Aquaculture 418–419 (2014) 165–170

Contents lists available at ScienceDirect

Aquaculture

j ourna l homepage: www.e lsev ie r .com/ locate /aqua-on l ine

High corticosterone and sex reversal in common carp (Cyprinus carpio L.)with adrenal hyperplasia caused by P450c17a2 deficiency

Mohammad A. Nematollahi a,c,⁎, Hilde van Pelt-Heerschap b,d, Hans Komen b

a Aquaculture and Fisheries Group, Wageningen Institute of Animal Sciences (WIAS), Wageningen University, P.O. Box: 338, 6700 AH Wageningen, The Netherlandsb Animal Breeding and Genomics Center, Wageningen Institute of Animal Sciences (WIAS), Wageningen University, P.O. Box: 338, 6700 AH Wageningen, The Netherlandsc Dept. of Fisheries, Faculty of Natural Resources, University of Tehran, Chamran St., P.O. Box: 31585-4314, Karaj, Irand IMARES-WUR, P.O. Box: 57, 1780 AB Den Helder, The Netherlands

⁎ Corresponding author at: Dept. of Fisheries, Faculty ofTehran, Chamran St., P.O. Box: 31585-4314, Karaj, Iran. Te

E-mail address: malahi@ut.ac.ir (M.A. Nematollahi).

0044-8486/$ – see front matter. Crown Copyright © 2013http://dx.doi.org/10.1016/j.aquaculture.2013.10.011

a b s t r a c t

a r t i c l e i n f o

Article history:Received 24 December 2012Received in revised form 28 August 2013Accepted 5 October 2013Available online 14 October 2013

Keywords:17α-hydroxylaseSex reversalAdrenal hyperplasiaCorticosteroneCortisolCommon carp

The aim of this study was to investigate the inheritance of adrenal hyperplasia, caused by 17α-hydroxylasedeficiency, and its association with female to male sex reversal in common carp. Inbred strains used in thisexperiment were E4 (XX, female, normal), E5 (XX, male, adrenal hyperplasia) and E7 (XX, male, adrenalhyperplasia). F1 females (E4E5: XX, normal) were crossed to E5 and E7 males to produce backcross progeny(BC5 and BC7), or reproduced by gynogenesis to produce doubled haploid progeny (DH). Fish (age 6months)were subjected to a net confinement stressor and sacrificed to collect blood plasma for analysis of cortisol andcorticosterone, and to determine sex. Values for plasma corticosterone showed a clear segregation pattern,consistent with a single recessive gene model. There was a highly significant difference between high and lowcorticosterone responders in BC and DH progeny groups. Mean corticosterone values for high corticosteroneresponders in the three groups BC5, BC7 and DH were, respectively: 1706, 1760 and 1366 ng/ml. For lowcorticosterone responders values were 1.5, 1.1 and 0.6 ng/ml. High corticosterone responders had on averagelow levels of cortisol: 21.3, 24 and 17.3 ng/ml for BC5, BC7 and DH. Low corticosterone responders had normallevels (66.7, 105.1 and 65.2 ng/ml for BC5, BC7 and DH, respectively). Sex ratios were significantly differentbetween high and low corticosterone responders. High corticosterone responders were predominantly male,while low corticosterone responders were female or intersex. High corticosterone responders had welldeveloped testis and were fertile indicating that in the gonads, 17α-hydroxylase is not impaired. The observedeffect on sex shows that 17α-hydroxylase deficiency most likely augments sex reversal in common carp, eitherdirectly through high levels of plasma corticosterone, or through paracrine effects of elevated ACTH production.

Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.

1. Introduction

Congenital Adrenal Hyperplasia (CAH) represents a group ofinherited disorders that are all characterized by impaired productionof cortisol by the adrenal gland (Hall et al., 2004). CAH occurs becauseof a deficiency in any one of the enzymes involved in the conversionof cholesterol to cortisol in human (Miller, 1991; New, 1998) andother mammals (Deaton et al., 1999; Gotoh et al., 1988; Migeon andDonohoue, 1991; Pang et al., 1992).

In teleost fish, cortisol is the main glucocorticoid and exerts a widerange of physiological effects; the cortisol producing cells are locatedin strands surrounding the cardinal veins in the head kidney, theadrenal homolog in fish (Milano et al., 1997; Ruane et al., 2005). Thefirst case of “adrenal” hyperplasia in fish was described in twohomozygous inbred strains, E5 and E7, of common carp, Cyprinus carpio

Natural Resources, University ofl./fax: +98 26 32245909.

Published by Elsevier B.V. All rights

L. (Ruane et al., 2005). Fish from these strains produce significantly lesscortisol during a net confinement stress test than normal carp and headkidneys are usually enlarged and have more and larger steroidproducing cells than head kidneys of normal carp (Ruane et al., 2005).The reduced production of cortisol is caused by a reduced conversionof progesterone into 17α-hydroxy progesterone, a precursor of cortisol(Ruane et al., 2005). This conversion is mediated by 17α-hydroxylase(P450c17). The reduced conversion results in accumulation ofprogesterone and subsequent conversion by 21-hydroxylase intodeoxy-corticosterone and corticosterone (Nematollahi et al., 2012).Corticosterone is a steroid not normally found in teleost fish and itsphysiological effects are unknown (Wendelaar Bonga, 1997).

There are twodistinct P450c17 enzymes, P450c17a1 and P450c17a2.Of these, P450c17a1 is expressed only in the gonad and has both 17α-hydroxylase activity and 17, 20 lyase activity. P450c17a2 possessesonly 17α-hydroxylase activity and is required for the synthesis ofcortisol in the head kidney and is expressed in the testes and ovaries(Nematollahi et al., 2009; Zhou et al., 2007). The mRNA levels ofP450c17a2 are significantly lower in E5 head kidney tissue compared

reserved.

166 M.A. Nematollahi et al. / Aquaculture 418–419 (2014) 165–170

to normal carp. This suggests that adrenal hyperplasia is caused by areduced expression of P450c17a2, and not by a functional mutation inthe gene itself (Nematollahi et al., 2012).

At present it is not knownwhether this case of adrenal hyperplasia isheritable as data are only available from inbred strains. Recently weshowed that carp from inbred strains E5 and E7 produce copiousamounts of corticosterone while fish from the inbred strain E4 and theF1 cross E4xE5 do not (Nematollahi et al., 2012). Using corticosteroneas a marker trait, the first objective of the present study was toinvestigate the inheritance of adrenal hyperplasia in backcross anddoubled haploid progeny of E5 and E7 carp.

Fish from the E5 and E7 strain are female but change their sex tomale during early gonadal differentiation (Komen et al., 1992a,1992b). This female tomale sex reversal originally led to the postulationof a recessive mutation in a putative autosomal sex determining gene,termedmas−1. In this model, homozygous XX (mas−1/mas−1) animalsdevelop as males (e.g. E5, E7) while heterozygous XX, (mas+/mas−1)animals develop as normal females. Normal XY males are not affected(Komen et al., 1992a, 1992b). The fact that E5 carp are sex reversedXX males, and have adrenal hyperplasia, suggests that these twophenomena might be related. The second objective of this study wastherefore to examine the co-segregation of high corticosterone, amarker for adrenal hyperplasia, with female to male sex reversal.

2. Materials and methods

2.1. Animals

The homozygous inbred strains used in this experiment were E4(XX, female, normal plasma cortisol response after stress), E5 and E7(XX, male, adrenal hyperplasia, low plasma cortisol response afterstress). Both E5 and E7 clones are reproduced by androgenesis(Komen et al., 1991). Heterozygous E4E5 animals had been producedearlier by conventional crosses between E4 and E5. In order to producethe E4E5 × E5 and E4E5 × E7 backcross progeny, the mature E4E5animals were backcrossed with E5 and E7 individuals (BC5 and BC7).These E4E5 females were also reproduced by homozygous gynogenesisto produce doubled haploids (DH; Komen and Thorgaard, 2007).

All crosses E4 × E5, BC5 and BC7 were produced by conventionalfertilization techniques by mixing freshly ovulated eggs with undilutedmilt before adding tap-water for activation and fertilization. Doubledhaploid (DH) offspring were produced by fertilizing eggs withirradiated sperm and giving eggs a heat shock (at 40 °C for 2 min,30 min after fertilization) at the prometaphase of the first mitosis, toinduce endomitosis (EM gynogenesis: Komen et al., 1991). Fish larvaefrom BC5, BC7 and DH groups were reared in 25-l tanks and fed withfreshly hatched Artemia nauplii 4–5 times a day for the first 3–4 weeks. After this period fish were distributed in groups of 100 fishover six experimental rectangular tanks (two tanks/genetic group)with a water volume of 120 l, and a water flow of 5 l/min. Fish werefed with pelleted food (Provimi) at a feeding level of 20g/kg metabolicfish weight per day (g/kg0.8/d) until 120 dph (days post hatching) andthen 10 g/kg0.8/d until 200 dph. All tanks were connected to the samerecirculation system equipped with a biofiltration unit and received aconstant flow of O2 saturated water at 25 °C under a 12-h light/darkphotoperiod. NH4

+ and NO2 were monitored daily and never exceeded1 ppm; pH was maintained at 6.5. Stocking density was decreased to70 fish per tank on 100dph and to 48 fish on 150dph.

2.2. Experimental procedure

The net confinement stressor was performed according to theprotocol developed by Ruane et al. (2001). Groups of 24 fish werecaught in one sweep and distributed over five 5-l nets (four to five fishper net) suspended in a confinement tank (140 l). Confinement lastedfor 1 h, after which all fish were sampled for analysis. Confinement

tanks received a flow of fresh water at 5 l/min and were well aeratedto avoid oxygen shortage in the nets.

In total 96 BC5 fish, 48 BC7 fish and 96 DH fish were sampled. Toavoid sampling effect of one group of fish on cortisol response ofremaining fish in the same tank, repeated sampling from the sametank was done on different days. During the experiment, tanks weredisconnected from the recirculating system to avoid cross-over effectsof cortisol and other metabolites released in the water.

In addition to the main experiment, we also confined and sampled10 E4E5 females, five E5 males and five E7 males (age 9 months to1 year) to determine stress levels of cortisol and corticosterone in theparental strains.

2.3. Blood sampling and steroid measurements

After 1 h net confinement fish were removed from the tank andquickly anesthetized with a mixture of tricaine methane sulfonate andsodium bicarbonate (3 g MS222 + 6 g NaHCO3/10 l water, CrescentResearch Chemicals). Blood was collected by heparin-saturatedhypodermic syringe from the caudal blood vessel within 3 min persample group. The collected blood was placed into cooled 1.5ml plastictubes containing 3mg Na2EDTA, mixed and centrifuged (2400 g; 4 °C)for 5 min. The collected plasma was stored at −20 °C for furtheranalyses.

Plasma cortisol was measured using an enzyme-linked immuno-sorbent assay (ELISA) kit (Neogen, Lexington, KY) validated forcommon carp plasma. Ten microliters (μl) of plasma was diluted with990 μl of EIA Buffer which was provided to dilute enzyme conjugateand cortisol standards. The reaction was measured with a microplatereader (Lab systems Multiskan MS, Diagnostic Merlin system BV) witha dual wave length at 450 nm and 690 nm. Cross reactivity of theantiserum to other steroids in the cortisol production pathway are:cortisone (8.1%), 11-deoxycorticosterone (15%), corticosterone (4.5%),progesterone (0.1%), 17α-hydroxy progestrone (1%), 17α-hydroxypregnenolone (1.1%).

Plasma corticosterone was measured using an ELISA kit (AssayDesigns, Ann Arbor, MI) validated for common carp plasma. Plasma wasdiluted 1:100 with a mixture of assay buffer and Steroid DisplacementReagent. The reaction was measured with a micro plate reader as before,using a dual wave length at 420 nm and 570 nm. Cross reactivity of theantiserum to other steroids in the corticosterone production pathwayare: 11-deoxycorticosterone (19%), Progesterone (0.4 %), 17α-hydroxyprogestrone (1%), 17α-hydroxy pregnenolone (1.1%). cortisol (4.5%),cortisone (0.01%). All assays were performed in duplicate (Nematollahiet al., 2012).

2.4. Determination of sex ratios

After blood collection, fish were killed with a mixture of tricainemethane sulfonate and sodium bicarbonate (6 g MS222 + 12 gNaHCO3/10 l water, Crescent Research Chemicals), standard lengthand weight of fish were measured and then the fish were dissected.

At the age of 200 dph common carp males contain well-developedwhite testes with mature spermatogonia; females have pale reddish,transparent ovaries with numerous vitellogenic oocytes. Intersexgonads are clearly identifiable as gonads having both ovarian tissueand white testicular lobes (Fig. 1). Due to the large size, a precisequantification of the areas of testicular and ovarian tissue is difficult.Therefore gonads were scored by visual inspection using five differentclasses: 1) testis, 2) IS-95; intersex gonads with more than 95% testistissue, 3) IS-50 gonads with more or less equal amount of ovarian andtesticular tissue, 4) IS-10 ovaries with less than 10% testis tissue, and5) ovaries. Filliform gonads were scored as sterile. A typical exampleof an IS-10 and IS-50 gonad is shown in Fig. 1.

Fig. 1.Carp gonads showing different degrees of intersexuality. Left: intersex gonad consistingmainly of ovarian tissuewith oocytes (orange) and isolated lobes of testicular tissue (white).This gonadwas scored as IS-10. Right: carp gonadswith bilateral intersexual gonads. Each gonad is composed of one ovarian sidewith oocytes (orange) and one sidewith testicular tissue(white). This gonad was scored as IS-50.

45L < 17 ng/ml

H > 800 ng/ml

167M.A. Nematollahi et al. / Aquaculture 418–419 (2014) 165–170

2.5. Data analysis

Due to a very clear segregation of corticosterone values, fish weregrouped in high (H) and low (L) corticosterone responders. A chi-square test for goodness of fit was used to compare observed andexpected frequencies of H and L numbers in each genetic group.

Data were analyzed using general linear model to determinedifferences between H and L fish in BC5, BC7 and DH groups. The effectsof corticosterone (L andH)nestedwithin strain (1, 2, and 3), sample day(1, 2 and 3), tank number (1–6), and body weight on cortisol valueswere tested for significance. Analysis of variance was performed usingproc GLM (SAS Institute, 1989) and the least significant factors wereremoved from the model in a stepwise manner until all remainingeffects were significant (P b 0.05). The final model used in for theanalysis was:

Yijk ¼ μ þ St: j CSð Þkþ β1� LogCSijkþ ε ijk ð1Þ

where, Yijk is an observation of cortisol, μ is the overallmean, St.j (CS)k iseffect of the kth corticosterone responder group (1, 2) nested within thejth strain (1, 2, 3), β1 is the regression coefficient of Log Corticosterone,LogCS ijk is observation of corticosterone, ε ijk is the random residualterm. Least square means (LSM) were calculated for the fixed effectclasses using the LSMEANS option of proc GLM.

3. Results

3.1. Steroid measurements

Plasma cortisol and corticosterone levels of E4E5 fish after 1hourconfinement were 177.6 ± 56.6 ng/ml and 2.6 ± 1.2 ng/ml,respectively. Cortisol levels for E5 fish were 22.02±4.18 ng/ml andfor E7 fish 39.86 ± 13.37 ng/ml. Corticosterone levels for E5 were1585.60±214.91 ng/ml and for E7 1339.04±92.70 ng/ml.

Table 1Mean (±SD) of corticosterone (ng/ml) and cortisol (ng/ml) in backcross (BC5, BC7) anddoubled haploid (DH) progeny groups of common carp (Cyprinus carpio L.).

Progeny group Corticosterone Cortisol

N Mean SD Mean SD

BC5 L 54 1.5 2.9 66.7 46.9H 42 1706 393 21.3 10.2

DH L 51 0.6 0.6 65.2 39.7H 45 1366 275 17.3 7.1

BC7 L 26 1.1 0.8 105 43.0H 22 1760 360 24.0 13.7

H=high corticosterone responders, L= low corticosterone responders.

Results of plasma corticosterone levels in BC5, doubled haploids andBC7 fish after 1hour confinement are shown in Table 1 and Fig. 2.

Corticosterone values could be classified into two groups: High (H)with a range of 832–3003 ng/ml and low (L) with a range of 0.14–16.70 ng/ml. The two distributions are not overlapping. Observedfrequencies of H and L fish were not significant different from anexpected 1:1 ratio in each of the three groups, using a χ2-test forgoodness of fit (P b 0.05, df 1). There were no significant differencesbetween strains for either high or low responding fish.

Results of plasma cortisol levels in BC and DH groups are shown inTable 1 and Fig. 3. Plasma cortisol values showed a more or lesscontinuous distribution in each group (not shown), but mean plasmacortisol values of high corticosterone responders were significantlylower than plasma cortisol values in low corticosterone responders(Fig. 3).

There was a significant effect of strain on plasma cortisol (Table 2).For H fish, mean values differed from 17.3 ± 7.1 in DH to 24.0± 13.7in BC7 groups; for L fish, mean values ranged from 65.2 ± 39.7 in DHto 105 ± 43 in BC7 groups. Type III mean squares of the effects inModel (1) are summarized in Table 2. All effects were highly significant(P b 0.05). Corticosterone and strain together explain 50% of thevariation in cortisol values.

3.2. Sex ratios

Females in general had clearly distinguishable pale reddish, trans-parent ovaries, with numerous small vitellogenic oocytes. Males hadwell developed white testes which contained mature sperm. Intersexfish had gonads composed of white opaque areas (testicular tissue)and pale transparent areas (ovarian tissue). Intersex gonads withmore than 95% testicular tissue resembled testes except for a small

0

5

10

15

20

25

30

35

40

-0.5 0 0.5 1 1.5 2 2.5 3 3.5

Log Corticosterone

No.

Fis

h

Fig. 2. Frequency distribution of plasma corticosterone (Log ng/ml) levels in BC5 (□), DH( ) and BC7 (■) progeny groups.

0

20

40

60

80

100

120

140

160

E45xE5 E45(DH) E45xE7

Cor

tisol

(ng

/ml)

(22)

(45)(42)

(26)

(51)(54)

A

BBB

AA

Fig. 3. Plasma cortisol (ng/ml) of low (□) and high ( ) corticosterone responders inBC5, BC7 and DH progeny groups of common carp. Values are mean±standard deviation(N in parenthesis); different letters indicate a significant difference (P b 0.001) between Land H responders from the same progeny group.

Table 3Percentage of males, intersex and females in H and L corticosterone responders in BC andDH progeny groups of common carp, Cyprinus carpio L.

Progeny group N Male IS-95 IS-50 IS-10 Female ST

BC5 L 54 7.4 59.3 11.1 14.8 7.4 0H 42 97.6 2.4 0 0 0 0

DH L 51 2 11.8 17.6 9.8 45.1 13.7H 45 51.1 26.7 11.1 2.2 4.4 4.4

BC7 L 26 7.7 42.3 3.8 11.5 30.8 3.8H 22 86.4 13.6 0 0 0 0

H = high corticosterone responder, L = low corticosterone responder; IS-95: intersexgonad with predominantly testicular tissue and some oocytes; IS-50: intersex gonadwith more or less equal amounts of testicular and ovarian tissue; IS-10: intersex gonadwith predominantly ovarian tissue and small areas of testicular tissue; ST: filliformgonad, presumably sterile.

168 M.A. Nematollahi et al. / Aquaculture 418–419 (2014) 165–170

cleft running the length of the gonad. Sex ratioswere grouped accordingto high or low corticosterone levels (Table 3), and tested for significanceusing a χ2-test for goodness of fit. The effect of corticosterone onmaleness was significant (P b 0.05) in all progeny groups. Fish withhigh corticosterone levels in BC5 and BC7 groups were almostexclusively male. Low corticosterone BC progeny groups containedlarge numbers of intersex fish with more than 95% testis development.In BC5, only few females (7.4%) were observed which were all lowcorticosterone fish. The BC7 group contained 30.8% females, again onlyamong the low corticosterone fish. The highest numbers of femalesand lowest number of males were observed in low corticosterone fishof the DH group: 45.1% and 2%, respectively. In this progeny group theeffect of corticosterone on maleness was highly significant (P b 0.01):in the high corticosterone group 4.4% females and 51.1% males wereobserved.

4. Discussion

In humans, CAH caused by P450c17 deficiency is relatively rare.P450c17 is responsible for the conversion of progesterone into 17α-hydroxyprogesterone, which is subsequently converted into cortisol.When17α-hydroxylase is impaired, the excess progesterone is convertedby 21β-hydroxylase into deoxy-corticosterone and corticosterone.Patients with P450c17 deficiency have decreased ability to producecortisol and elevated ACTH (adrenocorticotropic hormone) causesincreased serum levels of corticosterone; this feature is used in hormonaldiagnosis (New, 1998). In this study, both parental strains E5 and E7, andapproximately half of the backcross and DH haploid progenies had veryhigh values of plasma corticosterone. Thisfinding corroborates our earlierobservation that 17α-hydroxylase deficient fish have high plasmacorticosterone and as such share many of the characteristics as seen inhigher vertebrates (Nematollahi et al., 2012; Ruane et al., 2005). Thevery high levels of corticosterone observed in common carp suggestthat corticosterone turn-over rates are low. This might explain why fishwith even moderate levels of cortisol, could be identified by their

Table 2Type III mean squares (MS), degree of freedom (DF) and significance of effects for plasmacortisol levels (Model (1)).

Source DF MS

Log corticosterone 1 5264.9⁎

Strain (CS) 5 12055.4⁎⁎⁎

Error 227 1022R2 0.50

⁎ P b 0.05.⁎⁎⁎ P b 0.001.

corticosterone levels. The levels of cortisol observed in this study werehighly variable and comparable to those found in other studies on carp(Ruane et al., 2001; Tanck et al., 2000). However, grouping of fish inhigh and low corticosterone responders showed that the two groupsindeed have different mean cortisol responses but with overlappingdistributions. This is in contrast with earlier studies, using clonal strains(Ruane et al., 2005, 2007; Tanck et al., 2002) where all fish of the sameclone showed a fairly homogenous response to confinement stress. Inthis study, backcrosses and DH progenies were tested, i.e. groups of fishwhich were not genetically uniform, but segregating for heterozygousmaternal alleles. The large observed variation in cortisol responsesmight be due to the increased genetic variation in these backcross andDH progenies as was reported earlier by Tanck et al. (2001).

The production of high levels of corticosterone as a marker for 17α-hydroxylase deficiency is inherited as a monogenic recessive trait. First,both E4 and E4E5 fish produce normal physiological levels of cortisoland corticosterone after stress. Second, both backcross progeniesof E4E5 fish with either E5 or E7 (high corticosterone producers)contained nearly 50% fish which exhibited a normal stress responseand 50% fish which had very high plasma corticosterone levels afterstress. Furthermore, high and low corticosterone producers were alsofound in the DH group, again in equal numbers. When E4E5 fish areheterozygous for the recessive trait they should produce DH progenywhich are homozygous for either the wild type allele or the mutantallele. Together, these observations show that 17α-hydroxylasedeficiency is inherited as a single recessive trait, with Mendelian ratiosin backcross and DH offspring.

At present, we do not knowwhere themutation responsible for 17α-hydroxylase deficiency is located. The expression of P450c17a2 mRNA isreduced in head kidneys of E5 fish (Nematollahi et al., 2012). Thisindicates that the genetic defect is in the promoter or one of the controlelements of P450c17a2 expression. Transcript levels of P450c17 enzymesare regulated in a tissue and temporal specific pattern. In humans, ACTHbinds to melanocortin-2 receptor which activates a signaling cascadewhereby multiple transcription factors, including steroidogenic factor-1(SF-1), GATA-6, and sterol regulatory binding protein 1 (SREBP1), arerecruited to the promoter during activated transcription (Peter andDubuis, 2007; Sewer and Jagarlapudi, 2009). Head kidneys of E5 carprespond to ACTH (Ruane et al., 2005) which indicates that themelanocortin-2 receptor for ACTH (Metz et al., 2005) is functional(Greenwood et al., 2003). However, the transcription factors thatsubsequently bind to the promoter and control P450c17a2 expressionin common carp are not known.

The most important finding in this study was the observation thathigh corticosterone responders were predominantly male, while lowresponders were mainly intersex and female. This raises the questionwhether the two traits (i.e. “high corticosterone” and “male sexreversal”) are genetically identical or connected through pleiotropiceffects of the mutation responsible for 17α-hydroxylase deficiency.

Sex determination in common carp is primarily controlled by a, asyet unidentified, male dominant gene, termed “Y”, as in mammals

169M.A. Nematollahi et al. / Aquaculture 418–419 (2014) 165–170

(Bongers et al., 1999; Komen and Thorgaard, 2007). Males areheterozygous “XY” while females are homozygous recessive “XX”. Sexdetermination is labile however (Matsuda, 2003), and sex reversal canbe induced by endocrine intervention (Gimeno et al., 1996) and other,as yet unknown, environmental factors (reviewed in Devlin andNagahama, 2002). The present case of sex reversal was first describedin 1992, as being caused by a putative recessivemutationmas−1 causingfemale to male sex reversal in homozygous condition (Komen et al.,1992a, 1992b). The inheritance of this mutation was described in thesame clones and backcrosses as used in the present study. Crossing amas+/mas−1 (E4E5) female with a homozygous mas−1/mas−1 (E5)male resulted in approximately 50% male mas−1/mas−1 and 50%(pooled) intersex fish and females (mas+/mas−1). The results fromthe present study corroborate these earlier observations: all lowcorticosterone groups contained high numbers of intersexes havinggonads with more than 95% testis tissue indicating that also normalanimals were sensitive to sex reversal.

The most likely explanation is that the mutation responsible for17α-hydroxylase deficiency causes a shift towards maleness throughpleiotropic endocrine (paracrine) effects (Layman, 2002; Sato et al.,2008). First, corticosterone could be directly involved in sex reversal.High yolk concentrations of corticosterone, have been shown to affectoffspring sex ratios in quail (Pike and Petrie, 2006), and recently itwas shown that corticosterone injections in yolk increased the malesex ratio in Bassiana duperreyi, a lizard species with temperaturedependent sex determination (Warner et al., 2009). However, in thesespecies, corticosterone is the main glucocorticoid and has the samephysiological role as cortisol in teleost fish. There are no reports oninduced sex reversal with (deoxy) corticosterone in teleost fish. Inteleost species, cortisol is the main glucocorticoid produced and highlevels of cortisol have been shown to cause masculinization in rainbowtrout (Van denHurk and vanOordt, 1985), and in pejerrey and Japaneseflounder, two species with temperature-dependent sex determination(Hattori et al., 2009; Yamaguchi et al., 2010). Furthermore, cortisoltreated animals had elevated 11-keto testosterone and testosteronelevels and showed down regulation of cyp19-1a (aromatase).

In our carp model, sex reversal is associated with low levels ofcortisol. However, fish with 17α-hydroxylase deficiency have headkidneys that have more and larger steroid producing cells than headkidneys of normal carp (Ruane et al., 2005). A more likely explanationis therefore that ACTH, the factor causing a proliferation and hyperplasiaof corticosteroidogenic cells in the 17α-hydroxylase deficient headkidney, by paracrine action, causes a reduction of estrogen productionand increased androgen production during sex differentiation, therebyaugmenting the sex reversal of the genetically female gonad (Yu et al.,2003). In carp, early developing gonads are directly adjacent to theinterrenal tissue of the mesonephros and male sex reversal in femalecarp typically starts with the appearance of small spermatogenic cystsat an age when oocytes first enter meiotic prophase (Komen et al.,1992a, 1992b). In mice, adrenocortical cells and fetal Leydig cellspresumably share a common origin in an adreno-genital primordium(O'Shaughnessy et al., 2006), and fetal Leydig cells produce testosteronein response to ACTH challenge (O'Shaughnessy et al., 2003). In Rainbowtrout, both ovaries and testes express MCR-2, the receptor for ACTH(Aluru and Vijayan, 2008). In zebrafish, Alsop et al. (2009) showed thatACTH suppressed hCG-stimulated estradiol secretion from ovarianfollicles. In zebrafish and in Nile tilapia, the expression of the twoP450c17a enzymes in adrenals and gonads is probably controlled bydifferent transcription factors,which play a decisive role in the regulationof steroid synthesis (Ijiri et al., 2008;Wang andGe, 2004) Carpwith 17α-hydroxylase deficiency have well developed testes and are fully fertile,indicating that 17α-hydroxylase is functional in gonads and supportingthe idea of differential expression in adrenals and gonads; hence theelevated ACTH levels could cause increased production of 11 hydrox-ylated androgens in themas−1 gonad and simultaneously suppress FSHstimulated estrogen production. Experiments to describe the ontogeny

of steroid production in the developing carp gonad are currently inprogress.

In conclusion,we showed that 17α-hydroxylase deficiency in carp isheritable, and that affected individuals show all the classical symptomsof congenital adrenal hyperplasia as seen in humans, including lowamount of plasma cortisol after stress and high levels of plasmacorticosterone. Results suggest that 17α-hydroxylase deficiency causessex reversal through paracrine action on steroidogenic cells in thedeveloping gonad. The results from the carp model may help inunderstanding the role of chronic stress in the ontogeny of male-biased sex ratios as frequently observed in other species in aquaculture,e.g. in Senegalese sole (Solea senegalensis; Blanco-Vives et al., 2011) andseabass (Mylonas et al., 2005).

Acknowledgments

We would like to thank R. Booms, M. ter Veld and S. Leenstra fortheir assistance during fish rearing, experimentation and sampling.M.A. Nematollahiwas supported by a grant from theMinistry of Science,Research and Technology of the Islamic Republic of Iran.

References

Alsop, D., Ings, J., Vijayan, M.M., 2009. Adrenocorticotropic hormone suppressesgonadotropin-stimulated estradiol release from zebrafish follicles. PLoS ONE 4 (7),e64663.

Aluru, N., Vijayan, M.M., 2008. Molecular characterization, tissue-specific expression, andregulation of melanocortin-2 receptor in rainbow trout. Endocrinology 149,4577–4588.

Blanco-Vives, B., Vera, L.M., Ramos, J., Bayarri, M.J., Mañanós, E., Sánchez-Vázquez, F.J.,2011. Exposure of larvae to daily thermocycles affects gonad development, sexratio, and sexual steroids in Solea senegalensis, Kaup. J. Exp. Zool. 315A, 162–169.

Bongers, A.B.J., Zandieh Doulabi, B., Richter, C.J.J., Komen, J., 1999. Viable androgenetic YYgenotypes of common carp, Cyprinus carpio L. J. Hered. 90, 195–198.

Deaton, M.A., Glorioso, J.E., McLean, D.B., 1999. Congenital adrenal hyperplasia: not reallya zebra. Am. Fam. Physician 60, 1107.

Devlin, R.H., Nagahama, Y., 2002. Sex determination and sex differentiation in fish: anoverview of genetic, physiological, and environmental influences. Aquaculture 208,191–364.

Gimeno, S., Gerritsen, A., Bowmer, T., Komen, J., 1996. Feminization of male carp. Nature384, 221–222.

Gotoh, H., Sagai, T., Hata, J., Shiroishi, T., Moriwaki, K., 1988. Steroid 21-hydroxylasedeficiency in mice. Endocrinology 123, 1923–1927.

Greenwood, A.K., Butler, P.C., White, R.B., Demarco, U., Pearce, D., Fernald, R.D., 2003.Multiple corticosteroid receptors in a teleost fish: distinct sequences, expressionpatterns, and transcriptional activities. Endocrinology 144, 4226–4236.

Hall, C.M., Jones, J.A., Meyer-Bahlburg, H.F.L., Dolezal, C., Coleman, M., Foster, P., Price, D.A.,Clayton, P.E., 2004. Behavioural and physical masculinization are related to genotypein girls with congenital adrenal hyperplasia. J. Clin. Endocrinol. Metab. 89, 419–424.

Hattori, R.S., Fernandino, J.I., Kishii, A., Kimuar, H., Kinno, T., Oura, M., Somoza, G., Yokota,M., Strussmann, C.A., Watanabe, S., 2009. Cortisol-induced masculinisation: doesthermal stress affect gonadal fate in pejerrey, a teleost fish with temperature-dependent sex determination? PLoS ONE 4 (8), e6548.

Ijiri, S., Kaneko, H., Kobayashi, T., Wang, D., Sakai, F., Paul-Prasanth, B., Nakamura, M.,Nagahama, Y., 2008. Sexual dimorphic expression of genes in gonads during earlydifferentiation of a teleost fish, the Nile tilapia Oreochromis niloticus. Biol. Reprod.78, 333–341.

Komen, J., Thorgaard, G.H., 2007. Androgenetic and gynogenetic clones of fishes: a review.Aquaculture 269, 150–173.

Komen, J., Bongers, G., Richter, C.J.J., Huisman, E.A., 1991. Gynogenesis in common carp(Cyprinus carpio L.): II. The production of homozygous gynogenetic clones and F1hybrids. Aquaculture 92, 127–142.

Komen, J., Yamashita, M., Nagahama, Y., 1992a. Testicular development induced by arecessive mutation during gonadal differentiation of female common carp. Develop.Growth Differ. 34 (5), 535–544.

Komen, J., de Boer, P., Richter, C.J.J., 1992b. Male sex reversal in gynogenetic XX females ofcommon carp (Cyprinus carpio L.) by a recessive mutation in a sex-determining gene.J. Hered. 83, 431–434.

Layman, L.C., 2002. Human genemutations causing infertility. J. Med. Genet. 39, 153–161.Matsuda, M., 2003. Sex determination in fish: lessons from the sex determining gene of

the teleost medaka, Oryzias latipes. Develop. Growth Differ. 45, 397–403.Metz, J.R., Geven, E.J.W., van den Burg, E.H., Flik, G., 2005. ACTH, α-MSH and control of

cortisol release: cloning, sequencing and functional expression of the melanocortin-2 and melanocortin-5 receptor in Cyprinus carpio, L. Am. J. Physiol. 289, 814–826.

Migeon, C.J., Donohoue, P.A., 1991. Congenital adrenal hyperplasia caused by 21-hydroxylase deficiency. Endocrinol. Metab. Clin. North Am. 20, 277–296.

Milano, E.G., Basari, F., Chimenti, C., 1997. Adrenocortical and adrenomedullary homologsin eight species of adult and developing teleosts: morphology, histology andimmunochemistry. Gen. Comp. Endocrinol. 108, 483–496.

170 M.A. Nematollahi et al. / Aquaculture 418–419 (2014) 165–170

Miller, W., 1991. Congenital adrenal hyperplasia. Endocrinol. Metab. Clin. North Am. 20,721–749.

Mylonas, C.C., Anezaki, L., Divanach, P., Zanuy, S., Piferrer, F., et al., 2005. Influence ofrearing temperature during the larval and nursery periods on growth and sexdifferentiation in two Mediterranean strains of Dicentrarchus labrax. J. Fish Biol. 67,652–668.

Nematollahi, M.A., Van Pelt-Heerschap, H., Komen, J., 2009. Transcript levels of fiveenzymes involved in cortisol production in common carp: relationship with cortisol.J. Gen. Comp. Endocrinol. 164, 85–90.

Nematollahi, M.A., Van Pelt-Heerschap, H., Atsma, W., Komen, J., 2012. High level ofcorticosterone, and gene expression of star, cyp17a2, hsd3b, cyp21, hsd11b2 duringacute stress in common carp with interrenal hyperplasia. J. Gen. Comp. Endocrinol.176, 252–258.

New,M.I., 1998. Diagnosis andmanagement of congenital adrenal hyperplasia. Annu. Rev.Med. 49, 311–328.

O'Shaughnessy, P.J., Fleming, L.M., Jackson, G., Hochgeschwender, U., Reed, P., Baker, P.J.,2003. Adrenocotricotropic hormone directly stimulates testosterone production bythe fetal and neonatal mouse testis. Endocrinology 144, 3279–3284.

O'Shaughnessy, P.J., Baker, P.J., Johnston, H., 2006. The foetal Leydig cell— differentiation,function and regulation. Int. J. Androl. 29, 90–95.

Pang, S., Yang, X., Wang, M., Tissot, R., Nino, M., Manaligod, J., Bullock, L., Mason, J., 1992.Inherited congenital adrenal hyperplasia in the rabbit, absence cholesterol side-chaincleavage cytochrome P450 gene expression. Endocrinology 131, 181–186.

Peter, M., Dubuis, J., 2007. Transcription factors as regulators of P-450 enzymes. Eur.J. Clin. Invest. 30, 14–20.

Pike, T.W., Petrie, M., 2006. Experimental evidence that corticosterone affects offspringsex ratios in quail. Proc. R. Soc. B 273, 1093–1098.

Ruane, N.M., Huisman, E.A., Komen, J., 2001. Plasma cortisol and metabolite levelprofiles in two isogenic strains of common carp during confinement. J. Fish Biol.59, 1–12.

Ruane, N.M., Lambert, J.G.D., Goos, H.J.T., Komen, J., 2005. Hypocorticism and adrenalhyperplasia are not directly related to masculinization in XX mas−1/mas−1carp,Cyprinus carpio. Gen. Comp. Endocrinol. 143, 66–74.

Ruane, N.M., Goos, H.J.T., Komen, J., 2007. Stress induced facilitation of the cortisol responseoccurs in 17α-hydroxylase deficient, but not in normal carp (Cyprinus carpio, L.). Gen.Comp. Endocrinol. 150, 473–479.

SAS Institute, 1989. User's Guide; Statistics, Version 6.12 Edition. SAS institute, Cary, NC.Sato, T., Suzuki, A., Shibata, N., Sahaimuzi,M., Hamaguchi, S., 2008. The novelmutant scl of

themedaka,Oryzias latipes, shows no secondary sex characters. Zool. Sci. 25, 299–306.Sewer, M.B., Jagarlapudi, S., 2009. Complex assembly on the human CYP17 promoter. Mol.

Cell. Endocrinol. 300, 109–114.Tanck, M.W.T., Booms, G.H.R., Eding, E.H., Wendelaar Bonga, S.E., Komen, J., 2000. Cold

shocks: a stressor for common carp. J. Fish Biol. 57, 881–894.Tanck, M.W.T., Claes, T., Bovenhuis, H., Komen, J., 2002. Exploring the genetic background

of the stress response using isogenic progenies of common carp selected for high orlow stress related cortisol. Aquaculture 204, 419–434.

Van denHurk, R., van Oordt, P.G.W.J., 1985. Effects of natural androgens and corticosteroidson gonad differentiation in the rainbow trout (Salmo gairdneri). Gen. Comp. Endocrinol.57, 216–222.

Wang, Y., Ge,W., 2004. Cloningof zebrafish ovarian P450c17 (CYP17, 17α-hydroxylase/17,20-lyase) and characterization of its expression in gonadal and extra-gonadal tissues.Gen. Comp. Endocrinol. 135, 241–249.

Warner, D.A., Radder, R.S., Shine, R., 2009. Corticosterone exposure during embryonicdevelopment affects offspring growth and sex ratios in opposingdirections in two lizardspecies with environmental sex determination. Physiol. Biochem. Zool. 82, 363–371.

Wendelaar Bonga, S.E., 1997. The stress response in fish. Physiol. Rev. 77, 591–625.Yamaguchi, T., Yoshinaga, N., Yazawa, Y., Gen, K., Kitano, T., 2010. Cortisol is involved in

temperature-dependent sex determination in the Japanese flounder. Endocrinology151 (8), 3900–3908.

Yu, H., Cheng, H., Guo, Y., Xia, L., Zhou, R., 2003. Alternative splicing and differentialexpression of P450c17 (CYP17) in gonads during sex transformation in the ricefield eel. Biochem. Biophys. Res. Commun. 307, 165–171.

Zhou, L., Wang, D., Kobayashi, T., Yano, A., Paul-Prasanth, B., Suzuki, S., Sakai, F.,Nagahama, Y., 2007. A novel type of P450c17 lacking the lyase activity is responsiblefor c21-steriod synthesis in the fish ovary and head kidney. Endocrinology 148,4282–4291.