SEARCH

SEARCH BY CITATION

Contents

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References

Male felids frequently show teratospermia. At least in the domestic cat model, teratospermia is accompanied by impaired regulation of testicular apoptosis. We hypothesize that this phenomenon is caused by dysregulations in oestrogen signalling pathways. Both classical oestrogen receptors (ESR1 and 2) are expressed in species and/or tissue-specific manners and display different variants, inter alia, caused by alternative splicing. In vitro studies showed that exon deleted transcripts are translated into proteins and that some of the variants modify the effects of the full-length ERs. It has been proposed that some of the functional and morphological dysregulations, for example, during spermatogenesis, could directly derive from this phenomenon. In the present basic study, we investigated the expression pattern of ESR1 splicing variants in the gonads of domestic cats. Testicular, epididymal as well as ovarian tissue samples were collected from routine castrations. ESR1 variants were detected by means of RT-PCR using primers spanning one to three exons. We detected the variants Δ4 and Δ7 in all tissue samples investigated. Additionally, the testicular parenchyma expressed the variant Δ6 and double exon deletions of ESR1 (Δ4/6 and Δ6/7). Using an antiserum recognizing all previously identified ESR1 splicing variants, we revealed ESR1 proteins being expressed in nearly all cells of the testicular and ovarian parenchyma. ESR1 Δ6 protein, however, detected by an antiserum specifically raised against the Δ6 variant, was predominantly located in Sertoli cells. As the exon deletion variants are significantly expressed and show a distinct expression pattern, they could specifically modulate the cellular responsiveness to hormonal stimuli within the gonads.


Introduction

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References

Teratospermia is relatively common among felids and might be causal for reduced fertility especially in some endangered species (Pukazhenthi et al. 2006). In cats, teratospermia does not only compromise sperm quality, but is also characterized by disruption of testicular function. Teratospermic cats produce more sperm by virtue of more sperm-producing tissue, more germ cells per Sertoli cell and reduced germ cell loss during spermatogenesis (Jewgenow et al. 2009). Apart from testosterone, oestrogen also regulates testicular functionality. The exact molecular mechanisms by which oestrogens regulate the functions of the seminiferous epithelium are, however, still not fully elucidated (Carreau and Hess 2010). One more general role for oestrogens in spermatogenesis is the regulation of apoptosis, as they act as positive or negative regulators of programmed cell death (Lewis-Wambi and Jordan 2009). As apoptosis plays a crucial role in eliminating defect spermatozoa during spermatogenesis, and oestrogens are involved in apoptosis regulation, we hypothesize a link between impaired oestrogen signalling and the occurrence of teratospermia.

Oestrogen receptors (ERs) belong to the nuclear hormone receptor superfamily and comprise of different functional domains, inter alia a ligand-binding domain (LBD), a DNA-binding domain (DBD) and two activation sites (AF1/2, activation function one and two; Fig. 1; Aranda and Pascual 2001). The open reading frames of nuclear receptors cover eight exons (1–8); exons 2–3 encode the DBD, exons 4–8 the LBD. Different types of oestrogen receptor variants are known to exist (Hirata et al. 2003). Although splicing variants showing deletions of one or more exons of ERs are known since the early 1990s, the physiological role of these proteins is far from being understood. Many of the variants are translated into proteins and seem to modify the effects of the full-length ERs (Taylor et al. 2010). In human testicular cells, ERbeta (ESR2) splicing variants show a cell-type-specific expression pattern (Aschim et al. 2004). Therefore, they might be able to differentially influence hormone responsiveness of these cells and affect regulation of spermatogenesis.

image

Figure 1. Scheme of the functional domains of ESR1 and the corresponding exon structure of the mRNA. Arrowheads indicate the localization of the peptide used for ESR1Δ6 antiserum production

Download figure to PowerPoint

The aim of the present basic study was to identify the exon deletion splicing variants of ESR1 expressed in the testis, epididymis and ovary of the domestic cat. Furthermore, we present the localization of ESR1 in cat testicular tissue using a cat-specific antibody recognizing all possible splicing variants and show testis-specific variants within the parenchyma.

Material and Methods

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References

Tissue collection

Testes of adult tomcats (Felis silvestris catus, >1 year, n = 8) were received after routine castration from the animal shelter in Berlin, Germany. All samples were collected in late spring (May to June). Epididymides and testes were carefully separated immediately after castration. One epididymis was segmented into caput, corpus and cauda. Testicular and epididymal tissue was snap frozen in liquid nitrogen and stored at −80°C prior to RNA and protein isolation. The remaining testis was fixed in chilled Bouin's solution overnight at 4°C. For immunohistochemistry, fixed tissues were washed briefly in 70% ethanol, dehydrated and embedded in paraffin. Sections (~3 μm thickness) were cut and mounted on Superfrost ® slides (Gerhard Menzel; Glasbearbeitungswerk GmbH&Co. KG, Braunschweig, Germany). One slide of each animal was stained with hemalum/eosin and evaluated by light microscopy. The samples were included in the study only if spermatogenesis was fully active (n = 5 animals).

Ovaries from three queens (>1 year) were processed as described above (freezing and fixation/embedding).

RNA isolation and cDNA synthesis

Testicular and ovarian tissue was homogenized with Lysing Matrix D tubes (MP Biomedicals Germany GmbH, Eschwege, Germany) and epididymal tissue via IKA® Ultra-Turrax T8 (IKA®-Werke GmbH & Co.KG, Staufen, Germany). RNA samples were then extracted with InviTrap® Spin Cell RNA Mini kit (Invitek GmbH, Berlin, Germany) corresponding to the manufacturer's instruction. RNA integrity was confirmed by electrophoresis and ethidium bromide staining. Reverse transcription (RT) reactions were carried out using RevertAid reverse transcriptase and random hexamers (Fermentas GmbH, St. Leon-Rot, Germany). Standard reaction mixture contained 1 μg of total RNA, 200 U of reverse transcriptase, dNTPs 200 μm each, 2 μm random hexamers mixed with 5x RT buffer in a total volume of 20 μl. Reverse transcription reactions were performed for 1.5 h at 42°C.

PCR

Polymerase chain reaction was performed with 2.5 μl of cDNA in a final volume of 25 μl containing forward and reverse primers (Table 1), 1 Unit of IMMOLASE (Bioline GmbH, Luckenwalde, Germany), 3 mm MgCl2, dNTPs 200 μm each, mixed with 10x IMMOLASE buffer. The conditions for amplification were as follows: initial denaturation at 95°C for 10 min followed by 40 cycles comprising of denaturation at 95°C for 30 s, annealing at 60°C for 10 s and elongation for 30 s at 72°C. The final extension was at 72°C for 2 min.

Table 1. Primer sequences used for RT-PCR
PrimerSequence (5′ – 3′)
ex 2 fwCAGGTGCCCTATTACCTGGA
ex 4 fwGGCCTTCTTCAAGAGAAGTATTCA
ex 6 fwCCTATTTGCTCCCAACTTGC
ex 5 revAGATCTCCACCATCCCCTCT
ex 6 revTCCAGAGACTTCAGGGTGCT
ex 7 revTACAGGTGCTCCATGCCTTT
ex 8 revAGGGATTCTCAGGACCTTGG

ESR1 splice variant amplification was performed with exon flanking primers (Table 1) based on published oestrogen receptor sequences (Cardazzo et al. 2005).

Detection and sequencing of PCR products

The amplified ESR1 splice variant products were separated by electrophoresis in 1.5% agarose gels in Tris–acetate EDTA buffer and detected by ethidium bromide staining. To determine the identity of the variants, they were purified individually using the QIAquick gel extraction kit (QIAGEN GmbH, Hilden, Germany). The purified products were sequenced with BigDye® (Applied Biosystems, Darmstadt, Germany) at GATC Biotech AG, Konstanz, Germany.

Detection of ESR1 proteins

Based on ESR1 splice variants detected by means of PCR, two polyclonal antisera were manufactured by BioGenes, Gesellschaft für Biopolymere mbH, Berlin, Germany. One antiserum was produced by immunization of two rabbits with a peptide present in all splicing variants of ESR1 (C-TSDNRRQSGRERLA). The other antiserum was raised against a peptide only present in the variant Δ6 and Δ4/6 (C-RHLDPPDGQSGPV).

For immunohistochemistry, tissue sections were deparaffinised, rehydrated and washed in phosphate-buffered saline, pH 7.4 (PBS). Demasking of antigens was performed by boiling slides in citrate buffer for 2 min (pH 6.0) and additionally cooling down in buffer for another 20 min. Endogenous peroxidase activity was blocked by incubating the slides twice in H2O2/methanol (3%) for 15 min. Sera (rabbit anti-ESR1 WT or ∆6; diluted 1 : 100 in PBS) were incubated over night at 4°C. For negative control, slides were incubated with rabbit preimmune serum (1 : 100 in PBS) instead of the primary antibody.

After washing the slides three times for 15 min in PBS-T, sections were incubated with secondary antibody conjugated with horseradish peroxidase (Goat anti-rabbit poly-HRP; Thermo Fisher Scientific, Waltham, USA) diluted 1 : 50 in PBS containing 1%BSA. Peroxidase was visualized using diaminobenzidine (Thermo Fisher Scientific; 1–4 min, under visual control).

Results

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References

Analysis of ESR1 transcript profiles

The expression pattern of ESR1 exon deletion variants in the feline reproductive tissues is depicted in Fig. 2. The lane descriptions and expected sizes of amplification products of the different variants are summarized in Table 2.

Table 2. Detailed lane description for Fig. 2
Lane numberAmplified exonsMolecular weight wild type (bp)Molecular weight deletion variants (bp)Only amplified and sequenced in testicular tissue
12–3438  
22–4739Δ4 403 
33–4511Δ4 175 
43–5667Δ4 331 
54–5656Δ4 320 
64–6792Δ4 456 
Δ6 658X
Δ4/6 322X
75–6342Δ6 208X
85–7510Δ6 376X
Δ7 326 
Δ6/7 192X
96–7372Δ6 238X
Δ7 188 
Δ6/7 54X
106–8619Δ6 485X
Δ7 435 
Δ8 385 
Δ6/7 301X
Δ7/8 201 
117–8491Δ7 307 
Δ8 257 
image

Figure 2. Analysis of ESR1 transcript profiles in testicular parenchyma, epididymal and ovarian tissue by RT-PCR spanning one or more exons. See also Table 2 for detailed lane descriptions. Marker lane: Fermentas 1-kb ready-to-use ladder

Download figure to PowerPoint

In all tissue samples under investigation, we could show the single exon deletion variant Δ4 (lane 1–5) and Δ7 (lane 11). In addition to these two variants, we found the single exon deletion variant Δ6 (lane 7) and the double deletion variants Δ4/6 (lane 6) and Δ 6/7 (lane 8–10) in testicular tissue only. A band possibly representing the single exon deletion variant Δ8 was visible in some samples (lane 10 and 11), but could not be confirmed by sequencing.

Immunohistochemical detection of ESR1 and ESR1 variant Δ6

Immunohistochemistry showed signals for ESR1 variants in nearly all cells of the testicular and ovarian parenchyma. The signals were nuclear as well as cytoplasmatic. A clear nuclear staining was detected in Leydig cells. ESR1Δ6, however, could only be detected in testicular tissue. The staining was predominantly located in the nuclei of Sertoli cells. A weak cytoplasmic signal was present in nearly all testicular cells. The ovarian tissue did not show ESR1Δ6 protein expression (Fig. 3).

image

Figure 3. Immunohistochemical detection of ESR1 variant proteins in feline testicular (a/b) and ovarian (c/d) tissue. (a/c) Antiserum against all ESR1 variants, arrowheads point towards strongly nuclear stained Leydig cells; (b/d) antiserum against ESR1 Δ6, arrowheads point towards strongly nuclear stained Sertoli cells. Counterstain: hemalum, small pictures: negative control (preimmune serum)

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References

This study represents the first report of ESR1Δ6 being expressed on mRNA and protein level in mammalian testicular tissue.

The appearance of various splicing variants of ESR1 mRNA is a well-known phenomenon for many tissues. This is true for physiological conditions as well as for pathology, where especially estradiol-dependent cancers brought up the necessity to clarify the role and diagnostic value of these splicing variants (for review see Taylor et al. 2010). For ESR1 exon deletion variants are the most common type of alternative splicing variants (Ferro et al. 2003). Some of the exon deletion variants (especially Δ5 and 3) seem to have a direct role in classical oestrogen signalling. The functional relevance and tissue- or cell-type-specific effects are still under discussion (Bollig and Miksicek 2000; Horvath et al. 2002; Garcia Pedrero et al. 2003; Bryant et al. 2005). Even though most of the proteins resulting from exon deletions are expressed in vitro, reports about the detection of the translated proteins in mammalian tissue ex vivo are rare (Ishunina and Swaab 2008). In concordance to other publications, where deletion variants 4 and 7 were the most frequently found single exon deletion variants (Ishunina and Swaab 2008), we found these variants also to be widely expressed in feline reproductive tissues. One further deletion variant, ESR1Δ6, however, was identified in the testicular parenchyma only. Also double deletion variants missing exon 6 (Δ4/6 and 6/7) could only be shown in the testicular samples. To our knowledge, this is the first report about the expression of ESR1 Δ6 in testicular cells. Cardazzo et al. (2005)detected the mRNA of this variant also in feline ovarian tissue using the ‘splice targeted approach’. This result could neither on mRNA nor on protein level be verified in our study. Using an antiserum raised against a peptide exclusively present in the feline ESR1Δ6 variant, we detected the truncated protein in testicular, but not in ovarian cells. Interestingly, we found the protein of ESR1Δ6 to be predominantly located in the nuclei of Sertoli cells.

ESR1Δ6 comprises the first five exons (coding for AF1, the DBD and parts of the LBD) of the wild-type variant. The deletion of exon six causes a shift of the reading frame. After exon five, 61 amino acids different from the wild-type amino acid sequence are coded by the spliced mRNA before an internal stop codon occurs. Therefore, this variant might theoretically be able to dimerize and bind to DNA. However, protein folding patterns are surely heavily altered by the new C-terminus of the protein.

In vitro studies including ESR1 Δ6 variant did not yet reveal a function for this protein. Expression of ESR1 Δ6 in Cos 7 cells showed that the protein was neither able to bind estradiol nor DNA (Bollig and Miksicek 2000) and therefore at least in these cells, does not take part in the classical oestrogen signalling pathway. The same authors report that due to the impossibility to dimerize, the ESR1 Δ6 variant will possibly not enter the nucleus. In contrary, Picard et al. showed in CV1 cells that ERs (in contrast to glucocorticoid receptors) are able to enter the nucleus without ligand binding and dimerization. A truncated form of the oestrogen receptor only containing aa 1–302 and therefore (as ESR1Δ6) lacking the estradiol-binding domain as well as the ability to dimerize was found to be strongly nuclear (Picard et al. 1990).

The testis-specific expression of ESR1Δ6 and its distinct localization suggest a particular role of this splice variant within Sertoli cell function.

We hypothesize that in Sertoli cells, which are the key transit points of hormonal signals between somatic and germ cells, alternative splicing of ERs is one more regulating mechanism to specifically modulate the cellular responsiveness to hormonal stimuli. The splicing event as such and hence the perhaps selective inactivation of single receptor molecules might represent the biological role of some splice variants like Δ6.

These results build the basis for future investigations comparing the ESR1 and ESR1 deletion variant expression patterns in teratospermic and normospermic felids.

Acknowledgements

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References

This study was supported by the German Research Council (DFG, Je 163/9-1 and Scho 1231/2-1).

Author contributions

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References

JS and KJ conceived of the study, coordinated and supervised the study performance, and drafted and revised the manuscript. SS supervised the molecular biological experiments and revised the manuscript. JR performed the molecular biological experiments and drafted parts of the manuscript. JS and JR carried out the immunohistochemistry. All authors have approved the final version of the manuscript.

References

  1. Top of page
  2. Contents
  3. Introduction
  4. Material and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflicts of interest
  9. Author contributions
  10. References