochemical reactions blocked the primary antibody binding to the tissue, as shown in Fig. 25E (control) and Fig. 25F (EDS day 10), demonstrating the specificity of the primary antibody. Rat spleen and uterine tissues were used as negative and positive controls, respectively, for the immunohistochemical studies (data not shown). Immunohistochemical detection of estrogen receptor in the isolated Leydig cells and precursor cells The results of the ·immunohistochemical studies on the testis localized the testicular Figure 25. Localization of the testicular estrogen receptor by immunohistochemical analysis in the controls and day 10 EDS-treated rats. In the sections (A and B) from control tissue, estrogen receptor (brown stain) is confined to the interstitial tissue. In the sections (C and D) from day 10 EDS treated rats, the interstitial cell number is decreased significantly. Estrogen receptor is still detectable and is confined to the interstitial cells. In the sections (E, control and F, day 10 post-EDS) that were exposed to the synthetic receptor peptide, antibody binding was blocked and staining was absent. The results represent four to six testis per group and ten fields per testis were observed under a light microscope. Magnification: A, C, E and F, 40X; B and D,lOOX. 96 A 97 98 99 estrogen receptor in the interstitium. To further confirm the cellular localization of the estrogen receptor, Leydig cells and precursor cells were isolated from the control and EDS treated rats and subjected to immunohistochemical studies. The results, shown in Fig. 26, demonstrate that estrogen receptor was detected in both Leydig cells (A) and precursor cells (C) of the control rats. The addition of the synthetic estrogen receptor peptide blocked the primary antibody binding to Leydig cells (B) and precursor cells (D). Estrogen receptor also was detected in the precursor cells at day I 0 post-EDS treatment (E) and the immunostaining was blocked by the synthetic peptide (F). Detection of estrogen receptor mRNA in Leydig cells and precursor cells by reverse transcription-polymerase chain reaction (RT-PCR) The initial studies were designed to detect the presence of the gene expression of estrogen receptor in the Leydig cells and precursor cells. Using the primer set which flanks the ligand binding domain of the rat estrogen receptor eDNA, a PCR fragment of382 bp in length was generated. Total RNA extracted from the Leydig cell and precursor cell fraction was reverse transcribed and subjected to PCR. As shown in Fig. 27, an estrogen receptor RT PCR product of the expected size was found to be present in both Leydig cells and precursor cells in the control testis. To further demonstrate that estrogen receptor mRNA is expressed in the precursor cells when the testis is devoid of Leydig cells, total RNA was extracted from the precursor cell fraction afler EDS injection and subjected to RT -PCR reaction. The results show that estrogen receptor mRNA was detected in the precursor cells between 2 and 24 days post-EDS treatment (Fig. 27). For the purpose of the quantitative analysis of estrogen receptor mRNA in Leydig Figure26. Detection of the estrogen receptor by immunohistochemical analysis in the Leydig cell and precursor cell fractions from the ,controls and day I 0 EDS-treated rats. Estrogen. receptor staining is present in the Leydig cell fraction of the control testis (A). Addition of the synthetic peptide blocked the antibody binding to Leydig cells (B). In the precursor cell fraction, estrogen receptor is detected in both control (C) and day I 0 EDS-treated (E) rats. Figures D and F show no staining for estrogen receptor in the presence of the synthetic peptide in the precursor cell fraction of the control and day IO EDS-treated rat testis, respectively. The results represent three experiments, using two rats per group. Magnification: I OOX. 100 B c D Figure 27. RT-PCR detection of estrogen receptor mRNA in Leydig cells and precursor cells from controls and precursor cells from EDS-treated animals. Amplification products were electrophoresed in a 2. 5% agarose gel and viewed under U.V. light. The primer set flanks the estrogen binding domain of the rat estrogen receptor eDNA and amplified a 382 bp fragment. These results are representative of four to five experiments. LC: Leydig cell PLC: precursor Leydig cell w CJ1 0 0 0 0 cr cr "0 "0 I I s: PCR marker ...... LC (control) "' PLC (control) w PLC (EDS d2) .j:::.. PLC (EDS d1 0) CJ1 PLC (EDS d16) 0'> PLC (EDS d20) -..J PLC (EDS d24) -0 102 cells and precursor cells, an internal standard for RT-PCR was included. Rabbit P-globin mRNA was added into the reverse transcription reaction as an exogenous sequence to control the variations of the RT-PCR. Two primer sets were utilized and coamplification of the rat estrogen receptor eDNA and the rabbit P-globin eDNA was conducted. The estrogen receptor primer set yielded a product of 3 82 bp in length and the rabbit P-globin primer set generated a product of324 bp in length. To verity the sp.ecificity of the RT-PCR products generated by the proper primer sets, the rabbit P-globin primer set was used to amplifY the eDNA reverse transcribed from the testicular RNA and the estrogen receptor primer set was used to amplifY the eDNA reverse transcribed from the rabbit P-globin mRNA. As shown in Fig. 28, no products were generated from either rabbit P-globin primer set/testicular RNA or estrogen receptor primer set/rabbit P-globin mRNA. These results confirmed the specificity of the primer sets and the products. For each RT-PCR tube, 1 pg of the rabbit p-globin mRNA was added and reverse transcribed together with 1 ug of total RNA extracted from precursor cells or Leydig cells. The coamplification of the rat estrogen receptor and the rabbit P-globin eDNA was carried out in the same reaction tube. The two amplification products were different in length by 58 bp and clearly separated in a 2. 5% agarose gel, as shown in Fig. 29. To demonstrate the validility of the quantitative results ofRT-PCR, a set of standard line reactions was performed as quality control. In this standard line set, 0, 0.0625, 0.125, 0.25, 0.5, 1. 0, 2, and 4 ug of total testicular RNA was included in each reaction tube, while 1 pg of the rabbit P-globin mRNA was added to every standard line tube. The reverse Figure 28. Specificity of the RT-PCR products amplified by the primer sets. Total testicular RNA or the rabbit ~-globin mRNA was reverse transcribed and subjected to PCR with different primer set. The product generated by the estrogen receptor primer set is 3 82 bp in length and the rabbit ~-globin primer generated a product of324 bp. Lane 1. Estrogen receptor primer set/rabbit ~-globin mRNA Lane2. Estrogen receptor primer set/testicular RNA Lane3. Rabbit ~-globin primer set/rabbit ~-globin mRNA Lane4. Rabbit ~-globin primer set/testicular RNA 103 M 1 2 3 4 500 bp- 300 bp- Figure29. Gel electrophoresis of the coamplification products of the rat estrogen receptor and rabbit P-globin cDNAs. Total RNA extracted from Leydig cells or precursor cells was subjected to reverse transcription together with the p-globin rnRNA as an internal standard. Coamplification was conducted by utilization of two sets of primers. One set flanked the rat estrogen receptor eDNA and generated a product of 382 bp. The other flanked the rabbit P-globin eDNA and generated a 324 bp product. These results are representative of four to five experiments. LC: Leydig cell .. PLC: precursor Leydig cell w 01 0 0 0 0 =0'" =0'" I I :5': PCR marker __. LC (control) N I PLC (control) w I PLC (EDS d2) ~ ) " PLC (EDS d1 0) 01 PLC (EDS d16) en I PLC (EDS d20) --.J I PLC (EDS d24) 00 PLC (EDS d36) (.!) LC (EDS d36) )\ :::0 m Pol en c::r=i 0'" 0 ;:::;: tC "t;OCD I :::J tC ...... - CD 0 (") 0'" CD ::::!-· =...... 0...... ..... 0 ~ 105 transcription reaction was conducted and coamplification was carried out using two sets of primers under the same conditions as the sample tubes. Fig. 30A demonstrates the gel electrophoresis result of the standard line RT-PCR. The upper 324 bp bands represent rabbit 13-globin amplification fragment and were constant among different reaction tubes. The lower bands are 382 bp in length and represent rat estrogen receptor RT-PCR products generated from different amount of testicular RNA. One ul of each RT-PCR products was electrophoresed on 2.5% agarose gel. The gel was denatured in 0.5 N NaOH and transferred to a nylon membrane. Hybridizations of the membrane with 32P-Iabeled rat estrogen receptor and rabbit 13-globin eDNA probes were conducted and the data are shown in Fig. 30B. The blots were scanned on a densitometer and the amount of estrogen receptor was corrected by that of the 13-globin in the same reaction tube. Correlation analysis demonstrated a correlation coefficient between the amount of RNA used for RT-PCR analysis and the amount of amplified eDNA to be 0.95 (p<0.05), as shown in Fig. 30C. Quantification of estrogen receptor mRNA levels was then performed using this coamplification and Southern blonechnique. Fig. 31A shows the southern blot data of the RT-PCR products using total RNA extracted from the precursor cells of the control and EDS-treated rats. Estrogen receptor mRNA levels in the precursor cell fraction, corrected by 13-globin, decreased after EDS injection and reached a nadir at day 16 post-EDS treatment, as shown in Fig. 31B. Thereafter, precursor cell estrogen receptor mRNA levels increased significantly at day 20 and returned to the control precursor cell level around seven weeks post-EDS treatment. Total RNA extracted from Leydig cells in the control and 16 days or longer post-EDS Figure 30. Quality control of the RT-PCR. Zero to four ug of the total te-sticular RNA was transcribed with 1 pg of rabbit p-globin mRNA and then subjected to PCR reaction. The estrogen receptor and P-globin primer sets were utilized to amplifY the cDNAs. Gel electrophoresis of the RT PCR products is shown in Fig. A. The RT-PCR products were then analyzed by Southern blotting. Hybridization of the membrane with the estrogen receptor and P-globin eDNA probes were performed, as shown in Fig. B. The amount of the estrogen receptor amplification products was standardized to that of the P-globin and a standard line was obtained (Fig. C). ER: estrogen receptor 106 A ...... _ C) ~...... _ 0> 0> :i. ro 0> :i. LO 0> E :i. LO 0> 0> 0> :i. (.0 0> a: LO "'0 c..:> :i. :i. :i. LO "'~ :i. a... ~ "' ~ 0 "'0 0 0 0 500 bp- ~ ER 300 bp- "--p-globin B 0> 0> :i. 0> :i. 0> LO :i. LO 0> 0> 0> :i. LO (.0 0> ~ "' :i. :i. :i. LO "' 0 :i. ~ N ~ 0 "'0 0 0 0 ER p-globin 107 c 10.0 .---~-----~----'--:..._:_....:...... :-'----, r = 0.95086 --.. c ..c 0 0> I ~ """- • 0::: UJ ••• 1.0 +-' -c :J a.> .> +-' ctl • a.> 0:: 0 d RNA Concentration (1-19) Figure 31. Southern blot and quantitative analysis of the RT-PCR products generated from precursor cells of the control and EDS-treated rats. Coamplification products were electrophoresed in 2.5% agarose gel and transferred onto a nylon membrane. Hybridizations with the estrogen receptor eDNA and P globin eDNA were performed (A). Estrogen receptor mRNA was standardized relative to P-giobin mRNA The arbitrary unit 1.0 was assigned to the control precursor cells and the amount of the EDS-treated groups were expressed relative to 1 (B). Statistical analysis by ANOVA showed a significant difference among the control and EDS-treated groups (p<0.01). Each EDS treated group was then compared by Dunnett's t-test to the controls (* p<0.05). These data are the mean± SEM from three to five experiments, each performed using five to six rats. PLC: precursor cell 108 A s ;;;;- s :;;;- G' g "0 '0 '0 "0"" """0 """"0 c: "'(/) (/) (/) (/) (/) (/) · 0 Cl Cl Cl Cl Cl Cl ~ ':!::- ':!::- ':!::- ':!::- ':!::- ':!::- '-'_J '-'_J '-'_J '-'_J '-'_J '-'_J '-'_J "- "- "- "- "- "- "- Estrogen receptor Rabbit ~-globin 2 3 4 5 6 7 B ,.----... 1.4 <( z 0:::: 1.2 E c :.0 0 0) co._ 0.8 ...... 0:::: ..__..,UJ 0.6 ...... (J) ·c :::J 0.4 c ,_ro ...... 0.2 :.0,_ <( 0 0 2 10 16 20 24 45 Days after EDS 109 treatment were utilized to conduct RT -PCR and quantitative analysis, as shown in Fig. 32. In vitro testosterone production confirmed that no Leydig cells were present before 16 days post-EDS treatment and therefore the Leydig cell fraction was not analyzed prior to this time. Estrogen receptor mRNA levels in Leydig cells were also standardized by rabbit 13-globin. Statistical analysis demonstrated no significant difference in estrogen receptor mRNA levels in the Leydig cell fraction among the control and EDS-treated groups during Leydig cell regeneration. Since the same amount of total RNA extracted from the Leydig cell or precursor cell fraction was utilized for RT-PCR and the RT -PCR and Southern blot analysis were conducted under the same conditions, we were able to compare estrogen receptor mRNA levels between the precursor cell and Leydig cell fractions. As described above, estrogen receptor levels were corrected by those of the 13-globin in the same reaction tubes. As demonstrated clearly in Fig. 31A and Fig. 32A, estrogen receptor mRNA level was higher in the precursor cells than in Leydig cells. In the control mature rats, estrogen receptor mRNA level was 20 ± 1.34 fold higher in the precursor cells (Fig. 31, Lane 1) than in Leydig cells (Fig. 32 Lane 1). Comparative study of estrogen receptor levels in the precursor cells and Leydig cells by immunofluorescent analysis The above RT-PCR results demonstrated that estrogen receptor mRNA is expressed in Leydig cells and at a higher level in the precursor cells. It also was confirmed by immunohistochemical studies that estrogen receptor is present in these two cell populations. To compare the amount of estrogen receptor in Leydig cells and precursor cells, cells were isolated from Percoll gradient and immunofluorescent studies were conducted. By using a Figure 32. Southern blot and quantitative analysis of the RT-PCR products generated from Leydig cells of the control and EDS-treated rats. Coamplification products were electrophoresed in 2.5% agarose gel and transferred onto a nylon membrane. Hybridizations with the estrogen receptor eDNA and ~ globin eDNA were performed (A). Estrogen receptor mRNA was standardized relative to ~-globin mRNA. The arbitrary unit 1. 0 was assigned to the control Leydig cells and the amount of the EDS-treated groups were expressed relative to 1 (B). Statistical analysis by one-way ANOVA showed no significant difference among the control and EDS-treated groups. These data are the mean± SEM from three experiments, each performed using five to six rats. LC: Leydig cell 110 A Estrogen receptor ·• Rabbit p-globin 2 3 4 5 B ...--.. z<( 1.4 0:::: E 1.2 c ...0 0 0') ~ 0.8 ...... 0:::: ...... u.J 0.6 U) .-t= c :::s 0.4 c ,__ro 0.2 .-t= ...0,__ <( 0 0 16 20 24 45 Days after EDS Ill linker and ClonoPAP antibody and then goat anti-HRP/FITC in the reaction, the fluorescent signals were amplified. The synthetic estrogen receptor peptide was included in one set of slides to demonstrated the specificity of the primary antibody binding. As shown in Fig. 33A, the Leydig cell fraction isolated from the control testis demonstrated fluorescence of estrogen receptor antibody binding. The inclusion of the synthetic peptide in the immunofluorescent reaction blocked the antibody binding to the cells, as demonstrated in Fig. 33B. The results also demonstrated the presence of estrogen receptor in the precursor cell fraction of the control (Fig. 33C) and EDS day 10 (Fig. 33E) rats. The intensity of the fluorescence in the cells was analyzed on the Metamorph program and summarized in Table 1. The average gray value, which reflects the intensity of the fluorescence, was significantly higher in the precursor cells ofboth the control (136 ± 30) and day 10 post-EDS treatment rats ~135 ± 25) than that in the control Leydig cells (1 07 ± 24). In addtion, the distribution of the cells in each group versus the average gray value is shown in Fig. 34. Histogram data also demonstrat~her average gray value in the precursor cell fractions. Regulation of precursor Leydig cell differentiation by LH/hCG and other possible factors The initial experiments were designed to investigate the differentiation of the precursor Leydig cells in the immature rats. Precursor Leydig cells were isolated on Percell gradients from immature rats (21 days of age) and incubated in vitro with or without the addition of 50 miU hCG. The media were collected every 24 hr for 6 days and testosterone production by precursor cells was measured. The results show~at during the first day in cell culture, hCG administration resulted in a 1.5 ± 0.2 fold increase in testosterone production Figure 33. Comparative study of the estrogen receptor levels in Leydig cells and precursor cells of the control and day 10 post-EDS rats. The fluorescence of estrogen receptor is present in both Leydig cells (A) and precursor cells (C) in the control rats. Estrogen receptor also was detected in the precursor cells at day 10 post-EDS treatment (E). The addition of the ~ynthetic peptide in the reaction blocked the primary antibody (ER 715) binding in the control Leydig cells (B), control precursor cells (D) and EDS day 10 precursor cells (F). The results represent two experiments. 112 Table I. Comparison of estrogen receptor levels by immunofluorescent study in Leydig cells and precursor cells of the control and day I 0 post-EDS rats. Fluorescence in the cells was analyzed oil the Metamorph program and the intensity of the fluorescence was measured by the average gray value. Ten fields per group from two experiments were measured and the results represent the mean ± SEM. Fisher's LSD t-test showed a significant difference between groups a and b. 113 Table 1. Comparison of estrogen receptor levels by immunofluorescent study in the Leydig cells· and precursor cells of the control and day 10 post EDS rats Average gray value :-·. . . ·.- ... -.. . . ·. - . . . . Precursor cell{(corifrol) ·. 136±30 a Leydig cells (control) 107 ±24 b Precursor cells (EDS day 10) 135 ± 25 a Figure 34. Average gray value, indicative of estrogen receptor immunofluorescence, in the Leydig cell and precursor cell populations. Leydig cell and precursor·cell fractions isolated from the control rats and precursor cell fraction isolated from the day 10 post-EDS treatment rats were subjected to immunofluorescent analysis. Fluorescence of estrogen receptor in the cells was analyzed on the Metamorph program .and represented by the average gray value. The results represent two experiments, each performed using two rats. Ten fields per group were measured. Figure A: Leydig cells (control) FigureB: Precursor cells (control) Figure C: Precursor cells (day 20 post-EDS treatment) 114 A ..... 800 Q) ..0 ·600 E ::l 400 c 200 (i) u 0 . '10 60 . 80. 1 00 120 140 1 60 180 . 200 220 B ; ... a; i oci u o.+ -+-- 40 60 80 1 00 120 140 160 1 80 200 220 c ..... 1000 Q) ..0 800 E ::l 600 c 400 Q) 200 u 0 +.--t--- 40 W ~ 1001201~160180~2W I Average Gray Value 115 (Fig. 35). hCG stimulated testosterone production in the precursor cells by 3.1 ± 0.4 fold during the second day in culture. Thereafter, the responsiveness of precursor cells to hCG increased gradually and reached the highest point, 13.2 ± 0. 75 fold stimulation, at day 5 in culture. To investigate the ability of the precursor cells in the testis of the EDS-treated rats to differentiate under the control of hCG in vitro, the precursor cell fraction was isolated from the rat testes at day 20 post-EDS, a time when precursor cells are postulated to be differentiating into Leydig cells. In vitro incubation of these precursor cells was carried out and testosterone production was measured by RIA During the first 24 hr in culture, hCG caused a 7.05 ± 0.95 fold increase in testosterone production by precursor cells. Thereafter, I the responsiveness of precursor cells to hCG stimulation decreased and by day 4 testosterone production by these cells was not detectable, even in the presence of hCG stimulation (Fig. 35). The results obtained from the precursor cells of the EDS-treated rats indicate that LH/hCG alone was not sufficient to stimulate precursor cell differentiation process, suggesting that other factors are involved in precursor cell differentiation after EDS treatment. It has been reported by different investigators that growth factors can regulate Leydig cell development and function, and among which insulin like-growth factor I (IGF-I) plays an important role. Based upon these previous studies, we conducted another experiment in which the effect ofiGF-I on precursor cell differentiation was investigated. As show in Fig. 36, neither IGF-I by itself nor IGF-I with hCG was able to maintain the production of testosterone-by precursor cells during the in vitro incubation. Figure 35. Stimulation of testosterone production by hCG in the precursor cells of the immature and day 20 post-EDS rats. Precursor cells were isolated from animals and incubated in culture in the absence or presence of 50 miU hCG. Basal and stimulated testosterone production per 24 hr was measured and the results were expressed as stimulation fold by hCG. The results represent two experiments and are expressed as the mean ± SEM. Each experiment was performed using fifteen immature rats and six day 20 post-EDS rats. 116 16 c immature 14 -• EDS- 12 "'0 0 4-- 1 c .8. 0 - .+-' - ·ro- 6 ::l E .._. 4 (/) 1 2 3 4 5 6 _ Days 1n culture Figure 36. Effects of hCG or/and IGF-1 on the in vitro testosterone production by precursor cells at day 20 post-EDS treatment. Precursor cells were isolated at day 20 post-EDS treatment and incubated in culture. hCG (50 rniU) or/and IGF (50 ng/ml) was added to the media and the basal and stimulated testosterone production per 24 hr was measured. The results were expressed as stimulation fold by hCG or/and IGF-1. The experiment was performed one time. 117 7 6 .c.hCG 1!1 IGF-1 • hCG+IGF-1 -o 5 ...... 0 4 c ·.0 -~ ··. ro 3 :J E 2 ·+-' (/) 1 0 1 2 3 Days in culture DISCUSSION Characterization of Leydig cell degeneration and regeneration after EDS treatment The destructive effect of EDS on testicular Leydig cells in the rat was well documented from the results ofthe present study. Within 2 days after a single intraperitoneal injection ofEDS, serum testosterone levels decreased dramatically from 2.325 ± 0.098 ng!ml in the control rats to the level of0.025 ± 0.003 ng!ml (Fig. 6). This suggests that Leydig cells are destroyed by EDS treatment. The specific cytotoxicity ofEDS on Leydig cells has been demonstrated by a series of experiments. Jackson (1966) first reported the antifertility effect of EDS. Later work of Bu'Lock and Jackson (1974) showed that testicular androgen synthesis was suppressed by EDS treatment, suggesting that Leydig cell function was impaired. In the last 10 to 15 years, it has been clearly demonstrated that this antifertility effect ofEDS is due to the specific destruction of Leydig cells in the rat testis. Histological studies revealed that most Leydig cells undergo destructive changes 12 hr after EDS injection and no histologically recognizable Leydig cells are present in the testis 2 days post-EDS treatment (Morris et al, 1986; Jackson and Jackson, 1984; Jackson eta!, 1986; Kerr et al, 1987; Kerr et al, 1988). The mechanism by which EDS destroys Leydig cells is still unknown. As an alkylating agent, the cytotoxic effects of EDS may be a result of alkylation of proteins that are 118 119 specifically present in the mature Leydig cells and that play an essential role in the maintenance of their cellular functions (Rommerts and Hoogerbrugge, 1987). Experiments conducted by Edwards et al (1988) showed that structural changes in the EDS molecule caused the total loss oftoxicity in Leydig cells with the concomitant appearance of a general systemic toxicity. This indicates that the specificity of the EDS effect on Leydig cells depends on its precise molecular structure. The results also demonstrated that a potential metabolite of EDS was inactive. This made it unlikely that EDS is converted to an active metabolite which exerts the cytotoxic effects. However, EDS may undergo activation which is unique to Leydig cells to affect this specific target cell. It has been reported that Leydig cells cultured in the presence ofbuthionine sulfoximine, a specific inhibitor of glutathione synthesis, were far less sensitive to EDS effect on testosterone production than Leydig cells cultured in the medium alone. This protective effect of buthionine sulfoximine was abolished by restoring intracellular glutathione levels with glutathione ethyl ester (Kelce and Zirkin, 1993). These results suggest that the mechanism by which EDS selectively kills adult rat Leydig cells may involve Leydig cell glutathione. Another interesting hypothesis concerning the mechanism ofEDS specific effect on Leydig cells was raised by the observation that abundant macrophages are present in the interstitium between 2 and 14 days post-EDS treatment. It has been suggested that the destruction of Leydig cells elicits an immune response involving macrophage-lymphocyte interaction (Kerr et al, 1985). EDS may be an allergen specific to the macrophages and precipitates their phagocytic activities. As antigen-presenting cells, macrophages interact with and activate lymphocytes. Activated lymphocytes produce lymphokines and attract more macrophages to the testis. The augmentation of macrophage 120 phagocytic activity is suggested to be responsible for the destruction of Leydig cells. The precise mechanism ofEDS effect on Leydig cells, however, still remains unproven. Most studies concerning the cytotoxic effect of EDS on Leydig cells have been conducted in vivo. A few studies, nevertheless, have been carried out in vitro. Klinefelter et a1 (1991) reported that the in vitro EC50 ofEDS for hCG-stimulated testosterone production by Leydig cells was 370 uM while the in vivo EC50 was 60 mg!Kg for the rat. When hCG stimulated testosterone production decreased by 50% following in vitro or in vivo exposures, the morphological integrity of the Leydig cells appeared to be normal. This suggests that functional impairment of Leydig cells takes place at a lower dosage of EDS than morphological changes. It was further demonstrated in their studies that the biosynthesis of testosterone is compromised between the cAMP activation of protein kinase A and the P- 450""' enzyme (Klinefelter et al, 1991 ). It was concluded from another study conducted by Rommerts eta! (1988) that the in vitro effects ofEDS on Leydig cells are dependent on species and age ofthe animal. Their results showed that EDS exerted a direct inhibitory effect on the rat Leydig cells, but had no effect on LH-stimulated steroid production by mouse Leydig cells. They also .showed that a cytotoxic response to EDS developed in rat Leydig cells during maturation. It has also been demonstrated in other studies that immature rat Leydig cells are intrinsically less sensitive to EDS effects than adult Leydig cells. Total androgen production by testes perfused in vitro with 94 ug EDS/ml was dramatically reduced in adult, but not in immature rats. Adult rat Leydig cells were far more sensitive to the effects of EDS on LH-stimulated androgen production and on 35S methionine incorporation in comparison to the immature Leydig cells (Kelce et al, 1991). This lack of response of 121 immature Leydig cells to EDS might explain the resistance of Leydig cells to multiple doses of EDS. When a second challenge of EDS is introduced to rats with a interval of 2 to 3 weeks, the Leydig cell recovery is not affected. Responses of testis to multiple doses ofEDS are similar to those after a single dose, indicating that the regenerating Leydig cells are functionally different from the mature Leydig cells (Edwards et al, 1989). The pattern of the changes in serum testosterone levels after EDS treatment in the present study is similar to that in other reports (Molenaar et al, 1986; Vreeburg et ·at, 1988). Serum testosterone levels decreased to the lowest point between 2 to 10 days post-EDS treatment and increased gradually thereafter. The changes in serum testosterone levels reflect the degeneration and regeneration of Leydig cells after EDS treatment. As demonstrated in histological studies, Leydig cells undergo degeneration 24 hr post-EDS and no histologically recognizable Leydig cells are present in the testis by 48 hr post-EDS treatment (Kerr et al, 1985; Morris et al, 1986; Jackson et al, 1986). This morphological degeneration ofLeydig cells parallels the functional index of the mature Leydig cells, i.e., the production and secretion of testosterone into blood. When the histological regeneration of Leydig cells in the testis of the EDS-treated rats started around 2 weeks post-EDS treatment (Jackson et al, 1986; Kerr et al, 1987), serum testosterone levels increased gradually. Serum testosterone levels were lower yet not significantly different from those of the control rats at day 24 post EDS treatment in this study and other studies (Molenaar et al, 1986; Vreeburg et al, 1988). However, the morphological development ofLeydig cells is not complete and the number of Leydig cells in the interstitium observed in the histological studies has not returned to the control levels until 5 to 7 weeks post-EDS treatment (Jackson et al, 1986; Kerr et al, 1987; 122 Myers and Abney, 1991). This discrepancy of serum testosterone level and Leydig cell morphological development suggests that the functional development of Leydig cells after EDS treatment precedes the morphological maturation of Leydig cells. Testis weight decreased significantly at day 10 post-EDS treatment in comparison to the age-matched controls and reached the lowest point at day 16 after EDS treatment (half of the control value). Since Leydig cells represent only 3-4% of the total testicular volume in the mature rat testis (Mori and Christensen, 1980), the decline in testis weight probably reflects the damage of seminiferous tubules due to the deprivation of testosterone support after EDS treatment. This has been demonstrated in previous studies. Twenty four to 48 hours after EDS treatment, most Leydig cells exhibited various degrees of disintegration, yet the seminiferous tubules appeared qualitatively and quantitatively normal (Morris et al, 1986; Kerr et al, 1985). Between 4 and 7 days post-EDS treatment, degenerating germ cells were observed in the testis, especially in the basal aspects of seminiferous tubules at stage Vll and Vlll of the spermatogenic cycle. By 14 days post-EDS treatment, marked changes occured in the seminiferous epithelium. The seminiferous tubules were regressed and most of the germ cells were degenerated. No late spennatids were observed. The seminiferous tubules appeared normal at day 45 post-EDS when Leydig cells were regenerated in the testis. The changes of germ cells after EDS treatment are identical to those after hypophysectomy-induced deprivation of testosterone reported previously by Russell and Clermont (1977). Supplementation ofEDS-treated rats with increasing doses of a long-acting preparation of testosterone ester prevented the degenerative changes in the seminiferous epithelium in a dose-dependent manner (Jackson and Jackson, 1984; Sharpe et al, 1988). These results 123 support the conclusion that testosterone secreted from Leydig cells is absolutely required for spermatogenesis. Interstitial cell numbers, another quantitative index for cell degeneration and regeneration, were also measured in this study. In comparison to whole testis weight, interstitial cell number decreased significantly at day 2 post-EDS treatment. This dramatic decline in interstitial cell number may reflect the degeneration of Leydig cell and other cell types in the interstitium which are influenced by Leydig cell death. This is a reasonable interpretation considering that the recoverable interstitial cell number decreased from 89.9 X 106/testis in the control rats to the lowest point , 39.5 X Id /testis, at day 16 post-EDS treatment. This decline in interstitial cell number exceeds the total number ofLeydig cells in the rat testis which is approximately 25 X 106/testis (Saez, 1994). Therefore, degeneration of Leydig cells cannot explain the total loss of interstitial cells in the rat testis after EDS treatment. It is possible that Leydig cells are required to maintain the normal structure and function of other cell populations in the interstitium, through either testosterone or other regulatory factors secreted by Leydig ,cells. Testicular interstitial cell number increased between day 16 and day 36 post-EDS treatment and returned to the control level at day 36 post-EDS. This subsequent increase in interstitial cell number correlates with the regeneration of Leydig cells. The changes of the Leydig cell and precursor cell number after EDS treatment were also investigated in the present study. Leydig cell number decreased sharply at day 2 post EDS treatment. This is in agreement with the histological evidence that Leydig cells are degenerated within 2 days after EDS injection. Leydig cell number was lowest at day I 0 and 124 began to increase around two to three weeks post-EDS treatment. This increase in the Leydig cell number coincided with the rise in serum testosterone levels. By seven weeks after EDS treatment, Leydig ceils had completely regenerated and the number ofLeydig ceils returned to the control v!ilues. The pattern of changes in the precursor ceil number after EDS treatment is interesting. A slight increase in the precursor ceil number at day 2 post-EDS treament is probably associated with the proliferation process. Teerds et al (1990) reported that a peak of 3H thymidine incorporation in the mesenchymal ceils occured at day 2 post EDS treatment, indicating active ceil proliferation. However, the study ofMyers and Abney (1991) suggested that this initial peak of3H incorporation represents proliferation of other connective tissue ceils directly or indirectly destroyed by EDS treatment. This proliferative activity may not be associated with Leydig cell development. On the other hand, a smail peak of 3H incorporation in the interstitial ceils around day 20 post-EDS treatment is related to Leydig ceil development. The pattern of changes in the precursor ceil number after EDS treatment demonstrated in the present study may also result from the concurrent proliferation and differentiation processes. Therefore, no dramatic changes of the precursor ceil number were observed during Leydig ceil regeneration after EDS treatment. Histochemical studies of the precursor cell and Leydig cell fraction by 313- hydroxysteroid dehydrogenase (313-HSD) staining 313-HSD is one ofthe key enzymes in testosterone synthesis. It catalyzes the reaction which converts 313-hydroxy-5-ene steroids into 3-keto-4-ene steroids. Histochemical studies with 313-HSD staining using etiocholane as the substrate were utilized in this study to identify 3 13-HSD positive ceils in the testis. Leydig ceiis and precursor ceils were isolated from 125 Percoll gradient according to their densities. In the Leydig cell fraction of the control rats, more than 90% of the cells were intensely stained as 3j3-HSD positive cells. This indicates that' the purity of the Leydig cell fraction is high, with more than 90% Leydig cells in this fraction. In contrast to the Leydig cell fraction , the majority of the precursor cell fraction isolated from the control rats were only faintly stained. Less than 5% of this cell fraction were stained positive for 3j3-HSD. It is possible that these few 3j3-HSD positive cells in the precursor cell fraction were due to cross-contamination of Leydig cells which migrated from the heavier density fraction during the isolation process. However, it is also possible that some of the precursor cells were undergoing differentiation process to replace the normal physiological loss ofLeydig cells in the testis. Apoptotic changes have been observed at a low level in Leydig cells under physiological conditions (Tapanainen et al, 1993). Leydig cells are considered virtually a non-mitotic cell type. The number ofLeydig cells in the testis, however, remains relatively constant for a long period, if not throughout life (Christensen and Peacock, 1980). The maintenance of the stable Leydig cell number can only be achieved through differentiation from its precursors. Indeed, it has been demonstrated that LH/hCG induced an increase in Leydig cell number due to the stimulation of precursor cell differentiation (Christensen and Peacock, 1980). The interpretation that the precursor cells undergo differentiation is more plausible when we consider the results of the histochemical studies obtained from the day 20 EDS-treated rats. At day 20 post-EDS treatment, serum testosterone level began to increase and the in vitro testosterone production by the testicular interstitial cells began to increase (Myers and Abney, 1990). All the data from the present and previous studies indicate that Leydig cells are regenerating from the precursor cells around 126 this time. In the precursor cell fraction of the day 20 EDS-treated rats in the present study, overall93% of the cells are stained, of which 33% were as darkly stained as those observed in the control Leydig cell fraction and the rest were faintly stained. In contrast to less than 5% darkly stained cells in the precursor cell fraction of the control testis, a significant increase in the number of 313-HSD positive cells was observed in the precursor cell fraction at day 20 post-EDS treatment. This. provides direct evidence that precursor Leydig cells in the testis undergo an accelerated differentiation process to become the steroidogenic cells after EDS treatment. In the Leydig cell fraction at day 20 post-EDS treatment, 65% of the cells were 313-HSD positive, comparable to the precursor cell fraction at this time. These regenerated Leydig cells are derived from the differentiation and further maturation of the precursor cells. Similar results of 313-HSD histochemical studies were reported for immature rats (Hardy et al, 1990; Shan and Hardy, 1992). In their studies, precursor cells were isolated from the immature rats (21 days old) and the majority of these cells were faintly stained like those in our study. After 3 days in vitro incubation with hCG and dihydrotestosterone, most of the cells became positive for 313-HSD staining. These results indicate that precursor cells in the testis of the immature rats differentiate under the influence of LH/hCG and possibly androgens. The differentiation of the precursor cells during puberty gives rise to the adult Leydig cell population .. Edwards et al (1987) characterized the regenerating Leydig cell population after EDS treatment using 313-HSD staining. 313-HSD activity in the interstitial spaces and in the individual Leydig cells of the unfixed fresh-frozen testis sections was determined. About 40% of the interstitial spaces contained 313-HSD positive cells at day 21 post-EDS treatment. It is interesting to note that they also found 313-HSD activity in the 127 cytoplasm of the individual Leydig cells at day 21 post-EDS treatment was significantly higher than that of the controls. They postulated that this increase in the steroidogenic enzyme activities per Leydig cell in the EDS-treated rats is responsible for the relatively normal serum testosterone levels around this time. In another study using testicular tissue homogenates to measure the steroidogenic enzyme activities (O'Shaughnessy and Murphy, 1991), the results showed that 3 f3-HSD was the most active enzyme in the control testis. Three days after EDS treatment, 3f3-HSD activity in the testis homogenates decreased to 0.04% of the control level. 3f3-HSD was the first enzyme to increase in activity after EDS treatment. Between 7 and 14 days post-EDS, 3 f3-HSD activity increased more than tenfold while the other steroidogenic enzyme activities did not change (P-450,"', P-45017J or continued to decline (17-ketosteroid reductase). Therefore, histochemical studies using 3 f3-HSD staining is a reliable index for Leydig cell development. Analysis of the precursor cell and Leydig cell fractions from the control and EDS treated rats by flow cytometry In contrast to the classical histological methods, flow cytometry is an investigational technique which provides multiple objective simultaneous measurements at the single cell levels. This advanced technique was utilized in our study to characterize Leydig cell and precursor cell fractions isolated from the Percoll gradient regarding their physical and biochemical features. Cell fractions from the control and EDS day 20 rats were subjected to flow cytometry. Precursor cell and Leydig cell fractions at day 20 post-EDS treatment were analyzed and compared to the controls because at that time precursor cells are undergoing the differentiation process to give rise to a new Leydig cell population. 128 Of all the data display and handling procedures in flow cytometry, counting within gates is the simplest and most efficient. The gates were chosen according to the distribution ofthe cell population and applied to each group. Data collected by this method show that a higher percentage of cells in the precursor cell fraction lie in the larger cell range compared to the Leydig cell fraction, indicating cells in the precursor cell fraction are generally larger than those in the Leydig cell fraction. It is difficult to explain this observation considering that precursor cells are quiescent and functionally undifferentiated and Leydig cells are active steroidogenic cells in the adult rats. It is unlikely that this is due to the purity of the cell populations, because the results of the histochemical studies indicate that the purity of these two cell fractions were greater than 90%. This significant difference in the cell size between precursor cells and Leydig cells remains unclear at this time and awaits further studies at higher morphological and biochemical levels, e.g., utilization of electron microscopy for the ultrastructural studies .. At day 20 post-EDS treatment, a time when precursor cells are differentiating, cell size increased significantly in the precursor cell fraction in comparison to the controls. This increase in cell size of precursor cells is predictable. Specific genes are turned on and proteins are synthesized during cell differentiation. Cells in the Leydig cell fraction also increased in size at day 20 post-EDS treatment when compared to the control Leydig cells. These cells were regenerated from the differentiation of the precursor cells and the maturation process was still continuing at this time. It is reasonable to interpret that these differentiating and maturing cells were greater in size than the fully mature Leydig cells in the control testis. This temporary increase in cell size during differentiation is probably due to a more active gene 129 expression and protein synthesis in the precursor cell and maturing Leydig cell fractions at day 20 post-EDS treatment. Granularity is an index which reflects the density of internal particles. Flow cytometric analysis reveal that at day 20 post-EDS treatment, when precursor cells were differentiating, granularity in this cell fraction increased significantly in comparison to the controls. This increase in granularity suggests that active cellular events were turned on during precursor cell differentiation, such as an increase in protein synthesis and in the number of organelles. It is interesting to note that when precursor cells further differentiate into Leydig cells, which were isolated from a heavier density range on the Percell gradient at day 20 post-EDS treatment, granularity of the cells returned to the control Leydig cell level. This indicates that the activation of the genes, and in turn the protein synthesis, is developmentally regulated. Certain genes involved in development are expressed only at a specific time during the precursor cell differentiation. It is well known that steroidogenic cells contain abundant mitochondria. Leydig cells are characterized by prominent smooth endoplasmic reticulum, Golgi apparatus and large mitochondria containing tubular cristae. Fluorescent probes are now available and serve as a useful tool to study organelles in living cells. The mitochondria-specific probe, cyanine dye DiO, belongs to a class called distributional probes. The cationic cyanine dye is accumulated in compartments with a negative potential, such as mitochondria, and thus is regarded as an indicator of mitochondrial activity. It is used as a selective stain for mitochondria electronegativity and/or number in viable cells by flow cytometric analysis. The data in the present study indicate that DiO-mitochondria fluorescence was greater in the precursor cell 130 fraction than that in the Leydig cell fraction. Precursor cells at day 20 post-EDS treatment exhibited higher DiG-mitochondria fluorescence compared to those in the control rats. This difference in DiD-mitochondria fluorescence can be explained by either an increase in the mitochondria activity or an increase in the number of mitochondria. We can't distinguish between these two indices in the present study. Nevertheless, when the DiG-mitochondria fluorescence is corrected by cell size, the difference between precursor cells and Leydig cells is no longer significant. Similarly, cells in the control and EDS-treated rats do not demonstrate significant difference in DiD-mitochondria fluorescence when it is corrected by cell size. It is therefore postulated that the difference in the mitochondria fluorescence among different cell population is due to changes in mitochondria number in individual cells. During differentiation, precursor cells increase in cell size and thus the number of the mitochondria is expected to increase. Abundance of lipid droplets in the cytoplasm is a characteristic of the fetal-type Leydig cells (Magre and Jost, 1980; Narbaitz and Adler, 1967). The measurement of lipid droplets in this study was based upon the metachromatic fluorescence of nile red. Nile red exhibits properties of a near-ideallysochrome. It is almost nonfluorescent in water and polar solvents. However, it is strongly fluorescent in the presence of a hydrophobic environment and undergoes large blue shifts in both absorption and emission. Lipid droplets are neutral lipids, containing mainly cholesteryl esters or triacylglycerols. By examining the fluorescence of nile red-stained cells at wavelengths of,;; 570 nm, the fluorescence of nile red interacting with an extremely hydrophobic environment (i.e., neutral lipid droplets) is detected preferentially whereas the fluorescence of nile red interacting with cellular membranes is 131 minimized (Greenspan et al, 1985). Therefore, nile red fluorescence was measured at 488 nm in this study and the fluorescence in different cell populations was compared. Data from the flow cytometric analysis in the present study show that cells in the precursor cell fraction demonstrated a higher lipid droplet level compared to the Leydig cell fraction. This is comparable to the fetal Leydig cells. It ha~ been illustrated by different investigators that the precursor and immature Leydig cells share similarities with the fetal-type Leydig cells. Histological studies using EDS-treated rats clearly indicate that the newly formed Leydig cells after EDS treatment contain a large amount of lipid droplets (Vreeburg et al, 1988; Jackson et al, 1986). The significance of the abundant lipid droplets in the fetal type and immature Leydig cells is unclear. Nevertheless, fetal-type Leydig cells of different species demonstrate this distinct characteristic, including the rat and the human (Byskov, f 1986). This abundance in lipid droplets during Leydig cell development may result from an increase in lipid transportation and storage in the cytoplasm for subsequent steroidogenesis. When the cells further mature, the capacity to synthesize steroid hormone increases greatly. Cytoplasmic lipid is efficiently transported into mitochondria for the synthesis of steroid hormone and this may result in the decrease in the cytoplasmic lipid droplets. Characterization of in vitro testosterone production by Leydig cells and precursor cells during differentiation Precursor cells and Leydig cells were isolated from testes of the control and EDS treated-rats. These cells were incubated in vitro for 24 hr in the absence or presence of 50 miU hCG. Testosterone production, a main functional characteristic of Leydig cells, was measured by RIA to determine the differentiated state of the precursor cells and the 132 regeneration of the Leydig cells. In the control testis, cells isolated from the Leydig cell fraction produced testosterone and exhibited a high responsiveness to hCG stimulation (7.5 fold increase in testosterone production). Similar results have been reported using Leydig cells prepared by the same cell isolation system (Klinefelter et al, 1987). Precursor cells isolated from testes of the control rats showed little capacity for testosterone production and hCG responsiveness. The presence of such a precursor population for Leydig cells in the testis has been reported by several studies. Hardy et a! (1990) isolated precursor Leydig cells from immature rats (21 days old) and demonstrated the differentiation of these precursor cells in vitro under the influence of LH/hCG and androgens. Human Leydig .cell mesenchymal precursors have also been isolated from patients with the androgen insensitivity syndrome (Chemes et a!, 1992). These precursor cells were cultured in vitro for 11 days. Electron microscopic studies showed a homogenous population of poorly differentiated, fibroblastic type mesenchymal cells in the absence of hCG. Cultures stimulated with hCG demonstrated a marked increase in organelles, particular the number of mitochondria cisternae and smooth endoplasmic reticulum. The addition ofhCG in the culture also resulted in a increase in 3P HSD positive cells and testosterone production by these precursor cells. · Precursor Leydig cells are a progenitor cell type that have the capacity to differentiate into Leydig cells under appropriate circumstances. Although quiescent and non-functional, these precursor cells in the testis have probably been determined for their further development. Determination involves the establishment of characteristics which distinguish these cells from others, e.g., expression of the truncated form of the LH receptor mRNA transcript. Determination is essentially an irreversible change. In this case, the precursor 133 Leydig cells reside in the interstitium of the testis and have been comitted to a particular developmental pathway, which is to differentiate into Leydig cells under appropriate circumstances (EDS treatment). The equilibrium between Leydig cell and precursor cell populations in the testis of the mature rats is maintained by LH and local regulators. After Leydig cells are destroyed by EDS injection, this equilibrium is altered. The most dramatic change is the rapid rise of the pituitary gonadotropins in the blood. The serum UI Ievel·sharply increases between 24 and 48 hr after EDS treatment, reaches the highest point between 7 and 14 days after EDS treatment. The peak value of serum LH level is around 500 ng/ml, which is similar to that of the castrated rats (Molenaar et a!, 1986; Keer et al, 1987). Thereafter, as a consequence of Leydig cell regeneration and a rapid rise in serum testosterone level, the serum LH level sharply declines to the baseline level of about 100 ng!ml. Serum FSH level shares a similar pattern as that of UI after EDS treatment. However, unlike LH, the increase in FSH after EDS treatment does not reach the castration level. This is due to the effect of inhibin produced by Sertoli cells in the EDS-treated rats. Administration of antiserum against inhibin further increases serum FSH to the castration level (Sharpe et al, 1990). The high level of serum LH is essential for Leydig cell regeneration post-EDS treatment, as demonstrated in previous studies. In rats with testosterone implants that suppressed UI levels but maintained normal FSH levels, no repopulation of Leydig cells was observed until 35 days post-EDS treatment (Molenaar et al, 1986). The results also showed Leydig cell regeneration after EDS injection was suppressed in hypophysectomized rats. Daily injection of hCG restored Leydig cell regeneration in both hypophysectomized rats and the 134 rats with testosterone implants. Teerds et al (1989) suggested, based upon their data, that early proliferation of precursor cells in the testis of the EDS-treated rats is regulated by both LH and local factors, whereas the differentiation of precursor cells is dependent upon LH stimulation. The present study suggests that within 2 days after EDS treatment, Leydig cells are completely destroyed. This is demonstrated by the absence of testosterone production in this cell fraction. Precursor cells exhibited no capacity to produce testosterone at day 2 and day 10 post-EDS treatment, indicating that the differentiation process had not started in the precursor cell fraction. Precursor cells began to produce testosterone and respond to hCG stimulation around 2-3 weeks post-EDS treatment. During the period of 3-6 weeks after EDS injection, this precursor cell population underwent further differentiation and increased in steroidogenic capacity. Testosterone production in the precursor cell fraction during this period resulted from the functional differentiation of these cells. Our data suggest that the initial stages of differentiation probably involve alterations. in biochemical features, i.e., steroidogenic enzyme activity. These functional changes occur prior to the morphological changes. Once the differentiating precursor cells undergo further maturation, they change morphologically and become heavier. These cells became Leydig cells and were then isolated on the Percoll gradient in the Leydig cell fraction at a density higher than 1.070 g. Consequently, when all differentiated precursor cells have matured into Leydig cells, the precursor cell fraction again contained only the undifferentiated cells and lacked the capacity to produce testosterone. It has been reported that during fetal development the rat testis demonstrates 3 13-HSD activity and the capacity to convert progesterone into testosterone in 135 vitro between days 13 and 15 of fetal life, whereas fetal Leydig cells are not histologically observed until day 15.5 of fetal life (Niemi and Ikonen, 1961; Noumura et al, 1966; Gangnerau and Picon, 1987). This suggests that the functional differentiation precedes the structural maturation during Leydig cell development in fetal life. The results of the present study indicate that the regeneration of Leydig cells after EDS treatment shares the same pattern. Risbridger and Davies (1994) utilized this precursor ceii!Leydig cell isolation system to study Leydig cell development in the EDS-treated rats. In vitro testosterone production by these two cell fractions was measured in their studies. Their data also demonstrated that precursor cells differentiated into Leydig cells after EDS treatment, which is in agreement with ours. Coinciding with the differentiation of the precursor cells, a new population of Leydig cells were formed after EDS treatment. Between days 16 and 20 post-EDS treatment, functional steroidogenesis was demonstrated in the Leydig cell fraction. Testosterone production and hCG responsiveness in this Leydig cell fraction further increased between days 24 and 60 after EDS treatment, but was not statistically different from that ofthe control. These data suggest that the newly formed Leydig cells are derived from the differentiation of the precursor cells in the testis. It is likely that the early differentiation process in the precursor cells, including the acquisition of the steroidogenic capacity, is more prominent than the further maturation of the newly formed Leydig cells. LH receptor and steroidogenic enzyme (P-450,.,., and P-45017J messenger RNA changes during the differentiation process LH is the preeminent tropic factor regulating Leydig cell function. LH receptor has 136 been demonstrated to be exclusively present in Leydig cells in the male rat. Thus, LH receptor can be used as a functional index for testicular Leydig cells. The level of LH receptor is modulated by a variety of hormones, growth factors and second messenger analogs. The regulation of LH receptor density in Leydig cells can occur at transcriptional, post transcriptional (i.e., splicing) and receptor protein levels. Multiple species ofLH receptor mRNA were first detected by McFarland et al (1989) using Northern blot analysis and these initial findings have been confirmed by different investigators. These transcripts vary in size from 1.2 Kb to 7.6 Kb. Since the full length LH receptor eDNA is 2.1 Kb in size, the truncated' forms of 1.2 and 1.8 Kb transcripts obviously can't be translated into the whole receptor. Of all known transcripts, the 1.2 Kb and 7.6 Kb transcripts are least abundant and not always detectable (Segaloffand Ascoli, 1993). The major specie ofLH receptor in both testis and ovary is the 7.0 Kb transcript. The amount of the ·1.0 Kb transcript is well correlated with hCG binding capacity in the testis (Vihko et al, 1992), suggesting that this transcript is responsible for the synthesis of the receptor protein. In the present study, all four major transcripts ofLH receptor (7.0, 4.2, 2.5 and 1.8 Kb) were detected in the Leydig cell fraction of the mature control rats using Northern blot analysis. On the other hand, in the undifferentiated precursor cell fraction ofthe control rats, only the truncated 1.8 Kb transcript was detected. This truncated form was also the only transcript present in the precursor cell fraction at day 2 post-EDS treatment. Studies using EDS-treated rats have shown that hCG binding capacity in the testis declined sharply within 48 hr to very low or negligible levels (Jackson et al, 1986; Myers and Abney, 1990). It has been demonstrated that this 1.8 Kb transcript of LH receptor encodes a truncated form of receptor corresponding to the 137 extracellular domain of the receptor protein. This truncated form ofLH receptor is capable ofbinding IR/hCG. However, it remains trapped within the cells rather than being secreted to the cell surface (Segaloffand Ascoli, 1993). This explains the observation that the testis devoid ofLeydig cells after EDS treatment lacks the capacity to bind LHihCG in spite of the expression of a truncated form ofLH receptor transcript in the precursor cells. The persistent expression of a truncated form ofLH receptor mRNA in the testis of the EDS-treated rats was also demonstrated by Tena-Sempere eta! (1994). Their results showed that only the 1.8 Kb transcript was detected in the testis by both Northern blot analysis and RT-PCR at days 5 and 15 post-EDS treatment. The whole testes, instead of isolated precursor cells, were utilized in their experiment. It was concluded from their observations that the Leydig cell precursors constitutively express a truncated form ofLH receptor mRNA. Northern blot results in the present study show that the precursor cells expressed all major transcripts ofLH receptor at day 10 post-EDS treatment, whereas Tena-Sempere et al (1994) reported that only the truncated 1.8 Kb transcript was detected in the testis at day 15 post-EDS treatment. This discrepancy probably results from the lower sensitivity of northern blot analysis when the whole testicular tissue was used instead of isolated cell fractions. The pattern ofLH receptor mRNA levels correlates to the in vitro testosterone production by cells in the precursor cell fraction after EDS treatment. The level of the full length LH receptor transcript (7.0 Kb) increased in the precursor cells between days 16 and 45 after EDS treatment. Concomitantly, precursor cells exhibited an increase in testosterone production and hCG responsiveness. It is interesting to note that the truncated form ofLH receptor mRNA in the precursor cells decreased dramatically during this period. These results 138 suggest that alternative splicing is important in the regulation ofLH receptor mRNA levels. At day 60 post-EDS treatment, the regeneration of Leydig cells was complete and the differentiated precursor cells had given rise to a new Leydig cell population. Cells isolated in the precursor cell fraction, therefore, no longer exhibited the capacity of producing testosterone or expressing LH receptor mRNA other than the truncated transcript. These results indicate that precursor Leydig cells in the rat testis constitutively express the truncated form ofLH receptor mRNA. Full length transcripts of the LH receptor gene are associated with the differentiated function ofLeydig cells. The regulation ofLH receptor mRNA during Leydig cell development is probably mainly due to the alternative splicing. Hence, the truncated' form ofLH receptor mRNA is characteristic of the precursor cells and might be related to the early differentiation ofthe precursor cells. Ofthe steroidogenic enzymes, P-450.., and P-4501701 are considered of great interest in the investigation of Leydig cell development. P-450"' is the rate limiting enzyme during fetal-type Leydig cell development and, as a labile steroidogenic enzyme, P-4501701 is regulated by many factors via different mechanisms. Therefore, both P-450.., and P-4501701 are viewed as important markers in the Leydig cell degeneration and regeneration process after EDS treatment. The results show that neither of these two steroidogenic enzymes was expressed in the precursor cells in the control rats. Mature Leydig cells expressed both the enzymes as expected. As early as 10 days post-EDS treatment, the same time when the full length LH receptor mRNA transcripts were detected, both P-450.,. and P-4501701 mRNA were expressed in the precursor cell fraction. It has been reported that the enzymatic activities of both P-450.,. and P-4501701 increased markedly between days 14 and 21 post-EDS treatment 139 (O'Shaughnessy and Murphy, 1991). This increase in the enzymatic activiy results from the increase in the amount of the messenger RNA. The marked increase of serum LH level is essential for the development of the steroidogenic enzymes. In both mouse (Payne and Sha, 1991) and rat (Payne, 1990) Leydig cells, expression of P-45017,. mRNA is absolutely dependent on LH acting via cAMP. LH also stimulates the expression ofP-450"', even though this enzyme is constitutively expressed in Leydig cells at a low level. In the present study, between days 24 and 36 post-EDS treatment, P-450,,. mRNA levels reached a peak in the precursor cell fraction. The peak values ofP-450... mRNA level in the precursor cells were similar to that in the control Leydig cells. Thereafter, at day 45 post-EDS treatment, P- 450,,. mRNA level in the precursor cell fraction decreased significantly. The pattern ofP- 45017,.mRNA changes in the precursor cell fraction during differentiation is similar but not identical to that ofP-450... mRNA The peak values ofP-45017,. mRNA in the precursor cell fraction were higher than that of the control Leydig cells. Unlike P-450.., P-45017,. mRNA level remained high at day 45 post-EDS treatment. This overexpression ofP-45017,. mRNA is probably important for the differentiation of the precursor cells, especially when the labile nature ofP-45017,. is considered. Overexpression would ensure that sufficient levels of the enzyme are synthesized. Again, these two steroidogenic enzymes were not expressed in the precursor cell fraction when the differentiation process was completed. These results demonstrate the differentiation of the precursor cells after EDS treatment at the mRNA level and are correlated to other results in the present study, e. g., serum testosterone levels and in vitro testosterone production by the isolated cells. 140 Estrogen receptor and its mRNA levels in the testis and in isolated precursor cells and Leydig cells Previous studies have suggested that the estrogen binding capacity of the rat testis is located in the interstitium (Stumpf, 1969; Mulder eta!, 1973). In fact, studies using purified Leydig cells (Nozu eta!, 1981; Lin et al, 1982) and the TM3-mouse Leydig cell line (Nakhla eta!, 1984) have led to the conclusion that one characteristic ofLeydig cells is the expression of the estrogen receptor. In the present study, estrogen binding capacity was demonstrated in the testis of the mature rats. This is in agreement with the current knowledge that estrogen receptor is present in Leydig cells. Yet, when Leydig cells were destroyed by EDS treatment, a lower but significant level of estrogen binding was detected in the testis. This strongly suggests that estrogen receptor is present in other cell types in the testis. A previous report from our laboratory showed that estradiol directly inhibits the regeneration of Leydig cells in the EDS treated rat testis (Abney & Myers, 1991). It is plausible to propose a hypothesis from these data that estrogen receptor is present in the precursor Leydig cells in the testis. Estrogen receptor binding data in the present study illustrated the presence of estrogen receptor in the testis devoid ofLeydig cells. To test the hypothesis that estrogen receptor is present in the precursor cells, further localization of estrogen receptor in the testis was investigated. Although it is generally accepted that estrogen receptor is localized in the interstitium of the testis, studies ofNakhla et al (1984) showed that Sertoli cells contained estrogen receptor. It therefore became imperative that localization of estrogen receptor in the testis of the EDS-treated rats by immunohistochemical analysis be performed to resolve this question. Our data demonstrate clearly that the receptor was localized within the interstitial 141 cells. In the control testis, in which a population of mature functional Leydig cells is found, the immunohistochemical staining was restricted to the interstitial cells. No staining was observed in the seminiferous tubules. In the day 10 EDS-treated animal, which is devoid of Leydig cells but contains precursor cells, immunostaining also was exhibited in the interstitium and not in the tubules. Staining was more intense in the nucleus, but was not confined to the nucleus. It is generally accepted that the estrogen receptor is localized in the nucleus of target tissues. Previous studies (De Boer et al., 1977), however, have suggested that the estrogen receptor is not as tightly bound in the cell nucleus in the testis as in the uterus. This weaker nuclear binding might explain the observation in the present study that the estrogen receptor appeared to be diffilsed to the cytoplasm during immunohistochemical analysis. The data obtained with the synthetic receptor peptide, nevertheless, demonstrated that the immunohistochemical staining was specific for estrogen receptor. Furthermore, the lack of the immunostaining of estrogen receptor in spleen in this study verifies the specificity of estrogen receptor localization in the testis. To further demonstrate that estrogen receptor is localized in the precursor cells, immunohistochemical studies were performed in the isolated cells from testis. The results indicate that estrogen receptor was present in both Leydig cells and precursor cells in the control rats. Precursor cells isolated at day 10 post-EDS treatment also were shown to contain estrogen receptor. The present study therefore clearly demonstrated that both Leydig cells and the precursor cells contain estrogen receptor. RT -PCR was utilized to detect estrogen receptor expression in the testis at the messenger RNA level in the present study. Estrogen receptor mRNA has been identified in 142 many tissues by using RT-PCR technique (Wu et at., 1993; Hou and Gorski, 1993). This technique is especially useful in detecting the presence of specific mRNAs of low abundance. This is the case in the present study where Leydig cells and precursor cells represent only a small portion, less than 4%, of the whole testicular mass. By simultaneous reverse transcription and amplification of the target estrogen receptor mRNA and the exogenous rabbit ~-globin mRNA as the internal standard in the same reaction tube, variations of the reverse transcription and the eDNA amplification are controlled. Quantification of the mRNAs using RT-PCR with an exogenous sequence was first described by Wang et al (1989). In their study, a synthetic RNA internal standard was included in the PCR to quantify several cytokine mRNA levels in stimulated macrophages. The RNA standard used in their approach shared primer binding sites with the target RNA, therefore the variation in amplification efficiency was controlled. This made it possible to calculate the absolute amount of the target RNA In the present study, the differences in levels of estrogen receptor mRNA between Leydig cells and precursor cells, and the pattern of changes in the estrogen receptor mRNA in these cells during Leydig cell development were emphasized, rather than the ab.solute amount ofthe mRNA The internal standard was used to control the variable effects of the conditions in the reverse transcription and amplification procedures among different reaction tubes. The differences in amplification efficiencies between the target RNA and the internal standard RNA were of no concern in this case. A similar coarnplification method also was performed by Chamberlain eta! (1988) and Kellogg eta! (1990), with an endogenous RNA, ~-actin, as the internal standard. The quantitative RT-PCR data show that in the control rat, estrogen receptor mRNA 143 levels were 20 fold higher in the precursor cells than in Leydig cells. Estrogen receptor mRNA levels decreased after EDS treatment and reached the lowest point around day 16 post-EDS, a time when precursor cells were beginning to differentiate into functional Leydig cells. In the course of precursor cell differentia!ion, estrogen receptor mRNA levels increased gradually and when the precursor cells finished the differentiation process, estrogen receptor mRNA returned to the control precursor cell levels. The amount of estrogen receptor mRNA in Leydig cells was low compared to the precursor cells, and was constant during the newly formed Leydig cell maturation period. The results suggest that the relatively high level of estrogen receptor gene expression may be a characteristic of the precursor cells and the decrease in estrogen receptor mRNA might be important for precursor cells to differentiate into Leydig cells. The quantitative analysis of estrogen receptor by immunofluorescence study further confirmed that a significant higher level of estrogen receptor was present in the precursor cells. The present study therefore suggests that the role of estradiol in the testis may extend beyond that as a modulator of steroidogenesis. Indeed, estradiol might serve as a paracrine factor in regulating Leydig cell ontogeny and maturation. One could postulate that estradiol produced from testosterone by aromatase in the Leydig cells (Valladares et al., 1981; Payne et al., 1987) might be secreted and act locally in the interstitium to regulate either the proliferation or differentiation process of the precursor cells. Regulation of precursor Leydig cell differentiation by LH/hCG and other possible factors This set of experiments was initially designed to examine the regulatory effect of 144 IB/hCG and other factors on the precursor cell differentiation process. It is well known that the development of Leydig cells during puberty is absolutely dependent upon LH as demonstrated by in vivo studies (Lee et al, 1975, Tapanainen et al, 1984). In the present study, a Percoll gradient technique was utilized to isolate precursor cells from the testis. Precursor cells were cultured in vitro and testosterone production was determined to examine the differentiation status of the precursor cells. The results show that precursor cells isolated from the immature rats (21 days old) differentiated in vitro under the stimulation ofhCG, as demonstrated by increasing testosterone production and hCG responsiveness. This hCG dependent differentiation of precursor cells in the immature rats has been reported previously by Hardy et al (1990). However, in spite of the similarity of Leydig cell development in the immature and EDS-treated rats, precursor cells isolated at day 20 post-EDS treatment failed to differentiate in vitro in the presence ofhCG. The introduction ofiGF-I in addition to hCG in the media also failed to stimulate the functional differentiation of precursor cells after EDS treatment. These results suggest that the development of Leydig cells after EDS treatment is somehow not identical to that during puberty. LH/hCG is the primary regulator on Leydig cell regeneration after EDS treatment, yet not the only one. Testicular local factors might be involved in the regulation of the precursor cell differentiation process. This area warrants further investigation in future studies. Table 2 summarizes the differential changes of precursor cells and Leydig cells after EDS treatment. Table 2. Summary of precursor cell differentiation and Leydig cell regeneration after EDS treatment. Changes during differentiation of precursor cells and regeneration ofLeydig cells after EDS treatment were summarized. Each value represents the mean from different experiments. mRNA levels were standardized by the house keeping gene, GAPD mRNA, and expressed as relative units. a: the lowest value after EDS treatment. b: the first significant increase in comparison to the lowest value. c: the first value which shows no significant difference from that of the controls. i: the first time when a biochemical feature appears or reappears after EDS treatment. u: the first significant increase in comparison to i. m: the first value which shows no significant difference from that of the controls. V') ·.·. -' The present study was designed to characterize precursor Leydig cells and· their differentiation process. EDS-treated rats were used as the animal model and the following objectives were examined: 1) the development of testosterone production and hCG responsiveness in the precursor cell and the regenerated Leydig cell populations after EDS treatment. 2) LH receptor and steroidogenic enzyme gene expression during Leydig cell development. 3) determination and localization of estrogen receptor and its mRNA in the testis. Several conclusions are drawn based upon the results: 1). The regenerated Leydig cells after EDS treatment are derived from a precursor cell population in the interstitium of the testis. 2). All four major transcripts ofLH receptor mRNA are expressed in the mature Leydig cells, while the precursor cells express only a truncated form ofLH receptor mRNA (1.8 Kb). The fuU length 7.0 Kb transcript and the other two transcripts are expressed in the precursor cells during differentiation. The level of the full length transcript increases, whereas that ofthe truncated form decreases, in the precursor cells during the differentiation process. The expression ofLH receptor returns to the control undifferentiated levels in the precursor cell fraction when Leydig cell regeneration is complete. 3). Gene expression of the steroidogenic enzymes, P-450... and P-45017,., is not 146 147 present in the precursor cells in the control rats. Precursor cells acquire the expression of these steroidogenic enzymes when differentiation is initiated. mRNA levels ofP-450"' and P- 45017~ increase in the precursor cell fraction during the differentiation process and return to the control levels when the differentiated precursor cells have formed a new Leydig cell population. 4). Estrogen receptor and its mRNA is present in both Leydig cells and precursor cells in the testicular interstitium. A higher level of estrogen receptor is characteristic of the precursor cells. Estrogen receptor mRNA levels in the precursor cell fraction decreases dramatically around 2 weeks post-EDS treatment when the differentiation process begins. Thereafter, the levels of estrogen receptor mRNA in the precursor cell fraction increase and return to that of the control. It is therefore concluded from the present study that the expression of a truncated form ofiJI receptor mRNA and a higher level of estrogen receptor are characteristics of the precursor Leydig cells. In addition, a decrease in the estrogen receptor mRNA level in the precursor cells might be associated with the differentiation process. LITERATURE CITED Abney TO 1976 A cytoplasmic estradiol receptor in rat testicular tissue. Endocrinology 99: 555-556 Abney TO, Carswell LS 1986 Gonadotropin regulation of Leydig cell DNA synthesis. Mol Cell Endocrinol 45: 157-165 Abney TO, Myers RB 1991 17J3-estradiol inhibition of Leydig cell regeneration in the ethane dimethylsulfonate-treated mature rat. J Androl 12: 295-304 Anderson CM, Mendelson CR 1985 Regulation of steroidogenesis in rat Leydig cells in culture: effects of human chorionic gonadotropin and dibutyryl cyclic AMP on the synthesis of cholesterol side chain cleavage cytochrome P450 and adrenodoxin. Arch Biochem Biophys 238: 378-387 Agelopoulou R, Magre S, Patsavoudi E, Jost A 1984 Initial phases of the rat testis differentiation in vitro. J Embryo! Exp Morpho! 83: 15-31 Aquilano AR, Dufau ML 1984 Functional and morphological study on isolated Leydig cells: purification by centrifugation, elutriation and metrizamide fractionation. Endocrinology 114: 499-510 Ascoli M 1981 Chracterization of several clonal lines of cultured Leydig tumor cells: gonadotropin receptors and steroidogenic responses. Endocrinology 108: 88-95 148 149 Ascoli M, Segaloff DL 1989 On the structure of the luteinizing hormone/chorionic gonadotropin receptor. Endocr Rev 10: 27-44 Aubert ML, Begeot M, Winiger BP, Morel G, Sizonenko PC, Dubois PM 1985 Ontogeny of hypothalamic luteinizing hormone-releasing hormone (GnRH) and pituitary GnRH receptors in fetal and neonatal rats. Endocrinology 116: 1565-1575 Avallet 0, Vigier M, Perrard-Sapori MH, Saez JM 1987 Transforming growth factor- 131 inhibits Leydig cell function. Biochem Biophys Res Commun 146: 575-581 Avallet 0, Vigier M, Leduque P, Dubois PM, Saez JM 1994 Expression and regulation of transforming growth factor 131 mRNA and protein in cultured porcine Leydig and Sertoli cells. Endocrinology 134: 2079-2087 Bain PA, Yoo M, Clarke T, Hammond SH, Payne AH 1991 Multiple forms of mouse 313-hydroxysteroid dehydrogenaseh. '-~>4 isomerase and differential expression in gonads, adrenal glands, liver and kidneys of both sexes. Proc Nat! Acad Sci ' 88: 8870-8874 Bardin CW, Cheng CY, Musto NA, Gunslus GL 1988 The Sertoli cell. In: Knobil E, Neill D, Ewing LL, Greenwald GS, Markert CL, PfaffDW (eds) The Physiology of Reproduction. Raven Press, New York, pp 933-974 Bartke A, Haflez AA, Bex FJ, Dalterio S 1978 Hormonal interactions in regulation of androgen secretion. Bioi Reprod 18: 44-54 Benahmed M, Reventos J, Tabone E, Saez JM 1985 Cultured Sertoli cell mediated FSH stimulatory effect on Leydig cell steroidogenesis. Am J Physiol 248: E 1 7 8- E181 150 Bergh A 1982 Local differences in Leydig cell morphology in the adult rat testis: evidence for a local control of Leydig cells by adjacent seminiferous tubules. lnt J Androl 5: 325- 330 Bradshaw KD, Waterman MR, Couch RT, Simpson ER, Zuger MX 1987 Characterization of complementary deoxyribonucleic acid for human adrenocortical17a.-hydroxylase: a probe for analysis of 17a.-hydroxylase deficiency. Mol Endocrinol 1: 348-354 Braun T, Schofield PR, Sprengel R 1991 Amino-terminal leucine-rich repeats in gonadotropin receptors determine hormone selectivity. EMBO J 10: 1885-1890 Brinkmann AO, Mulder E, Lamers-Stahlhofen GJM, Mechielsen MJ, van der Molen HJ 1972. An estradiol receptor in rat testis interstitial tissue. FEBS Lett 26: 301-305 Brinkmann AO, Leemborg FG, Roodnat EM, de Jong FH, van der Molen HJ 1980 A specific action of estradiol on enzymes involved in testicular steroidogenesis. Bioi Reprod 23: 801-809 Bu'Lock DE, Jackson CM 1974 Suppression of testicular androgen synthesis in the rat by ethane 1,2, dimethanesulphonate. J Steroid Biochem 6: 1181-1185 Burgoyne PS, Buehr M, Koopman P, Rossant J, McLaren A 1988 Cell-autonomous action of the testis-determining gene: Sertoli cells are exclusively XY in XX XY chimaeic mouse testes. Development 102: 443-450 Byskov AG, Yding Anderson C, Westergaard L 1983 Dependence of the onset of meiosis on the internal organization of the gonad. In: McLaren A, Wyllie CC ( eds) Current Problems in Germ Cell Differentiation. Cambridge University 151 Press, Cambridge, pp 215-224 Byskov AG 1986 Differentiation of mammalian embryonic gonad. Physiological reviews 66: 71-117 Caileau J, Vermeire S, Verhoeven G 1990 Independent control of the production of insulin-like growth factor I and its binding protein by cultured testicular cells. Mol Cell Endocrinol 69: 79-89 Canuck JA, Makris A, Gunsalus GL, Ryan KJ 1979 Testicular aromatization in immature rats: localization and stimulation after gonadotropin administration in vivo. Endocrinology 104: 285-288 Chamaberlain JS, Gibbs RA, Ramier JE, Nguyen PN, Casky CT 1988 Deletion screening of the Duchenne muscular dystrophy locus via multiplix DNA amplification. Nucleic Acids Res 16: 11141-11156 Chemes HE, Gottlieb SE, Pasqualini T, Domenichini E, Rivarola MA, Bergada C 1985 Response to acute hCG stimulation and steroidogenic potential of Leydig cell fibroblastic precursors in humans. J Androl6: 102-112 Chemes H, Gigorraga S, Bergada C, Schteingart H, Rey R, Pellizzari E 1992 Isolation of human Leydig cell mesenchymal precursors from patients with the androgen insensitivity syndrome: testosterone production and response to human chorionic gonadotropin stimulation in culture. Bioi Reprod 46: 793-801 Chen YDI, Shaw MJ, Payne AH 1977 Steroid and FSH action on LH receptors and LH-sensitivity testicular responsiveness during sexual maturation of the rat. Mol Cell Endocrinol 8: 291-299 152 Chen YI, Kraemer FB, Reaven GM 1980 Identification of specific HDL-binding sites in rat testis. J Bioi Chern 265: 9162-9169 Cheng CY, Morris PC, Zwain I, Carreau S, Drosdowsky MA, Bardin CW 1993 A Leydig cell stimulator from Sertoli cells. In: Grave GD, Cutler GB (eds) Sexual Precocity: Etiology, Diagnosis and Management. Raven Press, New York, pp 221- 227 Christensen AK, Peacock KC 1980 Increase in Leydig cell number in testes of adult rats treated chronologically with an excess ofhCG. Bioi Reprod 22: 383-391 Cigorraga SB, Sorrell S, Bator J, Catt JK, Bufa ML 1980 Estrogen dependence of a gonadotropin-induced steroidogenic lesion in rat testicular Leydig cells. J Clin Invest 65: 699-705 Claus R, Schopper D, Hoang-vu C 1985 Contribution of individual compartments of the genital tract to oestrogen and testosterone concentrations in ejaculates of the boar. Acta Ecdocrinol109: 281-288 Conn MP, Tsuruhara T, Dufau ML, Catt KJ 1977 Isolation of highly purified Leydig cells. Endocrinology 101: 639-642 Damber JE, Bergh A 1980 Decreased testicular response to acute LH stimulation and increased intratesticular concentrations of oestradiol-1713 in the abdominal testes in cryptorchid rats. ActaEndocrinol95: 416-421 Davis JS, Weakland LL, Farese RV, West LA 1987 Luteinizing hormone increases 2 inositol triphosphate and cytosolic free Ca + in isolated bovine luteal cells. J Bioi Chern 262: 8515-8521 153 De Boer W, De Vries J, Muller E, van der Molen, IU. 1977 Kinetics of in vitro binding of oestradiol in subcellular fractions of testicular and uterine tissue; characterization of oestradiol binding in testicular nuclei. J Steroid Biochem 8: 859-865. Dhar JD, Setty Bs 1976 Epididymal response to exogenous testosterone in rats sterilized neonatally by estrogen. Endokrinologie 68: 14-21 Dintinger T, Gaillard JL, Zwain I, Bouchamide R, Silberzahn P 1989 Synthesis and aromatization of 19-Norandrogens in the stallion testis. J Steroid Biochem 32: 537-544 Dive C, Yoshida TM, Simpson DJ, Marrone BL 1992 Flow cytometric analysis of steroidogenic organelles in differentiating granulosa cells. Bioi Reprod 47: 520-527 Dahlman HG, Thorner J, Caron MG, Lefkowwitz RJ 1991 Model systems for the study of seven transmembrane segment receptor. Annu Rev Biochem 60: 653-688 Dym M, Raj HGM, Chemes HE 1977 Response of the testis to selective withdrawal of LH or FSH using antigonadotropic sera. In : Troen P, Nankin HR (eds) The Testis in Normal and Infertile Men. Raven Press, New York, pp 97-124 Edwards G, Robertson WR, Morris ID 1988 Characterization of the regenerated Leydig cell population of the rat after destruction by ethylene-1,2-dimethanesulphonate. J Endocr 117: 11-18 Edwards G, Fox BW, Jackson H, Morris ID 1988 Leydig cell cytotoXicity of putative male antifertility compounds related to ethane-1,2-dimethanesulphonate. In: Cooke BA, Sharpe RM (eds), The molecular and cellular endocrinology of the testis. Serono Symposia. Raven Press 154 Edwards G, Lendon R, Morris ID 1989 Leydig cell recovery following a second challenge with ethane-1,2-dimethanesulphonate is enhanced by short interval between cytotoxin adminstration to rats. J Endocrinol 123: 197-203 Eik-nes KB 1975 Biosynthesis and secretion of testicular steroids. Handbook ofPhysiology. pp 95-115. Eds Greep RO, Astwood EB, Hamilton DW & Geiger S. American Physiol Soc. Washington D.C. Ewing LL, Cochrari C, Zirkin B, Chubb CE 1980 Steroid synthesis and secretion by in vivo perfused rat, rabbit, dog, guinea pig and hamster testes. In: Steinberger A, Steringerger E ( eds) testicular Development, Structure and Function. Raven Press, New York, pp 117-127 Fevold HR, Lorence MC, McCarthy JL, Trant JM, Kagimoto M, Waterman MR, Mason ll 1989 rat P45017a. from testis: characterization of a full length eDNA encoding a unique steroid hydroxylase capable of catalyzing both "'4 and .1. -steroid-17,20- lyase reaction. Mol Endocrinol 3: 968-975 Furlow JD, Ahrens H, Mueller G, Gorski J. 1990 Antiserum to a synthetic peptide recognize native and denatured rat estrogen receptors. Endocrinology 127: 1028-1032. Gangnerau MN, Funkenstein B, Picon R 1982 LH/hCG receptors and stimulation of testosterone biosynthesis in the rat testis: changes during foetal development in vivo and in vitro. Mol Cell Endocrinol 28: 499-512 Gangnerau MN, Picon R 1987 Onset of steroidogenesis and differentiation of functional LH receptors in rat fetal testicular cultures. Arch Androl18: 215-224 155 Giber SJ, Hardy MP, Charnindrani-Handagama SML, Casella SJ 1992 Effects of insulin-like growth factor-! on androgen production by highly purified pubertal and adult rat Leydig cells. J Andrei 13: 125-13 0 George FW, Wilson JD 1988 Sex determination and differentiation. In: Knobil E, Neill D, Ewing LL, Greewald GS, Market CL, PfaffSW (eds) The Physiology of reproduction. Raven Press, New York, pp 67-118 Gondos B 1980 Development and differentiation of the testis and male reproductive tract. In: Sterinberger A, Sterinberger E (eds) Testicular Development, Structure, and Function. Raven Press, New York, pp 3-20 Greenspan P, Mayer EP, Fowler SD 1985 Nile red: a selective fluorescent stain for intracellular lipid droplets. J Cell Bioi 100: 965-973 Gudermann T, Birbaumer M, Bimbaumer L 1992 (a) Evidence of dual coupling of the murine luteinizing hormone receptor to adenylyl cyclase and phosphoinositide 2 breakdown and Ca + mobilization. J Bioi Chern 267: 4479-4488 Gudermann T, Nichols C, Levy FO, Bimbaumer M, Bimbaumer L 1992 (b) Ca2+ mobilization by the LH receptor expressed in Xenopus oocytes independent of 3', 5'-cyclic adenosine monophosphate formation: evidence for parallel activation of two signaling pathways. Mol Endocrol 6: 272-278 Habert R, Picon R 1982 Control of testicular steroidogenesis in foetal rat: effect of decapitation on testosterone and plasma luteinzing hormone-like activity. Acta Endocrinol99:466-473 Habert R, Picon R 1984 Testosterone, dihydrotestosterone and estradiol 17-~ levels 156 in maternal and in fetal testes in the rat. J Steroid Biochem 21: 193-198 Habert R, Picon R 1990 Attempts for identification of a chorionic gonadotropin-like bioactivity in the rat placental which stimulates the testosterone secretion of the fetal testis in vitro. Bioi Neonate 58: 24-31 Habert R, Brignaschi P 1991 Development changes in 'in vitro' testosterone production by dispersed Leydig cells during fetal life in rats. Arch Androl27: 65-71 Hall PF, Charponnier C, Nakamura M, Gabbiani G 1979 The role of microfilaments in the response of Leydig cells to luteinizing hormone. L Steroid Biochem 11: 1361-1366 Hall PF 1985 Trophic stimulation of steroidogenesis: in search for the elusive trigger. Recent Prog Horm Res 41: 1-59 Hall PF 1988 Testicular steroid synthesis: Organization and Regulation. In: Knobil E, Neil S (eds) The Physiology of Reproduction. Raven Press, New York, pp 975-988 Hall PF 1991 Cytochrome P450 C21,.,. : one enzyme with two actions: hydroxylase and lyase. J Steroid Biochen Mol Bio 40: 527-532 Handelsman DJ, Spoliviero JA, Scott CD, Baxter RC 1985 Identification of insulin like growth factor I and its receptor in the rat testis. Acta Endocrinol 109: 543- 549 Hanudoglu I 1992 Steroidogenic enzymes: structure, function and role in regulation of steroid hormone biosynthesis. J Steroid Biochem Mol Bioi 43: 779-804 Hardy MP, Zirkin BR, Ewing LL 1989 Kinetic studies on the development of the adult 157 population ofLeydig cells in testes of the pubertal rats. Endocrinology 124: 762-770 Hardy MP, Kelce WR, Klinefelter GR, Ewing LL 1990 Differentiation of Leydig cell precursors in vitro: a role for androgen. Endocrinology 127: 488-490 Henricks DM, Hoover JLB, Gimenez LW 1988 A study of the source of estradiol-17~ in the bull. Horm Metab Res 20: 494-497 Hou Q, Gorski J Estrogen receptor and progesterone receptor genes are expressed differentially in mouse embryos during preimplantation development. Proc Nat! Acad Sci 1993 90: 9460-9464 Huhtaniemi I, Pelliniemi LJ 1992 Fetal Leydig cells: cellular origin, morphology, life span, and special functional features. Proc Soc Exp Bioi Med 201: 125-140 Hutson JC 1990 Changes in the concentration and size of testicular macrophages during development. Bioi Reprod 43: 885-890 Jackson AE, O'Leary PC, Ayers MM, de Krestser DM 1986 The effects of ethylene dimethane sulphonates (EDS) on rat Leydig cells: evidence to support a connective tissue origin ofLeydig cells. Bioi Reprod 35: 425-43} Jackson CM 1966 Antifertility compounds in the male and female. Thomas, Springfield, IL Jackson CM, Jackson H 1984 Comparative protective actions of gonadotrophins and testosterone against the antispermatogenic action of ethane dimethanesulfonate. J Reprod Fertil 71: 393-401 Jennes L 1990 Prenatal development of gonadotropin-releasing hormone receptors in the rat anterior pituitary. Endocrinology 126: 942-947 Ji I, Ji T 1991 Exons 1-10 of the rat LH receptor encode a high affinity hormone binding site 158 and exon 11 encodes a G-protein modulation and a potential second hormone binding site. Endocrinology 128: 2648-2650 Johnson BH, Ewing LL 1971 Follicle-stimulating hormone and the regulation of testosterone secretion in rabbit testis. Science 173: 635-637 Johnson L, Neaves WB 1981 Age-related changes in the Leydig cell population, seminiferous ) tubules and sperm production in stallions. Bioi Repro 24:703-712 J ost A, Vigier M, Prepin J, Perchellet JP 1973 Studies on sex differentiation in mammals. Recent Prog Horm Res 29: 1-33 Jost A, Magre S, Agelopoulou R 1981 Early stages of testicular differentiation in the rat. Hum Genet 68: 59-63 J ost A, Periman S, Valentino 0, Castanier M, Scholler R, Magre S 1988 Experimental control of the differentiation of Leydig cells in the rat fetal testis. Proc Nat! Acad Sci 85: 8094-8097 Juenemann KP, Kogan BA, Abozeid MG 1986 Fertility in cryptorchidism: an experimental model. J Urol136: 214-216 Kafatos FC, Efstratiadis A, Forget BG, Weissman SM 1977 Molecular evolution of human and rabbit beta-globin mRNA Proc Nat! Acad Sci 74: 5618-5622 Keeney DS, Mendis-Handagama SMLC, Zirkin BR, Ewing LL 1988 Effect of long term deprivation of luteinizing hormone on Leydig cell volume, Leydig cell number and steroidogenic capacity of the rat testis. Endocrinology 123: 2906- 2915 Kelce WR, Zirkin BR, Ewing LL 1991 Immature rat Leydig cells are intrinsically less 159 sensitive than adult Leydig cells to ethane dimethanesulfonate. Toxicol Appl Pharmacol 111: 189-200 Kelce WR, Zirkin BR 1993 Mechanism by which ethane dimethanesulfonate kills ad u It rat Leydig cells: involvement of intracellular glutathione. Toxicol Appl Pharmacol 120: 80-88 Kelch RP, Jenner MR, Weinstein R, Kaplan SL, Grumbach MM 1972 Estradiol and testosterone secretion by human, simian and canine testes in males with hypogonadonism and in male pseudohermaphrodites with the feminizing testes J Clin Invest 51: 824-83 0 Kellogg DE, Sninsky JJ, Kwok S 1990 Quantitation of HIV-1 proviral DNA relative to cellular DNA by the polymerase chain reaction. Pro Nat! Acad Sci 87: 2725- 2729 Kerr JB, Sharpe RM 1985 (a) Follicle-stimulating hormone induction of Leydig cell maturation. Endocrinology 116: 2592-2604 Kerr JB, Donachie K, Rommerts FFG 1985 (b) Selective destruction and regeneration of rat Leydig cells in vivo. Cell Tissue Res 242: 145-156 Kerr JB, Donachle K 1986 Regeneration of Leydig cells in unilaterally cryptorchid rats: evidence for stimulation by local testicular factors. Cell Tiss Res 245: 649-655 Kerr JB, Bartlett JMS, Donachie K, Sharpe RM 1987 Origin of regenerating Leydig cells in the testis of the adult rat. Cell Tissue Res 249: 367-377 Kerr JB; Knell CM 1988 The fetal Leydig cells during the development of the fetal and postnatal rat testis. Development 103: 535-544 160 Ketelslegers JM, Hetzel WD, Sherins RH, Catt KJ 1978 Development changes in testicular gonadotropin receptors: plasma gonadotropins. Endocrinology 103: 212-222 Khan S, Teerds K, Darrington J 1992 Growth factor requirements for DNA synthesis by Leydig cells from the immature rat. Bioi Reprod 46: 335-341 Klinefelter GR, Laskey JW, Roberts NL 1991 In vitro/in vivo effects of ethane dimethanesulfonate on Leydig cells of adult rats. Toxicol Appl Pharmacol107: 460471 Koike S, Sakai M, Muramatsu M 1987 Molecular cloning and characterization of rat estrogen receptor eDNA. Nucleic Acids Res 15: 2499-2513 Kuopio T, Paranko J, Pelliniemi LJ 1989 Basement membrane and epithelial features offetal-type Leydig cells in rat and human testis. Differentiation 40: 198-206 Kuopio T, Tapanainen J, Pelliniemi LJ, Huhainiemi I 1989 Developmental stages of fetal-type Leydig cells in prepubertal rats. Development 107:213-220 Labrie F, Simard J, Luu-The V, Belanger A, Pelletier G 1992 Structure, function and tissue-specific gene expression of 3 f3-hydroxysteroid dehydrogenase/ 5-ene-4- ene isomerase enzymes in classical and peripheral intracrine steroidogenic tissues. J Steroid Biochem Mol Bioi 43: 805-820 Labrie F, Simard J, Luu-The V, Pelletier G, Belanger A 1992 Cloning, expression and 5 4 regulation of tissue-specific expression of 3 P-hydrqxysteroid dehydrogenaseh -.11 isomerase. In: Saez JM, Brownie AL, Capponi A, Chambaz EM, Mantero F (eds) Cellular and Molecular Biology of the Adrenal Cortex. INSERM/John Libbey Eurotext, Paris, vol 222: 89-109 LaPolt PS, Hia XC, Sincich C, Hsueh AJW 1991 Ligand-induced down-regulation of I6I testicular and ovarian luteinizing hormone (LH) receptor is preceded by tissue specific inhibition of alternatively processed LH receptor transcripts. Mol Endocrinol 5: 397-403 LecerfL, Rouiller-Fabre V, Levacher C, Gaurier C, Saez JM, Habert R I993 Stimulatory effects of follicle stimulating hormone on basal and LH stimidated testosterone secretion by the fetal rat testis in vitro. Endocrinology 133: 2313-23I8 Lee VWK, deKrester DM, Hudson B, Wang C I975 Variations in serum FSH, LH and testosterone levels in male rats from birth to sexual maturity. J Repord Fert 42: I2I-I26 Liebrman S, Prasad VVK I990 Heterodox notions on pathway of steroidogenesis. Endocr Rev II: 469-493 Lei ZM, Rao ChV, Komyer JL, Licht P, Hiatt ES I993 Novel expression of human chorionic gonadotropin/luteinizing hormone receptor gene in brain. Endocrinology 132: 2262-2270 Lejune H, Skalli M, Sanchez P, Avellet 0, Saez JM I993 Enhancement of testosterone secretion by normal adult human Leydig cells by coculture with enriched preparations of normal adult human Sertoli cells. Int J Androii6: 27-34 Leydig F I850 Zur Anatomie der Miinnlichen Geschlechtsorgane und Analdrusen der Siingethier. Z Wiss Zool 2: I-3 Lin SX, Yang F, Jin JZ, Breton R, Zhu DW, Luu-The V, Labrie F I992 Subunit identity of the dimeric I7~-hydroxysteroid dehydrogenase from human placenta. J Bioi Chern 267: 16I82-I6I87 162 Lin T, Chen GCC, Murono EP, Osterman J, Nankin HR 1982 Estradiol receptors of two populations of rat Leydig cells. Steroids 40: 53-63 Lin T, Haskell J, Vinson N, Terracio L 1986 Characterization of insulin and insulin like growth factor I receptors of purified Leydig cells and their role in steroidogenesis in primary culture: a comparative study. Endocrinology 119: 1641-1647 Lin T, Wang D, Calkins JH, Cuo H, Chi R, Housley PR 1990 Regulation of insulin like growth factor-I messenger ribonucleic acid expression in Leydig cells. Mol Cell Endocrinol73: 147-152 Loosfelt H, Misrahi M, Atger M, Salesse R, Vu Gai-Luu Thi MT, Joliver A, Guiochon Mantel A, Sar S, Jallal B, Gamier J, Milgrom E 1989 Cloning and sequencing of porcine LH-hCG receptor eDNA: variants lacking transmembrane domain. Science 245: 525-528 Lording DW, De Krester SM 1972 Comparative ultrastructural and histochemical studiesof the interstitial cells of the rat testis during fetal and postnatal development. J Reprod Fertil298: 261-269 Magre S, Jost A 1980 The initial phases of testicular organogenesis in the rat. Arch Anat Microsc Morpho! Exp 69: 297-318 Magre S, J ost A 1984 Dissociation between testicular organogenesis and endocrine cytodifferentiation ofSertoli cells. Proc Nat! Acad Sci 81: 7831-7834 Martel C, Rheaume E, Takagashi M, Trudel C, Couet J, Luu-The V, Simard J, Labrie F 1992 Distribution of 1713-hydroxysteroid dehydrogenase gene expression and activity in rat and human tissues. J Steroid Biochem Mol Bioi 41: 597-603 163 Mcfarland KC, Sprengel R, Phillips HS, Kohler M, Rosemblil N, Nikolics K, Segaloff SL, SeeburgPH 1989 Lutropinchorio-gonadotropin receptor: an unusual member of the G protein-coupled receptor family. Science 245:"494-499 McLaren A 1990 What makes a man a man? Nature 346: 216-217 McLaren A 1991 Sex determination-The making of male mice. Nature 351: 96 Mendis-hangdagama SKLC, Risbridger GP, De Kretser DM 1987 Morphometric analysis of the components of the neonatal and the adult rat testis interstitium. Int J Androl10: 525-534 Merchant-Larios H 1975 The onset of testicular differentiation in the rat: An ultrastructural study. Am J Anat 145: 319-330 Miyashita Y, Hirose T, Kishimoto S, Matsumoto K, Sato B 1990 Identification of unoccupied but transformed nuclear estrogen receptor in cultured mouse Leydig cells. J Steroid Biochem 35: 561-567 Moger WH 1980 In vitro inhibitory effect of estradiol on testosterone production. J Steroid Biochm 13:61-66 Molenaar R, de Rooij DG, Rommerts FFG, van der Molen HJ 1986 Repopulation of Leydig cells in mature rats after selective destruction of the existent Leydig cells with ethylene dimethane sulfonate is dependent on luteinizing hormone and not follicle-stimulating hormone. Endocrinology 118: 2546-2554 Moon TS, Hardy MH 1973 The early differentiation of the testis and interstitial cells in the fetal pig, and its duplication in organ culture. Am J Anat 138: 253-268 Morera AM, Cachet C, Keramidas M, Chauvin MA, De Peretti E, Benahnied M 1988 164 Direct regulating effects of transfonning growth factor-13 on the Leydig cell steroidogenesis in primary culture. J Steroid Biochem 30: 443-447 . Mori H, Christensen AK 1980 Morphometric analysis of Leydig ceils in the nonnal rat testis. J Ceil Bioi 84: 340-354 Morris ID, Phollips DM, Bardin CW 1986 Ethylene dimethanesulfonate destroys Leydig ceils in the rat testis. Endocrinology 118: 709-719 Muiianey BP, Skinner MK 1993 Transforming growth factor-13 (131, 132and 133) gene expression and action during pubertal development of the seminiferous tubule: potential role at the onset ofspennatogenesis. Mol Endocrinol 7: 67-76 Myers, RB, Abney TO 1988 The effect of reduced 0 2 and antioxidants on steroidogenic capacity of cultured rat Leydig ceils. J Steroid Biochem 31: 305-309 Myers RB, Abney TO 1990 Testosterone and androstanediol production by regenerating Leydig ceiis in the ethylene dimethane sulphonate-treated rat. Int J Androl 13: 4-16 Myers RB, Abney TO 1991 Interstitial ceil proliferation in the testis of the ethylene dimethane sulfonate-treated rat. Steroids 56: 91-96 Nagpal ML, Wang D, Calkins lli, Chang W 1991 Human chorionic gonadotropin up regulates insulin-like growth factor-I receptor gene expression of Leydig ceils. Endocrinology 129: 2820-2826 Nakhla AM, Mather JP, Janne OA, Bardin CW 1984 Estrogen and androgen receptors in. Sertoli, Leydig, myoid and epithelial ceils: effect of time in culture and ceil density. Endocrinology ll5: 121-128 Nakajin S, Shively J, Yuan PM, Hail PF 1981 Microsomal cytochrome P-450 from 165 neonatal pig testis: two enzymatic activities (17o:-hydroxylase and C17_20 lyase) associated with one protein. Biochemistry 20: 4037-4042 Nakajin S, Shinoda M, Hanui M, Hanui M, Shively JE, Hall PF 1984 The C21 steroid side-chain cleavage enzyme from porcine adrenal microsome: purification and characterization of the 17o:-hydroxylase C11-2o lyase cytochrome P-450. J Bioi Chern 259: 3971-3976 Namiki M, Kitamura M, Buczko E, Dufau ML 1988 Rat testis P450 17o: eDNA: the deduced amino acid sequence, expression, and secondary structural configuration. Biochem Biophys Res Comrnun 157: 705-712 Narbaitz R, Adler R 1967 Submicroscopical aspects in the defferentiation of rat fetal Leydig cells. Acta Physiol Latinoam 4: 286-291 Neuman I, Solano AR, Paz C, Mete P, Maciel FC, Lemos JR, Fermamdez HN, Podesta EJ 1991 The action of luteinizing hormone on the testis. J Steroid Biochem Molec Bioi 40: 441-451 Niemi M, Ikonen M 1961 Steroid-3 P-ol-dehydrogenase activity in foetal Leydig's cells. Nature 189: 592-593 Noumura T, Weisz J, Lloyd CW 1966 In vitro conversion of 7-3H progesterone to androgens by the rat testis during the second half of fetal life. Endocrinology 78: 245-253 Odell DW, Swerdloff RS 1975 The role of testicular sensitivity to gonadotropins in sexual maturation of the male rat. J Steroid Biochem 6: 853-858 Orth JM 1982 Proliferation of Sertoli cells in fetal and postnatal rats: a quantitative 166 autoradiographic study. Anat Rec 203: 458-492 Papadopoulos V, Carreau S, Szerman-Joly E, Drosdowsky MA 1986 Rat testis 17~ estradiol: Identification by gas chromatography-mass spectrometry and age related cellular distribution. J Steroid Biochem 24: 1211-1216 Papadopoulos V, Kamtchouing P, Drosdowsky MA, Carreau S 1987 Spent media from immature seminiferous tubules and Sertoli cells inhibit adult rat Leydig cell aromatase activity. Horm Metab Res 19: 62-64 Parvinen M, Nikula H, Huhtaniemi I 1984 Influence of rat seminiferous tubules on Leydig cell testosterone production in vitro. Mol Cell Endocrinol 37: 311-336 Patsavoudi E, Magre S, Castanier M, Scholler R, Jost A 1984 Dissociation between testicular morphogenesis and functional differentiation of Leydig cells. J Endocrinol 105:235-238 Payne AH, Downing JR, Wong KL 1980 Luteinizing hormone receptors and testosterone synthesis in two distinct populations ofLeydig cells. Endocrinology 106: 1424-1429 Payne AH, Perkins LM, Georgiou M, Quinn PG 1987 Intratesticular site of aromatase activity and possible function of testicular estradiol. Steroids 50: 435-448 Payne AH 1990 Hormonal regulation of cytochrome P450 enzymes, cholesterol side chain cleavage and 17a;-hydroxylase/C17-20 lyase in Leydig cells. Bioi Reprod 42:399-404 Payne AH, Sha LL 1991 Multiple mechanisms for regulation of 3~-hydroxysteroid 5 4 dehydrogenasel.o. -L>. -isomerase, 17a;-hydroxylase/C11•20 lyase cytochrome P450,arrl cholesterol side-chain cleavage cytochrome P450 messenger ribonucleic acid levels 167 in primary cultures of mouse Leydig cells. Endocrinology 1991 129: 1429-1435 Payne AH, Youngblood GL, Sha L, Burgos-Trinidad M, Hammond SH 1992 Hormonal regulation of steroidogenic enzyme gene expression in Leydig cells. J Steroid Biochem Mol Biol43: 895-906 Pelliniemi U, Niemi M 1969 Fine structure of the human foetal testis. I. The interstitial tissue. Z Zellforsch 99: 507-522 Peltoketo H, Isomaa V, Maentausta 0, Vihko R 1988 Complete amino acid sequence of human placental 17P-hydrosysteroid dehydrogenase deduced from eDNA. FEBS Lett 239: 73-77 Pollard I, Williamson S, Magre S 1990 Influence of caffeine administration during pregnancy on the early differentiation offetal rat ovaries and testes. J Dev Physiol 13: 59-65 Ponglikitmonglol M, Green S, Chambon P 1988 Genonic organization of the human oestrogen receptor gene, EMBO J 7: 3285-3388 Prince FP 1984 Ultrastructure of immature Leydig cells in the human prepubertal testis. Anat Rec 209: 165-176 Prins GS 1992 Neonatal estrogen exposure induces lobe-specific alterations in adult rat prostate androgen receptor expression. Endocrinology 130: 3703-3714 Privalle CT, Crivello n, Jefcoate CR 1983 Regulation of intramitochondrial cholesterol transfer to side-chain cleavage cytochrome P4,_in rat adrenal gland. Proc Nat!. Acad Sci 80: 702-706 Puris JL, Canick JA, Latif SA, Rsxenbaum JH, Hologgitas J, Menard RH 1973 Lifetime of microsomal cytochrome P 450 and steroidogenic enzymes in rat testis as influenced by 168 human chorionic gonadotropin. Arch Biochern Biophys 159: 39-49 Raeside JI, Renaud RL, Friendship RM 1988 Differences in arornarase activity between Leydig cells from the scrotal and abdominal testis in the naturally unilateral cryptorchid boar. Bioi Reprod 39: 566-572 Rebois RV 1982 Establishment of gonadotropin-responsive murine Leydig tumor cell line. J Cell Biol94: 70-76 Rebois RV, Fishman PH 1984 Down-regulation of gonadotropin receptors in a murine Leydig tumor cell line. J Bioi Chern 259: 3096-3101 Risbridger GP, Davies A 1994 Isolation of rat Leydig cells and· precursor forms after adrninstration of ethane dirnethane sulfonate. Am J Physiol266: E975-E979 Rivenson A, Ohmori T, Harnazaki M, Madden R 1981 Cell surface recognition: Spontaneous identification mouse Leydig cells by lymphocytes, rnacrophages and eosinophils. Cell Mol Bioi 27: 49-56 Romrnerts FFG, Hoogerbrugge JW 1987 In vitro effects of ethylene dirnethane sulfonate (EDS) on isolated Leydig cells. Annals New York Academy of Sciences 5 13: 401-404 Romrnerts' FFG, Cooke BA 1988 The mechanism of action of luteinizing hormone II. Transducing systems and biological effects. In: Cooke BA, King RJB, van der Molen HJ (eds) New Comprehensive Biochemistry: Hormones and Their Action. Elsevier, Amsterdam, vol2: 163-180 Rornrnerts FFG, Teerds KJ, Hoogerbrugge JW 1988 (b) In vitro effects of ethylene dirnethane sulfonate (EDS) on Leydig cells: inhibition of steroid production and cytotoxic effects are dependent on species and age of rat. Mol Cell Endocrinol 169 55: 87-94 Roosen-Runge EC, Anderson D 1959 The development of the interstitial cells in the testis of the albino rat. Acta Anal 37: 125-137 Russell ill, Clermont Y 1977 Degeneration of germ cells in normal hypophysectomized and hormone treated hypophysectomized rats. Anat Rec 187: 347-366 Russell ill, Corbin TJ, Ren HP, Amador A, Bartke A, Ghosh S 1992 Structural changes in rat Leydig cells posthypophysectomy: a morphometric and endocrine study. Endocrinology 131: 498-508 Saez JM, Haour F, Loras B, Sanchez P, Cathiard AM 1978 Oestrogen induced Leydig cell refractoriness to gonadotropin stimulation. Acta Endocrinol 89: 379-392 Saez JM, Avallet 0, Naville D, Perrard-Sapori MH, Chatelain PG 1989 Sertoli-Leydig cell communications. Ann NY Acad Sci 564:210-231 Saez JM 1994 Leydig cells: endocrine, paracrine, and autocrine regulation. Endoc Rev 15: 574-625 Sambrook J, Fritsch EF, Maniatis T 1989 Molecular Cloning. Cold Spring Harbor · -Laboratory Press, New York Sanchez-Yagiie J, Rodriguez MC, SegaloffDL, Ascoli M 1991 Truncation of the cytoplasmic tail ofthe lutropin choriogonadotropin receptor prevents agonist-induced uncoupling. J Bioi Chern 267: 7217-7220 Sato B, Maeda Y, Nakao M, Noma K, Kishimoto S, Matsumoto K 1985 Multiple estrogen binding sites in malignant ·mouse Leydig cells and their role in cell proliferation. Eur J Cancer Clin Oncol21: 199-205 170 Scatchard G 1949 The attraction of proteins for small molecules and ions. Ann N.Y. Acad Sci 51: 660-672 SegaloffDL, Sprengel R, Niklics K, Ascoli M 1990 The structure of the lutropin/choriogonadotropin receptor. Recent Prog Horm Res 46: 261-303 Segaloff DL, Ascoli M 1993 The lutropin/choriogonadotropin receptor... 4 years later. Endoc Rev 14: 324-347 Shan LX, Hardy MP 1992 Developmental changes in levels of luteinizing hormone receptor and androgen receptor in rat Leydig cells. Endocrinology 131: 1107- 1114 Sharpe RM, Donachie K, Cooper I 1988 Re-evaluation of the intratesticular level of testosterone required for quantitative maintenance of spermatogenesis in the rat. J Endocrinol 117: 19-26 Sharpe RM 1990 Intratesticular control of steroidogenesis. Clin Endocrinol (Oxf) 33: 787-807 Simpson ER, Waterman MR 1988 Regulation of the synthesis of steroidogenic enzymes in adrenal cortical cells by ACTH. Annu Rev Physiol 50: 427-440 Skinner MK 1991 Cell-cell interactions in the testis. Endocr Rev 12: 45-72 Smith EP, Svoboda ME, Van WYK JJ, Kierzenbaum AL, Tress LL 1987 Partial characterization of a somatomedin-like peptide from the medium of cultured rat S ertoli cells. Endocrinology 120: 186-193 Sokka T, Hiimalainen T, Huhtaniemi I 1992 Functional LH receptor appears in the neonatal rat ovary after changes in the alternative splicing pattern of the LH 171 receptor mRNA Endocrinology 130: 1738-1740 Sprangers SA, Brenner RM, Bethea CL 1989 Estrogen and progestin receptor immunocytochemistry in lactotropes versus gonadotropes of monkey pituitary c e I 1 culture. Endocrinology 124: 1462-1470 Stocco DM, Teerds KJ, Van Noort M, Rommerts FFG 1990 Effects of hypophysectomy and human chorionic gonadotropin on Leydig cell function in mature rats. Endocrinology 126: 367-375 Sugiura K, Stock CC 1957 Studies in a tumor spectrum. V. The effect of esters of sulfonic acids on the growth of a variety of mouse and rat tumors. Cancer 10: 596-605 SwerdloffRS, Walsh PC, Odell WD 1971 Serum LH and FSH during sexual maturation in the male rat: effect of castration and cryptorchidism. Endocrinology 88: 120-128 Swerdloff RS, Heber D 1981 Endocrine control of testicular function from birth to puberty. In: Burger H, De Kretser D (eds) The testis. Raven Press, New York, pp 108-126 Tahja KM 1989 Local control mechanisms in the testis. Int J Dev Biol33: 141-148 Tapanainen J, Kupoio T, Pelliniemi U, Huhtaniemi I 1984 Rat testicular endogenous steroids and number ofLeydig cells between the fetal period and sexual maturity. Bioi Reprod 31: 1027-1035 Tapanainen JS, Tilly JL, Vihko KK, Hsueh AJW 1993 Hormonal control of apoptotic cell death in the testis: gonadotropins and androgens as testicular cell survival factors. Mol Endocrinol 7: 643-650 172 Teerds KJ, Clossert J, Rommerts FFG, De Rooij DG, Stocco DM, Colenbrander B, Wensing GJG, Hennen G 1989 (b) The effects of pure FSH and LH preparations on the number and function of Leydig cells in immature hypophysectomized rats. J Endocrine! 120: 97-106 Teerds KJ, de Rooij DG, Rommerts FFG, van den Hurk R, Wensing CJG 1989 (c) Pfoliferation and differentiation of possible Leydig cell precursors after destruction of the existing Leydig cells with ethane dimethyl sulphonate: the role of LH/human chorionic gonadotrophin. J Endocrine! 122: 689-696 Teerds KJ, de Rooij DG, Ronnerts FFG, Wensing CJG 1990 Development of a new Leydig cell population after the destruction of existing Leydig cells by ethane dimethane sulphonate in rats: an autoradiographic study. J Endocrine! 126: 229-236 Teerds KJ, Derrington JH 1993 Localization of transforming growth factor-~! and ~2 during testicular development in the rat. Bioi Reprod 48: 40-45 Tena-Sampere M, Zhang FP, Huhtaniemi I 1994 Persistent expression of a truncated form of the luteinizing hormone receptor messenger ribonucleic acid in the rat testis after selective Leydig cell destruction by ethylene dimethane sulfonate. Endocrinology 13 5: 1018-I 024 Tsai-Morris CH, Knox GF, Dufau ML 1987 Acquisition of hormone-mediated mechanisms regulating testicular steroidogenesis during development. Ann NY Acad Sci 513: 40- 57 Tsai-Morris CH, Buczko E, Wang W, Xie XZ, Dufau ML 1991 Structural organization of 173 the rat luteinizing hormone (LH) receptor gene. J Bioi Chern 266: 11355-11359 Vakkadares LE, Payne AH 1979 Acute stimulation of aromatization in Leydig cells by human chorionic gonadotropin in vitro. Proc Natl Acad Sci USA 76: 4460- 4463 Van Noort M, Rommerts FFC, Van Amerougen A, Wrirtz KWA 1988 Localization and hormonal regulation of the nonspecific transfer protein (sterol carrier protein 2) in the rat testis. J Endocrinol109: R13-R16 Verhoeven G, Caileau J 1985 A factor in spent media from Sertoli cell -enriched cultures that stimulates steroidogenesis in Leydig cells. Mol Cell Endocrinol40: 57-68 Verhoeven G, Caileau J 1987 A Leydig cell stimulatory factor produced by human testicular tubules. Mol Cell Endocrinol49: 137-148 Verhoeven G, Caileau J 1990 Influence of co culture with Sertoli cells on steroidogenesis in immature rat Leydig cells. Mol Cell Endocrinol 71: 239-251 Vigier B, Forest MG, Eychenne B, Bezard J, Garrigou 0, Robel P, Josso N 1990 Anti Miillerian hormone produces endocrine sex reversal offetal ovaries. Proc Natl Acad Sci 86: 3684-3688 Vihko KK, Lapolt P, Nishimori K, Hsueh AJW 1991 Stimulatory effects of recombinant follicle-stimulating hormone on Leydig cell function and spermatogenesis in immature hypophysectomized rats. Endocrinology 129: 1926-1932 Vreeburg JTM, Ooms MP, Rommerts FFG, Teerds KJ 1988 Functional properties of developing rat Leydig cell~ after treatment with ethylene dimethanesulphonate (EDS). J Reprod Fert 84: 63-69 174 Wang AM, Doyle MV, Mark DF 1989 Quantitation of mRNA by the polymerase chain reaction. Proc Nat! Acad Sci 86: 9717-9721 Wang H, Ascoli M, SegaloffDL 1991 Multiple luteinizing hormone/chorionic gonadotropin receptor messenger ribonucleic acid transcripts. Endocrinology 129: 133-138 Warren DW, Dufau ML, Catt KJ 1984 Hormonal regulation of gonadotropin receptors and steroidogenesis in cultured fetal rat testes. Science 218: 375-377 Warren DW, Huhtaniemi IT, Tapanainen J, Dufau MJ, Catt KJ 1984 Ontogeny of gonadotropin receptors in the fetal and neonatal rat testis. Endocrinol 114: 470-476 Wartenberg H 1981 Differentiation and development of the testis. In: Bruger H, de Kretser D (eds) The testis, Comprehensive Endocrinology, New York: Raven, pp 39-80 Wartenberg H 1989 Differentiation and development of the testis. In: Burger H, de Kretser D (eds) The Testis, ed 2. Raven Press, New York, pp 67-118 Weuestem JJSM, Smals AGH, Hofinan JA, Kloppenborg PWC, Benraad TJ 1987 Early time sequence in pregnenolone metabolism to testosterone inhomogenates of human and rat testis. Endocrinology 120: 1909-1913 Wu T -CJ, Wang L, Wan Y -JY 1993 Detection of estrogen receptor messenger ribonucleic acid in human oocytes and cumulus-oocyte complexes using reverse transcriptase polymerase chain reaction. Fert Steril 59: 54-59 Wurzel JM, Curatola LM, Gurr JA, Goldsmidt AM, Kourides IA 1983 The luteotropic activity of rat placenta is not due to a chorionic gonadotropin. Endocrinology 113: 1854-1857 175 Xie YB, Wang H, SegloffDL 1990 Extracellular domain oflutropin/choriogonadotropin receptor expressed in transfected cells binds choriogonadotropin with high affinity. J Bioj Chern 265: 21411-21414 Yamamoto R, Kallen CB, Babalola GO, Rebbert H, Biiiherimer JT, Stauss JF 1991 Cloning and expression of a eDNA encoding human sterol carrier protein 2. Proc Nat! Acad Sci 88: 463-467 Y annaihara T, Troen P 1972 Studies of the human testis. Biosynthetic pathways from androgen formation in human testicular tissue in vitro. J Clin Endocrinol Metab 34: 783-792