(INSL3) but not testosterone is reduced in ... - Wiley Online Library

Clinical Endocrinology (2015) 82, 242–247

doi: 10.1111/cen.12500

ORIGINAL ARTICLE

Cord blood Insulin-like peptide 3 (INSL3) but not testosterone is reduced in idiopathic cryptorchidism nichel*,†,1, Najiba Lahlou‡,1, Patrick Coquillard§, Patricia Pana€ıa-Ferrari¶, Kathy Wagner-Mahler** Patrick Fe and Francßoise Brucker-Davis*,† *Department of Endocrinology, Diabetology and Reproductive Medicine, CHU Nice, †Institut National de la Recherche Medicale, UMR U1065, Universite Nice-Sophia Antipolis, Nice, ‡Department of Hormonology and Metabolic Disorders, H^opital Cochin, APHP, Paris-Descartes University, Paris, §Institut Sophia-Agrobiotech, [INRA-CNRS- Nice University], Sophia-Antipolis, ¶Department of Biochemistry, CHU Nice, and **Department of Paediatrics, CHU Nice, Nice, France

descent along with genetic and anatomical factors. Whether foetal environment (nutritional and/or toxicological) interferes with INSL3 secretion in humans remains to be confirmed.

Summary Background Cryptorchidism, the most frequent congenital malformation in full-term male newborns, increases the risk of hypofertility and testicular cancer. Most cases remain idiopathic but epidemiological and experimental studies have suggested a role of both genetic and environmental factors. Physiological testicular descent is regulated by two major Leydig hormones: insulin-like peptide 3 (INSL3) and testosterone. Objectives To study the endocrine context at birth as a reflection of late pregnancy in isolated idiopathic cryptorchidism and to analyse the possible disruptions of INSL3 and/or testosterone. Methods From a prospective case–control study at Nice University Hospital, we assessed 180 boys born after 34 weeks gestation: 52 cryptorchid (48 unilateral, 4 bilateral; 26 transient, 26 persistent), and 128 controls matched for term, weight and time of birth. INSL3 and testosterone were measured in cord blood and compared in both groups as were other components of the pituitary-gonadic axis: LH, HCG, FSH, AMH and SHBG. Results INSL3 was decreased in cryptorchid boys (P = 0031), especially transient cryptorchid (P = 0029), while testosterone was unchanged as were the other hormones measured. INSL3 was significantly decreased (P = 0018) in the group of 20 with nonpalpable testes compared with the group of 21 with palpable testes (15 suprascrotal, five inguinal, one high scrotal) according to Scorer classification. In the whole population, INSL3 correlated positively with LH and negatively with AMH, but with no other measured hormones. Conclusions INSL3 but not testosterone is decreased at birth in idiopathic cryptorchidism, especially in transient forms. This hormonal decrease may contribute to the impaired testicular

Correspondence: Patrick Fenichel, MD, PhD, Department of Endocrinology, Diabetology and Reproductive Medicine, l’Archet Hospital 2, CHU Nice, 151 route de Saint-Antoine, 06200 Nice, France. Tel.: +33 4 92 03 55 19; Fax: +33 4 92 03 54 25; E-mail: [email protected] 1

Contribution of both authors was equivalent.

242

(Received 10 January 2014; returned for revision 2 February 2014; finally revised 14 April 2014; accepted 9 May 2014)

Introduction Undescended testis (UDT), also called cryptorchidism, is the most frequent congenital malformation in males, occurring in 2–5% of full-term male births.1–3 In young adults, it is the bestcharacterized risk factor for male infertility and testicular cancer.4,5 With the exception of complex syndromes with multiple congenital abnormalities,6 most cases of UDT remain idiopathic.6 Physiological descent of the testes during foetal development7 is directed by the cranial suspensory ligament (CSL) and the gubernaculum, and it proceeds in two successive phases. During the trans-abdominal phase (10–23 weeks gestation in humans), the testis migrates from the uro-genital ridge to the inguinal region, following the regression of the CSL associated with gubernaculum growth. The inguino-scrotal phase occurs between 28 weeks gestation and birth8 due to the regression of the gubernaculum. Two Leydig hormones regulate these two phases, INSL3 and testosterone, as demonstrated in genetically modified rodents.9–11 INSL3, a peptide hormone belonging to the relaxin family, is produced in the testis by the differentiated Leydig cells12 and acts through the relaxin family peptide receptor 2 (RXFP2), which is developmentally expressed in the gubernaculum.13 INSL3 classically regulates the abdominal phase,6 but experimental data support also its participation in the inguinoscrotal phase.4 INSL3 and RFXP2 knockout models present disturbed testis descent resulting in bilateral UDT with severely affected gubernaculum development,9,10 but these genetic syndromes in humans are uncommon.6,14,15 Androgens induce the regression of the gubernaculum during the inguino-scrotal phase,16 which has been well characterized in both animal models and human studies.10,11 In humans, both

© 2014 The Authors. Clinical Endocrinology published by John Wiley & Sons Ltd. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

INSL3 and idiopathic cryptorchidism 243 congenital hypogonadotropic hypogonadism (lack of testosterone) and androgen insensitivity (impaired androgen action)11 are associated with UDT, but they are also very rare.6,15 The incidence of UDT seems to have increased over the past few decades,17–19 although significant geographical differences have been noted,20 while the increased incidence of related diseases such as infertility or testicular cancer is better documented.21,22 One explanation could be the influence of environmental factors, especially chemicals acting as environmental endocrine disruptors (EEDs) during testicular foetal development and descent,16 as Skakkebaeck et al.23 hypothesized for the testicular dysgenesis syndrome. This hypothesis is supported by experimental data in rodents showing that exposure to several EEDs in foetal life can prevent testicular descent by inhibiting INSL3 gene expression24,25 and testosterone production or action.26 However, until now, impaired secretion and/or action of foetal Leydig cell hormones involved in testicular descent has been very seldom reported in human cryptorchidism. Human clinical studies are scarce. Bay et al.27 who led the only study so far on INSL3 observed a decreased INSL3 in idiopathic UDT but unchanged levels of testosterone. Taking advantage of a prospective case–control clinical study, we have performed on UDT in the area of Nice, France, between 2002 and 2005,28,29 during which we collected cord blood samples; we planned to verify and analyse the INSL3 modifications reported by Bay et al.,27 and to measure testosterone with a reference method (UPLC-MS/MS) as well as the other hormones of the pituitary-gonadal axis. INSL3 and testosterone, the two Leydig cell secreted hormones involved in testicular descent, were measured in cord blood to verify whether they were altered during late pregnancy in idiopathic UDT.

Material and methods Study design We designed a 3-year prospective study1,3 and screened newborns for cryptorchidism. The research was approved by the ethics committee of our institution. Between 2002 and 2005, 6246 boys were born alive after 34 weeks gestational age (GA) in the maternity wards of Nice University Hospital and the nearby Grasse General Hospital. Neonatal examination was standardized for the two wards, and diagnosis of cryptorchidism was accepted after at least two concordant examinations by senior paediatricians before hospital discharge. Testicular position was determined according to the Scorer criteria completed by Hack et al.,28 after firm but unforced traction of the testis to the most distal position along the pathway of normal descent. After parental consent was obtained, boys with nonpalpable, inguinal, supra-scrotal, or high scrotal testes were included in the undescended testis (UDT) group. Retractable testes were excluded from both the UDT and control groups. All the boys were examined again at three and twelve months by the same team to determine whether UDT was persistent or transient, as secondary descent

may occur before 3 months in half of the UDT cases diagnosed at birth.8 Population Over the 3-year period, 102 out of 6246 eligible newborn boys were diagnosed with UDT, and 95 of them were included after parental consent was obtained.1 Two control boys, born on the same ward at around the same time, were recruited for each case; they were matched for gestational age, birth weight and, when possible, parental geographical origin. INSL3 immunoassay was not available when the study was designed, but we performed this additional assay later, using samples with sufficient remaining volumes. They were available in 52 of the 95 cryptorchid newborns and 128 of the 188 controls. The boys in this study did not differ from the previously described boys,1 in terms of gestational age, birth weight or other parameters. All 52 newborns were nonsyndromic, isolated cases of UDT with normal karyotype, including 48 unilateral and four bilateral forms, with 26 persistent and 26 transient forms. Clinical characteristics are shown in Table 1. Hormone measurements Cord blood (CB) was drawn prospectively just after delivery, in all male newborn before checking testis position. Cord blood was drawn within minutes after delivery through a metallic needle and a glass syringe into two 10-ml glass tubes, following as closely as possible a standardised arterial blood sampling protocol. Samples were then centrifuged, aliquoted and stored at 70 °C. Steroid and polypeptide hormones were measured in CB. INSL3 was assayed by means of a modified EIA (Phoenix Pharmaceuticals, Belmont, CA, USA).29,30 According to the manufacturer, the antiserum does not exhibit any significant cross-reactivity with other members of the relaxin family or with insulin, C-peptide and inhibin B. The regression curve of observed values in a nest serum pool spiked with increasing amounts of purified INSL3 (62–500 pmol/l) was: y = 13 + 106x. The ordinate intercept does not differ from zero. Sensitivity was 11 ng/l. Intra-assay coefficients of variation at the Table 1. Clinical characteristics of cryptorchid and control boys Cryptorchid Controls Transient n = 26

Persistent n = 26

Total n = 52

n = 128

P

Gestational 385 17 391 17 388 17 391 14 NS age Birth 3145 513 3173 540 3159 526 3220 494 NS weight (g) C section % 244% 187% 215% 201% NS Results are expressed as means SEM. Gestational age at birth is expressed in weeks of amenorrhoea. C section corresponds to the percentage of delivery after Caesarian section.

© 2014 The Authors. Clinical Endocrinology published by John Wiley & Sons Ltd. Clinical Endocrinology (2015), 82, 242–247

244 P. Fenichel et al. levels of 36 and 115 ng/l were 97 and 5%, respectively. Testosterone was measured after diethyl-ether extraction by means of ultrapressure liquid chromatography-tandem mass spectrometry using Waters Acquity-Quattro Premier equipment (Saint-Quentin en Yvelines, France). In this method, there was no cross-reactivity of any steroid. The recovery of testosterone from spiked sera was 102 to 112% from 1 to 35 nmol/l. Serial dilutions of high-level samples paralleled the calibration curve from 3467 to 014 nmol/l (r = 099). The limit of quantitation at 20% CV was 009 nmol/l. The intra-assay coefficients of variation at levels 15, 89 and 359 nmol/l were 63, 39, and 29%, respectively. The inter-assay coefficients of variation at levels 162, 867 and 347 nmol/l were 52, 41 and 50%, respectively. Other hormones assayed in CB included LH, FSH (Bayer Diagnostics, Cergy Pontoise, France), HCG (Bayer Diagnostics), inhibin B (Diagnostic Systems Laboratories, Webster, TX, USA), AMH (Immunotech, Marseille, France) and SHBG (Biomerieux, Lyon, France). Statistical analysis Because of sample volume limitations, an exhaustive study of the whole cohort1 was not possible. However, the subgroup analysed here did not differ from the whole cohort for clinical parameters. Logistic regressions (using a binomial distribution of errors and a first-level risk a = 005) were used to test the correlation between hormones and UDT in each of the subsamples. Correlations between variables and the 3-level testis migration status (control, transient or persistent UDT) were tested using a polytomic regression followed by a Wald ratio calculation and a chi² test. In order to test differences between nonpalpable testis and palpable ones, we used a Wilcoxon unilateral test. Similarly, we validated the previous result using a unilateral Student test. To test both direct and combined effects of hormones on testis status (dependent variable), the polytomic regression was followed by a classical two-directional stepwise method in order to select the best among all possible models on the basis of the AIC criterion. Each regression (either logistic or polytomic) was followed by a likelihood ratio calculation and next by a chi² test to estimate the significance level of the models. Statistical regressions and tests were performed using the environment for statistical computing R 2.15.2 with the MASS and net packages (R, 2012).

Table 2. Cord blood hormones levels in cryptorchid and control boys Cryptorchid

Controls

Hormones

n

mean SEM

Insl3 ng/l Testosterone nmol/l LH UI/l HCG UI/l AMH μg/l Inhibin ng/l SHBG nmol/l

43 73 43 43 73 43 69

2257 047 122 732 3173 198 439

193 045 027 295 185 296 37

n

mean SEM

103 164 109 102 166 92 162

2714 041 135 624 297 2128 411

184 031 03 207 141 313 144

P 003 NS NS NS NS NS NS

P value corresponds to the significance level for the logistic regression test.

of the cryptochid groups (control, transient and permanent) was performed. The model obtained was satisfying (P = 0049; LR = 59), and we next tested the significance of the model coefficients. We found that the cryptorchid subgroup (P = 22) had a prominent role (P = 0029) rather than the persistent cryptorchid subgroup (n = 21; P = 034 (Fig. 1). In addition, we deduced from a confusion matrix that the positive predictive value of INSL3 was 72% (transient and persistent UDT combined). The low number of bilateral cases (four including two in each subgroup) did not allow a statistical comparison. The position of the undescended testes following Scorer classification was available in 41 cases: 20 nonpalpable testes and 21 palpable testes (15 suprascrotal, five inguinal, one high scrotal). Comparison of palpable vs nonpalpable cryptorchid groups revealed that the nonpalpable group exhibited significantly lower cord blood INSL3 levels: 179 94 vs 243 107 ng/l (P = 0018). Strikingly, testosterone concentrations (Table 2, Fig. 2) were not different between the cryptorchid and control groups, 047 045 nmol/l and 041 031 nmol/l, respectively (P = 07). Figure 2 shows that this was true for transient as well

Results Cord blood INSL3 but not testosterone concentrations were decreased in cryptorchid newborns The cord blood steroid and polypeptide hormones were compared in the two groups of newborns (Table 2). INSL3 was significantly decreased in the cryptorchid group: mean SEM: 2257 194 ng/l (n = 43) vs 2714 184 ng/l in the control group (n = 103, P = 0031). This significant difference was due to the transient cryptorchid subgroup. To determine such a difference, a polytomic regression

Fig. 1 INSL3 cord blood levels in UDT and control groups. Box-andwhisker diagram of INSL3 cord blood levels in controls, transient and persistent cases of cryptorchidism. The ‘total group’ corresponds to the sum of transient and persistent cases; *P = 0029, **P = 0031 when compared to control group (logistic regression test).


INSL3 and idiopathic cryptorchidism 245

Fig. 2 Testosterone cord blood levels in UDT and control groups. Boxand-whisker diagram of testosterone cord blood levels in controls, transient and persistent cases of cryptorchidism. The ‘total group’ corresponds to the sum of transient and persistent cases. No significant differences between any group and control (logistic regression test).

as persistent UDTs. No other hormones or SHBG were significantly correlated with UDT (Table 2). The analysis of interactions between hormones (data not shown) revealed that INSL3 was significantly and positively correlated with LH (R2 = 006, P = 0033) and negatively with AMH (R2 = 003, P = 0039), but with no other hormones. Using a bidirectional stepwise method model, the best model we obtained was: UDT approximately INSL3 + LH + INSL3 9 LH (P = 0026). This model showed that in UDT, INSL3 was very significantly decreased (P = 0006), but not LH (P = 007). Moreover, testosterone was not correlated with INSL3, LH, HCG or SHBG (data not shown). Lastly, there was no correlation of INSL3 with the studied clinical parameters (term, birth weight, mode of delivery).

Discussion Cord blood INSL3, one of the two Leydig cell factors regulating foetal testicular descent, was significantly decreased in transient UDT in this prospective case–control study assessing the two Leydig cell hormones in human ‘idiopathic’ UDT. In contrast, foetal testosterone, the other major Leydig hormonal regulator, appeared unchanged at birth. Cord blood INSL3, which was measured to assess foetal Leydig production during late pregnancy, showed variability even in the control group. The difficulties in drawing true arterial cord blood may have contributed to this variability; however, although INSL3 maternal production and trans-placental transfer have been recently reported,31 it is unlikely to have been a significant factor at that stage of pregnancy. Furthermore, the fact that INSL3 was found to be lower in the nonpalpable UDT group when compared to the palpable UDT group enhances the reliability of our results. In 2007, Bay et al.27 reported the only work so far on INSL3 in human UDT. They found, using a ‘home-made’ fluorescent immunoassay, lower INSL3 levels at birth. Using an enzyme-

immunoassay,29,30 we here confirm significantly lower INSL3 levels at birth in patients with idiopathic, isolated UDT, especially in those with transient UDT. This point differed somewhat in this respect, from the Danish results. The fact that in our study, INSL3 was decreased mostly in the transient subgroup was an unexpected, but indeed a novel and original finding. Bay and coll.27 reported a significant decrease both in transient and persistent UDT with a normalization at 3 months of age for the transient cases, but not for the persistent one.27 Physiologically, the neonatal LH wave has been shown to stimulate INSL3 secretion during the first 3 months of life.12 In cases of transient UDT, this wave may overcome the impact of low foetal/neonatal INSL3 secretion and contribute, with increasing testosterone secretion, to spontaneous postnatal descent. This hypothesis, in agreement with our results, was already suggested by Bay et al.,27 who showed that INSL3 blood levels normalized at 3 months in transient UDT.27 In our persistent group, INSL3 levels were lower although not significantly and were associated with a wider dispersion of the values and a higher mean, suggesting a greater heterogeneity of the causal factors. Indeed, even in our transient subgroup, although highly significant, the relatively low-positive predictive value of 72% indicates that INSL3 impairment should not be considered as the only mechanism involved. In our study, we also confirmed, using the gold-standard reference method (UPLC-MS/MS) avoiding matrix effect and cross-reactivity, the normal ranges of testosterone concentrations and LH/testosterone ratio in cryptorchid newborns, as reported by others.32,33 A few authors have reported an increased LH/testosterone ratio at 3 months of age33 or a decreased testosterone at 6 months, suggesting secondary Leydig cell dysfunction. The normal testosterone level observed at birth does not exclude an antagonistic action at the receptor level mediated by an antiandrogenic EED, which could indirectly impair the testosterone effect. Normal testosterone levels at birth contrasted with lower INSL3 which appeared at this time as a specific and sensitive marker of foetal Leydig cell impairment. A positive correlation has been found between LH and INSL3, contrasting with the lack of correlation between LH and testosterone. This correlation at birth is a novel but not an unexpected result, as it has been shown at other times of development that INSL3 was regulated by LH but in a different way compared with the regulation of testosterone by LH.12 Decreased INSL3 in UDT: cause or consequence? Decreased cord blood INSL3 could be either a cause or a consequence of UDT. Classically INSL3 is considered as a Leydig hormone regulating the first phase of testis migration during the second trimester of pregnancy. In our study, INSL3 was measured in cord blood, closer to the second phase, and we cannot extrapolate to the concentrations corresponding to the first phase. However, although the role of INSL3 was first well-established during the abdominal phase, its participation during the inguino-scrotal phase has more recently been suggested in animal models. Indeed, in the LH receptor knock-out mouse, testosterone administration


246 P. Fenichel et al. causes an up-regulation of gubernaculum RXFP2 expression in an androgen receptor-dependent fashion34,, and the presence of INSL3 antagonist diminishes the testosterone-induced inguinoscrotal descent.34 Moreover, strong expression of RXFP2 has been reported in adult mouse cremaster.13 All these data support a role for INSL3 in both the abdominal and the inguino-scrotal phases. Converging evidence indicates that our results, taken together with some others reported in the literature, support the INSL3 decrease being a causal factor, rather than a consequence. First, experimental induction of cryptorchidism in mice does not significantly alter the expression of INSL3 mRNAs in the testis.34 Secondly, testosterone, another Leydig hormone, was not affected in our cohort. Thirdly, if reduced cord blood INSL3 was a consequence of UDT, then the extent of the decrease might have been similar or even more marked in the persistent UDT boys, and this was not the case in our study. When compared with the Bay results,27 our data concerning the transient group brings an additional light to the aetiopathology of UDT, showing that lower reversible INSL3 levels may be important mainly in the transient forms of UDT that can be corrected after birth. This suggests a functional effect with foetal down-regulation of expression of INSL3 rather than a true testicular injury, while in the permanent forms, sustained, lower levels of INSL3 suggest testicular injury. What could be the mechanism leading to INSL3 decrease? The understanding of the fundamental role of INSL3 in testicular descent in mice and humans6,7 has prompted the active search for mutations or polymorphisms14 in the INSL3 and its receptor genes, in human patients with idiopathic UDT and normal karyotype (thereby excluding Klinefelter syndrome). These mutations seem in fact to be very rare as in a study involving 600 isolated cryptorchid infants, Ferlin et al. found only 11% of either INSL3 or RXFP2 mutations, these, mostly, in bilateral, persistent cases.15 Although these results make very unlikely the presence of such mutations in our subgroup of transient unilateral, cryptorchid newborns, this might be verified by a systematic genetic study. The lack of mutations in most cases of idiopathic UDT does not rule out a decrease in INSL315 in other nongenetic causes. INSL3 gene expression is negatively regulated by oestrogens and positively by androgens as shown in Leydig cells in vitro.35,36 In mice, maternal exposure to xenoestrogens, including the potent synthetic oestrogen diethylstilbestrol (DES), results in down-regulation of INSL3 mRNA expression level in Leydig cells37 and is associated with intra-abdominally located testes.37 In humans, an increased risk of UDT has been reported in the male offspring of women who took DES during pregnancy to prevent miscarriage.38 There are robust data in rodents26 and more recently in humans39 supporting the deleterious effects of phthalates on testicular descent26 and function.39 They may act on INSL3 gene expression/ action, steroid hormone production or as an androgen antagonist.26,34 Thus, foetal exposure to oestrogenic or antiandrogenic EEDs may interfere with INSL3 secretion. However, while it has been clearly shown that maternal exposure to oestrogenic or antiandrogenic EEDs could induce cryptorchidism in rodents, it remains unproven that such

environmental factors are operating in human idiopathic UDT, even if epidemiological20 or statistical correlations do exist.1,40 Moreover, as noted in a recent review by Virtanen et al. on EEDs and UDT,40 it is also hazardous to link one clinical endpoint – here UDT – with a single chemical product. Foetuses are indeed exposed to thousands of these chemicals, which are likely to interfere through different pathways and may as well present additive, antagonistic and/or synergistic effects known as the ‘cocktail effect’. This was shown for UDT by Damgaardt et al.,41 who found a correlation with a ‘cocktail’ of several pesticides, but not with any single pesticide, and by our team using exposure scores for PCB153 alone or associated with DDE.1

Conclusion INSL3 but not testosterone is decreased at birth in idiopathic UDT, especially in transient forms. This hormonal decrease may contribute to the impaired testicular descent along with genetic and anatomical factors. Whether the foetal environment (nutritional and/or toxicological) interferes with INSL3 secretion in humans remains to be confirmed.

Acknowledgements This project was sponsored by the Clinical Research Board of Nice University Hospital and supported by a grant from the French Research Ministry and by Institut de Recherche Endocrinienne et Metabolique, H^ opital Cochin, Paris.

Conflict of interest statement There was no conflict of interest to declare.

Contributors PF: conceived and wrote the paper; NL: performed INSL3 and testosterone assay and discussed the results; PC: made the statistical study; PPF: performed hormonal assays; KWM: supervised the clinical study; and FBD: directed the prospective study and wrote the paper.

References 1 Brucker-Davis, F., Wagner-Mahler, K., Delattre, I. et al. (2008) Cryptorchidism at birth in Nice area (France) is associated with higher prenatal exposure to PCBs and DDE, assessed by colostrum concentrations. Human Reproduction, 23, 1708–1718. 2 Lee, P.A. & Houk, C.P. (2013) Cryptorchidism. Current opinion in Endocrinology, Diabetes, and Obesity, 3, 210–213. 3 Wagner-Mahler, K., Kurzenne, J.Y., Delattre, I. et al. (2011) Prospective study on the prevalence and associated risk factors of cryptorchidism in 6246 newborn boys from Nice area, France. International Journal of Andrology, 34, 499–510. 4 Hadziselimovic, F. (2002) Cryptorchidism: its impact on male infertility. European Urology, 41, 121–123. 5 Cook, M.B., Akre, O., Forman, D. et al. (2010) A systematic review and meta-analysis of perinatal variables in relation to the


INSL3 and idiopathic cryptorchidism 247

6

7

8 9

10 11

12

13

14

15 16

17

18

19

20

21

22

23

24

risk of testicular cancer. International Journal of Epidemiology, 39, 1605–1618. Foresta, C., Zucarello, D., Garolla, A. et al. (2008) Role of hormones, genes, and environment in human cryptorchidism. Endocrine Reviews, 29, 560–580. Virtanen, H.E., Cortex, D., Rajpert-De Meyts, E. et al. (2007) Development and descent of the testis in relation to cryptorchidism. Acta Paediatrica, 96, 622–627. Hughes, I.A. & Acerini, C.L. (2008) Factors controlling testis descent. European Journal of Endocrinology, 2159, 975–982. Zimmermann, S., Steding, G., Emmen, J.M.A. et al. (1999) Targeted disruption of the insl3 gene causes bilateral cryptorchidism. Molecular Endocrinology, 13, 681–691. Nef, S. & Parada, L.F. (1999) Cryptorchidism in mice mutant for Insl3. Nature Genetics, 22, 295–299. Hutson, J.M. (1986) Testicular feminization: a model for testicular descent in mice and men. Journal of Pediatric Surgery, 21, 195–198. Bay, K. & Anderson, A.M. (2010) Human testicular insulin-like factor 3: in relation to development, reproductive hormones and andrological disorders. International Journal of Andrology, 34, 97–109. Feng, S., Bogatcheva, N.V., Truong, A. et al. (2007) Developmental expression and gene regulation of insulin-like 3 receptor RXFP2 in mouse male reproductive organs. Biology of Reproduction, 77, 671–680. Feng, S., Coressis, V.K., Hwang, A. et al. (2004) Mutation analysis of INSL3 and GREAT/LGR8 genes in familial cryptorchidism. Urology, 64, 1032–1036. Ferlin, A., Zucarello, D., Zucarello, B. et al. (2008) Genetic alterations associated with cryptorchidism. JAMA, 300, 2271–2276. Bay, K., Main, K.M., Toppari, J. et al. (2011) Testicular descent: INSL3, testosterone, genes and the intrauterine milieu. Nature Reviews Urology, 4, 187–196. Chilvers, C., Pike, M.C., Forman, D. et al. (1984) Apparent doubling of frequency of undescended testis in England and Wales in 1962-1981. Lancet, 2, 330–332. Toppari, J., Kaleva, M. & Virtanen, H.E. (2001) Trends in the incidence of cryptorchidism and hypospadias, and methodological limitations of registry-based data. Human Reproduction Update, 7, 282–286. Acerini, C.L., Miles, H.L., Dunger, D.B. et al. (2009) The descriptive epidemiology of congenital and acquired cryptorchidism in a UK infant cohort. Archives of Disease in Childhood, 94, 868–872. Boisen, K.A., Kaleva, M., Main, K.M. et al. (2004) Difference in prevalence of congenital cryptorchidism in infants between two Nordic countries. Lancet, 363, 1264–1269. Swan, S.H., Elkin, E.P. & Fenster, L. (2000) The question of declining sperm density revisited: an analysis of 101 studies published 1934-1996. Environmental Health Perspectives, 108, 961– 968. Pauniaho, S.L., Salonen, J., Helminen, M. et al. (2012) The incidences of malignant gonadal and extragonadal germ cell tumors in males and females: a population-based study covering over 40 years in Finland. Cancer Causes and Control, 23, 1921–1927. Main, K.M., Skakkebaeck, N.E. & Toppari, J. (2009) Cryptorchidism as part of the testicular dysgenesis syndrome: the environmental connection. Endocrine Reviews, 14, 167–173. McKinnell, C., Sharpe, R.M., Mahood, K. et al. (2005) Expression of insulin-like factor 3 protein in the rat testis during fetal

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41


and postnatal development and in relation to cryptorchidism induced by in utero exposure to di (n-butyl) phthalate. Endocrinology, 146, 4536–4544. Nef, S., Shipman, T. & Parada, L.F. (2000) Molecular basis for estrogen-induced cryptorchidism. Developmental Biology, 224, 354–361. Gray, L.E., Ostby, J., Furr, J. et al. (2001) Effects of environmental antiandrogens on reproductive development in experimental animals. Human Reproduction Update, 7, 248–264. Bay, K., Virtanen, H.E., Hartung, S. et al. (2007) Insulin-like factor 3 levels in cord blood and serum from children: effects of age, post-natal hypothalamic-pituitary-gonadal axis activation and cryptorchidism. Journal of Clinical Endocrinology and Metabolism, 92, 402064027. Hack, W.W., Meijer, R.W., Bos, S.D. et al. (2003) A new clinical classification for undescended testis. Scandinavian Journal of Urology and Nephrology, 37, 43–47. Cabrol, S., Ross, J.L., Fennoy, I. et al. (2011) Assessment of Leydig and Sertoli cell functions in infants with nonmosaic Klinefelter syndrome: insulin-like peptide 3 levels are normal and positively correlated with LH levels. Journal of Clinical Endocrinology and Metabolism, 96, E746–E753. Hirsch, H.J., Eldar-Geva, T., Gross-Tsur, V. et al. (2013) Normal insulin-like peptide-3 levels despite low testosterone in adult males with Prader-Willi syndrome: variations in Leydig cell function from infancy to adulthood. Journal of Clinical Endocrinology and Metabolism, 98, E135–E143. Anand-Ivell, R. & Ivell, R. (2014) Regulation of the reproductive cycle and early pregnancy by relaxin family peptides. Molecular and Cellular Endocrinology, 382, 472–479. Suomi, A.M.L., Main, K.M., Kaleva, M. et al. (2006) Hormonal changes in 3-month-old cryptorchidic boys. Journal of Clinical Endocrinology and Metabolism, 91, 953–958. Toppari, J., Virtanen, H., Skakkebaeck, N.E. et al. (2006) Environmental effects on hormonal regulation of testicular descent. Journal of Steroid Biochemistry and Molecular Biology, 102, 184– 186. Yuan, F.P., Li, X., Lin, J. et al. (2010) The role of RXFP2 in mediating androgen-induced inguino-scrotal testis descent in LH receptor knockout mice. Reproduction, 139, 759–769. O’Shaughnessy, P.J., Johnston, H., Willerton, L. et al. (2002) Failure of normal adult Leydig cell development in androgenreceptor deficient mice. Journal of Cell Science, 115, 3491–3496. Lague, E. & Tremblay, H. (2009) Estradiol represses insulin-like peptide 3 expression and promoter activity in MA-10 Leydig cells. Toxicology, 258, 101–105. Emmen, J.M.A., McLuskey, A. et al. (2000) Involvement of insulin-like factor 3 (Insl3) in diethylstilbestrol-induced cryptorchidism. Endocrinology, 141, 846–849. Palmer, J.R., Herbst, A.L., Noller, K.L. et al. (2009) Urogenital abnormalities in men exposed to diethylstilbestrol in utero: a cohort study. Environmental Health, 8, 37–42. Desdoits-Lethimonier, C., Albert, O. et al. (2012) Human testis steroidogenesis is inhibited by phthalates. Human Reproduction, 27, 1451–1459. Virtanen, H.E. & Adamson, A. (2012) Cryptorchidism and endocrine disrupting chemicals. Molecular and Cellular Endocrinology, 355, 208–220. Damgaard, I.N., Skakkebaeck, N.E., Toppari, J. et al. (2006) Persistent pesticides in human breast milk and cryptorchidism. Environmental Health Perspectives, 114, 1133–1138.