0021-972X/03/$15.00/0 Printed in U.S.A.

The Journal of Clinical Endocrinology & Metabolism 88(9):4273 4279 Copyright 2003 by The Endocrine Society doi: 10.1210/jc.2003-030359

The INSL3-LGR8/GREAT Ligand-Receptor Pair in Human Cryptorchidism

ALBERTO FERLIN, MAURO SIMONATO, LUCIA BARTOLONI, GIORGIA RIZZO, ANDREA BETTELLA, TANIA DOTTORINI, BRUNO DALLAPICCOLA, AND CARLO FORESTA
University of Padova, Department of Medical and Surgical Sciences, Clinica Medica 3, Center for Male Gamete Cryopreservation (A.F., M.S., L.B., G.R., A.B., C.F.), 35128 Padova, Italy; and University of Rome La Sapienza, Institute of Medical Genetics and Institute CSS-Mendel (T.D., B.D.), 00161 Rome, Italy
Testicular descent is a complex multistep embryonic process requiring the interaction between anatomical and hormonal factors. Failure in any of these steps results in cryptorchidism, the most frequent congenital anomaly of the urogenital tract in human males. Evidence for a genetic cause for cryptorchidism is numerous and supported by animal models. In particular, INSL3 and LGR8/GREAT proteins seem to act as ligand and receptor, respectively, and to have a role in gubernaculum development involved in testicular descent. In a cohort of 87 ex-cryptorchid patients and 80 controls, we looked for mutations in INSL3 and LGR8/GREAT genes by sequencing. Patients were classified on the basis of seminal, hormonal, and testicular cytological analyses. We found three mutations in the INSL3 gene in four patients and one LGR8/ GREAT mutation in four patients (8 of 87, 9.2%). The eight patients show different phenotypes, ranging from normozoospermia to complete azoospermia, and from bilateral cryptorchidism to retractile testes. Furthermore, the endocrine function of the testis appears normal in all subjects. The findings of our study demonstrate that INSL3-LGR8/GREAT mutations are frequently associated with human cryptorchidism and are maternally inherited. The only clinical consequence of alterations of the INSL3-LGR8/GREAT system seems to be failure of the testis to normally descend in the scrotum during embryonic development, without affecting the spermatogenic and endocrine components of the testis itself. (J Clin Endocrinol Metab 88: 4273 4279, 2003)

RYPTORCHIDISM (failure of testicular descent) is the most frequent congenital birth defect of the urogenital tract, occurring in 12% of full-term boys. The surgical correction of cryptorchidism (orchidopexy) needs to be performed as soon as possible after birth to reduce the risk of infertility and testicular cancer in adulthood. Etiology of cryptorchidism remains for the most part unknown, reflecting our faint knowledge of the mechanisms regulating testicular descent from abdomen to scrotum during embryonic development. This complex, multistage process requires the interaction of anatomical and hormonal factors and consists of two major phases, the transabdominal and the inguinoscrotal descent (1). Two mesenteric ligaments play a critical role in this two-stage process: the cranial suspensory ligament (CSL) and the caudal genitoinguinal ligament, or gubernaculum. The outgrowth of gubernaculum and the regression of the CSL result in the transabdominal migration of the testes into the inguinal region. The following descent of the testes to the scrotum is due to the shortening of the gubernacular cord and the outgrowth of the gubernacular bulb. A critical role in testicular descent has been attributed to hormones (1, 2), such as androgens and Mullerian inhibiting substance (MIS) (3, 4), and this is reflected by the evidence of cryptorchidism in congenital disorders that cause hypogonadism or androgen resistance (5). In particular, androgens induce the involution of the CSL and stimulate the

Abbreviations: CSL, Cranial suspensory ligament; FNAC, fine needle aspiration cytology; Insl3, insulin-like factor 3; MIS, Mullerian inhibiting substance; SCOS, Sertoli cell-only syndrome.

development of Wollfian derivates, whereas MIS causes involution of the Mullerian ducts. Evidence supporting a genetic cause of cryptorchidism is numerous. In some cases of unilateral cryptorchidism, the contralateral, normally descended testis may be altered too (6), and testicular cancer may originate from the contralateral, not retained testis (7). Furthermore, familial cases have been described (8, 9), and more recently, animal models suggested candidate genes for cryptorchidism. In particular, a role for a novel testicular hormone, Insl3 (insulin-like factor 3, whose product is also called relaxin-like factor) has been provided (10, 11). Insl3 is a member of the relaxin-insulin family, and it is expressed in pre- and postnatal testis Leydig cells and at reduced levels in postnatal ovarian thecal cells (12, 13). Insl3/ male mice exhibit bilateral cryptorchidism due to impaired development of the gubernaculum (10, 11). The receptor for Insl3 has been recently identified and named Lgr8 (leucine-rich repeat-containing G protein-coupled receptor 8) or Great (G protein-coupled receptor affecting testis descent) (14). The correspondence between Insl3 as ligand and Lgr8/Great as receptor was suggested by the phenotype similarity between the Great mutant mice and the Insl3 knockout mutants (10, 11, 15, 16). Moreover, relaxin, a hormone belonging to the same family of Insl3, activates Lgr8/ Great signaling through a cAMP-dependent pathway (14), and interactions between Insl3 and Lgr8/Great were demonstrated by ligand-receptor cross-linking (14, 17). The expression of Lgr8/Great is restricted to the testis, brain, and skeletal muscles, with the highest level of expression in the gubernaculum (14, 15). The INSL3-LGR8/GREAT system is supposed to influence the growth and thickening of the gu-

4273

4274

J Clin Endocrinol Metab, September 2003, 88(9):4273 4279

Ferlin et al. INSL3-LGR8/GREAT in Human Cryptorchidism

bernaculum, allowing the testis to be retained in the inguinal region. Several polymorphisms in both INSL3 and LGR8/GREAT genes have been reported previously, whereas only a few mutations specifically associated with cryptorchidism have been found (9, 16, 18 25). Actually, cryptorchidism is a heterogeneous disorder, and testicular function may be altered at different degrees, with seminal quality ranging from normozoospermia to complete azoospermia (6). Therefore, it is difficult to derive a clear phenotype associated with mutations of INSL3 or LGR8/GREAT from these previous studies. In the present work we report a more comprehensive analysis by systematically looking at mutations in both of these genes in the same well characterized group of cryptorchid men. We show that the clinical consequence of alterations of this system is failure of the testis to descend correctly in the scrotum, without apparently affecting the spermatogenic and endocrine components of the testis itself.
Patients and Methods Patients and controls
Eighty-seven consecutive subjects (age, 24 45 yr) with history of maldescended testes referred to our center for fertility evaluation were recruited. Thirty-six men had bilateral cryptorchidism, and 46 had unilateral cryptorchidism. Age at orchidopexy ranged from 127 yr. The remaining five subjects reported failure of the testes to normally descend in the scrotum at birth with spontaneous descent in first years of life, and they still were affected by bilateral retractile testes. All 87 subjects were Caucasian from different regions of Italy. A careful history and physical examination were performed in all patients. At least two semen analyses were performed, following WHO recommendations (26), and in all subjects ultrasound examination of the testes was performed (for testis morphology and volume), and plasma FSH, LH, and testosterone concentrations were determined. When semen analysis repeatedly revealed azoospermia (absence of sperm) or oligozoospermia with sperm concentrations less than 10 million/ml, a bilateral testicular fine needle aspiration cytology (FNAC) (27) was performed. A total of 80 Italian men with absence of a clinical history of maldescended testes were used as controls. Informed consent was obtained from each subject, and the authors institutional ethical committee approved the study.

Mutation analysis of INSL3 and LGR8/GREAT

Genomic DNA was extracted from peripheral blood of each patient and control. We used the LGR8/GREAT cDNA to blast search the chromosome 13 human sequence at the University of California at Santa Cruz Genome Bioinformatics (http://genome.ucsc.edu). We designed 17 pairs of intronic primers to amplify the 18 LGR8/GREAT exons (Table 1). Genomic DNA was subjected to PCR, with the following conditions: denaturation at 94 C for 1 min, followed by 20 cycles of 94 C for 30 sec, 55 C (decreasing 0.5 C in each cycle) for 30 sec, and 72 C for 30 sec and 15 cycles of 94 C for 1 min, 45 C for 1 min, and 72 C for 1 min. The PCR products were submitted to single strand conformational polymorphism and/or were purified using shrimp alkaline phosphatase and exonuclease and direct sequenced using the same primers as for PCR. The single strand polymorphism analysis was performed in precast gel (ExcelGel, Amersham Pharmacia Biotech, Indianapolis, IN) following the manufacturers instructions. Direct sequencing was preferred for the amplicons with two exons (because of the size) and for the PCR fragments with frequent polymorphisms. The same conditions were used to amplify the two INSL3 exons with the primers listed in Table 1.

Molecular modeling analysis of INSL3 and LGR8/ GREAT proteins

Comparative molecular modeling analysis was performed with the SWISS-MODEL and the Swiss-Pdb Viewer (28). For LGR8/GREAT we used the PDB-Brookhaven crystal structure coordinates of internalin (1D0B) (29), porcine ribonuclease inhibitor (2BNH) (30), and von Willebrand factor-binding domain of glycoprotein Ib (1M0Z) (31) as structural references. For INSL3 we used insulin (1G7A) (32) and Cdk6P16Ink4A tumor suppressor complex (1BI7) (33), suggested by the program 3D-PSSM (34) as a folding cognate protein, as possible protein model candidates. Other templates, such as relaxin and IGF, did not produce enough data to be analyzed due to the short stretch of amino acids present in these structures. Evaluation of the structures obtained was performed with the programs WHAT IF (35), PROSITE (36), and SAPS (37).

Results

In total, 18 of 87 subjects (20.7%) showed normozoospermia, 16 (18.4%) showed mild oligozoospermia (sperm count, 520 million/ml), and 53 (60.9%) showed azoospermia or severe oligozoospermia (sperm count, 5 million/ml). Clinical and testicular bilateral FNAC allowed us to relate cryp-

TABLE 1. Sequences of intronic primers used to amplify LGR8/GREAT exons (top) and INSL3 exons (bottom)
Forward Reverse

LGR8/GREAT Exon 1 Exon 2 Exon 3 Exon 4 Exon 5 Exon 6 Exon 7 Exon 8 Exon 9 Exon 10 Exon 11 Exon 1213 Exon 14 Exon 15 Exon 16 Exon 17 Exon 18 INSL3 Exon 1 Exon 2

CAGGCTGCATAGTGACAAG CTCAACTCATATAGCTCTTG GTTTCTAGAATAAGTGACAC CAGTTATTAAAGGCAAAGTC GCATGTGGCTTCTAGGGAAG CAACATGAGAATAGGACTTCG CAGCCGAATACTTTCAGACG ACACAAAGGAGTAGGCCAGG AAAGACTGTCCTAAAAGTGTC GGAAAGATAGTAACAACTGG GAAGACTTCAAAGGAGGC GGATGATGATAATTGTGAGG CATTCTAATCTGCATTGG GTGAAGTTGTATACCTAGC CTCGCTTCTTGGTACTCAG GTAAGGGATACTCAACCTG GTCTCCCTCTCATAAATAGC CTCGTTGCCCAGTGCTCCC GGATTCGTGCACATGCAGC

ACAACTGCCATAGAACTAC CTACAGAACACCAAATCAGC CAGACAGCAGACCTGTTAGC GCTAATGCACAAGCAACAC CATCTTACACCACCACGAG CCTACAATCAAGTCGAGCC GGAAACTACATTAGCACTG CTAGTGCTAGATGTCATTGC GGAAGGTAATATGATCAG TAACTTCCAAATATATAGGG GGCAAAGCAGTCTCACTAG CTGAAATGCATGCTCCTGTG GCATAGAAACACATTGCAG GTCTAGTGCACAACTCACAG GTCTTGACTATGTTATCTC CCTAGACTCTGGCCTTGTC GTAACCTGGTTGGAAGTTCTG CATGCAAACCTGCCCACC GGCCTGTAGATGCGAGAC

Ferlin et al. INSL3-LGR8/GREAT in Human Cryptorchidism

J Clin Endocrinol Metab, September 2003, 88(9):4273 4279 4275

torchidism to the spermatogenic function of both testes. FNAC is fundamental above all in the case of unilateral cryptorchidism to analyze the function of the ex-cryptorchid testis and the contralateral normally descended testis. As summarized in Table 2, bilateral FNAC of patients affected by azoospermia or severe oligozoospermia (29 with bilateral and 20 with unilateral cryptorchidism) showed Sertoli cellonly syndrome (SCOS; complete absence of germ cells) or bilateral severe hypospermatogenesis (strong reduction of germ cells),respectively. These patients can therefore be classified as severe bilateral testiculopathy, and this means, therefore, that in unilateral cryptorchid subjects the normally descended testis was damaged like the ex-cryptorchid one. Among patients with unilateral cryptorchidism and mild oligozoospermia, six of 11 (13.0%) showed SCOS or severe hypospermatogenesis in the ex-cryptorchid testis, but normal spermatogenesis in the contralateral testis. As discussed below, four patients were found to carry INSL3 mutations, and four had LGR8/GREAT mutations (8 of 87, 9.2%).
INSL3 mutations

Mutation analysis of the INSL3 gene identified three common polymorphisms in both cryptorchid and control subjects: L9L (27G3 A), in 3% of cryptorchid men and 5% of controls, A60T (178G3 A) in 42.5% of cryptorchid men and 47% of controls, and L42L (126G3 A) in 57% of cryptorchid men and 58% of controls. On the contrary, three nucleotide substitutions (278C3 T, 304C3 T, and 305G3 A) leading to amino acid change (respectively, P93L, R102C, and R102H) were found in the heterozygous condition in four cryptorchid men (4 of 87, 4.6%). These three mutations were not found in any of the 80 controls. Furthermore, no mutations were found in a cohort of 150 noncryptorchid infertile men.

R102H was found in two subjects (no. 1440 and O-58) classified as bilateral cryptorchidism with bilateral severe testiculopathy. In particular (Table 3), patient 1440 underwent bilateral orchidopexy at 1 yr of age, a second surgery at 5 yr and still the right testis is located at the edge of the inguinal canal. He also reported a bilateral testicular trauma at 15 yr and was affected by left varicocele. His testiculopathy could, therefore, have different causes. Patient no. O-58 underwent bilateral orchidopexy at 6 and 11 yr of age. Seminal analysis showed in both patients severe oligozoospermia (sperm count, 0.5 million/ml) with a testicular cytological analysis resembling severe hypospermatogenesis. This bilateral severe testiculopathy involving only the spermatogenic component was in accordance with plasma concentrations of gonadal hormones, which showed high FSH levels with normal LH and testosterone. No other cases of cryptorchidism were present in their family, and no blood samples were available for relatives of these subjects. R102H has previously been reported only in a female control (20), with apparently no phenotypic effect, but our findings suggest a role for this mutation in testis descent. R102C was found in a patient (no. 958) previously reported (22) with a history of unilateral cryptorchidism orchidopexied at 5 yr of age. Seminal analysis showed a sperm concentration between 2.5 and 7.2 million/ml, and hormonal plasma concentrations were in the normal range. He was therefore classified as unilateral cryptorchid with mild oligozoospermia. No cases of cryptorchidism were reported in his family. Mutation analysis of INSL3 in his parents revealed the same mutation in heterozygous condition in the mother, whose history was completely negative for both reproductive function and any other pathological conditions. P93L was found in a patient (no. 1372) affected by uni-

TABLE 2. Results of INSL3 and LGR8 mutation analysis in the 87 ex-cryptorchid subjects, classified on seminal and testicular phenotypes
n INSL3 mutations LGR8 mutations Total

Bilateral cryptorchidism Bilateral severe testiculopathy (azoospermia/severe oligozoospermia with bilateral SCOS/SH) Mild oligozoospermia Normozoospermia Subtotal Unilateral cryptorchidism Bilateral severe testiculopathy (azoospermia/severe oligozoospermia with bilateral SCOS/SH) Mild oligozoospermia with severe testiculopathy (SCOS/SH) in the ex-cryptorchid testis Mild oligozoospermia Normozoospermia Subtotal Retractile testes Bilateral severe testiculopathy (azoospermia/severe oligozoospermia with bilateral SCOS/SH) Mild oligozoospermia Subtotal Total

29 4 3 36 20 6 5 15 46 4 1 5 87

2 (6.9) 0 0 2 (5.6) 0 0 1 (20.0) 1 (6.7) 2 (4.3) 0 0 0 4 (4.6)

1 (3.4) 0 0 1 (2.6) 0 1 (16.7) 0 1 (6.7) 2 (4.3) 1 (25.0) 0 1 (20.0) 4 (4.6)

3 (10.3) 0 0 3 (8.3) 0 1 (16.7) 1 (20.0) 2 (13.3) 4 (8.7) 1 (25.0) 0 1 (20.0) 8 (9.2)

Azoospermia, absence of sperm in the ejaculate; severe oligozoospermia, sperm count 5 million/ml; and mild oligozoospermia, sperm count 520 million/ml. SH, Severe hypospermatogenesis. Percentages are in parentheses.

4276

J Clin Endocrinol Metab, September 2003, 88(9):4273 4279

Ferlin et al. INSL3-LGR8/GREAT in Human Cryptorchidism

TABLE 3. Clinical features of the eight subjects with mutations in INSL3-LGR8/GREAT

Patient no. Cryptorchidism Seminal analysis Testicular cytology Hormonal pattern Notes

INSL3 mutations R102H 1440

Bil, orchidopexy at 1 and 5 yr

0.5

Bilateral SH

FSH, 21; LH, 4.8; T, 6.2 FSH, 19.0; LH, 3.9; T, 5.8 FSH, 4.4; LH, 2.3; T, 3.9 n.e. FSH, 22.0; LH, 9.2; T, 5.9 FSH, 5.8; LH, 3.8; T, 5.7; inhibin B, 162 FSH, 4.3; LH, 5.1; T, 7.8; inhibin B, 248 FSH, 16.9; LH, 4.2; T, 3.2

R102H R102C P93L

O-58 958 1372

Bil, orchidopexy at 6 and 11 yr Uni (left), orchidopexy at 5 yr Uni, orchiectomy at 9 yr Bil, orchidopexy at 26 yr Uni (right), spontaneous descent at 13 yr Uni (left), orchidopexy at 4 yr

0.5 2.57.2 Normo Azoo 5.6 7.8

Bilateral SH Only right, MH n.e. SCOS with few loci of SH Right, SCOS; left, MH

Right testis still at the edge of inguinal canal; bilateral testicular trauma at 15 yr; left varicocele; no familial cases of cryptorchidism No familial cases of cryptorchidism Mother: same (R102C) mutation; no familial cases of cryptorchidism Mother has same (P93L) mutation; paternal cousin with Bil No familial cases of cryptorchidism Left varicocele; fertile after varicocelectomy; no familial cases of cryptorchidism Mother has same (T222P) mutation; nephew with right cryptorchidism Left varicocele; no familial cases of cryptorchidism

LGR8 mutation T222P 466 T222P B77

T222P

B85

Normo

n.e.

T222P

1030

Bil, spontaneous descent at puberty, still retractile testes

1.13.0

Bilateral SH

Bil, Bilateral cryptorchidism; uni, unilateral cryptorchidism; azoo, azoospermia; SH, severe hypospermatogenesis; MH, mild hypospermatogenesis; n.e., not examined. Seminal pattern is expressed as sperm concentration in millions per milliliter; FSH and LH plasma concentrations are in international units per liter (normal ranges, 1 8 and 1 6, respectively), testosterone is in nanograms per milliliter (normal range, 39), and inhibin B is in picograms per milliliter (normal range, 150 250). T, testosterone.

lateral cryptorchidism who underwent orchidectomy of the undescended testis at 9 yr of age because of atrophy. Seminal analysis repeatedly revealed normozoospermia, and hormonal concentrations were all in the normal range. Analysis of the parents DNA showed in the mother the same P93L mutation that, like R102C, had no phenotypic effect. He has a paternal cousin affected by bilateral cryptorchidism, but a DNA sample was not available for analysis. In the putative protein model the amino acid R102 is within a long coil, whereas P93 is inside an -helix, in agreement with data obtained from the secondary structure prediction programs. No variations from the secondary structure prediction was observed for the mutated R102C/H proteins compared with the native protein. However, protein analysis revealed that the segment encompassing the region of R102 is exhibiting a relatively high charge concentration centered precisely in amino acid R102. This behavior is lost in both mutated proteins R102C and R102H. No new disulfide bonds with other cysteins are predicted to exist. Analysis of the amino acid stretch QPLPQ containing P93 revealed that it is quite conserved among different species. The mutant P93L shows an evident secondary structure rearrangement in the amino acids preceding the mutation. Thus, the mutation seems to determine a sterical problem in the protein causing a loss of an entire -helix.

LGR8/GREAT mutations

The LGR8/GREAT mutation analysis identified one deleterious mutation (in heterozygosity), T222P, in four patients. This missense mutation has been described in one patient among 60 subjects affected by cryptorchidism (16) and was never found in controls from different regions (193 in the Gorlovs study and 80 in the present study). The four subjects with the T222P mutation have different phenotypes (Table 3). One patient (no. 466) was affected by bilateral cryptorchidism orchidopexied at 26 yr of age, and he has bilateral severe testiculopathy with very small testicles, azoospermia, high FSH with normal LH and testosterone plasma levels, and bilateral Sertoli cell-only syndrome, with few loci of severe hypospermatogenesis. Two patients (no. B77 and B85) were affected by unilateral cryptorchidism. In one case (patient B77) the right testis spontaneously descended in the scrotum at 13 yr, he was affected by mild oligozoospermia, hormonal levels were in the normal range, and FNAC showed SCOS in the ex-cryptorchid testis and only mild hypospermatogenesis in the contralateral testis that presented an important varicocele. Correction of varicocele led to an improvement of seminal parameters, and the patient become a father naturally. The other patient (patient B85) underwent left orchidopexy at 4 yr of age and repeatedly presented normozoospermia, with normal plasma con-

Ferlin et al. INSL3-LGR8/GREAT in Human Cryptorchidism

J Clin Endocrinol Metab, September 2003, 88(9):4273 4279 4277

centrations of gonadal hormones. Analysis of the parents DNA showed in the mother the same T222P mutation in the heterozygous condition that, like INSL3 mutations, had no phenotypic effect. He has a nephew (son of his sister) affected by right cryptorchidism, but a DNA sample was not available for analysis. The last patient (no. 1030) experienced bilateral cryptorchidism at birth with spontaneous descent during puberty, and at present his testes are retractile. He shows severe oligozoospermia with a bilateral severe hypospermatogenesis and high FSH levels. The analysis of a protein model reveals that in the T222P mutation the introduced proline seems to cause possible bending or kinking problems to the adjacent secondary structure elements, determining some locally protein structure instabilities. This is also confirmed by a secondary structure prediction database (PSIpred) (38). The amino acid T222 is conserved within the LRR domain and seems to be one of the amino acids important for determining the consensus of the domain as LRR. Therefore, T222 appears to be a possible key residue for the activity role of the protein, and alteration of one of these units probably is the cause of the reduction in protein binding strength. The introduction of a proline at position 222 cannot determine a big change in structure, and therefore, the mutations seems to influence only the folding of the LRR in which it is located. Several polymorphisms in LGR8/GREAT exonic and intronic sequences were identified in patients and controls (Table 4), including a stretch of adenine at the intron 12-exon 12 junction that was shown to be polymorphic in length, varying from 9 13. Analysis of exon 12 and intron 12 polymorphisms allowed us to discriminate six different haplo-

types (Table 4), whose frequency was not different between controls and cryptorchid subjects. It is worth noting that all patients with the LGR8/GREAT T222P mutation show haplotype GA13, suggesting that T222P occurred once on the chromosome bearing the GA13 haplotype. According to this hypothesis a founder effect could be present, and all patients in our group might share a common ancestor.
Discussion

Genetic research in reproductive medicine has expanded in last years so that, for example, the percentage of male infertile subjects diagnosed as idiopathic is progressively decreasing (39). Specific genetic alterations have been identified and shown to be associated with severe damage of the testis and severe infertility. For example, Y chromosome microdeletions are now recognized as an important cause of nonobstructive azoospermia and severe oligozoospermia (40) and may also be found in patients affected by cryptorchidism (41). However, Y chromosome microdeletions do not seem to be directly implicated in the pathogenesis of cryptorchidism itself, but, rather, they cause only severe testiculopathy involving the spermatogenic component. In these cases cryptorchidism is probably the result of altered testicular responses to mechanisms regulating testicular descent (41). On the other hand, a variety of congenital syndromes, due to either chromosomal aberration or monogenic disease, may present with cryptorchidism, but in these cases cryptorchidism is part of a more complex syndrome. Insl3 and Lgr8/Great knockout mice are very important animal models, above all because they represent the first

TABLE 4. Polymorphisms of LGR8/GREAT exons and introns (top), construction of haplotypes for intron 12 and exon 12 polymorphisms (middle), and their frequency distribution in controls and cryptorchid subjects (bottom)
Location Nucleotide Amino acid Note

Intron 2 Exon 3 Intron 4 Exon 12 Exon 12 Intron 12 Exon 17 Exon 18

Patient no.

INV2(15)CT 261GA INV4(17)TA 957GA 993AG INV12(2)A9A13 1810AG Utr 3(176)AG

Exon 12 polymorphism 957GA 993AG

A87A E319E L331L I604V

Intron 12 polymorphism INV12(2)A9A13

No new splicing site No new splicing site No new splicing site No new splicing site No new splicing site No new splicing site No new splicing site New splicing site possible
Haplotypes

466 B77 B85 1030 1440 O-58 958 1372

Haplotype

GG GA GG GG GG GG GG AA

GA AA AA AA GA GA GA AA
Cryptorchids (%)

11/13 10/13 11/13 11/13 11/13 11/13 11/12 10/10

GG11 AA10 GA11 GA11 GG11 GG11 GA12 AA10

Controls (%)

GA13 GA13 GA13 GA13 GA13 GA13 GG11 AA10

AA9 AA10 GA11 GG11 GA12 GA13

0 30 7 19 31 13

1 37 7 16 29 10

Haplotypes were determined by the presence of homozygous individuals.

4278

J Clin Endocrinol Metab, September 2003, 88(9):4273 4279

Ferlin et al. INSL3-LGR8/GREAT in Human Cryptorchidism

description of a specific mutation causing cryptorchidism as the only phenotype. Both Insl3 and Lgr8/Great homozygotes male mice show the same cryptorchid phenotype due to failure of gubernaculum differentiation (10, 11, 15, 16, 42). Interestingly, the mice show absence of spermatogenesis, but surgically descended testes have normal spermatogenesis (16), suggesting that the mutation impairs only testicular descent without directly affecting testis function, whose alteration would be the result of the testis malposition. This was confirmed by the finding that in Inls3 knockout mice orchidopexy can rescue the cryptorchid testes from their fate of Sertoli cell-only syndrome and lead to fertility (43). Our results seem to support this hypothesis. We found a high incidence of INSL3-LGR8/GREAT mutation in cryptorchid patients (9.2%), showing that the clinical consequence of alterations of this system seems to be failure of the testis to correctly descend in the scrotum without apparently affecting the spermatogenic and endocrine components of the testis itself. In fact, INSL3 and LGR8/GREAT mutations could be associated with different seminal patterns, including normozoospermia, and are compatible with fertility, and the eventual damage of tubular function (up to Sertoli cell-only syndrome) seems to be secondary to the abnormal position of the testis or other causes (surgical trauma, varicocele, etc.). Therefore, the timing of orchidopexy is crucial for the phenotype of cryptorchid testes in the adult. It is important to note that the phenotype of subjects with INSL3-LGR8/GREAT mutations may vary from bilateral cryptorchidism, to unilateral maldescent, to spontaneous descent of the testes after birth with subsequent retractile testes. All of these phenotypes agree well with an alteration of gubernaculum development. Although INSL3 is expressed in the steroidogenic Leydig cells, and LGR8/GREAT too is expressed in testis, the endocrine function of the testis of subjects with the INSL3LGR8/GREAT mutation is not damaged, as documented by normal LH, testosterone, and (when available) inhibin B plasma concentrations. All of the INSL3 or LGR8/GREAT mutations we found were heterozygous. We have never found homozygous mutations, nor have we found patients with two mutant alleles or patients compound heterozygote for mutations in LGR8/ GREAT and INSL3. This is in contrast to that observed in mice, where homozygotes knockout mice are cryptorchid, but heterozygotes are normal. There are several hypotheses to explain how these heterozygote mutations might be responsible for causing cryptorchidism. Mutations may be dominant negative characters where the mutated allele may compete with the wild type for the ligand/receptor. This could explain the difference between humans and knockout mice, as in heterozygous male mice, one allele is completely missing. It is possible that another unidentified mutation is present (for example, in the promoter region), but no heterozygous mutation has been never found in unaffected controls (9, 16, 18 25). Another hypothesis is related to haploinsufficiency, where a threshold effect exists, and the absence of one allele is enough to cause the phenotype. In the three cases (two INSL3 and one LGR8/GREAT) in which parents DNA was available, we found that the mutation was transmitted from the mothers, who have no phenotypic abnormalities. This is in agreement with that found in mouse,

where homozygous Great(Lgr8)/ or Insl3/ females seem to be normal (16). Therefore, in humans, as in mice, female carriers are normal. The finding that mutations seems to be maternally inherited allows us to suppose that these genes are subject to imprinting, so that only the maternal allele is expressed. However, in the genomic regions where INSL3 and LGR8/GREAT map, no imprinted gene is described. Another very remote possibility is that the mutations are activating and somehow interfere with normal INSL3 action. We also have to consider that other factors (genetic or environmental) may contribute to the phenotype, and this may explain why identical mutations were associated with varying phenotypes. This is particularly evident for LGR8/ GREAT, where the same T222P mutation is found in four patients completely different one from the other. The last hypothesis is that the mutations detected are not associated with the cryptorchid phenotype, but represent a fortuitous finding. However, there are several elements supporting a pathogenetic role of these mutations. Firstly, adding the results of the present study with those of all previous published studies, these mutations have never been found in control subjects (almost 600 controls for INSL3 and almost 200 controls for LGR8/GREAT) (9, 16, 18 25). Secondly, in transfection assays the LGR8 T222P mutation has been shown to be unable to mediate relaxin-induced cAMP production (16). Finally, molecular modeling and comparative analyses of protein sequences revealed the importance of the mutated amino acids. In fact, results from the present and previous works revealed two important residues in the INSL3 amino acid sequence (93 and 102) and only one in the LGR8/GREAT sequence (222). The mechanisms by which mutations of these residues alter protein function are hypothetical and different. The arginine at position 102 in INSL3 is the fourth before last of the C peptide, and it belongs to a conserved stretch of positively charged amino acids near the cleavage site between C peptide and A chain. Molecular modeling structure analysis revealed that R102C and R102H mutations do not determine structural changes, but an alteration of a charged zone of the protein. It is possible that the change in the amino acid sequence surrounding the cleavage site might abolish the processing of C peptide that will not be excised. Such protein will possibly not be able to stimulate its receptor (44). Proline 93 is part of a conserved specular sequence, QPLPQ, of C peptide, and P93L causes evident secondary structure rearrangement with loss of an entire -helix and therefore protein instability. Further functional analyses are in progress to assess the pathogenetic mechanism of INSL3 mutations. On the contrary, in vitro functional analysis of the T222P substitution has been already performed (16). Our analysis of putative protein model confirmed these results, showing that the T222P missense substitution causes a structural reorganization with partial unfolding in the fourth LRR and subsequent reduction in protein binding strength. The partial unfolding of this region is in agreement with clinical data and with the variable penetrance encountered. In conclusion, our data demonstrate that cryptorchidism may frequently be associated with mutations in INSL3 and its receptor LGR8/GREAT. These genes seem to have an important role in testicular descent without apparently affect-

Ferlin et al. INSL3-LGR8/GREAT in Human Cryptorchidism

J Clin Endocrinol Metab, September 2003, 88(9):4273 4279 4279

ing the spermatogenic and endocrine components of the testis itself. Further studies would clarify the molecular events involved in INSL3-LGR8/GREAT interaction as well as possible interaction with other factors implicated in testicular descent.
Acknowledgments
We thank Dr. Cinzia Vinanzi, Chiara Bedin, and Barbara Zecchin for excellent technical support and helpful comments. We thank patients and families for their courtesy and comprehension. Received February 28, 2003. Accepted May 28, 2003. Address all correspondence and requests for reprints to: Prof. Carlo Foresta, University of Padova, Department of Medical and Surgical Sciences, Clinica Medica 3, Center for Male Gamete Cryopreservation, Via Ospedale 105, 35128 Padova, Italy. E-mail: carlo.foresta@unipd.it. This work was supported by a University of Padova grant (to A.F.).

References
1. Hutson JM, Hasthorpe S, Heyns CF 1997 Anatomical and functional aspects of testicular descent and cryptorchidism. Endocr Rev 18:259 280 2. Nef S, Parada LF 2000 Hormones in male sexual development. Genes Dev 14:30753086 3. Emmen JM, McLuskey A, Adham IM, Engel W, Grootegoed JA, Brinkmann AO 2000 Hormonal control of gubernaculum development during testis descent: gubernaculum outgrowth in vitro requires both insulin-like factor and androgens. Endocrinology 141:4720 4727 4. Kubota Y, Temelcos C, Bathgate RA, Smith KJ, Scott D, Zhao C, Hutson JM 2002 The role of insulin 3, testosterone, Mullerian inhibiting substance and relaxin in rat gubernacular growth. Mol Hum Reprod 8:900 905 5. Buyse M, Feingold M 1979 Syndromes associated with abnormal external genitalia. In: Vallet HL, Porter IH, eds. Genetic mechanisms of sexual development. New York: Academic Press; 425 435 6. Foresta C, Ferlin A, Garolla A, Milani C, Oliva G, Rossato M 1996 Functional and cytological features of the contralateral testis in cryptorchidism. Fertil Steril 66:624 629 7. Giwercman A, van der Maase H, Skakkebaek NE 1993 Epidemiological and clinical aspects of carcinoma in situ of the testis. Eur Urol 23:104 114 8. Czeizel A, Erodi E, Toth J 1981 Genetics of undescended testis. J Urol 126: 528 529 9. Koskimies P, Virtanen H, Lindstrom M, Kaleva M, Poutanen M, Huhtaniemi I, Toppari J 2000 A common polymorphism in the human relaxin-like factor (RLF) gene: no relationship with cryptorchidism. Pediatr Res 47:538 541 10. Nef S, Parada LF 1999 Cryptorchidism in mice mutant for Insl3. Nat Genet 22:295299 11. Zimmermamm S, Steding G, Emmen JM, Brinkmann AO, Nayernia K, Holstein AF, Engel W, Adham IM 1999 Targeted disruption of the Insl3 gene causes bilateral cryptorchidism. Mol Endocrinol 13:681 691 12. Roche PJ, Butkus A, Wintour EM, Tregear G 1996 Structure and expression of Leydig insulin-like peptide mRNA in the sheep. Mol Cell Endocrinol 121: 171177 13. Zimmermann S, Schottler P, Engel W, Adam IM 1997 Mouse Leydig insulinlike (Ley I-L) gene: structure and expression during testis and ovary development. Mol Reprod Dev 47:30 38 14. Hsu SY, Nakabayashi K, Nishi S, Kumagai J, Kudo M, Sherwood OD, Hsueh AJ 2002 Activation of orphan receptors by the hormone relaxin. Science 295: 671 674 15. Overbeek PA, Gorlov IP, Sutherland RW, Houston JB, Harrison WR, Boettger-Tong HL, Bishop CE, Agulnik AI 2001 A transgenic insertion causing cryptorchidism in mice. Genesis 30:26 35 16. Gorlov IP, Kamat A, Bogatcheva NV, Jones E, Lamb DJ, Truong A, Bishop CE, McElreavey K, Agoulnik AI 2002 Mutations of the GREAT gene cause cryptorchidism. Hum Mol Genet 11:2309 2318 17. Kumagai J, Hsu SY, Matsumi H, Roh JS, Fu P, Wade JD, Bathgate RA, Hsueh AJ 2002 INSL3/Leydig insulin-like peptide activates the LGR8 receptor important in testis descent. J Biol Chem 277:3128331286

18. Krausz C, Quintana-Murci L, Fellous M, Siffroi JP, McElreavey K 2000 Absence of mutations involving the INSL3 gene in human idiopathic cryptorchidism. Mol Hum Reprod 6:298 302 19. Tomboc M, Lee PA, Mitwally MF, Schneck FX, Bellinger M, Witchel SF 2000 Insulin-like 3/relaxin-like factor gene mutations are associated with cryptorchidism. J Clin Endocrinol Metab 85:4013 4018 20. Lim HN, Raipert-de Meyts E, Skakkebaek NE, Hawkins JR, Hughes IA 2001 Genetic analysis of the INSL3 gene in patients with maldescent of the testis. Eur J Endocrinol 144:129 137 21. Marin P, Ferlin A, Moro E, Garolla A, Foresta C 2001 Different insulin-like 3 (INSL3) gene mutations not associated with human cryptorchidism. J Endocrinol Invest 24:RC13-RC15 22. Marin P, Ferlin A, Moro E, Rossi A, Bartoloni L, Rossato M, Foresta C 2001 Novel insulin-like 3 (INSL3) gene mutation associated with human cryptorchidism. Am J Med Genet 103:348 349 23. Takahashi I, Takahashi T, Komatsu M, Matsuda J, Takada G 2001 Ala/Thr60 variant of the Leydig insulin-like hormone is not associated with cryptorchidism in the Japanese population. Pediatr Int 43:256 258 24. Baker LA, Nef S, Nguyen MT, Stapleton R, Pohl H, Parada LF 2002 The insulin-3 gene: lack of a genetic basis for human cryptorchidism. J Urol 167: 2534 2537 25. Canto P, Escudero I, Soderlund D, Nishimura E, Carranza-Lira S, Gutierrez J, Nava A, Mendez JP 2003 A novel mutation of the insulin-like 3 gene in patients with cryptorchidism. J Hum Genet 48:86 90 26. WHO 1999 WHO laboratory manual for the examination of human semen and sperm-cervical mucus interaction. Cambridge: Cambridge University Press 27. Foresta C, Varotto A, Scandellari C 1992 Assessment of testicular cytology by fine needle aspiration as a diagnostic parameter in the evaluation of the azoospermic subject. Fertil Steril 57:858 865 28. Guex N, Peitsch MC 1997 SWISS-MODEL and the Swiss-Pdb viewer: an environment for comparative protein modelling. Electrophoresis 18:2714 2723 29. Marino M, Braun L, Cossart P, Ghosh P 1999 Structure of the Inlb leucine-rich repeats, a domain that triggers host cell invasion by the bacterial pathogen L. manocytogenes. Mol Cell 4:1063 30. Kobe B, Deisenhofer J 1996 Mechanism of ribonuclease inhibition by ribonuclease inhibitor protein based on the crystal structure of ist complex with ribonuclease A. J Mol Biol 26:1028 31. Huizinga EG, Tsuji S, Romijn RAP, Schiphorst ME, De Groot PG, Sixma JJ, Gros P 2002 Structures of glycoprotein Ib and its complex with von Willebrand factor A1 domain. Science 297:1176 32. Smith GD, Pangborn WA, Blessing RH 2001 Phase changes in T3R3F human insulin: temperature or pressure induced? Acta Crystallogr Sect D 57:1091 33. Russo AA, Tong L, Lee JO, Jeffrey PD, Pevletich NP 1998 Structural basis for inhibition of the cyclin-dependent kinase Cdk6 by the tumor suppressor p16INK4a. Nature 395:217221 34. Kelley LA, MacCallum RM, Sternberg MJE 2000 Enhanced genome annotation using structural profiles in the program 3D-PSSM. J Mol Biol 299:499 520 35. Vriend G 1990 What if: a molecular modelling and drug design program. J Mol Graph 8:5256 36. Bairoch A, Bucher P, Hofmann K 1997 The PROSITE database, its status. Nucleic Acids Res 25:217221 37. Brendel V, Bucher P, Nourbakhsh I, Blaisdell BE, Karlin S 1992 Methods and algorithms for statistical analysis of protein sequences. Proc Natl Acad Sci USA 89:20022006 38. McGuffin LJ, Bryson K, Jones DT 2000 The PSIPRED protein structure prediction server. Bioinformatics 16:404 405 39. Foresta C, Ferlin A, Gianaroli L, Dallapiccola B 2002 Guidelines for the appropriate use of genetic tests in infertile couple. Eur J Hum Genet 10:303312 40. Foresta C, Moro E, Ferlin A 2001 Y chromosome microdeletions and alterations of spermatogenesis. Endocr Rev 22:226 239 41. Foresta C, Moro E, Garolla A, Onisto M, Ferlin A 1999 Y chromosome microdeletions in cryptorchidism and idiopathic infertility. J Clin Endocrinol Metab 84:3660 3665 42. Ivell R, Bathgate RA 2002 Reproductive biology of the relaxin-like factor (RLF/INSL3). Biol Reprod 67:699 705 43. Nguyen MT, Showalter PR, Timmons CF, Nef S, Parada LF, Baker LA 2002 Effects of orchidopexy on congenitally cryptorchid insulin-3 knockout mice. J Urol 168:1779 1783 44. Bullesbach EE, Schwabe C 1999 Specific, high affinity relaxin-like factor receptors. J Biol Chem 274:22354 22358