Specificity of steroid hormone-induced sex determination
in
a
turtle
T. Wibbels and D. Crews Institute of Reproductive Biology, Department of Zoology, University of Texas at Austin, Austin, Texas 78712, U.S.A.
received
26 June 1991
ABSTRACT
The specificity of steroid hormone-induced sex determination was investigated in the red-eared slider, Trachemys scripta, a turtle with temperaturedependent sex determination. All eggs were incubated at either a female-producing temperature (31 \s=deg\C)or a male-producing temperature (26 \s=deg\C)and received control or experimental treatments at stage 17\p=n-\18 of embryonic development. A variety of treatments induced female sex determination at the male-
producing temperature. Oestradiol-17\g=b\, diethylstilboestrol (DES) (an oestrogen agonist) and norethindrone (NET) (a progestin with reputed oestrogenic as well as anti-oestrogenic properties) were the most effective in inducing female sex determination. Other reputed oestrogen antagonists/partial agonists (i.e. tamoxifen, nafoxidine and clomiphene citrate) were also capable of inducing female sex determination, but to a lesser extent. A high dosage of testosterone resulted in the production of some females (7 of 15 hatchlings) whereas dihydrotestosterone had no detectable effect on sex determination. This latter finding suggests that testosterone could be acting via aromatization to oestradiol-17\g=b\. A few females resulted from eggs that had been treated with
inhibitor, 1,4,6-androstatrien-3,17-dione (ATD) (3 of 97), the antiandrogen hydroxyflutamide (1 of 55) and progesterone (3 of 36), suggesting the possibilities of non-specific effects of these compounds when used in large dosages. Alternatively, metabolites of these compounds may be oestrogenic. Collectively, aromatase
the results at the male-producing temperature are consistent with the hypothesis that steroid-induced female sex determination is mediated via an oestrogen\x=req-\
specific receptor.
INTRODUCTION
In many
reptiles, sex is determined by the temperature the egg is incubated (i.e. temperature-
at which
With the exception of a single male turtle (produced from an egg treated with 4-androsten-4-ol-3,17-dione (4-OHA)), the various treatments were not able to induce male sex determination at a female-producing temperature. This suggests that either (i) sex determination has little or no sensitivity to steroid hormones at a female-producing temperature, (ii) the effects of steroid hormones are overridden by the effect of the female-producing temperature or (iii) the effects of steroid hormones are similar to the effects of the female-producing temperature and are thus masked. The effects of the treatments on M\l=u"\llerianduct differentiation and development were also recorded. NET exerted distinct and equivalent effects at both male- and female-producing temperatures, with many of the females exhibiting a hypertrophied cranial portion of the M\l=u"\llerianducts but lacking the caudal portion (i.e. M\l=u"\llerianduct agenesia). A similar effect was shown for the majority of females produced from eggs treated with DES and for one turtle produced from an egg treated with oestradiol-17\g=b\at the female-producing temperature. Certain dosages of oestradiol-17\g=b\and oestradiol-3-benzoate resulted in females with full length hypertrophied M\l=u"\llerianducts. Tamoxifen, nafoxidine and clomiphene citrate treatments did not stimulate M\l=u"\llerianduct agenesia, but they did appear to act as oestrogen agonists in blocking M\l=u"\llerianduct regression in some males. Further, although clomiphene citrate did not stimulate agenesia, it did produce hypertrophic M\l=u"\llerianducts in some males and females. The latter result is consistent with the hypothesis that different receptors could be involved in the differentiation versus hypertrophy of the M\l=u"\llerianducts. Journal of Endocrinology (1992) 133, 121\p=n-\129
dependent sex determination or TSD) (Bull, 1980; Raynaud & Pieau, 1985; Ewert & Nelson, 1991). The mechanism by which temperature affects sex determi¬ nation is not known, but it has been suggested that the
determination pathway may include the produc¬ tion of steroid hormones (Raynaud & Pieau, 1985; Crews, Wibbels & Gutzke, 1989). Several studies have shown that the effect of male-producing temperatures can be overriden by the administration of exogenous steroid hormones (reviewed by Raynaud & Pieau, 1985; also see Gutzke & Bull, 1986; Bull, Gutzke & Crews, 1988; Crews et al. 1989; Crews, Bull & Wibbels, 1991). Further, temperature and exogenous oestrogen exert a synergistic effect on sex deter¬ mination, suggesting that steroid hormones and sex
temperature may
pathway (Wibbels,
act on a common
Bull & Crews,
physiological
1991a). As such,
steroid hormones represent an avenue for probing the basis of sex determination in species with TSD. The present study investigates the speci¬ ficity of steroid hormone-induced sex determination in the red-eared slider, Trachemys scripta, a turtle with
physiological TSD.
MATERIALS AND METHODS
Freshly laid eggs were obtained commercially (Robert
Kliebert, Hammond, LA, U.S.A.). After transport to
laboratory, they were placed in containers with moistened vermiculite (vermiculite/water, 1:2) and placed in incubators set at either 31 °C or 26 °C. Pre¬ vious studies indicated that a continuous incubation temperature of 31 °C produces all female hatchlings whereas an incubation temperature of 26 °C produces all male hatchlings (Bull, Vogt & McCoy, 1982; Crews our
al. 1991; Wibbels, Bull & Crews, \99\b). Embryonic development was monitored by candling eggs and by dissecting two to four eggs approximately twice a week to verify specific developmental stages, based on criteria described by Yntema (1968). When embryos reached stage 17-18, the approximate midpoint in the temperature-sensitive period in this species (Wibbels et al. 1991 b), the eggs were randomized into control and experimental groups. Eggs from experimental et
a single treatment of a specific ligand (i.e. hormone, agonist or antagonist) suspended in 5 pi 95% ethanol. Oestradiol-17ß, diethylstilboestrol (DES), norethindrone (NET), nafoxidine, tamoxifen, clomiphene citrate, testosterone, 5a-androstan-17ßol-3-one (DHT), progesterone, corticosterone, 6 methyl-17a-hydroxyprogesterone (depo-provera) and 2-methyl-1,2-di-3-pyridyl-1 -propanone (metyrapone) were obtained from Sigma (St Louis, MO, U.S.A.), l,4,6-androstatrien-3,17-dione (ATD) from Steraloids (Wilton, NH, U.S.A.), hydroxyflutamide from Schering Plough (Kenilworth, NJ, U.S.A.), 4androsten-4-ol-3,17-dione (4-OHA) from Dr A. Brodie at University of Maryland (Baltimore, MD, U.S.A.),
groups received
cyanoketone from Sterling-Winthrop Research Insti¬ tute (Rensselaer, NY, U.S.A.), and R5020 and
RU38486 (RU486) from Roussel Uclaf (Romainville, Seine-St-Denis, France). The dosages chosen for each
ligand were based on previous studies with turtles (Gutzke & Bull, 1986; Bull et al. 1988; Crews et al. 1989, 1991) and on the availability and solubility of the specific agents. Eggs in control groups received a single treatment consisting of 5 pi 95% ethanol. All treatments were either injected into the egg or applied topically to the vascularized portion of the upper shell (Crews et al. 1991). Both techniques have been shown to be effective means of delivering steroid hormones to embryos in T. scripta eggs (Crews et al. 1991). Groups of approximately 30 eggs were used for each injected dosage group: topical administration groups consisted of approximately 15 eggs. After receiving treatments, all eggs were placed back into their respective incubators (31 °C or 26 °C) until they hatched. Turtles were killed approximately 2—4 weeks after hatching. Gonadal sex and develop¬ mental status of the Müllerian ducts were assessed by examination of the reproductive tracts under a dissec¬ tion microscope. The gonads of hatchling T. scripta are relatively well differentiated and, with rare excep¬ tion, appear distinctly testicular or ovarian when viewed under a dissection microscope (Crews et al. 1991; Wibbels et al. 1991e). Ovaries are long and flat whereas testes are shorter, more round, and have visible seminiferous tubules (see Crews et al. 1991 ). The developmental status of both the cranial and caudal halves of the Müllerian ducts was examined and scored as either absent, regressed but visible, normal (as in a typical female hatchling), or hypertrophied.
RESULTS
Gonadal sex and developmental status of Müllerian ducts for control and experimental turtles are shown in Tables 1-3. As shown previously for oestrogens (Crews et al. 1991), both modes of hormone adminis¬ tration were capable of delivering at least some of the treatments in this study. However, the effectiveness of some treatments varied with the mode of adminis¬ tration. Mortality of eggs after treatment varied widely. In general, during the first few weeks after treatment there was high mortality of injected eggs and low mortality of eggs treated topically. In addition, high mortality that did not appear to be related to hormone administration occurred in some groups near the time of pipping. We speculate that the humid environment created by the moist vermiculite together with the fluid from the egg provided an optimal environment for infectious organisms.
1. Effects of oestrogen and putative agonists and antagonists on sex determination and Müllerian duct development of red-eared sliders, Trachemys scripta, at a male-producing temperature (26 °C). Control treatments consisted of 5 pi 95% ethanol table
Dose*
Survival0
(Hg)
10 10 1
Diethylstilboestrol
Sat. Sat. 100 10
Status of Müllerian ducts'
(%)
Males
I
19 15
0 0
95 50
A A
I
0 0 5
18 11 11
90 37 80
11 females
0 0 0 2
6 3 10 6
40 10 75 40
Treatment Control
Oestradiol-17ß
Intersex
Females
Adm."
I
, 4 males A, 1 male R
=
=
=
H, 2 females HC,ACD HCACD HC,ACD 1 male N, 1 male R, 2 females A, 4 females HC,ACD 17 females HC,ACD, 1 female H, 4 females
=
=
=
=
=
=
10
63
=
1 female Sat. Sat.
Norethindrone
I
9 12
60 43
13 13 13
87 43 87
10
70
=
=
A
1 female H, 8 females HC,ACD 1 male A, 7 females NC.ACD, females HC,ACD all HC,ACD 10 females HC,ACD, 3 females A 4 females N, 7 females H, 2 females HC,ACD 9 males A, 9 females N, 1 female NC, ACD, 2 intersex =
=
=
=
=
100 100 10
I
=
=
=
=
=
=
10
=
=
=
Clomiphene citrate
100 100 10 10
Nafoxidine
I
13 2 14 19
I
10 14
I
100 100
I 11 0 0
=
93 43 93 63
13 males H, 1 female 1 male H, 1 male N, 11 females all A 16 males A, 3 males R
87 57
10 males N, 3 females 3 males A, 1 male R, 10 males 3 females 7 males A, 2 males R, 5 males female
=
=
=
=
=
=
=
=
=
=
=
=
=
N,
=
10
14
100
=
=
=
N,
=
Sat. 100
Tamoxifen
15 12
100 100
10
60 100
2 males A, 9 males R, 4 males 1 male A, 3 males R, 8 males N, 2 females 6 males R, 12 males 1 male A, 5 males R, 2 males N, 2 males H, 3 females N, 2 females 4 males A, 4 males R, 7 males N, 5 females =
=
=
=
=
=
=
100 10
I
=
=
=
=
=
=
=
=
10
15
67
=
=
=
=
"Dosages delivered in 5 µ ethanol. Sat. saturated solution. bAdministration techniques: topical administration, I injection. 'Percentage of embryos in each group which survived to hatching. All of the eggs in four treatment groups (1 \ig oestradiol- 17ß injected, 100 µg diethylstilboestrol injected, 10 µg nafoxidine injected and saturated tamoxifen injected) died and, therefore, these groups are not listed in the table. dN typical Müllerian ducts of control female hatchling (i.e. 'normal'), R Müllerian ducts visible but regressed, A Müllerian ducts absent, Müllerian ducts hypertrophied, HC cranial region of Müllerian ducts hypertrophied, NC cranial regions of Müllerian ducts normal, ACD caudal region of Müllerian ducts absent (see text for a more detailed description). =
=
—
=
=
=
Decreasing the ratio of water to vermiculite will hope¬ fully alleviate this problem in future studies. Previous studies indicate that differential mortality is not the means by which temperature or oestrogens produce biased sex ratio (Bull & Vogt, 1979; Bull, 1987; Crews et al. 1991).
=
=
=
=
Male-producing temperature Turtles
receiving control treatments all developed Oestrogen and oestrogen-related compounds were the most effective agents inducing female sex determination including both reputed agonist and testes.
The androgen DHT and the aromatase inhibitor 4-OHA did not appear to affect sex determi¬ nation (Table 2). However, testosterone and to a lesser extent the androgen antagonist hydroxyflutamide and the aromatase inhibitor ATD resulted in the production of some females. The progesterone antagonist RU486 and the progesterone agonist R5020 did not appear to affect sex determination, but the high dosage of progesterone (250 pg) resulted in the production of three females out of 36 hatchlings
antagonists.
(Table 2). Corticosterone antagonist metyrapone did determination (Table 2).
and the corticosteroid not appear to affect sex
Turtles that received control treatments lacked Müllerian ducts, whereas the development of Müllerian ducts in turtles treated with oestrogens and oestrogen-related compounds was highly variable (Table 1). Depending on the specific treatment, the status of the Müllerian ducts ranged from absent to hypertrophied. Additionally, NET and DES prevented
2. Effects of a variety of steroid hormones and putative agonists and antagonists on sex determination and Müllerian duct development of red-eared sliders, Trachemys scripta, at a male-producing temperature (26 °C). Control treatments consisted of 5 pi 95% ethanol (see Table 1) table
Doseb
(Pg) Androgens and putative androgen antagonist Testosterone
DHTa
Hydroxyflutamide
100 100 10 10
Sat. Sat. 100 100 10 10 Sat. 100 100
Survival11 Adm.c
Males
I
13
1
15 19
4-OHA"
Progesterone and putative agonist and antagonist Progesterone R5020
RU486
Sat. Sat. 100 100 10 10 100 100 10
Sat. Sat. 10
100 10 10
(%)
Status of Müllerian ducts'
(I
100 43 100 63
7 males A A A
0 0
10
o
8 10
0 0
I
0
I
14 13 25
I
13 14 12
0 0 0
15
0
13 13 14
0 0 0 0
I
I
0 o
75 27 75 47 87 83 87 100 40 100
A 14 males A A
87 43 100 87 93 53
8 males A 14 males A A 14 males
15 19 20
100 63 67
A A
1
12 5
12 males 5 males
16
80 20 100
6 7 2
40 47 7
13 6 12
87 20 80
133, 121-129
19 males
A, 1 female
=
=
I I
1
=
A A A A A A
1
0 0
A, 1 male R, 7 females
=
26 14 14
100 100 10
Journal of Endocrinology (1992)
Intersex
I
10
Putative aromatase inhibitors ATD"
Females
=
A, 5 males R =
=
A, 1 female
=
A, 2 females
=
A, 1 male
=
=
=
=
=
, 2 females , 1 female =
=
=
table
2. Continued Survival11
Dose'
(µ«)
Intersex
Females
Status of Müllerian ducts'
(%)
Adm.'
Males
T
14 12
93 40
A A
12 16 14 25 12 23
80
A A 13 males = A, 1 male A A A
Corticosterone and
putative antagonists Corticosterone
Metyrapone
10 10
I
Sat. Sat. 100 100 10
I I I
10
Hydroxysteroid dehydrogenase inhibitor
Cyanoketone
100 100 10 10
53
93 83 80 77
15 17 15 14
I I
0 0 0 0
100 57 100 46
=
R
A
A A A
dihydroxytestosterone, ATD 1,4,6-androstatrien-3,17-dione, 4-OHA 4-androsten-4-ol-3,17-dione.
aDHT
=
hDosages were delivered in 5 µ
ethanol. Sat.
=
saturated solution. 'Administration techniques: topical administration, I injection. dPercentage of embryos in each group which survived to hatching. All of the eggs in four treatment groups (saturate hydroxyflutamide injected, 10 µg hydroxyflutamide injected, 10 µg 4-OHA topically applied, 10 µg progesterone injected, 100 µg R5020 injected and 10 µg RU486 injected) died and, therefore, these groups are not listed in the table. CN typical Müllerian ducts of control female hatchling (i.e. -normal'), R Müllerian ducts visible but regressed, A Müllerian ducts absent. —
=
=
=
=
=
the differentiation of the caudal portion of the Müllerian ducts in some females (i.e. agenesia of the Müllerian ducts).
Female-producing temperature With a single exception, all individuals in all groups had ovaries (Table 3). The one hatchling with testes (confirmed by histological analysis) received a high dosage (lOOpg) of 4-OHA, an aromatase inhibitor; the other ten individuals ofthat group showed typical
female development. All treatments except oestradiol- 17ß, oestradiol-3benzoate (EB) and NET resulted in hatchlings with oviducts similar to those of the control females. The Müllerian ducts of turtles treated with oestradiol17ß, EB and NET were variable. The higher dosages of oestradiol-17ß and NET (lOOpg and saturated) resulted in some turtles in which the cranial portion of the Müllerian ducts was hypertrophied, but the caudal portion was absent. Lower dosages of these treat¬ ments resulted in some individuals with Müllerian ducts similar to those of individuals that received the higher dosages, whereas others had either normal Müllerian ducts, full length Müllerian ducts which were hypertrophied, or normal cranial portions of the Müllerian ducts, but lacking the caudal portion. The Müllerian ducts of individuals treated with EB appeared hypertrophied or normal.
=
DISCUSSION
Gonadal sex at
male-producing temperature
Past studies have shown that exogenous oestrogens, and to a lesser extent testosterone, can induce female sex determination at a male-producing incubation tem¬ perature in reptiles with TSD (Pieau, 1974; Gutzke & Bull, 1986; Bull et al. 1988; Crews et al. 1989, 1991; Wibbels et al. 1991 ). It has been hypothesized that steroid-induced feminization is an oestrogen-specific event and the feminization caused by exogenous testos¬ terone treatments acts through aromatization in the egg (Raynaud & Pieau, 1985; Crews et al. 1989). Consistent with this hypothesis, oestradiol-17ß and DES effectively induced female sex determination in the present study (Table 1). Additionally, NET, a synthetic progestin (Edgren & Sturtevant, 1976; Mauvais-Jarvis, 1983) that has been reported to be an oestrogen antagonist and agonist as well as an irrever¬ sible aromatase inhibitor (Aldercreutz & Tenhunen, 1970; Perez-Palacios, Chavez, Escobar et al. 1981; Osawa, Yarborough & Osawa, 1982; Hutson, Donahoe & MacLaughlin, 1985), was also very effec¬ tive in feminizing gonads. NET has additionally been shown to feminize partially the gonads of male chickens (Hutson et al. 1985). Tamoxifen, nafoxidine and clomiphene citrate were not as potent as oestradiol-17ß, DES and NET, but were capable of
3. Effects of steroids and reputed steroid agonist antagonists on sex determination and Müllerian duct development at a female-producing temperature (31 °C). Control treatments consisted of 5 pi 95% ethanol
table
Dose*
Survival11
(µß)
Males
Adm.'
Females 14 15
93 50
10000 1000 100
12 5
80
Sat. 10000 1000
7
Control Control
1
Oestrogens and putative antagonists/partial agonists Oestradiol-17ß Oestradiol-3-benzoate
II
6 10
100 001
9 14 12 15 14
Nafoxidine
10000 1000 100 010
15 14 12 10
Tamoxifen
Sat. 10000 1000 100 0-10 001
Norethindrone
Sat. 10000 1000
1
14
10 19 10 15 13
Status of Müllerian ducts'
(%)_
HC,ACD 1=N,3 H, 1=HC,ACD
33 73
=
23 20 33
HC.ACD HCACD 3 N, 1 =H, 1 =HC, 7 HC,ACD
30 47 40
=
93 93 100 93 80 67
13
=
=
N, 1=NRT,ALT
47 33 63
33 100 S7
Androgen and aromatase
inhibitors
Dihydrotestosterone
10000 5000 1000
5
ATD"
Sat. 10000 1000 100 010
11 21 9 13 14
87 93
10000 1000
10 11
67 73
Sat. 10000 1000
4 11 7
13 37
23
1000 010 001
10 14 6
67 93 40
Sat. 10000 1000
14 7 4
47 23 13
4-OHA'
Progestogens Progesterone
Depo-provera
3p-Hydroxysteroid dehydrogenase Cyanoketone "ATD
=
27 27 17
37 70
females
=
, male R =
1,4,6-androstatrien-3,17-dione, 4-OHA 4-androsten-4-ol-3,17-dione. =
bDosages were delivered in 5 µ
ethanol. Sat.
saturated solution. 'Administration techniques: topical administration, I injection. All injections at this temperature were conducted during the first year of this study, whereas topical administration was used during the second year to decrease mortality. ^Percentage of each treatment group which survived to hatching. All of the eggs in the norethindrone (010 µg topical application) group died and they are therefore not listed. 'N typical Müllerian ducts of control female hatchling (i.e. 'normal'), R Müllerian ducts visible but regressed, A Müllerian ducts absent, Müllerian ducts hypertrophied, HC cranial region of Müllerian ducts hypertrophied, ACD caudal region of Müllerian ducts absent (see the text for a more detailed description), NRT/ALT normal right Müllerian duct and absent left Müllerian duct. =
=
=
=
=
=
=
=
=
=
inducing female sex determination in some embryos (Table 1). All three of these agents are triphenylethylene derivatives and are reputed oestrogen antagonists/partial agonists (reviewed by Lerner & Jordan, 1990; Jordan & Murphy, 1990). The current findings indicate that these putative anti-oestrogens or their metabolites are capable of mimicking the feminizing effects of oestrogen in this reptilian system. Similarly, tamoxifen was recently reported to induce female sex determination in alligator embryos incubated at a male-producing temperature (Lance & Bogart, 1991). One of the high dosages of testosterone (lOOpg applied topically) resulted in female development in some embryos, whereas all individuals that received the same dosage by injection or the lower dosage by
either administration method were male. Adminis¬ tration of the lower dosage of testosterone (10 pg) by both administration methods produced only males. Past studies which have successfully produced females at male-producing temperatures have used relatively high dosages ( > 200 pg) of testosterone propionate (Pieau, 1974; Gutzke & Bull, 1986) in comparison with the dosage of oestrogen required to induce female sex determination (01 pg) (Crews et al. 1991). These findings are consistent with the hypothesis that testosterone may act via aromatization to oestradiol17ß. Contrary to the hypothesis of oestrogen-specific induction of female sex determination, a few females were produced from eggs treated with the androgen antagonist hydroxyflutamide (1 of 55 hatchlings), the aromatase inhibitor ATD (3 of 97 hatchlings) and progesterone (3 of 36 hatchlings). However, it is possible that the higher dosages of these treatments were metabolized into compounds with oestrogenic
activity. Collectively,
these results indicate that oestrogens
oestrogen-related compounds represent potent compounds for inducing female sex determination. It is presently unclear, however, exactly how this feminizing effect is mediated. One possibility is via a 'classic' high-affinity oestrogen receptor. Such a and
receptor has been isolated and characterized in the
painted turtle, Chrysemys pietà (Mak, Ho & Callard, 1983; Riley, Heisermann, MacPherson & Callard, 1987; Ho, Fehrer, Yu et al. 1988), and appears to be a high-affinity and low-capacity receptor. Unfortu¬ nately, those studies did not examine the binding and biological activity of compounds such as tamoxifen. However, tamoxifen has been shown to antagonize oestrogen-stimulated behaviour in a squamate reptile (M. T. Mendonca & D. Crews, unpublished data). Alternatively, other receptors could be involved. For example, it has been hypothesized that oestrogen type II binding sites may be involved in the sexual dif¬ ferentiation of birds and that NET may interact with
et al. 1985). Previous studies in mammals indicate that these and possibly other binding sites (with similar steroid specificity) may be involved in the regulation of cell growth and the production of transforming growth factor beta (TGF-ß) in certain cancer cell lines (Markaverich & Clark, 1979; Panko, Watson & Clark, 1981; Markaverich, Williams, Upchurch & Clark, 1981; Giguere, Yang, Segui & Evans, 1988; Densmore, Markaverich, O'Malley & Clark, 1989). Of particular interest, Müllerian inhibiting hormone is a member of the TGF-ß family (Cate, Mattaliano, Hession et al. 1986). While no studies have examined the presence of such a binding site in reptiles, the paradoxical effects of tamoxifen in alligators suggest that the receptor mediating steroid-induced female sex determination is different from the oestrogen receptor in the liver and oviducts (Lance & Bogart, 1991). However, if type II binding sites are mediating steroid-induced feminiza¬ tion, then the specificity of the receptor in this turtle must differ from that in birds and mammals, since nafoxidine does not appear to bind type II binding sites in birds and mammals (Markaverich et al. 1981;
type II binding sites (Hutson
Densmore et al.
Gonadal
sex at
1989). female-producing temperature
Previous studies have consistently failed to produce males at a female-producing temperature in response to steroid hormone treatments (reviewed by Raynaud & Pieau, 1985; also see Crews et al. 1989). With the exception of a single male, all turtles in the present study from treated eggs developed ovaries at the female-producing temperature. While it is possible that 4-OHA affected sex determination in the single male, it is also possible that this represents a spurious result. Regardless, the results indicate that either (i) sex determination has little or no sensitivity to steroid hormones at a female-producing temperature, (ii) the effects of steroid hormones are overridden by the effects of the female-producing temperature or (iii) the effects of steroid hormones are similar to the effects of the female-producing temperature and are thus masked. The apparent ability of steroids to drive sex determination in only the female direction appears consistent with the mammalian concept of an organiz¬ ing sex (or induced sex) versus a default sex (Jost, 1970). That is, oestrogens and related compounds could either block or stimulate the physiological cascade that would result in the induced sex. How¬ ever, a recent study indicates that DHT can induce male sex determination when an incubation regime that results in a 1:1 sex ratio in control groups is utilized (Wibbels, Bull & Crews, 1992). This finding supports the hypothesis that female-producing tem¬ peratures either prevent or override any masculinizing
effects of steroid hormones on the gonads. Further, this indicates that sex determination can be driven toward either sex by specific steroid hormones. DHT and oestrogens could be exerting opposite effects on the same physiological pathway. Alternatively, they could be acting independently on different physiologi¬ cal cascades which lead to opposite sexes. Thus it is possible that each sex could represent an induced
the R5020 and the RU38486, Schering Plough pro¬ vided the hydroxyflutamide and Sterling-Winthrop Research Institute provided the cyanoketone. This research was supported by NIH NRSA Fellowship HD-07319 to T. W., NIH NRSA training grant HD07264 to the Institute of Reproductive Biology and by NIH grant HD-24976 and NIMH Research Scientist Award 00135 to D.C.
state.
Effects of treatments on differentiation and
development of the Müllerian ducts Oestrogens and oestrogen-related compounds affected the differentiation and development of the Müllerian ducts. Many of the females treated with the higher dosages of oestradiol-17ß, DES or NET lacked the caudal regions of the Müllerian ducts (i.e. 'agenesia' of the Müllerian ducts), whereas the cranial regions were hypertrophied. Agenesia of the Müllerian ducts
in response to steroid hormone treatments has been observed in previous studies ofbirds (Stoll, Faucounau & Maraud, 1987,1990), and the hypertrophie effects of oestrogens on oviducts are well established (reviewed
by O'Malley, Woo, Harris et al. 1975; Schimke, McKnight, Shapiro et al. 1975). The ability of NET to stimulate hypertrophy of the Müllerian ducts con¬ trasts with the findings in chickens in which NET acts as an oestrogen antagonist (Hutson et al. 1985; Stoll et al. 1990). The oestrogen antagonists/partial agonists tamoxifen, nafoxidine and clomiphene citrate did
not stimulate Müllerian duct agenesia, but they did appear to prevent or decrease Müllerian duct regression in some males (Table 1). The ability of oestrogens to block Müllerian duct regression has been well documented in birds (Hutson, Ikawa & Donahoe, 1982; MacLaughlin, Hutson & Donahoe, 1983; Hutson et al. 1985; Osamu & Hutson, 1988). The present results suggest that tamoxifen, nafoxidine and clomiphene citrate again acted as oestrogen agonists in this turtle system. Of particular interest, the higher dosages of clomiphene citrate consistently resulted in the hypertrophy of the entire oviduct in many males and females. The ability of clomiphene citrate to stimulate hypertrophy but not agenesia suggests that these two events may be regulated by different receptors. If different receptors are involved, it will be of interest to determine if either or both are also involved in gonadal differentiation.
ACKNOWLEDGEMENTS
We should like to acknowledge the technical assistance of A. Alexander, L. Riemenschneider, S. Bell, F. Chan, P. Gideon and S. Sinclair. Dr A. Brodie pro¬ vided 4-OHA for this study, Roussel Uclaf provided
REFERENCES Aldercreutz, H. & Tenhunen, R. (1970). Some aspects of the
interaction between natural and synthetic female hormones and the liver. American Journal of Medicine 49,630-638. Bull, J. J. (1980). Sex determination in reptiles. Quarterly Review of
Biology SS, 3—21. Bull, J. J. (1987). Temperature-dependent sex determination in
reptiles: validity of sex diagnosis in hatchling lizards. Canadian Journal ofZoology 65,1421-1424. Bull, J. J., Gutzke, W. H. N. & Crews, D. (1988). Sex reversal in three reptilian orders. General and Comparative Endocrinology 70, 425-428. Bull, J. J. & Vogt, R. C. (1979). Temperature-dependent sex determination in turtles. Science 206, 1186-1188. Bull, J. J., Vogt, R. C. & McCoy, C. J. (1982). Sex determining temperatures in turtles: A geographic comparison. Evolution 36, 326-332. Crews, D., Bull, J. J. & Wibbels, T. (1991). Estrogen and sex reversal in turtles: A dose-dependent phenomenon. General and Comparative Endocrinology 81, 357-364. Crews, D., Wibbels, T. & Gutzke, W. H. N. (1989). Action of sex steroid hormones on temperature-induced sex determination in the snapping turtle (Chelydra serpentina). General and Comparative Endocrinology 76, 159-166. Cate, R. L., Mattaliano, R. J., Hession, C, Tizard, R., Farber, N. M., Cheung, ., Ninfa, E. G, Frey, . ., Gash, D. J., Chow, E. P., Fisher, R. ., Bertonis, J. M., Torres, G., Wallner, B. P., Ramachandran, K. L., Ragin, R. C, Manganaro, T. F., MacLaughlin, D. T. & Donahoe, P. K. (1986). Isolation of bovine and human genes for Müllerian inhibiting substance and expression of the human gene in animal cells. Cell 45,685-698. Densmore, C. L., Markaverich, . M., O'Malley, B. W. & Clark, J. H. (1989). Characterization and partial purification of an estrogen type II binding site in chick oviduct cytosol.
Biochemistry 28,7788-7796. Edgren, R. A. & Sturtevant, F. M. (1976). Potencies of oral contraceptives. American Journal of Obstetrics and Gynecology 125,1029-1038. Ewert, M. A. & Nelson, G. E. (1991). Sex determination in turtles: Diverse patterns and some possible adaptive values. Copeia 1991, 50-69.
Giguere, V., Yang, N., Segui, P. & Evans, R. (1988). Identification
of a new class of steroid hormone receptors. Nature 331,91-94. Gutzke, W. H. N. & Bull, J. J. (1986). Steroid hormones reverse sex in turtles. General and Comparative Endocrinology 64, 368-372. Ho, S„ Fehrer, S„ Yu, M., Liang, L. & Press, D. (1988). High affinity binding of [3H]estradiol-17ß by an estrogen receptor in the liver of the turtle. General and Comparative Endocrinology 70, 382-394. Hutson, J. Mi, Donahoe, P. K. & MacLaughlin, D. T. (1985). Steroid modulation of Müllerian duct regression in the chick embryo. General and Comparative Endocrinology 57,88-102. Hutson, J. M„ Ikawa, H. & Donahoe, P. K. (1982). Estrogen inhibition of Müllerian inhibiting substance in the chick embryo. Journal of Pediatrie Surgery 17,953-959.
Jordan, C. V. & Murphy, C. S. (1990). Endocrine pharmacology of antiestrogens as antitumor agents. Endocrine Reviews 11, 578-610.
Jost, A. (1970). Hormonal factors in the sex differentiation of the mammalian foetus. Philosophical Transactions of the Royal Society ofLondon B259,119-130. Lance, V. A. & Bogart, M. H. (1991). Tamoxifen sex reverses alligator embryos at male producing temperatures, but is
antiestrogen in female hatchlings. Experimentia 47, 263-266. Lerner, L. J. & Jordan, C. (1990). Development of antiestrogens and their use in breast cancer: Eighth Cain Memorial Award Lecture. Cancer Research 50,4177-4189. MacLaughlin, D. T., Hutson, J. M. & Donahoe, P. K. (1983). Specific estradiol binding in embryonic Müllerian ducts: A potential modulator of regression in the male and female chick. Endocrinology 113, 141-145. Mak, P., Ho, S. & Callard, I. P. (1983). Characterization of an estrogen receptor in a turtle testis. General and Comparative an
Endocrinology 52,182-189.
Markaverich, B. M. & Clark, J. H. (1979). Two binding sites for estradiol in rat uterine nuclei; relationship to uterotropic response. Endocrinology 105, 1458-1462. Markaverich, B. M., Williams, M., Upchurch, S. & Clark, J. H. ( 1981 ). Heterogeneity of nuclear estrogen-binding sites in the
A simple method for quantitation of Type I and Type II sites by [3H]estradiol exchange. Endocrinology 109, 62-69. Mauvais-Jarvis, P. (1983). Progesterone and progestins: A general overview. In Progesterone and Progestins, pp. 1-16. Eds C. W. Bardin, E. Milgrom & P. Mauvais-Jarvis. New York: Raven Press. O'Malley, B. W., Woo, S. L., Harris, S. E., Rosen, J. M., Comstock, J. P., Chan, L., Bordelon, C. B„ Holder, J. W., Sperry, P. & Means, A. R. (1975). Steroid action in animal cells. American Zoologist 15(Suppl. 1), 215-225. Osamu, D. & Hutson, J. M. (1988). Pretreatment of chick embryos with estrogen in ovo prevents Müllerian duct regression in organ culture. Endocrinology 122,2888-2891. Osawa, Y., Yarborough, C. & Osawa, Y. (1982). Norethisterone, a major ingredient of contraceptive pills is a suicide inhibitor of estrogen biosynthesis. Science 215,1249-1251. rat uterus:
Panko, W. ., Watson, C. S. & Clark, J. H. (1981). The presence of a second, specific estrogen binding site in human breast cancer. Journal of Steroid Biochemistry 14, 1311-1316. Perez-Palacios, G., Chavez, B., Escobar, N., Vilchis, F., Larrea, F., Lince, M. & Perez, . E. (1981). Mechanism of action of contra¬
ceptive synthetic progestins. Journal of Steroid Biochemistry 15, 125-130.
Pieau, C. (1974). Différenciation du sexe en fonction de la temperature chez les embryons d'Emys orbicularis L. (Chelonien); effets des hormones sexuelles. Annales d'Embryologie et de
Morphogenese 1,365-394. Raynaud, A. & Pieau, C. (1985). Embryonic development of the genital system. In Biology of the Reptilla, pp. 149-300. Eds C. Gans & F. Billet. New York: Wiley. Riley, D., Heisermann, G. J., MacPherson, R. & Callard, I. P. (1987). Hepatic estrogren receptor in the turtle, Chrysemys pida: Partial characterization, seasonal changes and pituitary dependence. Journal of Steroid Biochemistry 26,41-47. Schimke, R. T., McKnight, G. S., Shapiro, D. J., Sullivan, D. & Palacios, R. (1975). Hormonal regulation of ovalbumin synthesis in the chick oviduct. Recent Progress in Hormone Research 31, 175-208.
Stoll, R., Faucounau, N. & Maraud, R. (1987). Influence of an
antiestrogenic drug (tamoxifen) on Müllerian duct agenesia induced by various steroidal sex hormones in the female chick
embryo. General and Comparative Endocrinology 66,218-223. Stoll, R., Faucounau, N. & Maraud, R. (1990). Action of estradiol
Müllerian duct regressions induced by treatment with norethindrone of female chick embryos. General and Comparative Endocrinology 80,101 -106. Wibbels, T., Bull, J. J. & Crews, D. (1991a). Synergism of tempera¬ ture and estradiol: A common pathway in sex determination? Journal of Experimental Zoology 260, 130-134. Wibbels, T., Bull, J. J. & Crews, D. (19916). Chronology and morphology of temperature-dependent sex determination. Journal ofExperimental Zoology 260, 371-381. Wibbels, T., Bull, J. J. & Crews, D. (1992). Steroid hormoneinduced male sex determination in an amniotic vertebrate. Journal ofExperimental Zoology. (In Press.) Yntema,C. L. (1968). A series of stages in the embryonic development of Chelydra serpentina. Journal of Morphology 125, 219-252. on
in
a
turtle
T. Wibbels and D. Crews Institute of Reproductive Biology, Department of Zoology, University of Texas at Austin, Austin, Texas 78712, U.S.A.
received
26 June 1991
ABSTRACT
The specificity of steroid hormone-induced sex determination was investigated in the red-eared slider, Trachemys scripta, a turtle with temperaturedependent sex determination. All eggs were incubated at either a female-producing temperature (31 \s=deg\C)or a male-producing temperature (26 \s=deg\C)and received control or experimental treatments at stage 17\p=n-\18 of embryonic development. A variety of treatments induced female sex determination at the male-
producing temperature. Oestradiol-17\g=b\, diethylstilboestrol (DES) (an oestrogen agonist) and norethindrone (NET) (a progestin with reputed oestrogenic as well as anti-oestrogenic properties) were the most effective in inducing female sex determination. Other reputed oestrogen antagonists/partial agonists (i.e. tamoxifen, nafoxidine and clomiphene citrate) were also capable of inducing female sex determination, but to a lesser extent. A high dosage of testosterone resulted in the production of some females (7 of 15 hatchlings) whereas dihydrotestosterone had no detectable effect on sex determination. This latter finding suggests that testosterone could be acting via aromatization to oestradiol-17\g=b\. A few females resulted from eggs that had been treated with
inhibitor, 1,4,6-androstatrien-3,17-dione (ATD) (3 of 97), the antiandrogen hydroxyflutamide (1 of 55) and progesterone (3 of 36), suggesting the possibilities of non-specific effects of these compounds when used in large dosages. Alternatively, metabolites of these compounds may be oestrogenic. Collectively, aromatase
the results at the male-producing temperature are consistent with the hypothesis that steroid-induced female sex determination is mediated via an oestrogen\x=req-\
specific receptor.
INTRODUCTION
In many
reptiles, sex is determined by the temperature the egg is incubated (i.e. temperature-
at which
With the exception of a single male turtle (produced from an egg treated with 4-androsten-4-ol-3,17-dione (4-OHA)), the various treatments were not able to induce male sex determination at a female-producing temperature. This suggests that either (i) sex determination has little or no sensitivity to steroid hormones at a female-producing temperature, (ii) the effects of steroid hormones are overridden by the effect of the female-producing temperature or (iii) the effects of steroid hormones are similar to the effects of the female-producing temperature and are thus masked. The effects of the treatments on M\l=u"\llerianduct differentiation and development were also recorded. NET exerted distinct and equivalent effects at both male- and female-producing temperatures, with many of the females exhibiting a hypertrophied cranial portion of the M\l=u"\llerianducts but lacking the caudal portion (i.e. M\l=u"\llerianduct agenesia). A similar effect was shown for the majority of females produced from eggs treated with DES and for one turtle produced from an egg treated with oestradiol-17\g=b\at the female-producing temperature. Certain dosages of oestradiol-17\g=b\and oestradiol-3-benzoate resulted in females with full length hypertrophied M\l=u"\llerianducts. Tamoxifen, nafoxidine and clomiphene citrate treatments did not stimulate M\l=u"\llerianduct agenesia, but they did appear to act as oestrogen agonists in blocking M\l=u"\llerianduct regression in some males. Further, although clomiphene citrate did not stimulate agenesia, it did produce hypertrophic M\l=u"\llerianducts in some males and females. The latter result is consistent with the hypothesis that different receptors could be involved in the differentiation versus hypertrophy of the M\l=u"\llerianducts. Journal of Endocrinology (1992) 133, 121\p=n-\129
dependent sex determination or TSD) (Bull, 1980; Raynaud & Pieau, 1985; Ewert & Nelson, 1991). The mechanism by which temperature affects sex determi¬ nation is not known, but it has been suggested that the
determination pathway may include the produc¬ tion of steroid hormones (Raynaud & Pieau, 1985; Crews, Wibbels & Gutzke, 1989). Several studies have shown that the effect of male-producing temperatures can be overriden by the administration of exogenous steroid hormones (reviewed by Raynaud & Pieau, 1985; also see Gutzke & Bull, 1986; Bull, Gutzke & Crews, 1988; Crews et al. 1989; Crews, Bull & Wibbels, 1991). Further, temperature and exogenous oestrogen exert a synergistic effect on sex deter¬ mination, suggesting that steroid hormones and sex
temperature may
pathway (Wibbels,
act on a common
Bull & Crews,
physiological
1991a). As such,
steroid hormones represent an avenue for probing the basis of sex determination in species with TSD. The present study investigates the speci¬ ficity of steroid hormone-induced sex determination in the red-eared slider, Trachemys scripta, a turtle with
physiological TSD.
MATERIALS AND METHODS
Freshly laid eggs were obtained commercially (Robert
Kliebert, Hammond, LA, U.S.A.). After transport to
laboratory, they were placed in containers with moistened vermiculite (vermiculite/water, 1:2) and placed in incubators set at either 31 °C or 26 °C. Pre¬ vious studies indicated that a continuous incubation temperature of 31 °C produces all female hatchlings whereas an incubation temperature of 26 °C produces all male hatchlings (Bull, Vogt & McCoy, 1982; Crews our
al. 1991; Wibbels, Bull & Crews, \99\b). Embryonic development was monitored by candling eggs and by dissecting two to four eggs approximately twice a week to verify specific developmental stages, based on criteria described by Yntema (1968). When embryos reached stage 17-18, the approximate midpoint in the temperature-sensitive period in this species (Wibbels et al. 1991 b), the eggs were randomized into control and experimental groups. Eggs from experimental et
a single treatment of a specific ligand (i.e. hormone, agonist or antagonist) suspended in 5 pi 95% ethanol. Oestradiol-17ß, diethylstilboestrol (DES), norethindrone (NET), nafoxidine, tamoxifen, clomiphene citrate, testosterone, 5a-androstan-17ßol-3-one (DHT), progesterone, corticosterone, 6 methyl-17a-hydroxyprogesterone (depo-provera) and 2-methyl-1,2-di-3-pyridyl-1 -propanone (metyrapone) were obtained from Sigma (St Louis, MO, U.S.A.), l,4,6-androstatrien-3,17-dione (ATD) from Steraloids (Wilton, NH, U.S.A.), hydroxyflutamide from Schering Plough (Kenilworth, NJ, U.S.A.), 4androsten-4-ol-3,17-dione (4-OHA) from Dr A. Brodie at University of Maryland (Baltimore, MD, U.S.A.),
groups received
cyanoketone from Sterling-Winthrop Research Insti¬ tute (Rensselaer, NY, U.S.A.), and R5020 and
RU38486 (RU486) from Roussel Uclaf (Romainville, Seine-St-Denis, France). The dosages chosen for each
ligand were based on previous studies with turtles (Gutzke & Bull, 1986; Bull et al. 1988; Crews et al. 1989, 1991) and on the availability and solubility of the specific agents. Eggs in control groups received a single treatment consisting of 5 pi 95% ethanol. All treatments were either injected into the egg or applied topically to the vascularized portion of the upper shell (Crews et al. 1991). Both techniques have been shown to be effective means of delivering steroid hormones to embryos in T. scripta eggs (Crews et al. 1991). Groups of approximately 30 eggs were used for each injected dosage group: topical administration groups consisted of approximately 15 eggs. After receiving treatments, all eggs were placed back into their respective incubators (31 °C or 26 °C) until they hatched. Turtles were killed approximately 2—4 weeks after hatching. Gonadal sex and develop¬ mental status of the Müllerian ducts were assessed by examination of the reproductive tracts under a dissec¬ tion microscope. The gonads of hatchling T. scripta are relatively well differentiated and, with rare excep¬ tion, appear distinctly testicular or ovarian when viewed under a dissection microscope (Crews et al. 1991; Wibbels et al. 1991e). Ovaries are long and flat whereas testes are shorter, more round, and have visible seminiferous tubules (see Crews et al. 1991 ). The developmental status of both the cranial and caudal halves of the Müllerian ducts was examined and scored as either absent, regressed but visible, normal (as in a typical female hatchling), or hypertrophied.
RESULTS
Gonadal sex and developmental status of Müllerian ducts for control and experimental turtles are shown in Tables 1-3. As shown previously for oestrogens (Crews et al. 1991), both modes of hormone adminis¬ tration were capable of delivering at least some of the treatments in this study. However, the effectiveness of some treatments varied with the mode of adminis¬ tration. Mortality of eggs after treatment varied widely. In general, during the first few weeks after treatment there was high mortality of injected eggs and low mortality of eggs treated topically. In addition, high mortality that did not appear to be related to hormone administration occurred in some groups near the time of pipping. We speculate that the humid environment created by the moist vermiculite together with the fluid from the egg provided an optimal environment for infectious organisms.
1. Effects of oestrogen and putative agonists and antagonists on sex determination and Müllerian duct development of red-eared sliders, Trachemys scripta, at a male-producing temperature (26 °C). Control treatments consisted of 5 pi 95% ethanol table
Dose*
Survival0
(Hg)
10 10 1
Diethylstilboestrol
Sat. Sat. 100 10
Status of Müllerian ducts'
(%)
Males
I
19 15
0 0
95 50
A A
I
0 0 5
18 11 11
90 37 80
11 females
0 0 0 2
6 3 10 6
40 10 75 40
Treatment Control
Oestradiol-17ß
Intersex
Females
Adm."
I
, 4 males A, 1 male R
=
=
=
H, 2 females HC,ACD HCACD HC,ACD 1 male N, 1 male R, 2 females A, 4 females HC,ACD 17 females HC,ACD, 1 female H, 4 females
=
=
=
=
=
=
10
63
=
1 female Sat. Sat.
Norethindrone
I
9 12
60 43
13 13 13
87 43 87
10
70
=
=
A
1 female H, 8 females HC,ACD 1 male A, 7 females NC.ACD, females HC,ACD all HC,ACD 10 females HC,ACD, 3 females A 4 females N, 7 females H, 2 females HC,ACD 9 males A, 9 females N, 1 female NC, ACD, 2 intersex =
=
=
=
=
100 100 10
I
=
=
=
=
=
=
10
=
=
=
Clomiphene citrate
100 100 10 10
Nafoxidine
I
13 2 14 19
I
10 14
I
100 100
I 11 0 0
=
93 43 93 63
13 males H, 1 female 1 male H, 1 male N, 11 females all A 16 males A, 3 males R
87 57
10 males N, 3 females 3 males A, 1 male R, 10 males 3 females 7 males A, 2 males R, 5 males female
=
=
=
=
=
=
=
=
=
=
=
=
=
N,
=
10
14
100
=
=
=
N,
=
Sat. 100
Tamoxifen
15 12
100 100
10
60 100
2 males A, 9 males R, 4 males 1 male A, 3 males R, 8 males N, 2 females 6 males R, 12 males 1 male A, 5 males R, 2 males N, 2 males H, 3 females N, 2 females 4 males A, 4 males R, 7 males N, 5 females =
=
=
=
=
=
=
100 10
I
=
=
=
=
=
=
=
=
10
15
67
=
=
=
=
"Dosages delivered in 5 µ ethanol. Sat. saturated solution. bAdministration techniques: topical administration, I injection. 'Percentage of embryos in each group which survived to hatching. All of the eggs in four treatment groups (1 \ig oestradiol- 17ß injected, 100 µg diethylstilboestrol injected, 10 µg nafoxidine injected and saturated tamoxifen injected) died and, therefore, these groups are not listed in the table. dN typical Müllerian ducts of control female hatchling (i.e. 'normal'), R Müllerian ducts visible but regressed, A Müllerian ducts absent, Müllerian ducts hypertrophied, HC cranial region of Müllerian ducts hypertrophied, NC cranial regions of Müllerian ducts normal, ACD caudal region of Müllerian ducts absent (see text for a more detailed description). =
=
—
=
=
=
Decreasing the ratio of water to vermiculite will hope¬ fully alleviate this problem in future studies. Previous studies indicate that differential mortality is not the means by which temperature or oestrogens produce biased sex ratio (Bull & Vogt, 1979; Bull, 1987; Crews et al. 1991).
=
=
=
=
Male-producing temperature Turtles
receiving control treatments all developed Oestrogen and oestrogen-related compounds were the most effective agents inducing female sex determination including both reputed agonist and testes.
The androgen DHT and the aromatase inhibitor 4-OHA did not appear to affect sex determi¬ nation (Table 2). However, testosterone and to a lesser extent the androgen antagonist hydroxyflutamide and the aromatase inhibitor ATD resulted in the production of some females. The progesterone antagonist RU486 and the progesterone agonist R5020 did not appear to affect sex determination, but the high dosage of progesterone (250 pg) resulted in the production of three females out of 36 hatchlings
antagonists.
(Table 2). Corticosterone antagonist metyrapone did determination (Table 2).
and the corticosteroid not appear to affect sex
Turtles that received control treatments lacked Müllerian ducts, whereas the development of Müllerian ducts in turtles treated with oestrogens and oestrogen-related compounds was highly variable (Table 1). Depending on the specific treatment, the status of the Müllerian ducts ranged from absent to hypertrophied. Additionally, NET and DES prevented
2. Effects of a variety of steroid hormones and putative agonists and antagonists on sex determination and Müllerian duct development of red-eared sliders, Trachemys scripta, at a male-producing temperature (26 °C). Control treatments consisted of 5 pi 95% ethanol (see Table 1) table
Doseb
(Pg) Androgens and putative androgen antagonist Testosterone
DHTa
Hydroxyflutamide
100 100 10 10
Sat. Sat. 100 100 10 10 Sat. 100 100
Survival11 Adm.c
Males
I
13
1
15 19
4-OHA"
Progesterone and putative agonist and antagonist Progesterone R5020
RU486
Sat. Sat. 100 100 10 10 100 100 10
Sat. Sat. 10
100 10 10
(%)
Status of Müllerian ducts'
(I
100 43 100 63
7 males A A A
0 0
10
o
8 10
0 0
I
0
I
14 13 25
I
13 14 12
0 0 0
15
0
13 13 14
0 0 0 0
I
I
0 o
75 27 75 47 87 83 87 100 40 100
A 14 males A A
87 43 100 87 93 53
8 males A 14 males A A 14 males
15 19 20
100 63 67
A A
1
12 5
12 males 5 males
16
80 20 100
6 7 2
40 47 7
13 6 12
87 20 80
133, 121-129
19 males
A, 1 female
=
=
I I
1
=
A A A A A A
1
0 0
A, 1 male R, 7 females
=
26 14 14
100 100 10
Journal of Endocrinology (1992)
Intersex
I
10
Putative aromatase inhibitors ATD"
Females
=
A, 5 males R =
=
A, 1 female
=
A, 2 females
=
A, 1 male
=
=
=
=
=
, 2 females , 1 female =
=
=
table
2. Continued Survival11
Dose'
(µ«)
Intersex
Females
Status of Müllerian ducts'
(%)
Adm.'
Males
T
14 12
93 40
A A
12 16 14 25 12 23
80
A A 13 males = A, 1 male A A A
Corticosterone and
putative antagonists Corticosterone
Metyrapone
10 10
I
Sat. Sat. 100 100 10
I I I
10
Hydroxysteroid dehydrogenase inhibitor
Cyanoketone
100 100 10 10
53
93 83 80 77
15 17 15 14
I I
0 0 0 0
100 57 100 46
=
R
A
A A A
dihydroxytestosterone, ATD 1,4,6-androstatrien-3,17-dione, 4-OHA 4-androsten-4-ol-3,17-dione.
aDHT
=
hDosages were delivered in 5 µ
ethanol. Sat.
=
saturated solution. 'Administration techniques: topical administration, I injection. dPercentage of embryos in each group which survived to hatching. All of the eggs in four treatment groups (saturate hydroxyflutamide injected, 10 µg hydroxyflutamide injected, 10 µg 4-OHA topically applied, 10 µg progesterone injected, 100 µg R5020 injected and 10 µg RU486 injected) died and, therefore, these groups are not listed in the table. CN typical Müllerian ducts of control female hatchling (i.e. -normal'), R Müllerian ducts visible but regressed, A Müllerian ducts absent. —
=
=
=
=
=
the differentiation of the caudal portion of the Müllerian ducts in some females (i.e. agenesia of the Müllerian ducts).
Female-producing temperature With a single exception, all individuals in all groups had ovaries (Table 3). The one hatchling with testes (confirmed by histological analysis) received a high dosage (lOOpg) of 4-OHA, an aromatase inhibitor; the other ten individuals ofthat group showed typical
female development. All treatments except oestradiol- 17ß, oestradiol-3benzoate (EB) and NET resulted in hatchlings with oviducts similar to those of the control females. The Müllerian ducts of turtles treated with oestradiol17ß, EB and NET were variable. The higher dosages of oestradiol-17ß and NET (lOOpg and saturated) resulted in some turtles in which the cranial portion of the Müllerian ducts was hypertrophied, but the caudal portion was absent. Lower dosages of these treat¬ ments resulted in some individuals with Müllerian ducts similar to those of individuals that received the higher dosages, whereas others had either normal Müllerian ducts, full length Müllerian ducts which were hypertrophied, or normal cranial portions of the Müllerian ducts, but lacking the caudal portion. The Müllerian ducts of individuals treated with EB appeared hypertrophied or normal.
=
DISCUSSION
Gonadal sex at
male-producing temperature
Past studies have shown that exogenous oestrogens, and to a lesser extent testosterone, can induce female sex determination at a male-producing incubation tem¬ perature in reptiles with TSD (Pieau, 1974; Gutzke & Bull, 1986; Bull et al. 1988; Crews et al. 1989, 1991; Wibbels et al. 1991 ). It has been hypothesized that steroid-induced feminization is an oestrogen-specific event and the feminization caused by exogenous testos¬ terone treatments acts through aromatization in the egg (Raynaud & Pieau, 1985; Crews et al. 1989). Consistent with this hypothesis, oestradiol-17ß and DES effectively induced female sex determination in the present study (Table 1). Additionally, NET, a synthetic progestin (Edgren & Sturtevant, 1976; Mauvais-Jarvis, 1983) that has been reported to be an oestrogen antagonist and agonist as well as an irrever¬ sible aromatase inhibitor (Aldercreutz & Tenhunen, 1970; Perez-Palacios, Chavez, Escobar et al. 1981; Osawa, Yarborough & Osawa, 1982; Hutson, Donahoe & MacLaughlin, 1985), was also very effec¬ tive in feminizing gonads. NET has additionally been shown to feminize partially the gonads of male chickens (Hutson et al. 1985). Tamoxifen, nafoxidine and clomiphene citrate were not as potent as oestradiol-17ß, DES and NET, but were capable of
3. Effects of steroids and reputed steroid agonist antagonists on sex determination and Müllerian duct development at a female-producing temperature (31 °C). Control treatments consisted of 5 pi 95% ethanol
table
Dose*
Survival11
(µß)
Males
Adm.'
Females 14 15
93 50
10000 1000 100
12 5
80
Sat. 10000 1000
7
Control Control
1
Oestrogens and putative antagonists/partial agonists Oestradiol-17ß Oestradiol-3-benzoate
II
6 10
100 001
9 14 12 15 14
Nafoxidine
10000 1000 100 010
15 14 12 10
Tamoxifen
Sat. 10000 1000 100 0-10 001
Norethindrone
Sat. 10000 1000
1
14
10 19 10 15 13
Status of Müllerian ducts'
(%)_
HC,ACD 1=N,3 H, 1=HC,ACD
33 73
=
23 20 33
HC.ACD HCACD 3 N, 1 =H, 1 =HC, 7 HC,ACD
30 47 40
=
93 93 100 93 80 67
13
=
=
N, 1=NRT,ALT
47 33 63
33 100 S7
Androgen and aromatase
inhibitors
Dihydrotestosterone
10000 5000 1000
5
ATD"
Sat. 10000 1000 100 010
11 21 9 13 14
87 93
10000 1000
10 11
67 73
Sat. 10000 1000
4 11 7
13 37
23
1000 010 001
10 14 6
67 93 40
Sat. 10000 1000
14 7 4
47 23 13
4-OHA'
Progestogens Progesterone
Depo-provera
3p-Hydroxysteroid dehydrogenase Cyanoketone "ATD
=
27 27 17
37 70
females
=
, male R =
1,4,6-androstatrien-3,17-dione, 4-OHA 4-androsten-4-ol-3,17-dione. =
bDosages were delivered in 5 µ
ethanol. Sat.
saturated solution. 'Administration techniques: topical administration, I injection. All injections at this temperature were conducted during the first year of this study, whereas topical administration was used during the second year to decrease mortality. ^Percentage of each treatment group which survived to hatching. All of the eggs in the norethindrone (010 µg topical application) group died and they are therefore not listed. 'N typical Müllerian ducts of control female hatchling (i.e. 'normal'), R Müllerian ducts visible but regressed, A Müllerian ducts absent, Müllerian ducts hypertrophied, HC cranial region of Müllerian ducts hypertrophied, ACD caudal region of Müllerian ducts absent (see the text for a more detailed description), NRT/ALT normal right Müllerian duct and absent left Müllerian duct. =
=
=
=
=
=
=
=
=
=
inducing female sex determination in some embryos (Table 1). All three of these agents are triphenylethylene derivatives and are reputed oestrogen antagonists/partial agonists (reviewed by Lerner & Jordan, 1990; Jordan & Murphy, 1990). The current findings indicate that these putative anti-oestrogens or their metabolites are capable of mimicking the feminizing effects of oestrogen in this reptilian system. Similarly, tamoxifen was recently reported to induce female sex determination in alligator embryos incubated at a male-producing temperature (Lance & Bogart, 1991). One of the high dosages of testosterone (lOOpg applied topically) resulted in female development in some embryos, whereas all individuals that received the same dosage by injection or the lower dosage by
either administration method were male. Adminis¬ tration of the lower dosage of testosterone (10 pg) by both administration methods produced only males. Past studies which have successfully produced females at male-producing temperatures have used relatively high dosages ( > 200 pg) of testosterone propionate (Pieau, 1974; Gutzke & Bull, 1986) in comparison with the dosage of oestrogen required to induce female sex determination (01 pg) (Crews et al. 1991). These findings are consistent with the hypothesis that testosterone may act via aromatization to oestradiol17ß. Contrary to the hypothesis of oestrogen-specific induction of female sex determination, a few females were produced from eggs treated with the androgen antagonist hydroxyflutamide (1 of 55 hatchlings), the aromatase inhibitor ATD (3 of 97 hatchlings) and progesterone (3 of 36 hatchlings). However, it is possible that the higher dosages of these treatments were metabolized into compounds with oestrogenic
activity. Collectively,
these results indicate that oestrogens
oestrogen-related compounds represent potent compounds for inducing female sex determination. It is presently unclear, however, exactly how this feminizing effect is mediated. One possibility is via a 'classic' high-affinity oestrogen receptor. Such a and
receptor has been isolated and characterized in the
painted turtle, Chrysemys pietà (Mak, Ho & Callard, 1983; Riley, Heisermann, MacPherson & Callard, 1987; Ho, Fehrer, Yu et al. 1988), and appears to be a high-affinity and low-capacity receptor. Unfortu¬ nately, those studies did not examine the binding and biological activity of compounds such as tamoxifen. However, tamoxifen has been shown to antagonize oestrogen-stimulated behaviour in a squamate reptile (M. T. Mendonca & D. Crews, unpublished data). Alternatively, other receptors could be involved. For example, it has been hypothesized that oestrogen type II binding sites may be involved in the sexual dif¬ ferentiation of birds and that NET may interact with
et al. 1985). Previous studies in mammals indicate that these and possibly other binding sites (with similar steroid specificity) may be involved in the regulation of cell growth and the production of transforming growth factor beta (TGF-ß) in certain cancer cell lines (Markaverich & Clark, 1979; Panko, Watson & Clark, 1981; Markaverich, Williams, Upchurch & Clark, 1981; Giguere, Yang, Segui & Evans, 1988; Densmore, Markaverich, O'Malley & Clark, 1989). Of particular interest, Müllerian inhibiting hormone is a member of the TGF-ß family (Cate, Mattaliano, Hession et al. 1986). While no studies have examined the presence of such a binding site in reptiles, the paradoxical effects of tamoxifen in alligators suggest that the receptor mediating steroid-induced female sex determination is different from the oestrogen receptor in the liver and oviducts (Lance & Bogart, 1991). However, if type II binding sites are mediating steroid-induced feminiza¬ tion, then the specificity of the receptor in this turtle must differ from that in birds and mammals, since nafoxidine does not appear to bind type II binding sites in birds and mammals (Markaverich et al. 1981;
type II binding sites (Hutson
Densmore et al.
Gonadal
sex at
1989). female-producing temperature
Previous studies have consistently failed to produce males at a female-producing temperature in response to steroid hormone treatments (reviewed by Raynaud & Pieau, 1985; also see Crews et al. 1989). With the exception of a single male, all turtles in the present study from treated eggs developed ovaries at the female-producing temperature. While it is possible that 4-OHA affected sex determination in the single male, it is also possible that this represents a spurious result. Regardless, the results indicate that either (i) sex determination has little or no sensitivity to steroid hormones at a female-producing temperature, (ii) the effects of steroid hormones are overridden by the effects of the female-producing temperature or (iii) the effects of steroid hormones are similar to the effects of the female-producing temperature and are thus masked. The apparent ability of steroids to drive sex determination in only the female direction appears consistent with the mammalian concept of an organiz¬ ing sex (or induced sex) versus a default sex (Jost, 1970). That is, oestrogens and related compounds could either block or stimulate the physiological cascade that would result in the induced sex. How¬ ever, a recent study indicates that DHT can induce male sex determination when an incubation regime that results in a 1:1 sex ratio in control groups is utilized (Wibbels, Bull & Crews, 1992). This finding supports the hypothesis that female-producing tem¬ peratures either prevent or override any masculinizing
effects of steroid hormones on the gonads. Further, this indicates that sex determination can be driven toward either sex by specific steroid hormones. DHT and oestrogens could be exerting opposite effects on the same physiological pathway. Alternatively, they could be acting independently on different physiologi¬ cal cascades which lead to opposite sexes. Thus it is possible that each sex could represent an induced
the R5020 and the RU38486, Schering Plough pro¬ vided the hydroxyflutamide and Sterling-Winthrop Research Institute provided the cyanoketone. This research was supported by NIH NRSA Fellowship HD-07319 to T. W., NIH NRSA training grant HD07264 to the Institute of Reproductive Biology and by NIH grant HD-24976 and NIMH Research Scientist Award 00135 to D.C.
state.
Effects of treatments on differentiation and
development of the Müllerian ducts Oestrogens and oestrogen-related compounds affected the differentiation and development of the Müllerian ducts. Many of the females treated with the higher dosages of oestradiol-17ß, DES or NET lacked the caudal regions of the Müllerian ducts (i.e. 'agenesia' of the Müllerian ducts), whereas the cranial regions were hypertrophied. Agenesia of the Müllerian ducts
in response to steroid hormone treatments has been observed in previous studies ofbirds (Stoll, Faucounau & Maraud, 1987,1990), and the hypertrophie effects of oestrogens on oviducts are well established (reviewed
by O'Malley, Woo, Harris et al. 1975; Schimke, McKnight, Shapiro et al. 1975). The ability of NET to stimulate hypertrophy of the Müllerian ducts con¬ trasts with the findings in chickens in which NET acts as an oestrogen antagonist (Hutson et al. 1985; Stoll et al. 1990). The oestrogen antagonists/partial agonists tamoxifen, nafoxidine and clomiphene citrate did
not stimulate Müllerian duct agenesia, but they did appear to prevent or decrease Müllerian duct regression in some males (Table 1). The ability of oestrogens to block Müllerian duct regression has been well documented in birds (Hutson, Ikawa & Donahoe, 1982; MacLaughlin, Hutson & Donahoe, 1983; Hutson et al. 1985; Osamu & Hutson, 1988). The present results suggest that tamoxifen, nafoxidine and clomiphene citrate again acted as oestrogen agonists in this turtle system. Of particular interest, the higher dosages of clomiphene citrate consistently resulted in the hypertrophy of the entire oviduct in many males and females. The ability of clomiphene citrate to stimulate hypertrophy but not agenesia suggests that these two events may be regulated by different receptors. If different receptors are involved, it will be of interest to determine if either or both are also involved in gonadal differentiation.
ACKNOWLEDGEMENTS
We should like to acknowledge the technical assistance of A. Alexander, L. Riemenschneider, S. Bell, F. Chan, P. Gideon and S. Sinclair. Dr A. Brodie pro¬ vided 4-OHA for this study, Roussel Uclaf provided
REFERENCES Aldercreutz, H. & Tenhunen, R. (1970). Some aspects of the
interaction between natural and synthetic female hormones and the liver. American Journal of Medicine 49,630-638. Bull, J. J. (1980). Sex determination in reptiles. Quarterly Review of
Biology SS, 3—21. Bull, J. J. (1987). Temperature-dependent sex determination in
reptiles: validity of sex diagnosis in hatchling lizards. Canadian Journal ofZoology 65,1421-1424. Bull, J. J., Gutzke, W. H. N. & Crews, D. (1988). Sex reversal in three reptilian orders. General and Comparative Endocrinology 70, 425-428. Bull, J. J. & Vogt, R. C. (1979). Temperature-dependent sex determination in turtles. Science 206, 1186-1188. Bull, J. J., Vogt, R. C. & McCoy, C. J. (1982). Sex determining temperatures in turtles: A geographic comparison. Evolution 36, 326-332. Crews, D., Bull, J. J. & Wibbels, T. (1991). Estrogen and sex reversal in turtles: A dose-dependent phenomenon. General and Comparative Endocrinology 81, 357-364. Crews, D., Wibbels, T. & Gutzke, W. H. N. (1989). Action of sex steroid hormones on temperature-induced sex determination in the snapping turtle (Chelydra serpentina). General and Comparative Endocrinology 76, 159-166. Cate, R. L., Mattaliano, R. J., Hession, C, Tizard, R., Farber, N. M., Cheung, ., Ninfa, E. G, Frey, . ., Gash, D. J., Chow, E. P., Fisher, R. ., Bertonis, J. M., Torres, G., Wallner, B. P., Ramachandran, K. L., Ragin, R. C, Manganaro, T. F., MacLaughlin, D. T. & Donahoe, P. K. (1986). Isolation of bovine and human genes for Müllerian inhibiting substance and expression of the human gene in animal cells. Cell 45,685-698. Densmore, C. L., Markaverich, . M., O'Malley, B. W. & Clark, J. H. (1989). Characterization and partial purification of an estrogen type II binding site in chick oviduct cytosol.
Biochemistry 28,7788-7796. Edgren, R. A. & Sturtevant, F. M. (1976). Potencies of oral contraceptives. American Journal of Obstetrics and Gynecology 125,1029-1038. Ewert, M. A. & Nelson, G. E. (1991). Sex determination in turtles: Diverse patterns and some possible adaptive values. Copeia 1991, 50-69.
Giguere, V., Yang, N., Segui, P. & Evans, R. (1988). Identification
of a new class of steroid hormone receptors. Nature 331,91-94. Gutzke, W. H. N. & Bull, J. J. (1986). Steroid hormones reverse sex in turtles. General and Comparative Endocrinology 64, 368-372. Ho, S„ Fehrer, S„ Yu, M., Liang, L. & Press, D. (1988). High affinity binding of [3H]estradiol-17ß by an estrogen receptor in the liver of the turtle. General and Comparative Endocrinology 70, 382-394. Hutson, J. Mi, Donahoe, P. K. & MacLaughlin, D. T. (1985). Steroid modulation of Müllerian duct regression in the chick embryo. General and Comparative Endocrinology 57,88-102. Hutson, J. M„ Ikawa, H. & Donahoe, P. K. (1982). Estrogen inhibition of Müllerian inhibiting substance in the chick embryo. Journal of Pediatrie Surgery 17,953-959.
Jordan, C. V. & Murphy, C. S. (1990). Endocrine pharmacology of antiestrogens as antitumor agents. Endocrine Reviews 11, 578-610.
Jost, A. (1970). Hormonal factors in the sex differentiation of the mammalian foetus. Philosophical Transactions of the Royal Society ofLondon B259,119-130. Lance, V. A. & Bogart, M. H. (1991). Tamoxifen sex reverses alligator embryos at male producing temperatures, but is
antiestrogen in female hatchlings. Experimentia 47, 263-266. Lerner, L. J. & Jordan, C. (1990). Development of antiestrogens and their use in breast cancer: Eighth Cain Memorial Award Lecture. Cancer Research 50,4177-4189. MacLaughlin, D. T., Hutson, J. M. & Donahoe, P. K. (1983). Specific estradiol binding in embryonic Müllerian ducts: A potential modulator of regression in the male and female chick. Endocrinology 113, 141-145. Mak, P., Ho, S. & Callard, I. P. (1983). Characterization of an estrogen receptor in a turtle testis. General and Comparative an
Endocrinology 52,182-189.
Markaverich, B. M. & Clark, J. H. (1979). Two binding sites for estradiol in rat uterine nuclei; relationship to uterotropic response. Endocrinology 105, 1458-1462. Markaverich, B. M., Williams, M., Upchurch, S. & Clark, J. H. ( 1981 ). Heterogeneity of nuclear estrogen-binding sites in the
A simple method for quantitation of Type I and Type II sites by [3H]estradiol exchange. Endocrinology 109, 62-69. Mauvais-Jarvis, P. (1983). Progesterone and progestins: A general overview. In Progesterone and Progestins, pp. 1-16. Eds C. W. Bardin, E. Milgrom & P. Mauvais-Jarvis. New York: Raven Press. O'Malley, B. W., Woo, S. L., Harris, S. E., Rosen, J. M., Comstock, J. P., Chan, L., Bordelon, C. B„ Holder, J. W., Sperry, P. & Means, A. R. (1975). Steroid action in animal cells. American Zoologist 15(Suppl. 1), 215-225. Osamu, D. & Hutson, J. M. (1988). Pretreatment of chick embryos with estrogen in ovo prevents Müllerian duct regression in organ culture. Endocrinology 122,2888-2891. Osawa, Y., Yarborough, C. & Osawa, Y. (1982). Norethisterone, a major ingredient of contraceptive pills is a suicide inhibitor of estrogen biosynthesis. Science 215,1249-1251. rat uterus:
Panko, W. ., Watson, C. S. & Clark, J. H. (1981). The presence of a second, specific estrogen binding site in human breast cancer. Journal of Steroid Biochemistry 14, 1311-1316. Perez-Palacios, G., Chavez, B., Escobar, N., Vilchis, F., Larrea, F., Lince, M. & Perez, . E. (1981). Mechanism of action of contra¬
ceptive synthetic progestins. Journal of Steroid Biochemistry 15, 125-130.
Pieau, C. (1974). Différenciation du sexe en fonction de la temperature chez les embryons d'Emys orbicularis L. (Chelonien); effets des hormones sexuelles. Annales d'Embryologie et de
Morphogenese 1,365-394. Raynaud, A. & Pieau, C. (1985). Embryonic development of the genital system. In Biology of the Reptilla, pp. 149-300. Eds C. Gans & F. Billet. New York: Wiley. Riley, D., Heisermann, G. J., MacPherson, R. & Callard, I. P. (1987). Hepatic estrogren receptor in the turtle, Chrysemys pida: Partial characterization, seasonal changes and pituitary dependence. Journal of Steroid Biochemistry 26,41-47. Schimke, R. T., McKnight, G. S., Shapiro, D. J., Sullivan, D. & Palacios, R. (1975). Hormonal regulation of ovalbumin synthesis in the chick oviduct. Recent Progress in Hormone Research 31, 175-208.
Stoll, R., Faucounau, N. & Maraud, R. (1987). Influence of an
antiestrogenic drug (tamoxifen) on Müllerian duct agenesia induced by various steroidal sex hormones in the female chick
embryo. General and Comparative Endocrinology 66,218-223. Stoll, R., Faucounau, N. & Maraud, R. (1990). Action of estradiol
Müllerian duct regressions induced by treatment with norethindrone of female chick embryos. General and Comparative Endocrinology 80,101 -106. Wibbels, T., Bull, J. J. & Crews, D. (1991a). Synergism of tempera¬ ture and estradiol: A common pathway in sex determination? Journal of Experimental Zoology 260, 130-134. Wibbels, T., Bull, J. J. & Crews, D. (19916). Chronology and morphology of temperature-dependent sex determination. Journal ofExperimental Zoology 260, 371-381. Wibbels, T., Bull, J. J. & Crews, D. (1992). Steroid hormoneinduced male sex determination in an amniotic vertebrate. Journal ofExperimental Zoology. (In Press.) Yntema,C. L. (1968). A series of stages in the embryonic development of Chelydra serpentina. Journal of Morphology 125, 219-252. on
