Mol. Hum. Reprod. Advance Access originally published online on April 28, 2006
Molecular Human Reproduction 2006 12(6):357-365; doi:10.1093/molehr/gal033
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Tyrosine kinase B receptor and its activated neurotrophins in ovaries from human fetuses and adults
1Infertility and IVF Unit, 2Helen Schneider’s Women Hospital, Rabin Medical Center, Petach Tikva and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel, 3Department of Obstetrics and Gynecology, Royal Victoria Hospital, McGill University, Montreal, Quebec, Canada, 4Department of Pathology, Rabin Medical Center, 5The Felsenstein Medical Research Center, Petach Tikva and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel and 6Department of Human Genetics, McGill University, Montreal, Quebec, Canada
7 To whom correspondence should be addressed at: IVF Research Laboratory, Infertility and IVF Unit, Helen Schneider’s Women Hospital, Rabin Medical Center, Beilinson Campus, Petach Tikva 49100, Israel. E-mail: ronita{at}clalit.org.il
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The signals initiating the growth of primordial follicles are unknown. Growth factors such as neurotrophin 4/5 (NT-4/5) and brain-derived neurotrophic factor (BDNF) may play a role in this process. To investigate the expression of NT-4/5 and BDNF and their receptor tyrosine kinase B (TrkB) in the early developing follicles, we fixed and froze 12 ovarian samples from adolescents/adults and 31 ovaries from human fetuses. The fixed samples were prepared for immunohistochemical staining for NT-4/5, BDNF and the TrkB receptor. Total RNA was extracted from the frozen ovarian samples, and the expression of NT-4/5, BDNF and the TrkB receptor (full length and two truncated isoforms) was investigated by RT–PCR. Products were resolved by 1% agarose gel electrophoresis and image analysis. Immunohistochemical staining revealed the expression of NT-4/5 and BDNF mainly in oocytes and, in a minority of samples, also in the granulosa cells (GCs); TrkB receptor was identified in oocytes and GCs. Transcripts of NT-4/5, BDNF and all forms of TrkB receptor were identified in the samples. To elucidate whether indeed NT-4/5 and BDNF are involved in growth initiation of human primordial follicles, they should be added to the culture medium.
Key words: immunohistochemistry/neurotrophins/primordial follicles/RT–PCR/TrkB receptor
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
In the fetal ovary before follicular assembly, the germ cells are present in the form of small oogonia (Gosden, 1995
). Human follicular formation begins during gestational weeks (GWs) 20–22, when a single flat layer of cells surrounds the larger germ cells, the oocytes, to form primordial follicles. In mice and rats, however, follicular formation is initiated during the first post-natal days (Ojeda et al., 2000). Most of the human ovarian follicles in fetuses as well as in adults remain unilaminar (30–50 µm): mostly primordial follicles but also primary follicles with cuboidal granulosa cells (GCs) (Gougeon, 1996). The signals that trigger their growth into multilayered secondary follicles are unknown.
Understanding the signals responsible for the initiation of folliculogenesis is an important step towards developing a successful in vitro maturation system for primordial follicles—a putative technology that would restore fertility in cancer patients without the risks of reseeding the disease (Abir et al., 1998). It is possible that various neurotrophins (NTs) such as brain-derived neurotrophic factor (BDNF) and NT-4/5 (also known as NTF-4, NTF-5, NT-4, NT-5) might be involved in this process (Van den Hurk et al., 2000). To keep uniformity, we will refer to neurotrophin 4/5 as NT-4/5 throughout the manuscript (Hugo gene nomenclature website: http://www.gene.ucl.ac.uk/nomenclature).
NTs are small, homodimeric polypeptide growth factors that not only regulate the survival, maintenance and differentiation of neurons in the nervous system (Ibanez, 1995; Anderson et al., 2002; Paredes et al., 2004) but also act on non-neural cells, as in the reproductive system (Ibanez, 1995; Anderson et al., 2002; Paredes et al., 2004). NTs are initially synthesized as larger precursor molecules that undergo proteolytic cleavage to create the mature proteins. Both BDNF and NT-4/5 contain six cysteine residues, but NT-4/5 also includes an insertion of seven amino acids between its second and third cysteines (Ip et al., 1992). NT-4/5 shares a 95% amino-acid-sequence identity with BDNF, and it is the only family member that has a truncated precursor region. The gene encoding NT-4/5 is located on chromosome 19 in the human.
The NTs have two distinct receptor types, with no sequence similarity located on the cell surfaces: p75 (Chao and Hempstead, 1995) and tyrosine kinase (Trk) receptors. p75 is a receptor for all NTs (Chao and Hempstead, 1995; Fridman and Greene, 1999; Abir et al., 2005) and was previously identified by our group only during human follicular assembly in fetuses (Abir et al., 2005).
The Trk receptor family consists of three known, 140 kDa in size, high-affinity protein Trk receptors (Stoilov et al., 2002): TrkA, TrkB and TrkC (Ross, 1991). They differ from each other in their extracellular part, which provides them their ligand-binding specificity. It consists of a leucine-rich region, two cysteine-rich domains, two immunoglobulin-related domains and a consensus amino acid sequence of arginine, proline, any amino acid and tyrosine (NPXY) for internalization through coated pits. The receptors also contain a transmembrane part and a cytoplasmic region with a catalytic Trk domain (Chao and Hempstead, 1995). When NTs bind to their receptor, they activate a specific Trk domain, resulting in a rapid increase in the phosphorylation of second messengers and other specific cellular components (Chao and Hempstead, 1995; Fridman and Greene, 1999).
The TrkB receptor is activated by BDNF and NT-4/5 (Chao and Hempstead, 1995; Ibanez, 1995; Fridman and Greene, 1999; Spears et al., 2003), and the TrkB receptor locus is located on human chromosome 9 (Stoilov et al., 2002). In humans, the TrkB receptor is expressed in three major forms: a full-length form (TrkB-fl) and two truncated isoforms (TrkB-Shc and TrkB-T1) (Figure 1). The TrkB-fl receptor form contains all three types of receptor domains, with an Shc-binding site in the intracellular part and a tail region containing a phospholipase C-gamma-binding site (PLC-
). The truncated receptor forms lack the Trk domain and cannot signal to the cytoplasm. TrkB-T1 receptor form contains a short, specific cytoplasmic region in place of the kinase domain (Haapasalo et al., 2002; Stoilov et al., 2002). In rodents, a truncated receptor form similar to the human TrkB-T1 receptor form exists with a unique intracellular tail region (TrkB-T2 receptor form). TrkB-Shc receptor form contains the Shc-binding site and is expressed predominantly in the brain. Both the TrkB-fl and TrkB-Shc receptor forms are capable of initiating intracellular signalling.
|
Studies have shown that NT-4/5, BDNF and the TrkB receptor are expressed in oogonia (Spears et al., 2003), in oocytes (Spears et al., 2003; Paredes et al., 2004) and in GCs (Paredes et al., 2004) of mice (Spears et al., 2003; Paredes et al., 2004), of rats (Dissen et al., 1995) and of human fetuses aged 13–21 GWs (Anderson et al., 2002). Mice with deficiencies in the TrkB receptor genes (Dissen et al., 2002; Spears et al., 2003; Paredes et al., 2004) or in the NT-4/5 and BDNF genes had reduced numbers of follicles and oocytes (Spears et al., 2003; Paredes et al., 2004) and a decrease in follicular growth (Ojeda et al., 2000). When fetal (Spears et al., 2003) and neonatal (Dissen et al., 1995; Spears et al., 2003) mouse (Spears et al., 2003), rat (Dissen et al., 1995) and human fetal ovaries (Spears et al., 2003) were cultured with an inhibitor of the Trk receptors (Dissen et al., 1995; Spears et al., 2003) or with antibodies against NT-4/5 and BDNF (Spears et al., 2003), the survival of oogonia/oocytes (Spears et al., 2003) and primordial follicles (Dissen et al., 1995) decreased.
The aim of this study was, therefore, to investigate the expression of BDNF, NT-4/5 and the three forms of the TrkB receptor in human ovaries from adults and second-to-third trimester fetuses at the RNA-protein levels.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Human adult and fetal ovaries
Fetal ovarian samples were obtained from 31 aborted human fetuses aged 19–33 GWs (Table I). The abortions were induced by prostaglandins. All but four normal fetuses had anatomical and chromosomal abnormalities (Table I). Three of the normal fetuses were derived from legal terminations indicated for psychiatric reasons and the fourth from a termination due to premature preterm rupture of the membranes. As the departmental pregnancy-termination policy at our Center mandates feticide for all fetuses aged over 21 GWs, only fetal specimens shown to have no or very low apoptosis were used for the present analyses. The use of a deoxynucleotidyl transferase assay for the selection of viable fetal ovarian samples has been discussed in detail elsewhere (Abir et al., 2004
). Ovaries from the 13 fetuses aged 
21 GWs were obtained within 1 h from fetal expulsion. Those from fetuses aged 
22 GWs were collected within 24 h of the feticide, a time frame wherein the ovaries remain non-apoptotic (Abir et al., 2002).
|
In addition to the fetal ovarian samples, small ovarian biopsy samples were donated after informed consent from 12 premenopausal women and adolescents aged 13–38 years undergoing various gynaecological laparoscopic operations (Table II). Eight of the patients had various forms of cancer, and their operation was performed for ovarian biopsy for cryopreservation before chemotherapy (Abir et al., 1998). All the samples were handled in our laboratory within an hour of surgery. The institutional ethics committee at Rabin Medical Center approved the study protocol, and every woman or minor’s parents signed an informed consent. A portion of every sample was cut to be as uniform in size as possible (approximately 2 x 2 mm) and fixed immediately in neutral buffered formalin. The remainder was frozen for subsequent RNA extraction.
|
Cryopreservation of ovarian tissue
Five to seven tissue slices measuring 1–2 mm were placed in cryogenic vials (Nalge Nunc International, Roskilde, Denmark) filled with a solution of 1.5 M dimethylsulphoxide (DMSO) (Sigma, St Louis, MO, USA) and 0.1 M sucrose (Sigma) (Newton et al., 1998). Before freezing, the samples were kept on ice for 0.5 h. All samples were frozen gradually in a programmable freezer (Kryo 360-1/7, Planer Biomed, Sunbury on Thames, UK) and were immediately placed in liquid nitrogen. The slices were cryopreserved-stored for 3–24 months until RNA extraction.
Histological preparation and immunohistochemistry for BDNF, NT-4/5 and the TrkB receptor
The fixed specimens were dehydrated in a graded series of ethanol followed by paraffin embedding and sectioning (5 µm). Unstained sections were placed on OptiPlus positive-charged microscope slides (BioGenex Laboratories, San Ramon, CA, USA) and deparaffinized and rehydrated. Unless otherwise stated, all dilutions were performed with phosphate-buffered saline (PBS) at pH 7.4 (Biological Industries, Beit Ha’emek, Israel), which served also as the main rinsing solution. The incubations were carried out at room temperature. Solutions from two commercial kits were used for the immunohistochemical staining: an EnVision+ System, horse-radish peroxidase (HRP) 3-amino-9-ethylcarbazole (AEC) containing hydrogen peroxide anti-rabbit kit (DAKO Cytomation, Glostrup, Denmark) for BDNF (sc-446 and sc-2098: see Table III), NT-4/5 (sc-445: see Table III) and TrkB receptor (sc-8316: see Table III) identification and an LSAB+ System, HRP kit (DAKO) for TrkB receptor identification (sc-20542: see Table III).
|
To enhance antigen retrieval, we microwaved all the slides with citrate buffer at pH 6.0 (diluted in distilled water from a 10x citrate buffer solution at pH 6.0, CheMate, DAKO), first under full power for 15 min and then simmered for another 15 min. The antigen retrieval procedure was terminated by cooling and rinsing the sections in distilled water and PBS (Biological Industries). To block endogenous peroxidase activity, we quenched the slides for BDNF (sc-446 and sc-2098: see Table III), NT-4/5 (sc-445: see Table III) and TrkB receptor (sc-8316: see Table III) identification in 0.03% sodium azide (from an EnVision+ System, HRP-AEC kit, DAKO) for 5 min and quenched those for TrkB receptor (sc-20542: see Table III) staining in 3% H2O2 (Vitamed, Binyamina, Israel) for 5 min in the dark.
All sections were then incubated with the primary antibodies, which are described in Table III. To the best of our knowledge, no commercial antibodies recommended for immunohistochemistry that can identify exclusively NT-4/5 or specifically all three human TrkB receptor isoforms are available commercially. Negative control solutions for rabbit polyclonal antibodies for BDNF (sc-446: see Table III) and NT-4/5 (sc-445: see Table III) and for the goat polyclonal antibody for the TrkB receptor (sc-20542: see Table III) were prepared by absorbing equal volumes of the diluted primary antibodies with the corresponding blocking peptides (see Table III). All these negative control solutions were incubated for at least 1 h before application. A normal rabbit IgG antibody (see Table III) served as a negative control for the rabbit polyclonal antibodies for TrkB receptor (sc-8316: see Table III) and BDNF (sc-2098: see Table III).
Sections processed with rabbit polyclonal antibodies for BDNF (sc-446 and sc-2098: see Table III), NT-4/5 (sc-445: see Table III) and TrkB receptor (sc-8316: see Table III) detection were then incubated with peroxidase-labelled polymer conjugated to goat anti-rabbit immunoglobulins in PBS-containing carrier protein and 15 mM sodium azide (from an EnVision+ System, HRP-AEC anti-rabbit kit, DAKO) for 30 min. Sections processed with the goat polyclonal antibody for TrkB receptor (sc-20542: see Table III) identification were incubated with biotinylated anti-rabbit, anti-mouse and anti-goat immunoglobulins in PBS containing carrier protein and 15 mM sodium azide (link from DAKO LSAB+ System, HRP, DAKO) for 30 min. After further rinsing, sections processed with the goat polyclonal antibody for TrkB receptor (sc-20542: see Table III) were incubated with streptavidin conjugated to HRP containing carrier protein and antimicrobial agents (Streptavidin HRP from LSAB+ System, HRP, DAKO) for 30 min, followed by exposure to a diaminobenzidine urea H2O2 solution in distilled water (Sigma Fast tablets, Sigma) for 5 min (brown staining = expression of TrkB receptor). Sections processed with the rabbit polyclonal antibodies for BDNF (sc-446 and sc-2098: see Table III), NT-4/5 (sc-8316: see Table III) and TrkB receptor (sc-8316: see Table III) detection were incubated with AEC+substrate-Chromogen (from EnVision+ System, HRP-AEC kit, DAKO) for 30 min (red-brown staining = expression of BDNF or NT-4/5 or TrkB receptor). All the sections were counterstained with Mayer’s haematoxylin (Pioneer Research Chemicals, Colchester Essex, UK) (blue staining).
RNA extraction
RNA samples were extracted from the same ovarian samples used for immunohistochemistry. The frozen ovarian fragments were partially thawed at 37°C, rapidly removed from the semi-frozen DMSO solution and placed in TRizol Reagent (Pioneer Research Chemicals) at room temperature and homogenized (Abir et al., 2004). To obtain RNA fractions (supernatants) from these homogenates, we first added chloroform (Biolab, Jerusalem, Israel) and then a solution of phenol–chloroform–isoamyl alcohol (Sigma), followed by subsequent centrifugation at 4°C. To obtain RNA pellets, we mixed the RNA fractions with isopropanol (Biolab) and kept overnight at –20°C, followed by centrifugation at 4°C. Finally, ethanol (75%) was added to stabilize the pellet. The samples were kept at –80°C, until RT–PCR was performed.
RT–PCR
The cDNA amplification primers for BDNF, NT-4/5 and TrkB receptor were designed to span introns so as to eliminate genomic DNA contamination. The inner primers for TrkB-fl receptor form were selected from exon 22 and were located on the Trk region. The inner primers for TrkB-Shc receptor form were selected from exon 18 and the alternative terminating exon 19 and used only for the TrkB-Shc receptor isoform. The inner primers for TrkB-T1 receptor form were selected from exon 16 and used only for the TrkB-T1 receptor isoform (a schematic diagram of the primers design is shown in Figure 1).
Frozen total RNA samples from nine adult and nine fetal ovaries were centrifuged at 13 000 g for 30 min at 4°C. After the supernatant was completely removed, the pellets containing RNA were resuspended in 50 µl RNAse-free diethyl procarbonate (DEPC)-treated water. The concentration of each sample was measured by spectrophotometer (Cary UV 100, Varian, Mulgrave, Australia), and the sample was stored at –70°C. A total of 0.5 µg RNA was used for cDNA synthesis in the presence (RT+) or absence (RT–) of RT. The final reaction mixture, containing RNA 10 µl, Oligo DT (dT) 1 µl, 10 mM dNTP 1 µl, 5x RT buffer 4 µl, 0.1 M DTT 2 µl, RNAsin 1 µl and 100 units of Moloney murine leukemia virus (M-MLV) RT (BRL, Burlington, Ontario, Canada), was incubated at 37°C for 1 h; the reaction was terminated by heating at 95°C for 5 min. cDNA amplification was performed with two rounds of hemi-nested PCR primers to increase the specificity and yield of the PCR product. The total 50 µl PCR reaction contained 5 µl reverse-transcribed cDNA, 5 µl of 10xPCR buffer (QIAGEN, Ontario, Canada), 10 mM of each dNTP, 2.5 units of Taq polymerase and 0.4 µM of each primer. Each reaction was overlaid with 50 µl of mineral oil and heated to 95°C for 4 min. PCR was carried out essentially as previously described (Ao et al., 1994), for 25 cycles with outer primers and for 30 cycles with inner, hemi-nested primers at the appropriate annealing temperatures for each set (Table IV). For the hemi-nested PCR cycles, 2 µl of primary product was added to 28 µl of freshly prepared mix, as above. The amplified product was subjected to 1% agarose gel electrophoresis, using a 100 base-pair ladder as a reference for fragment size, and stained with ethidium bromide. The PCR products for BDNF, NT-4/5, TrkB-fl receptor form, TrkB-Shc receptor form and TrkB-T1 receptor form from human fetal and adult ovaries were purified with QIAquick Purification Kit (QIAGEN) and sequenced commercially.
|
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Immunohistochemical detection of BDNF, NT-4/5 and TrkB receptor
Similar results for BDNF immunohistochemical staining were obtained with both antibodies (sc-446 and sc-2098: see Table III). Figure 2A and B shows the typical immunohistochemical findings for BDNF staining in ovaries from fetuses and women/adolescents. In most samples, the positive staining in the ovarian cells was low to moderate and correlated with age: stronger in ovaries of younger women and in specimens from older fetuses. Full cytoplasmic oocyte staining for BDNF was identified immunohistochemically in fetal oocytes/oogonia of all samples but one (from a 22 GWs fetus with anencephaly and open neural tube which showed partial cytoplasmic staining). There was nuclear staining for BDNF only in 30% of the fetal oocytes/oogonia. Although nuclear staining for BDNF was detected in all the oocytes from women/adolescents, there were inconsistencies in the cytoplasm: half the specimens exhibited partial cytoplasmic staining, and half exhibited full cytoplasmic staining. We found staining for BDNF in the GCs of less than 30% of the samples from women and fetuses. These inconsistencies in cytoplasmic and nuclear staining and in GCs staining were unrelated to fetal or adult age, follicular class (primordial, primary or secondary), fetal normality/abnormality or type of fetal abnormality.
|
Figure 3A and B shows the typical immunohistochemical findings for NT-4/5 staining in ovaries from women/adolescents and fetuses. The general level of staining for NT-4/5 was low to strong in the samples from fetuses and low to moderate in those from women/adolescents. In more than 85% of the samples, the fetal oocytes/oogonia showed full cytoplasmic staining for NT-4/5 without nuclear staining. There was cytoplasmic staining without nuclear oocyte staining in the samples from women/adolescents, despite inconsistencies in the staining pattern (partial or full). GCs staining for NT-4/5 was detected in one-third of the samples from women and in 40% of the samples from fetuses. No relation was found between the staining pattern for NT-4/5 in oocytes and GCs and follicular class, fetal or adult age, fetal normality/abnormality or type of fetal abnormality.
|
Similar results for TrkB receptor immunohistochemical staining were obtained with both antibodies (rabbit and goat polyclonal antibodies: sc-8316 and sc-20542 respectively; Table III). TrkB receptor was identified in both the GCs and the oocytes in all the fetal samples (Figure 4A) and the samples from the women/adolescents (Figure 4B). The general level of staining for TrkB receptor was low to moderate in the samples from fetuses and low to strong in those from the women. Full cytoplasmic oocyte staining for TrkB receptor was identified immunohistochemically in all the fetal oocytes/oogonia. There was nuclear oocyte/oogonia staining in 89% of the samples from fetuses under 23 GWs of age but not in the samples from any of the older fetuses but one (a 27-GWs-old fetus with cardiac and pulmonary abnormalities). There was cytoplasmic staining in all the oocytes from women/adolescents, although in 50% of the samples, the staining was only partial. There were also inconsistencies in the presence or lack of nuclear staining. There was no relation between the staining pattern for TrkB receptor in the oogonia/oocytes and fetal or adult age, fetal normality/abnormality, type of abnormality or follicular class.
|
Some of the stroma cells in all the samples stained for BDNF, NT-4/5 and the TrkB receptor. The negative controls did not stain positively and were blue (Figures 2C, 3C and 4C).
Detection of BDNF, NT-4/5 and TrkB receptor transcripts
Hemi-nested RT–PCR analysis was performed on nine adolescents/adults and nine fetal ovarian samples. All genes studied yielded the expected fragment sizes (Table IV and Figure 5) in the adolescents/women and fetal samples. Constitutively expressed hypoxanthine phosphoribosyl transferase (HPRT) gene was used as the positive control for the RT reaction. There was no contamination of genomic DNA in any of the samples tested, and none of the negative controls processed without RT (RT–) yielded an amplification product. Sequence analysis of the product confirmed the identity of BDNF, NT-4/5, TrkB-fl receptor form, TrkB-Shc receptor form and TrkB-T1 receptor form (data not shown). Sequences corresponding to positions 449–735, positions 248–507, positions 1606–1884, positions 1643–1772 and positions 1358–1623 were identical to the human BDNF, NT-4/5, TrkB-fl receptor form, TrkB-Shc receptor form and TrkB-T1 receptor form mRNA sequences, respectively, published by the gene bank (http://www.ncbi.nlm.nih.gov/).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Our study reports the concurrent detection of BDNF, NT-4/5 and their TrkB receptor in human adult and fetal ovaries by both immunohistochemistry and RT–PCR. All three ligands were identified for the first time in human ovarian primordial, primary and secondary follicles from fetuses aged >21 GWs and from women/adolescents. Our immunohistochemical results localized the expression of the proteins for BDNF, NT-4/5 and TrkB receptor in oocytes and in GCs of human fetal and adult ovarian follicles from primordial stages onwards as well as in the stroma cells. However, BDNF staining and NT-4/5 staining were detected in GCs of a minority of the samples. Transcripts of BDNF, NT-4/5 and TrkB (fl, T1 and Shc receptor forms) were detected in all the fetal and adult ovaries tested. This is the first study demonstrating the existence of TrkB-Shc receptor form outside the brain.
The TrkB receptor gene can create at least 100 isoforms that encode 10 different proteins, of which three are considered major (Stoilov et al., 2002). The truncated TrkB-T1 and TrkB-Shc receptor forms probably act as dominant negative forms, suppressing NTs’ action by regulating the function of the catalytic form of TrkB receptor and differentially regulating TrkB receptor surface expression. TrkB-Shc receptor form has a longer juxtamembrane region than TrkB-T1 receptor form and is highly conserved between humans and mice. It is overexpressed in plasma membranes but is also expressed moderately in the cytoplasm. TrkB-fl receptor form can bind to TrkB-Shc receptor form but is unable to phosphorylate it. Studies have shown that coexpression of TrkB-fl receptor form with TrkB-Shc receptor form increased levels of TrkB-fl receptor form. However, coexpression of TrkB-fl receptor form with Trk-T1 receptor form led to significant decrease in TrkB-fl receptor form levels on cell surfaces and inhibited cell survival and TrkB receptor signalling. BDNF rapidly increased the cell-surface expression of TrkB-fl receptor form (Haapasalo et al., 2002).
Although most of the fetal ovaries in our series were derived from abnormal fetuses after feticide, the spectrum of abnormalities was very wide, and there were no differences in the expression pattern between ovaries derived from normal and abnormal fetuses. Thus, it seems highly unlikely that all the ovaries from the abnormal fetuses were connected with ovarian defects. Furthermore, seven ovarian specimens studied in the present investigation were also used in a previous study and survived in culture for 4 weeks, with an increase in 17ß-estradiol (E2) secretion (Biron-Shental et al., 2004), further supporting the viability of the ovarian follicles.
This study detected stronger positive staining for BDNF in specimens from older fetuses than from younger ones and lack of nuclear oocyte staining in fetuses 23 GWs, which might be related to fetal follicular and oocyte maturation processes. As most of the ovarian samples were also used to identify other growth factors and their receptors without similar immunohistochemical variations (Abir et al., 2004, 2005), it is highly unlikely that these and other current staining variations were a result of quality differences between the specimens. It is possible, however, that some of the immunoreactivity associated with NT-4/5 might be attributed to NT-6.
In the mouse, in situ hybridization localized both the TrkB-fl receptor form (Spears et al., 2003; Paredes et al., 2004) and its truncated form (Paredes et al., 2004) in oogonia/oocytes before (Spears et al., 2003), throughout and after (Spears et al., 2003; Paredes et al., 2004) follicular assembly and during early neonatal folliculogenesis. The proteins for NT-4/5 and BDNF were initially detected in oocytes of primordial follicles and their expression switched to the GCs during the beginning of follicular growth.
In the neonatal rat ovary, transcripts for TrkB receptor and NT-4/5 were predominantly expressed at the time of primordial follicular formation. Specifically, NT-4/5 mRNA was detected mainly in a subpopulation of oocytes (Dissen et al., 1995); NT-4/5 and TrkB receptor became barely detectable in 28-day-old rat ovaries; and BDNF was expressed at low levels during follicular assembly and early folliculogenesis (Dissen et al., 1995; Ojeda et al., 2000). All the discrepancies between our results in the human and those in rodents (Dissen et al., 1995; Ojeda et al., 2000; Spears et al., 2003; Paredes et al., 2004) can probably be explained by differences between species.
RT–PCR studies in human ovaries from normal fetuses aged 13–21 GWs revealed the expression of NT-4/5, BDNF and TrkB receptor (fl and T1 receptor forms) (Anderson et al., 2002). Immunohistochemical and immunoblotting studies confirmed the presence of the TrkB-fl receptor form protein and localized it to the cytoplasm of oogonia and oocytes of primordial follicles. Before follicular formation, NT-4/5 mRNA was predominantly located in stroma cells within clusters of oogonia; thereafter, NT-4/5 mRNA and protein were mostly expressed in the cytoplasm of GCs of primordial follicles. The mRNA levels of NT-4/5 and TrkB receptor increased during primordial follicular formation.
These results for the TrkB receptor correlate with our results, although we identified the TrkB receptor also in all the GCs. However, in contrast to Anderson et al. (2002), we identified NT-4/5 only in the minority of the GCs, probably due to the majority of our samples being derived from older subjects (fetuses and adults). Moreover, analysis of our 13 samples from fetuses aged 19–21 GWs yielded no connection between follicular assembly and NT-4/5/TrkB receptor expression, as detected in rats (Dissen et al., 1995) and humans (Anderson et al., 2002), possibly because we tested more ovaries at this stage or that our RT–PCR assay was unsuitable for localizing mRNA transcripts.
Mice carrying mutations for the TrkB receptor genes had grossly abnormal ovaries with greatly reduced numbers of oocytes and follicles (Spears et al., 2003), specifically secondary follicles (Dissen et al., 2002; Paredes et al., 2004). When ovaries from TrkB receptor-null mice were transplanted under the kidney capsule of recipient animals, follicular growth was impaired, and there was a widespread degradation of follicular organization and an increase in oocyte death (Paredes et al., 2004). Mice with mutations in both the NT-4/5 and the BDNF genes had follicular growth defects similar to those of the TrkB receptor-null mice (Ojeda et al., 2000).
Culture of fetal and neonatal mouse (Spears et al., 2003) and rat (Dissen et al., 1995) ovaries and human ovaries from fetuses aged 13–17 GWs (Spears et al., 2003) with an inhibitor of the Trk receptors resulted in a decrease in the survival of oocytes in newly formed murine primordial follicles, in the number of rat primordial follicles (Dissen et al., 1995) and in the survival of human oogonia (by 50%) (Spears et al., 2003). Similarly, anti-NT-4/5 and anti-BDNF antibodies significantly decreased the survival of murine oocytes and human oogonia.
Our detection of the TrkB receptor (specifically in the GCs) and its related NTs in human fetal and adult/adolescents ovaries suggests that these ligands may be involved in the activation of primordial follicles. However, in order to elucidate whether indeed NT-4/5 and BDNF are involved in growth initiation of human primordial follicles, they should be added to the culture medium. In addition, as the p75 receptor was identified previously (Abir et al., 2005) in human ovaries only during follicular assembly, it is possible that NTs participate in this process as well. Further studies are still needed to determine possible roles of other growth factors in early mammalian folliculogenesis.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The study was partially supported by a research grant from the Israel Cancer Association (R.A., B.F.). The authors are indebted to Dr G. Kessler-Icekson from Felsenstein Medical Research Center, Petach Tikva and Sackler Faculty of Medicine, Tel Aviv University for her assistance with the RNA extraction. We are also grateful to Ms G. Ganzach from the Editorial Board of Rabin Medical Center for the English editing, to the staff at the Gynaecology Ward for their help in locating suitable patients and to the Ultrasound Unit for identifying fetal gender.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Abir R, Fisch B, Raz Ah Nitke S and Ben Rafael Z (1998) Preservation of fertility in women undergoing chemotherapy—current approach and future prospects. J Assist Reprod Genet 15,469–478.[CrossRef][Web of Science][Medline]
Abir R, Orvieto R, Dicker D, Zukerman Z, Barnett M and Fisch B (2002) Preliminary studies on apoptosis in human fetal ovaries. Fertil Steril 78,259–264.[CrossRef][Web of Science][Medline]
Abir R, Fisch B, Jin S, Barnnet M, Freimann S, Van den Hurk R, Feldberg D, Nitke S, Krissi H and Ao A (2004) Immunocytochemical detection and reverse transcription polymerase chain reaction expression of leukemia inhibitory factor (LIF) and its receptor in human fetal and adult ovaries. Mol Hum Reprod 10,313–319.
Abir R, Fisch B, Jin S, Barnnet M, Ben-Haroush A, Felz C, Kessler-Icekson G, Feldberg D, Nitke S and Ao A (2005) Presence of NGF and its receptors in ovaries from human fetuses and adults. Mol Hum Reprod 11,229–236.
Anderson RA, Robinson LLL, Brooks J and Spears N (2002) Neurotropins and their receptors are expressed in the human fetal ovary. J Clin Endocrinol Metab 87,890–897.
Ao A, Erickson RP, Winston RML and Handyside AH (1994) Transcription of paternal Y-linked genes in the human zygote as early as the pronucleate stage. Zygote 2,281–287.[Medline]
Biron-Shental T, Fisch B, Van Den Hurk R, Felz C, Feldberg D and Abir R (2004) Survival of frozen-thawed human ovarian fetal follicles in long-term organ culture. Fertil Steril 81,716–719.[CrossRef][Web of Science][Medline]
Chao MV and Hempstead BL (1995) p 75 and Trk: a two receptor system. Trends Neurosci 18,321–326.[CrossRef][Web of Science][Medline]
Dissen GA, Hirshfield AN, Malamd S and Ojeda SR (1995) Expression of neurotropins and their receptors in the mammalian ovary is developmentally regulated: changes at the time of folliculogensis. Endocrinology 136,4681–4692.[Abstract]
Dissen GA, Romero C, Paredes A and Ojeda SR (2002) Neurotrophic control of ovarian development. Microsc Res Tech 59,509–515.[CrossRef][Web of Science][Medline]
Fridman WJ and Greene LA (1999) Neurotrophin signaling via Trks and p75. Exp Cell Res 253,131–142.[CrossRef][Web of Science][Medline]
Gosden RG (1995) Oocyte development throughout life. In Grudzinshes JG and Yovici JL (eds), Gametes—The Oocyte. Cambridge University Press, Cambridge, UK, pp. 119–149.
Gougeon A (1996) Regulations of ovarian follicular development in primates: facts and hypotheses. Endocrine Rev 17,121–154.
Haapasalo A, Sipola I, Larson K, Akerman KE, Stoilov P, Stamm S, Wong G and Castern E (2002) Regulation of TrkB surface expression by brain-derived neurotrophic factor and truncated TrkB isoforms. J Biol Chem 45,43160–43167.
Ibanez CF (1995) Neurotrophin factors: from structure–function studies to designing effective therapeutics. Trends Biotechnol 13,217–227.[CrossRef][Web of Science][Medline]
Ip NY, Ibanez CF, Nye SH, McClain J, Jones PF, Gies DR, Belluscio L, Le Beau MM, Epinosa R, Squinto SP et al. (1992) Mammalian neurotrophin-4: structure, chromosomal localization, tissue distribution, and receptor specificity. Neurobiology 89,3060–3064.
Newton H, Feisher J, Arnold JRP, Pegg DE, Faddy MJ and Gosden RG (1998) Permeation of human ovarian tissue with cryoprotective agents in preparation for cryopreservation. Hum Reprod 13,376–380.
Ojeda SR, Romero C, Tapia V and Dissen GA (2000) Neurotrophic and cell–cell dependent control of early follicular development. Mol Cell Endocrinol 163,67–71.[CrossRef][Web of Science][Medline]
Paredes A, Romero C, Dissen GA, DeChiara TM, Reichardt L, Cornea A, Ojeda SR and Xu B (2004) TrkB receptors are required for growth and oocyte survival in the mammalian ovary. Dev Biol 267,430–449.[CrossRef][Web of Science][Medline]
Ross AH (1991) Identification of tyrosine kinase Trk as a nerve growth factor receptor. Cell Regul 2,685–690.[Web of Science][Medline]
Spears N, Molinek MD, Robinson LLL, Fulton N, Cameron H, Shimoda K, Telfer EE, Anderson RA and Price DJ (2003) The role of neurotrophin receptors in female germ-cell survival in mouse and human. Development 130,5481–5491.
Stoilov P, Castern E and Stamm S (2002) Analysis of the human TrkB gene genomic organization reveals novel TrkB isoforms, unusual gene length, and splicing mechanism. Biochem Biophys Res Commun 290,1054–1065.[CrossRef][Web of Science][Medline]
Van den Hurk R, Abir R, Telfer EE and Bevers MM (2000) Preantral and antral follicles as possible source for fertilizable oocytes in human and bovine. Hum Reprod Update 6,457–474.
Submitted on January 11, 2006; accepted on February 28, 2006.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. Abir, B. Fisch, X.Y. Zhang, C. Felz, G. Kessler-Icekson, H. Krissi, S. Nitke, and A. Ao Keratinocyte growth factor and its receptor in human ovaries from fetuses, girls and women Mol. Hum. Reprod., February 1, 2009; 15(2): 69 - 75. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. A. Gook, D.H. Edgar, J. Borg, and M. Martic Detection of zona pellucida proteins during human folliculogenesis Hum. Reprod., February 1, 2008; 23(2): 394 - 402. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||