| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis; |
Institutes of Human Genetics,5 Cell and Molecular Pathology,6 Pathology,7 and Biometry,8 Hannover Medical School, D-30625 Hannover, Germany
Institute of Human Genetics,9 University of Göttingen, D-37073 Göttingen, Germany
Section of Endocrinology, Andrology and Internal Medicine,10 Department of Biomedical Sciences, and Master in Andrological and Human Reproduction Sciences, University of Catania, I-95123 Catania, Italy
Department of Anatomy and Cell Biology,11 Justus-Liebig-University of Giessen, D-35385 Giessen, Germany
ABSTRACT
TSPY (testis-specific protein, Y-encoded) genes are expressed in premeiotic germ cells and round spermatids. The topology and timing of TSPY expression, and also its homology to members of the TTSN-family, suggest that TSPY is a proliferation factor for germ cells. There is also evidence for a role of TSPY in the aetiology of testis cancer. TSPY is a candidate for GBY, the elusive gonadoblastoma locus on the human Y chromosome, which is thought to predispose dysgenetic gonads of 46, XY sex-reversed females to develop gonadoblastoma. We have previously generated a TSPY transgenic mouse line (Tg(TSPY)9Jshm) that carries approximately 50 copies of the human TSPY gene on the mouse Y chromosome. In order to elucidate TSPY expression under complete androgen insensitivity and to investigate a possible role of TSPY in gonadal tumorigenesis, we have now generated sex-reversed TSPY transgenic ArTfm mice hemizygous for the X-linked testicular feminization mutation (ArTfm). We can show that the TSPY transcript is aberrantly spliced in the testes of TSPY-ArTfm mice, and that TSPY expression is upregulated by androgen insensitivity in some but not all animals. TSPY transgenic mice showed significantly increased testes weights. In one TSPY transgenic ArTfm animal, spermatogenesis proceeded beyond meiotic prophase. No tumors of germ cell origin were found in the testes of TSPY-ArTfm mice. Five out of 46 TSPY transgenic ArTfm mice, and 3 out of 31 age-related NMRI-ArTfm controls developed Leydig cell tumors, whereas none of the age-matched ArTfm mice (n = 44) on a wild type background were affected by Leydig cell tumorigenesis.
Leydig cells, spermatogenesis, testis, testosterone, TSPY
Human TSPY (testis-specific protein, Y-encoded) is the product of a Y-chromosome-specific gene family located within the male-specific region of the Y chromosome [1]. TSPY expression is restricted to the prostate and the testis where expression is limited to embryonic germ cells, spermatogonia, spermatocytes, and round spermatids [2–4], and to epithelial cells of the prostate [5]. The TSPY expression pattern in embryonic and premeiotic germ cells [2, 3], its strong expression in testicular seminoma [6], nonseminomatous germ cell tumors [7], carcinoma in situ (CIS) [3], and gonadoblastoma [6, 8, 9], and the presence of a putative cyclin B-binding domain, which is similar to those of the SET oncogene (suppressor of variegation, enhancer of zeste and Trithorax) and the nucleosome assembly protein (NAP), all suggest a biological function of TSPY as an activator of germ cell proliferation [5]. Expression of TSPY in stably transfected HeLa and mouse NIH3T3 cells stimulates cell growth in vitro [10]. The fact that TSPY is overexpressed in human germ cell tumors and their precursor stages [5–9] and that its expression stimulates tumor growth in nude mice [10] suggests an oncogenic role of TSPY in germ cell tumorigenesis. TSPY is a candidate for GBY, which predisposes dysgenetic gonads of 46, XY sex-reversed females to develop gonadoblastoma, the precursor of tumors of the dysgenetic gonad that are mainly classified as dysgerminoma [2, 11]. GBY was mapped to the deletion intervals 3E-3G on Yp [12], the centromeric region 4B, and 5E on Yq [13], and in addition to TSPY, three other GBY candidate genes (RBM, PRY, and PRKY) are located in this region [6, 11]. TSPY expression in prostate cancers of high and low Gleason grades also suggests an involvement of TSPY in prostatic oncogenesis [5].
After TSPY was first discovered in man [14], orthologous genes have been found in many mammalian lineages [15–17]. While human TSPY is moderately repetitive with copy numbers ranging from 30 to 60 per genome [18] and is highly repetitive in the bull [19], no functional Tspy exists on the Y chromosome in species of the subgenus Mus, including the laboratory mouse [16, 17], or in the Mongolian gerbil [20]. We restored TSPY activity in a TSPY transgenic NMRI mouse line (Tg(TSPY)9Jshm) that carries a human TSPY transgene of approximately 50 copies on the mouse Y chromosome [21]. We here describe the creation of TSPY-ArTfm mice and investigate TSPY expression under complete androgen insensitivity [22] in order to contribute to understanding TSPY function in spermatogenesis and gonadal tumorigenesis.
TSPY Transgenic ArTfm, NMRI-ArTfm, and ArTfm Mice
The care and use of the research animals were within standard ethical guidelines. All procedures using mice were conducted according to the rules of the German animal welfare law and were licensed by the local authorities. This is in accordance with the International Guiding Principles for Biomedical Research Involving Animals. The ArTfm stock was derived from Dr. Susumo Ohno's line [23] and was kept in the Institute of Anatomy at the University of Tübingen, Germany. The ArTfm mutation in this stock is linked to the coat color marker Blotchy; thus, ArTfm hemizygous mice can principally be identified by their grey coat color. However, each ArTfm hemizygous male was genotyped by PCR using the primers Tfmf1 and Tfmr1 (Table1), and the amplified product was sequenced. The TSPY transgenic line Tg(TSPY)9Jshm [21] was founded on a NMRI background. Genotyping of TSPY transgenic males was performed as described by Schubert et al. [21]. TSPY transgenic males on a NMRI genetic background and NMRI males were mated with heterozygous ArTfm females in order to generate TSPY-ArTfm animals and NMRI-ArTfm mice, respectively. Only animals of the first filial generation of the respective mating were used in order to minimize the differences in genetic background. TSPY transgenic males, TSPY-ArTfm, NMRI-ArTfm, ArTfm mice, and heterozygous ArTfm females were kept three to four per cage after weaning at the age of 21 days in a room with controlled light and darkness cycle (12L:12D) and room temperature of 21°C ± 1°C. DNA was extracted from 0.1 g of tissue using the phenol-chloroform method as described in Hogan et al. [24].
RNA Isolation and Northern Blotting
Total RNA was isolated from testes using the RNeasy Protect Midi Kit (Qiagen, Hilden, Germany) according to the manufacturer's recommendation. Quantification and validation for integrity of the isolated RNAs were performed on an Agilent 2100 Bioanalyzer using an RNA 6000 NanoLabChip kit according to the manufacturer's protocol (Agilent Technologies, Böblingen, Germany). Ten micrograms testicular total RNA was separated by electrophoresis on a 1.2% agarose gel containing formaldehyde, then blotted to Hybond-N+ nylon membrane (Amersham Pharmacia Biotech, Freiburg, Germany), and hybridized with 
-32P-dCTP-labeled PCR product (High Prime Kit, Roche Diagnostics, Mannheim, Germany) generated with human TSPY primers TDGEX-2 (CTGGAAGCGAATTCATGCGCCCTGAGGGCTCGC) and TD-2 (AATCTGAAGCTTCATCATATTCAACTC) using pEGY9 as a template. The insert of pEGY9 is a human TSPY cDNA fragment covering the complete open reading frame of human TSPY (exons 1, 2, 3, 4, 5, and 6), which was amplified from human total testicular RNA by RT-PCR analysis with the primers TDGEX-2 and TD-2. The 961 bp cDNA was cloned into pGEX2-T (Amersham Pharmacia Biotech, Freiburg, Germany) [21]. Hybridization was performed in ExpressHyb hybridization solution (Clontech Laboratories, Palo Alto, CA) at 63°C overnight. After hybridization, membranes were washed at room temperature for 2 min in 2x saline sodium citrate (SSC), for 3 min in 2x SSC at 63°C, for 3 min in 0.5 x SSC/0.25% SDS, and finally for 3 min in 0.1x SSC/0.1% SDS at 63°C. The membranes were directly exposed to a Fuji BAS MP 2040S imaging plate for 1 day, and images were read on Fuji BAS 1000 phosphoimager (Fuji, Miyamodai, Japan). The relative intensity of the radioactive signals at each location was quantified using PCBAS 2 version software (TINA 2.0, raytest, Straubenhardt, Germany). Hybridization signals were also visualized by x-ray autoradiography. The integrity of RNAs was checked by rehybridization with a mouse glyceraldehyde 3-phosphate dehydrogenase (GAPDH)-cDNA probe, which was amplified with the GAPDH-F (GAAGCTTGTCATCAACGGGAAGCCC) and GAPDH-R (GCATCGAAGGTGGAAGAGTGGGAGT) primers using the recombinant plasmid p44–5 as a template. The recombinant plasmid p44–5 contains an mouse GAPDH-cDNA insert, which was isolated from mouse testicular RNA by reverse transcription (RT)-PCR analysis with the primers GAPDH-F and GAPDH-R.
PCR, RT-PCR, and Product Cloning
Amplification of genomic DNA or recombinant plasmid cDNA was performed by PCR with 10 pmol primer and 1 unit of Taq polymerase (Qiagen, Hilden, Germany) in a 30 µl reaction volume. Standard conditions were denaturation at 95°C for 5 min, 35 cycles of 1 min at 95°C, 1 min annealing (Table 1), 1–3 min incubation at 72°C, and a final extension at 72°C for 10 min.
|
First-strand cDNA syntheses were performed using the First Strand Synthesis Kit (Amersham Pharmacia Biotech, Freiburg, Germany) according to the manufacturer's recommendation. PCR- and RT-PCR-amplified products were cloned using the TOPO TA Cloning Kit Dual Promoter kit with the pCRII-TOPO plasmid (Invitrogen, San Diego, CA).
The quantification of human TSPY expression levels was performed using a SYBR-Green I-based real-time fluorescence detection method [25]. For real-time RT-PCR, primers for human TSPY were selected to span exon-exon junctions. Primers (300 µM of each), 1.5625–100 ng of total RNA, QuantiFast SYBR Green RT-PCR Kit (Qiagen, Hilden, Germany), including a mixture of the QIAGEN Omniscript and Sensiscript Reverse Transcriptase, and RNase inhibitor were added (one-step RT-PCR), and run under standard conditions in an ABI Prism 7900 sequence detector system (PE Biosystems, Foster City, CA). A standard curve was constructed for every primer pair (amplicon) with 1.5625, 6.25, 25, and 100 ng of TSPY transgenic testicular total RNA of an age-matched TSPY transgenic male (sample 1 in Fig. 3A). The fractional cycle number (Ct-value) was determined at which the amount of amplified target reaches a fixed threshold, which is directly related to the amount of starting RNA target. Absolute quantification of the amount of RNA in unknown samples was performed by determining its Ct value and by using the corresponding logarithmic standard curve plot for linear interpolation. Because GAPDH expression decreased in testes with maturation arrest [26], quantitative data were normalized by using the housekeeping gene beta-actin as an internal standard. For each gene and each sample, three independent analyses were performed. Real-time PCR primers for human TSPY were Ex3-forward (5'-CAAGGAATATCTGGTGAACATCACAG-3') and 4690–2-TSPY Ex4-reverse (5'-GAGAACCAGTTGAAGAAGTTAAGGCT-3'), and for mouse beta-actin, they were beta-actin-forward (CTTTGCAGCTCCTTCGTTGC) and beta-actin-reverse (ACGATGGAGGGGAATACAGC).
|
Sequencing of recombinant plasmid DNA and PCR fragments was carried out with the ABI PRISM 310 Genetic Analyzer (PE Applied Biosystems, Vaterstetten, Germany), using the ABI PRISM Big Dye Terminator Cycle Sequencing Ready Reaction Kit (PE Applied Biosystems, Vaterstetten, Germany).
Histological Examinations and Immunohistochemistry of Tissue Sections
Three-micron sections were generated after fixation of the testes in Bouin fixative, 10% buffered formalin fixative (Fischer, Saarbrücken, Germany), or 4% paraformaldehyde solution, followed by embedding in paraffin. Immunostaining of testis sections was performed using a modified peroxidase ABC detection protocol [27]. The TSPY antiserum 837/3 was used at 1:300 dilution ratio in blocking solution (6% goat serum, 1% BSA, 1 x PBS). The binding of biotinylated goat anti-rabbit IgG (1:200 dilution ratio in blocking solution) to the primary antibodies was increased by rabbit-PAP (peroxidase-linked anti peroxidase antibody, Dako, A/S, Denmark) (1:250 dilution ratio in blocking solution) and detected by ABC-peroxidase reagents (Vectastain Elite ABC kit, Vector Laboratories, Burlingame, CA). For the enzymatic detection, the DAB kit (Vector Laboratories, Burlingame, CA) was used. TSPY-specific immunostaining was assessed using antiserum blocked with excess recombinant TSPY protein.
For histological evaluations, testicular tissues were fixed in Bouin or 4% paraformaldehyde fixative and embedded in Technovit 7100 (Heraeus Kulzer, Wehrheim, Germany) or paraffin according to the manufacturer's recommendation. Testis sections of 2 µm were stained with haematoxylin (Gill) and eosin. The criteria for separating Leydig cell hyperplasias and tumors were made according to the international classification of rodent tumors by Mohr [28]. Tubular dilatation was classified as slight if at least two tubuli seminiferi of a testis showed diameter between 280 and 370 µm, as moderate if tubular diameters were ranging from over 370 to 465 µm, and as severe if the diameter from at least two tubules was above 465 µm. Measurements of tubular diameters were performed using a Zeiss ocular micrometer 5 +100/100 Y and a 10:100 objective micrometer scale (Zeiss, Jena, Germany), and only seminiferous tubules with approximately round diameter were included. Hyperplasia of peritubular myoid cells was classified as slight if double layers of peritubular myoid cells surrounded the seminiferous tubules, as moderate if three layers of peritubular myoid cells were present, and as severe if at least four layers of peritubular myoid cells were present. At least two slides of two different areas of each testis were analyzed.
Data were analyzed using the SPSS (SPSS 11.0 for Windows) statistical program (Pearson Chi-Square test, Kruskal-Wallis test, one-way ANOVA).
Generation of TSPY-ArTfm, NMRI-ArTfm, and ArTfm Mice
To define the roles of human TSPY in spermatogenesis and testicular tumorigenesis in the absence of androgen function, we generated sex-reversed TSPY transgenic ArTfm mice and NMRI-ArTfm and ArTfm animals as controls. TSPY transgenic males (founded on a NMRI background [21]), NMRI males, and wildtype males (derived from Dr. Susumo Ohno's ArTfm stock) were crossed with females heterozygous for ArTfm [23].
The ArTfm frameshift mutation [29] caused complete androgen insensitivity in hemizygous males and induced a phenotype that corresponded to the human testicular feminization syndrome [30]. Affected TSPY-ArTfm, NMRI-ArTfm, and ArTfm mice had female genitals, a blind ending vagina, inguinal testes, and no scrotum, and lacked the derivatives of the Wolffian ducts [31].
Transcription of the TSPY Transgene in TSPY-ArTfm Mice
To see if TSPY transcription and splicing were affected by androgen insensitivity in the testes of TSPY-ArTfm mice, we performed poly-dT-RT-PCR analyses with two primers from exon 1 (TDGEX-2.2) and exon 6 (TD-2.2), which span the open reading frame (ORF) of all known hTSPY transcripts. In addition to a predominant band of the size predicted by the human main transcript TSPYmajor (961 bp), several smaller cDNA fragments were amplified (Fig. 1). The amplified products were cloned and subsequently sequenced. In addition to full length products, which were spliced according to the splice pattern of TSPYmajor [3, 21], eight different types of clones of various lengths were identified (Fig. 2). Two transcripts corresponded to two previously published testicular human TSPY splice variants and were also identified in the testes of TSPY transgenic mice [21]. Both used a cryptic splice donor within exon 1 at position +88 of the human TSPY cDNA but different internal splice acceptor sites at positions +348 and +399 in exon 1 [3, 5]. This alternate RNA processing led to in-frame deletions of 87 (Exon1A transcript) and 104 (Exon1B transcript) amino acids, respectively, from the hTSPY ORF. Both spliced isoforms would be translated into shortened TSPY peptides harbouring the complete putative cyclin B binding domain [5]. The remaining six splice variants were identified only in TSPY transgenic ArTfm mice. The removal of 380 bp by the use of the cryptic splice donor in exon 1 at position +166 and an alternative splice acceptor site in exon 2 at position +545 shifted the protein coding ORF after amino acid residue 55. An in-frame deletion of 123 bp in exon 1 (deletion of amino acid residues numbered 48 to 88) in combination with the removal of 7 bp by the use of an alternative splice acceptor site in exon 4 [TSPYC, 21] led to a frameshift after amino acid residue 225. The resulting peptide would carry part of the NAP/SET domain [5]. Two truncated transcripts harbored a deletion of 520 and 541 bp, which destroyed the hTSPY-specific ORF, after amino acid residues 73 and 69, respectively (Fig. 2). One variant carried an extremely shortened TSPY cDNA with a deletion of 695 bp between the positions +52 in exon 1 and +746 in exon 4 that contained no apparent TSPY-specific ORF. One cryptically spliced transcript harbored this deletion in combination with a splice pattern of the type-2 transcript at the intron 4/exon 5 border [32].
|
|
Northern blot and real-time PCR expression analyses were performed in order to see whether the level of TSPY expression was affected by lack of androgen action. Northern hybridization of testicular RNA from 6-wk-old TSPY-ArTfm mice (n = 3) and genetic background-matched TSPY transgenic males (n = 3) detected a TSPY-specific band of 1.3 kb as predicted from known human TSPY transcriptional units [3, 32] and thus confirmed the presence of TSPY transcripts in the transgenic testes (Fig. 3). Our data showed that the majority of the transcripts were approximately 1.3 kb in size in the testes of TSPY-ArTfm mice, although a variety of alternatively spliced transcripts of shorter length were detected by RT-PCR (Fig. 1). Comparisons of the signal intensity of TSPY transgenic and TSPY-ArTfm testicular RNAs indicated that TSPY expression was approximately two-fold upregulated in the testes of two of three transgenic ArTfm mice (Fig. 3A), a result that was also confirmed by real-time-PCR analyses of the same samples (Fig. 3B). We quantified the transcription level of the human transgene in the testes of five TSPY-ArTfm animals (three of them were previously analyzed by Northern blot analysis) and two TSPY transgenic wildtype males using relative real-time RT-PCR procedures [25], and found a two-fold upregulated TSPY expression level in three out of five investigated androgen-insensitive mice.
Expression of the TSPY Transgene in Androgen-Insensitive Testes of Artfm Males
We then extended our expression studies to the protein level using a TSPY-specific polyclonal antiserum (837/3) that was raised against the N-terminal part of the TSPY peptide [3]. Within the undescended TSPY-ArTfm testes, TSPY expression was restricted to germ cells, and TSPY immunostaining was detectable in spermatogonia and to a lesser extent in prophase spermatocytes (Fig. 4). TSPY was mainly located in the cytoplasm, but the nucleoplasm was also usually positive. We further compared the proportion of TSPY-expressing germ cells in order to see whether the varying transcription levels of the human transgene were due to differences in the number of expressing cell types in noncryptorchid and cryptorchid TSPY transgenic mice. We immunostained the testes of four 2-mo-old TSPY transgenic males and five age-matched TSPY-ArTfm mice and found no differences in the proportion of TSPY positive germ cells.
|
Testicular Histology of TSPY-ArTfm, NMRI-ArTfm, and ArTfm Mice
We compared the testicular histology of 46 androgen insensitive TSPY transgenic ArTfm mice aged from 2 to 24 mo, with 31 age-matched NMRI-ArTfm controls and 44 age-matched ArTfm mice. In all investigated TSPY-ArTfm, NMRI-ArTfm, and ArTfm mice, spermatogenesis was drastically impaired and did not proceed past the meiotic prophase (Fig. 5) except for one 8-mo-old TSPY-ArTfm mouse in which round spermatids in nearly all tubules were found. The immature cryptorchid testes of TSPY-ArTfm and NMRI-ArTfm mice showed a significantly stronger hyperplasia of Leydig cells in comparison to ArTfm mice (Table 2, P < 0.001), whereas TSPY-ArTfm and NMRI-ArTfm testes did not significantly differ from each other. Multifocal tubular dilation was observed in testes of 15 TSPY-ArTfm (32.6%, n = 46) and 8 NMRI-ArTfm animals (25.8%, n = 31) but in none of the 44 investigated ArTfm-controls (Table 2, P < 0.001). Seminiferous tubules were multifocally enlarged in 4- to 24-mo-old mice, and the degree of the dysfunction was independent of age, albeit variable between TSPY- and nontransgenic NMRI-ArTfm mice (Table 2). Tubular ectasia of TSPY-ArTfm and NMRI-ArTfm mice were often combined with an enormous loss of germ cells. Tubules were filled with amorphous material, and often Sertoli cells and spermatogonia reduced in number were present along the basal membrane. A multifocal hyperplasia of peritubular myoid cells was observed in all investigated cryptorchid males (TSPY-ArTfm, NMRI-ArTfm, and ArTfm). The degree of peritubular thickening of the basal membrane ranged from one cell layer to over four cell layers of peritubular myoid cells, and hyperplasia of peritubular myoid cells was encountered in older ArTfm mice. Hyperplasia was significantly increased in TSPY-ArTfm and NMRI-ArTfm mice in comparison to ArTfm mice (Table 2, P < 0.001), but showed no significant difference between transgenic and age-matched NMRI-ArTfm controls. Testes weights of TSPY-ArTfm mice (n = 39) were significantly increased in comparison to NMRI-ArTfm mice (n = 30) (0.029 ± 0.02 vs. 0.018± 0.016g, P < 0.05).
|
|
None of 2–24-mo-old TSPY-ArTfm (n = 46), NMRI-ArTfm (n = 31), and ArTfm (n = 44) animals developed testicular germ cell tumors, but testicular Leydig cell tumors appeared spontaneously with a cumulative incidence of 10.9% in 17- to 24-mo-old TSPY-ArTfm mice. Also three 20- to 24-mo-old NMRI-ArTfm controls (9.7%, n = 31) were affected by Leydig cell tumorigenesis, but no such tumors were found in 44 age-matched ArTfm mice. TSPY immunostaining in these tumors showed no Leydig cells and tumor cells positive for the TSPY protein (data not shown).
The prevalence of germ cells tumors (gonadoblastoma, CIS, and/or invasive germ cell tumors) in patients with gonadal dysgenesis and Y chromosomal material is estimated at around 30%, and patients with androgen insensitivity syndrome are predisposed for germ cell tumor formation with an estimated life-time risk of 5%–10% [34]. The TSPY gene is regarded as the most likely candidate for the gonadoblastoma locus (GBY) on the human Y chromosome [8, 9, 11, 35, 36]. A role of TSPY in gonadal tumorigenesis and spermatogenesis has intensively been studied but not been clarified [3, 6, 7, 9, 37, 38]. We have recently generated a TSPY transgenic mouse line harbouring an 8.2 kb genomic fragment encompassing the entire human TSPY gene that recapitulates the human organization and expression pattern [21]. This animal model was applied in the present study to investigate TSPY expression and its supposed oncogenic function [39, 40] in mice with testicular feminization (ArTfm mice).
TSPY Splicing in the Androgen-Insensitive Testes
No data exist on the transcription of human TSPY in gonads with androgen insensitivity or deficiency. Analyses of human tissues (including testes, testicular seminoma, prostate, and prostatic cancer tissue) have found expression of various TSPY transcripts. However, there is currently no functional relevance for these transcripts in spermatogenesis or gonadal or prostatic tumorigenesis. It is yet unknown whether all of the various TSPY transcripts are derived from the same transcriptional unit. Our data clearly demonstrate that one transcriptional unit derived from Yp11.2 is able to generate type-1, type-2, Exon1A, Exon1B, and TSPYC transcripts [5, 21, 37] and various novel splice variants by alternative splicing.
The majority of the TSPY transcripts in the testes of TSPY-ArTfm mice are spliced according the pattern of TSPYmajor (type-1 transcripts), and several studies have shown the predominance of type-1 transcripts in prostatic, testicular, and seminoma samples [3, 5, 37]. Besides the preferentially expressed variants Exon1A and Exon1B that carry in-frame deletions of 87 and 104 amino acids, respectively, from the TSPY-specific ORF and harbor the complete putative cyclin B binding domain (Fig. 2), six novel alternative spliced transcripts were identified in the testes of mice with androgen insensitivity. It is noteworthy that we were not able to detect full-length type-2 transcripts in the testes of TSPY-ArTfm mice, suggesting that this variant is either very rare or absent in androgen-deficient gonads. It is unknown if these splice variants have any biological function, if they also occur in humans with complete or partial androgen insensitivity, or if they are involved in the process of gonadal tumorigenesis.
Aberrant Expression of TSPY in the Testes of TSPY-ArTfm Mice
TSPY transcripts are highly expressed in two precursor stages for human gonadal germ cell tumors, the carcinoma in situ of the testis and gonadoblastoma in dysgenetic gonads of XY-sex-reversed females [5, 9, 38], and it is hypothesized that aberrant TSPY expression may predispose to the formation of gonadal tumors through mitotic activation of germ cell and clonal expansion [6, 11]. TSPY is abundantly, yet variably, expressed in germ cells that stay in an unfavorable environment within dysgenetic gonads and gonads with androgen insensitivity syndrome as an attempt to survive and proliferate [3, 41, 42]. Such a TSPY expression profile is in line with the results from the animal model reported here. We also observed variations in the level of TSPY transcription in the testes of five age-matched TSPY-ArTfm mice, finding a two-fold upregulation of TSPY expression in three out of five of these in comparison to noncryptorchid controls. The observed increase in TSPY expression in maturation-delayed germ cells of ArTfm mice can be interpreted as an attempt of germ cells to compensate for the unfavorable environment of an undervirilized gonad [41, 42]. It should be noted that such variations of TSPY transcription levels were not observed in the original transgenic line at all [21], so variable copy numbers of the transgene are unlikely to account for the variations we observed. Neither can sequence differences of promoter and coding regions explain these variations in the investigated TSPY-ArTfm mice because all the mice were derived from the same transgenic founder who had integrated a single human transcriptional unit in approximately 50 copies at a single integration site on the mouse Y chromosome [21]. An important parameter that we were not able to correct for is the different genetic background of the analyzed TSPY-ArTfm and genetic background of matched noncryptorchid TSPY transgenic mice; we, therefore, cannot rule out completely that differences in TSPY expression level are due to the genetic background heterogeneity of TSPY-ArTfm and TSPY transgenic mice, and are not related to androgen insensitivity. In order to minimize these genetic differences, only mice of the first filial generation of the respective mating were used. Variations in TSPY immunostaining intensity in germ cells in androgen-insensitive gonads in comparison to normal adult and fetal testes have also been reported by Cools et al. [42], and our study shows that this is due to an upregulation of TSPY transcription in germ cells. It remains to be elucidated, whether variable transcriptional regulation of TSPY is also a feature in humans with undervirilization syndromes. In contrast to our finding, another study has reported an upregulation of TSPY expression after stimulation with androgens in the androgen-responsive prostatic cell line LNCaP [11, 43]. Nonetheless, a recent study has shown that TSPY expression is upregulated in androgen-independent LNCaP-C81 cells compared to androgen-dependent LNCaP-C33 prostate cancer cells, which is in line with our own finding [44].
Morphology and Histology of the Testes of TSPY-ArTfm Mice
The general morphology and histology of dysgenetic gonads in ArTfm mice are in line with previous studies [23, 45]. Spermatogenesis of nearly all TSPY-ArTfm mice ended at the pachytene spermatocyte stage or earlier, and thus it seems that human TSPY is not able to rescue the testicular phenotype of the androgen insensitivity syndrome in mice through mitotic stimulation of germ cells. This is not surprising because defective spermatogenesis in ArTfm mice is thought to be an unspecific secondary effect of cryptorchidism [23] and is likely an indirect effect of the lack of functional androgen receptors in Sertoli and peritubular myoid cells [46]. The observed multifocal hyperplasias of peritubular myoid and Leydig cells and multifocal tubular ectasia in TSPY-ArTfm and NMRI-ArTfm testes were also observed in undescended murine and human testes [23, 42, 45, 47], and thus also seem to be secondary effects of cryptorchidism, open to influence by unknown factors. Our observation that TSPY-ArTfm mice were more strongly affected by tubular dilatation than controls (P > 0.05, Table 2), points to an involvement of the TSPY gene in the pathogenesis of tubular ectasia in cryptorchid testes. The significant increase of testes weights in TSPY-ArTfm mice are in line with these findings.
Testicular Tumor Formation in TSPY-ArTfm and NMRI-ArTfm Mice
Androgen-insensitive gonads of TSPY transgenic ArTfm mice did not develop carcinoma in situ or malignant germ cell tumors, but testicular Leydig cell tumors with an incidence of 10.9% in 17- to 24-mo-old TSPY-ArTfm mice, and of 9.7% in 20- to 24-mo-old NMRI-ArTfm controls, did occur. The relatively high incidence of Leydig cell tumors in TSPY- and NMRI-ArTfm mice is noteworthy, as they normally occur in only 1.8% of ArTfm mice [48], suggesting a strong effect of the NMRI background on the genesis of Leydig cell neoplasm. The presence of these neoplasms in aging animals is in line with previous studies [49, 50]. Mice do not develop type-II germ cell tumors like testicular seminomas and nonseminomas, but testicular teratomas does arise spontaneously in mice of the 129/Sv inbred strain with an incidence of 1%–5% [51, 52]. Therefore, our strategy to generate TSPY transgenic ArTfm mice on a mixed genetic background [21, 23] could not be expected to lead to germ cell tumor formation by itself.
Although all patients with androgen insensitivity have an increased risk for the development of type-II germ cell tumor (seminomatous and nonseminomatous tumors), the risk is lower in patients with complete androgen insensitivity syndrome (CAIS) and higher in patients with the partial form (PAIS) [53]. It is assumed that the progressive loss of germ cells in the human CAIS gonads that begins at the age of 1 yr is responsible for this difference [34]. A depletion of germ cells was also observed in aging TSPY-ArTfm and NMRI-ArTfm mice, thus mimicking the human condition. At present, it is unknown whether TSPY is involved in the process of CIS and type-II germ cell tumor formation in patients with complete and partial androgen insensitivity, and further studies on the relationship between aberrant TSPY expression and its effect on gonadal tumor formation are necessary to support its putative oncogenic role in humans with the androgen insensitivity syndrome (CAIS and PAIS). The generation of a suitable mouse or rat model for PAIS would form an important basis to investigate such TSPY effects.
Currently, it is unknown whether TSPY acts as an oncogene in the pathogenesis of testicular germ cell tumors, and it is also uncertain whether mutations, aberrant expression, and/or differential allelic expression of the TSPY gene are responsible for its putative oncogenic potential [5, 11]. The present study has provided evidence for an upregulated transcription of the TSPY gene and aberrant splicing of TSPY transcripts in gonads with complete androgen insensitivity. It remains to be elucidated whether these aberrant TSPY expression profiles are present in gonads of humans with CAIS and PAIS and contribute to CIS formation and progression to malignant transformation in patients with undervirilization syndromes.
ACKNOWLEDGMENTS
We are grateful to Mr. Luke Lewis, Western General Hospital, MRC Human Genetics Unit (Edinburgh, UK), for revising a first version of the manuscript.
FOOTNOTES
1Supported by the Deutsche Forschungsgemeinschaft (SCHM 373/16-2 to J.S.). 
Correspondence: 2Stephanie Schubert, Institute of Human Genetics, Hannover Medical School, Carl-Neuberg-Str.1, D-30625 Hannover, Germany. FAX: 0049 511 532 5865; e-mail: schubert.steffi{at}mh-hannover.de
3Current address: Center of Human Genetics, D-79100 Freiburg, Germany.
4Current address: Institute of Neurology, Hannover Medical School, D-30625 Hannover, Germany.
Received: 5 December 2007.
First decision: 3 January 2008.
Accepted: 26 March 2008.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
