| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Testis |
Department of Human Genetics and Public Health,3 Graduate School of Proteomics, Faculty of Medicine, The University of Tokushima, Tokushima-City, 770-8503 Japan
Department of Urology,4 St. Marianna Medical University, School of Medicine, Kawasaki, 216-8511 Japan
Departments of Urology5
Anatomy and Cell Biology,6 School of Medicine, University of Tokushima, Tokushima-City, 770-8503 Japan
Core Research for Evolutional Science and Technology Corporation,7 Saitama, 332-0012 Japan
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A recent human genome project mapped more candidate genes for azoospermia in AZF regions [10]. Of the these genes, we focused on HSFY, which is mapped on AZFb [10] and is related to the heat shock protein transcriptional factor (HSF) family, because HSFs, such as HSF1 and HSF2, are shown to have important roles for spermatogenesis [13–16]. When we analyzed genomic DNA derived from men with azoospermia or oligospermia, we found that two males had deleted HSFY and RBMY, which is a candidate gene for azoospermia located on AZFb.
HSFY has a homologous gene, called LW-1, which is located on the X chromosome. Although LW-1 was originally cloned by Strausberg et al. in the project for sequencing 15 000 full-length human and mouse cDNA, it was not characterized [17].
HSFs are thought to regulate the expression of HSPs (heat shock proteins) through binding to the sequences located on the HSP genes, known as heat shock elements [18–22]. HSPs have been shown to have important roles in the repair of damaged proteins and survival of the cell by their chaperone function and have been suggested to be related to many pathological conditions, such as ischemic stress and inflammation [23, 24].
In testes, many different HSPs, such as HSP70-1, HSP70-2, HSFP70t, and HSP90, are expressed during spermatogenesis in a stage-specific manner and are postulated to have important roles in spermatogenesis [25–27]. Knock-out male mice without Hsp70-2 exhibited male-specific infertility by the dysfunction of synaptonemal complexes during spermatogenesis, suggesting the importance of HSPs for male reproduction [26, 27]. For humans, HSP70-2 is also suggested to have important roles in spermatogenesis [28]. Recently, transgenic mice for the active form of HSF1 and knockout mice for HSF2 were generated [13, 14, 29]. Both strains of mice exhibited anomalies in spermatogenesis in a stage-specific manner, indicating that HSFs have important roles in spermatogenesis, although a contradictory result has recently been reported for HSF2 [14, 29].
While preparing this manuscript, it was reported that HSFY has three alternative transcripts, although only alternative transcript 1 encodes a HSF-type DNA-binding domain [30]. However, it remains unclear whether those transcripts are actually translated and localized in spermatogenic cells [30]. In this study, we focus on HSFY that is derived from alternative transcription 1 because it has a HSF-type DNA-binding domain and seems to be important from the point of view of its structural similarity to HSFs. We show that HSFY located on AZFb and its homologous gene located on Xq28, LW-1, are predominantly expressed in testes and that they have novel characteristics compared with HSFs. Furthermore, we also discuss their biological functions and evolution.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Genomic DNA was extracted from peripheral white blood cells according to the standard method [31]. The infertile males analyzed in this study are described elsewhere [32]. In brief, 50 males with azoospermia and 13 with severe oligospermia (<5 x 106/ml) were diagnosed by standard seminal analysis and were shown to not have obstruction of the vas deferens. All karyotypes of the patients were cytogenetically normal. Among 50 azoospermic patients, 6 patients have been shown to have interstitial deletions on the Y chromosome (see Fig. 1b). Analysis of USP9Y and DBY was performed according to a previous report [33]. Genomic DNA was screened for deletions of HSFY by polymerase chain reaction (PCR) with the primers HSFYdel-F and HSFYdel-R (Table 1).
|
|
This study was approved by the Ethical Committee of The University of Tokushima. Every participant gave informed consent.
Mutation Analysis of HSFY and LW-1
For screening mutations of HSFY that correspond to alternative transcript 1 [30] and LW-1, DHPLC analysis was carried out according to a previous report [34]. In brief, genomic DNA derived from the azoospermic or oligospermic patients was amplified with the primers HSFYg1-F and HSFYg1-R or HSFYg2-F and HSFYg2R using a standard PCR technique. The PCR products were mixed with ones derived from fertile men and were used to generate heteroduplex PCR products. The heteroduplex PCR products were analyzed with WAVE system (Transgenomic, Inc. San Jose, CA) to detect mutations in the HSFY gene. For mutation analysis of LW-1, the primers, HSFXanal1F and HSFXanal1R or HSFXanal2F and HSFXanal2R were used to generate PCR products. The heteroduplex analysis for LW-1 was carried out the same as for HSFY. The primer sets for DHPC analysis of HSFY and LW-1 are listed in Table 1.
Construction of a Phylogenic Tree of HSFs
To generate a phylogenic tree of HSFs, the unweighted pair group method arithmetic average method was carried out by using Genetyx-SV/ R version 5.1.
RNA Extraction and Reverse Transcription-PCR
Total RNAs derived from specimens of testes and various tissue derived from 6-wk-old BDF1 mice were extracted using TRIzol (Invitrogen, NV Leek, The Netherlands) according to the manufacture's protocol. Each 2 µg total RNA treated with DNase I was subjected to synthesis of cDNA using a first strand cDNA synthesis kit (Invitrogen). This study conformed to the guideline of the care and the use of laboratory animals of the University of Tokushima.
To obtain cDNA containing entire coding regions of HSFY, LW-1, and mHSFYL, a PCR reaction was carried out in a total volume of 20µl with 2 µl cDNA synthesis mixture, 1 µl each primer tagged with Sal I site listed in Table 1, Tris-Cl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 0.001% gelatin, 200 mM each dNTP, and 0.5 units ExTaq (Takara Bio, Shiga, Japan). The condition for the PCR was as follows: initial denaturation at 94°C for 30 sec, annealing at 55°C for 30 sec, and extension at 72°C for 1 min. The final extension step was at 72°C for 10 min.
For detecting the expression of HSFY, LW-1, and mHSFYL in testes, the primer sets listed in Table 1 were used. For HSFY and LW-1, the primer sets encompassing the intron-exon boundary were used. For an inner control of reverse transcription-PCR (RT-PCR), beta actin cDNAs were amplified using the primer sets of ACT-1 and ACT-2 or M beta-actin F and M beta-actin R for human and mice, respectively. Each PCR product was dissolved using 2.5% agarose gel electrophoresis.
Plasmid Construction
To express the myc-tagged proteins, entire coding regions of the HSFY and LW-1 cDNA, which encode putative HSF-type DNA-binding proteins, were cloned in frame into a pCMV-Myc vector (Clontech, Palo Alto, CA).
Authenticity of each construct was confirmed using a DynamicET cycle sequencing kit (Amersham Biosciences, Piscataway, NJ).
Immunofluorescence Assay
NT2/D1 cell lines (ATCC, Manassas, VA) were cultured in glass-bottom dishes with a 35-mm diameter, using Dulbeco modified Eagle medium containing 4 g/L glucose and 10% fetal calf serum in 5% CO2 at 37°C. One microgram plasmid DNA was transfected into those cells using Fugene 6 (Roche, Mannheim, Germany), according to the manufacture's protocol. An immunofluorescence assay was carried out according to the previously described method of Toida et al. [35]. As the first antibody, a monoclonal antibody raised against myc-epitope (Clontech) was used at a 1:50 or 1:100 dilution. As the second antibody, anti-mouse IgG derived from goat (Sigma Aldrich Co., St. Louis, MO) was employed at a 1:200 dilution. To confirm the location of the nuclei, propidium iodide (Sigma) was used. The immunolabeled cells were mounted with Vectashield (Vector Laboratories, Burlingame, CA). The cells were analyzed with a fluorescence microscope (Olympus, Tokyo, Japan) and a confocal laser scanning microscope (Leica TCS-NT mounted on Leica light microscope DMRB; Leica AG, Wetzlar, Germany).
Northern Blot Hybridization
Multiple tissue Northern blots were purchased (Clontech) and were hybridized with HSF cDNA that corresponds to each entire coding region according to the manufacture's protocol. ExpressHyb was used as the hybridization solution. Finally, the blots were probed with a human beta-actin cDNA control probe (Clontech). The hybridized blots were exposed to imaging plates and scanned with MacBus1500 (Fujifilm, Tokyo, Japan).
Generation of an Antibody Against HSFY and Western Blot Analysis
A peptide specific for HSFY derived from alternative transcript 1, which consists of CSPLDKYHPNYN conjugated with KLH, was used for the immunization of rabbits. A polyclonal antibody raised against HSFY was purified with affinity chromatography. Fifty micrograms of proteins derived from testes (protein medley; Clontech), NT2/D1 cells, and lymphoblasts were separated using 10% SDS-polyacrylamide gel electrophoresis and were then transferred to a polyvinylidene fluoride membrane using a semidry method according to the manufacturer's protocol. For the immunodetection of HSFY, the first antibody against HSFY and the second antibody, horseradish peroxidase-linked anti-rabbit IgG, were diluted at 1: 2000 and 1:10 000, respectively. HSFY polypeptide was detected with ECL plus (Amersham Biosciences), which was used according to the manufacture's protocol. To confirm the specific band for HSFY, the first antibody, which was incubated with 10 times the amount of the peptide antigen of HSFY at 4°C overnight, was used for the immunodetection.
Immunohistochemistry
Nine samples showing spermatogenesis (mean Jonsen score 7.4–8.93) from nine patients, whose ages ranged between 35 and 50 yr old (six samples of nontumorous tissues orchiectomized from patients with testicular cancer, two biopsies of testicular tissues from patients with varicocele, and one biopsy of testicular tissue from one obstructive azoospermia patient), were obtained and used for immunohistochemistry. After 24-h fixation with Bouin solution or 10% formaldehyde at 4°C, the samples frozen in liquid nitrogen were dehydrated in ascending alcohols and embedded in paraffin. The 5- to 7-µm-thick sections were cut with a microtome and mounted on Matsunami adhesive silane (MAS)-coated slide glass. For frozen sections, after 24-h fixation with 4% paraformaldehyde in PBS, cryoprotection twice with 10% sucrose in PBS for 1 day, the sections were embedded with cryomold Tissue-Tek OCT compound (Sakura Co. Ltd., Tokyo, Japan) and frozen with dry ice acetone. The 3- to 5-µm-thick sections were cut with a cryostat (Leica, Bannockburn, IL), mounted on MAS-coated slide glass, dried for 2 h at room temperature, and kept at –80°C until use.
To retrieve the antigen, sections fixed with Bouin fluid were treated with 5% urea/distilled water in a microwave oven for 5 min [36] and washed in PBS according to the manufacturer's protocol. After blocking of endogenous peroxidase activity using 3% H2O2/MeOH, the slides were washed in PBS and then preincubated with 10% goat serum in PBS for 1 h to prevent the nonspecific binding of the first antibody at room temperature. Thereafter, anti-HSFY polyclonal antibody was applied as the first antibody at a 1:400 dilution in PBS/10% BSA at room temperature and incubated for 1 h. For the negative control of the first antibody, consecutive sections were incubated with anti-HSFY antibody preadsorbed with excess synthetic peptide HSFY at the same immunoglobulin concentration. After several washings in PBS, the sections were incubated with anti-rabbit IgG (Fab') labeled with amino acid polymer-peroxidase complex (Histofine simple stain MAX PO (R); Nichirei Co., Tokyo, Japan) for 1 h at room temperature. The sections were then washed and the peroxidase complex detected by incubation with aminoethyl-carbazol (AEC) (Histofine simple stain AEC solution; Nichirei Co.) following the manufacturer's instructions. Sections were counterstained with hematoxylin and mounted using aqueous permanent mounting solution (Nichirei Co.) with a cover glass.
This study was approved by the Ethical Committee/Institutional review board of St. Marianna University School Hospital and Oofuna-Chuo Hospital. To use samples for research, every participant gave informed consent.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
An analysis of the sequence database on the NCBI web site revealed that the HSFY gene is located on the human Y chromosome as multicopies. However, only HSFY located on the long arm of the Y chromosome has a significant long open-reading frame containing a presumed HSF-type DNA-binding domain that has 31% homology of amino acids compared with HSF1.
To know whether HSFY relates to azoospermia and oligospermia or not, we generated a PCR primer set that amplified only the HSFY genes located on the long arm of the Y chromosome (data not shown). We selected azoospermic or oligospermic males, who had been reported elsewhere [32], with and without deletions of AZFb or AZFc. For SMCY, RBMY, DYS7C, DAZ, and DYZ1 loci, using a PCR technique, we had already carried out deletion analysis in a previous study [32]. Using a primer set that detects only HSFY located on Yq, we screened genomic DNA derived from 57 males with azoospermia or with oligospermia for detecting deletions or mutations of HSFY. As a result, it was shown that two males with azoospermia had deleted HSFY, SMCY and RBMY on AZFb, while DAZ, which is suggested to be crucial for azoospermia on AZFc, was not deleted (Fig. 1, a and b). For these two males, we tested DBY and USP9Y loci on AZFa, which are suggested to be candidate genes for azoospermia [4, 5], and found that they were intact (data not shown). The results for deletion analysis of the Y chromosomes are summarized in Figure 1b. DHPLC analysis could not unveil any mutations of HSFY among the infertile males. For LW-1, we could not find any deletions or mutations of the coding region.
Relationship Between HSFY and LW-1
Presumed amino acid sequences of HSFY and its homologous gene on the X chromosome, LW-1, are shown in Figure 2. To elucidate the relationship between these two HSF-like factors on the human X and Y chromosomes, HSFY and LW-1, they were compared for their amino acid sequences (Fig. 3). As a whole, those two HSF-like factors exhibited 32% and 49% homology for amino acids and nucleotides of ORFs, respectively. Although HSFY and LW-1 show 53% homology of amino acids for their presumed DNA-binding domains, which are similar to HSF family members, the regions other than the DNA-binding domains show low similarity (Fig. 3a). A recent report on HSFY has shown that the HSFY gene has seven exons [30]. A database search revealed that the LW-1 gene is made up of three exons. Importantly, the introns are inserted into the similar positions of the presumed DNA-binding domains for HSFY and LW-1, suggesting that these two HSF-like factors are derived from a common ancestral gene (Fig. 3b).
|
|
Furthermore, we found an EST presumed to be a murine HSFY-like sequence (mHSFYL), 4933413G11Rik, on the NCBI web site, and compared its presumed amino acid sequence (see Fig. 2c) with those of HSFY and LW-1. The presumed DNA-binding domain of mHSFYL exhibited 70% homology compared with HSFY. As a result, it became clear that mHSFYL is more similar to HSFY than LW-1 as a whole (Fig. 3a). PCR analysis of mHSFYL, using genomic DNA derived from male and female mice, suggested that mHSFYL is located either on the X chromosome or an autosome (data not shown). Searching mouse genome database deposited at NCBI unveiled that mHSFYL is mapped on chromosome 1 and that it is an intronless gene.
Comparison of HSFY and LW-1 with HSFs for amino acid sequences confirmed that they are the most similar in their DNA-binding domains (Fig. 3c). All known HSFs in animals have two hydrophobic heptad repeat domains like leucine zippers, called HR-A/B and HR-C, except for HSF4, which lacks HR-C (Fig. 3c) [18–22]. However, although the three HSF-like factors analyzed in this study have some heptad repeats, they do not show apparent homology to HR-A/B and HR-C (see Figs. 2 and 3c). Moreover, those seem to lack typical leucine zipper motifs.
Although HSFY has a motif similar to the DNA-binding domain found in HSFs, gel shift assay with a classical HSE probe showed that HSFY could not bind to HSE (data not shown).
Generation of a Phylogenic Tree for HSFs
A phylogenic tree was constructed using the amino acid sequences of the DNA-binding domains of each HSF and presumed ones of the HSF-like factors (Fig. 4). As a result, the HSFs and the HSF-like factors were divided into two major clusters. One of the clusters included conventional HSFs such as HSF1, HSF2, and HSF4, while the other included HSFY, LW-1, and mHSFYL.
|
These data showed that HSFY, LW-1, and mHSFYL are similar to conventional HSFs, but belong to a class different from them.
RNA Blot Analysis of HSFY and LW-1 and Western Blot Analysis of HSFY
To assess the expression patterns of HSFY and LW-1 in various tissue, RNA blot analysis was carried out using the cDNA probes of HSFY and LW-1. As a result, it became clear that both HSFY and LW-1 genes are predominantly expressed in testes (Fig. 5a). To detect the expression of mHSFYL in various tissues of mice, RT-PCR analysis was performed. Like the results from the RNA blot analysis of HSFY and LW-1, it revealed that mHSFYL expression is also restricted to the testes (Fig. 5b).
|
To analyze the expression of the HSFY protein, a polyclonal antibody raised against HSFY was generated. The antibody was confirmed to recognize a GST-fused HSFY (data not shown). Using this antibody, we carried out Western blot analysis for testes, lymphoblasts, and NT2/D1 cell lines. As shown in Figure 5c, the antibody against HSFY could apparently detect a 45-kDa protein, the size of which is consistent with the predicted native one of HSFY derived from alternative transcript 1. Specificity of the antibody was confirmed using the antibody preadsorbed with a peptide of a HSFY antigen used for the immunization of rabbits.
RT-PCR Analysis for the Expression of HSFY and LW-1 in the Testes of an Azoospermic Patient
To know whether HSFY and LW-1 are expressed in a patient with azoospermia, the spermatogenic cells of which had been almost all deleted, we carried out RT-PCR analysis using total RNA derived from a testis specimen from an azoospermic patient who has no deletions for HSFY and LW-1 genes. RT-PCR analysis showed that HSFY and LW-1 were expressed in the azoospermic testes at very low levels, suggesting that HSFY and LW-1 are predominantly expressed in the spermatogenic lineage (Fig. 6).
|
Immunohistochemistry of HSFY in Human Testes
To elucidate the distribution of HSFY in testes, we carried out immunohistochemistry with an antibody raised against HSFY. The testis showing normal spermatogenesis revealed the expression of HSFY in germ cells and Sertoli cells (Fig. 7, A and B). The localization pattern of HSFY on germ cells and Sertoli cells was consistent in the testes showing normal spermatogenesis and the expression was observed in all seminiferous tubules.
|
In dark A spermatogonia, two localization patterns were observed. HSFY showed completely no expression over the whole of the cell (Fig. 7C-1a) or appeared at the nuclear region (Fig. 7C-1b). At the second step of type A spermatogonia, HSFY expression was observed at the perinuclear region (Fig. 7C-2b) or there were no signals over the whole region of the cell (Fig. 7C-2a) in pale A spermatogonia. The expression of HSFY on type A spermatogonia was not observed in all seminiferous tubules. During quiescence of HSFY expression, type B spermatogonia (Fig. 7C-3) divides to form preleptotene spermatocytes. At the beginning of the meiotic prophase, the preleptotene spermatocyte, and the leptotene spermatocyte phase, there were no localizations of HSFY in the spermatocytes (Fig. 7C-4). When the spermatocytes entered into the zygotene phase, however, strong expression of HSFY was observed around the chromatin cord in the nucleus (Fig. 7C-5). During the pachytene spermatocyte phase, transient expression of HSFY was maintained in the nucleus at first (Fig. 7C-6), then it was moved to the perinuclear region (Fig. 7C-7), and finally disappeared entirely from the spermatocytes at the end of the pachytene phase (Fig. 7C-8). After the second quiescence of HSFY expression, late pachytene spermatocytes completed the long prophase, then entered into the first division. After the division of first spermatocytes into secondary spermatocytes, the HSFY expression was observed transiently in the nucleus and in the cytoplasm around the nucleus (Fig. 7C-9), then secondary spermatocytes entered the second meiosis. After the end of meiosis, the remaining HSFY expressions in the nucleus and the cytoplasm (Fig. 7C-10) were translocated into the perinuclear region of the round spermatid (Fig. 7C-11). During the spermiogenic phase, HSFY was reexpressed in the nucleus of the younger elongated spermatid (Fig. 7C-12), and then it was translocated from the nucleus to the cytoplasm of the elongated spermatid (Fig. 7C-13), and finally disappeared and a complete spermatozoon was generated (Fig. 7C-14).
In contrast with the germ cells, Leydig cells (Fig. 7, A and E), myoid cells (Fig. 7, A and F), and lamina propria (Fig. 7A) did not exhibit expression of HSFY. Sertoli cells showed the HSFY localization only in the cytoplasm (Fig. 7, A and D) and did not show a change in the expression patterns in each seminiferous tubule (Fig. 7A).
The immunolocalization of the HSFY antibody gave good reproducibility independent of the fixative methods for the testicular samples used in this study.
Intracellular Localization of HSFY and LW-1 in the NT2/ D1 Cells
To know the intracellular localization of HSFY and LW-1 in the cell line, we carried out an immunofluorescent assay using myc-tagged HSFs in the NT2/D1 cell line, which was originally established from human testicular embryonal cell carcinoma. As a result, we observed that the myc-tagged HSFY and LW-1 were localized in the cytoplasm of NT2/D1 cells (Fig. 8). In our experiments, we could not find the tagged proteins located in the nucleus. These results are similar to conventional HSFs, except for HSF4, which is constitutively localized in the nucleus under no-stress conditions [20, 22].
|
It was observed that the HSFY fused with the GAL4-DNA-binding domain, which has a nuclear localization signal (NLS) that is retained in the cytoplasm of almost all NT2/D1 cells transfected with the fusion protein (data not shown). This indicates that HSFY has a crucial region for blocking its own translocation from the cytoplasm to the nucleus.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
HSFY is similar to the HSF family, which is crucial for the expression of HSPs. Because different HSPs are expressed in germ cell lineages during spermatogenesis [25], we were interested in the relationship between male infertility and deletions or mutations of HSFY. When we tested whether HSFY was deleted in azoospermic or oligospermic patients, we found that two azoospermic males lacked HSFY, although we could not find any point mutations for the other males. Ferlin et al. have also reported azoospermic males without HSFY in AZFb [37]. Because the males without HSFY in our study also lacked RBMY and maybe some other genes located on AZFb, we could not conclude that the deletion of HSFY lead to azoospermia. However, it is likely that HSFY has an important role in spermatogenesis. First, HSFY is expressed in spermatogenic cells and Sertoli cells in testes. Second, HSF members, HSF1 and HSF2, which are similar to the HSF-like factors analyzed in this study, are suggested to have important roles in spermatogenesis [13–16]. Third, as shown in this study, there are some patients with azoospermia who lack HSFY with other genes on AZFb. Recently, it was shown that the most frequent class of deletions on AZFb include the HSFY gene [38]. The mechanism that causes deletions of candidate genes for azoospermia on the Y chromosome seems to be related not to those genes themselves but to the repetitive sequences around them [10, 11].
The Structure of HSFY and LW-1 Are Considerably Different from Those of Conventional HSFs, HSF1, HSF2, and HSF4
Conventional HSFs have conserved domains that involve DNA-binding domains, HR-A/B, HR-C, and two presumed NLS, which are located on both sides of HR-A/B (Fig. 3c) [20, 22]. HR-A/B has an important role in the homotrimerization of HSFs and their retention in the cytoplasm [39, 40]. HR-C is also important for retention of HSFs in cytoplasm [20]. However, HSFY, LW-1, and the HSFY-related murine homolog, mHSFYL, seem to lack these conserved HR domains, and instead, they seem to have some heptad repeats without a leucine zipper-like motif (see Figs. 2 and 3c). HSF1 and HSF2 are located in the cytoplasm under no stimuli and are translocated into nucleus after stimulus [18, 19, 22]. HSF4 is constitutively localized in the nucleus because of its deletion of HR-C (Fig. 3c) [20].
In this study, we could not find any evidence that HSFY binds to HSE (data are not shown). For DNA-binding of HSFs, it is known that the helix-turn-helix motif in the DNA-binding domain is important [41]. Ahn et al. have already reported that the central helix-turn-helix motif in HSFs consists of 
-helix 2 and 3, where the latter is the DNA recognition helix [41]. However, although the three HSF-like factors analyzed in this study have presumed DNA-binding domains similar to ones of HSFs, -helix 2 and 3 conserved in HSFs are not conserved. Moreover, HSFY lacks a leucine zipper motif, which is important for the homotrimer formation of HSFs for binding to HSE. Therefore, it seems that HSFY does not bind to HSE from the point of view of its structure. So far, whether HSFY can work as a transcriptional regulator remains unclear.
The immunofluorecence assay with myc-tagged HSFY and LW-1 showed that they were localized in the cytoplasm of the NT2/D1 cell line (Fig. 8). So far, we do not know whether the heptad repeats observed in HSFY, LW-1, and mHSFYL contribute to homo or hetero oligomerization and retention of these HSF-like factors in the cytoplasm. For HSF1 and HSF2, some models for retention in the cytoplasm have already been provided [39, 40]. For the three HSF-like factors analyzed in this study, hydrophobic repeats may have important roles for their localization in the cytoplasm.
HSF1, HSF2, and HSF4 have two presumed NLSs, which are located on both sides of HR-A/B [22]. At least for HSF2, it has been reported that both NLSs are important for the translocation of HSF2 from the cytoplasm to the nucleus [40]. From the point of view of the similarity of their sequences, HSFY, LW-1, and mHSFYL seem to have only one NLS, which corresponds to the one located in the amino terminal of HR-A/B found in other HSFs (Fig. 3c). This may be important for the intracellular localization of HSFY, LW-1, and mHSFYL. Alternatively, the HSF-like factors in this study may require cargo proteins or posttranslational modifications to enter into the nucleus.
Function of HSFs in Testes
In the present study, we showed that HSFY was expressed in spermatogenic cells and Sertoli cells in testes. We also showed that HSFY was translocated from the cytoplasm to the nucleus in spermatogenic cells in a stage-dependent manner during spermatogenesis (Fig. 7). However, so far, biological meanings for this translocation of HSFY remain unclear. Moreover, in Sertoli cells, HSFY was located in the cytoplasm. This may suggest that HSFY has different roles for spermatogenic cells and Sertoli cells.
Recently, some studies for the function of HSFs in testes using transgenic or knockout mice have been reported [13, 14]. Nakai et al. revealed that the active form of HSF1 caused a blockade of spermatogenesis at the pachytene stage and increased the number of apoptotic spermatocytes, leading to male infertility [13]. It has been shown that HSF2 is localized in the nuclei of early pachytene spermatocytes and round postmeitotic spermatids [15]. Kallio et al. reported that many developing spermatocytes are eliminated via apoptosis in a stage-specific manner and that pachytene spermatocytes exhibited a structural defect in the synaptonemal complexes between homologous chromosomes in male mice without HSF2 [14]. The female mice without HSF2 exhibited multiple fertility defects [14]. However, recently, McMillan et al. showed that mice without HSF2 exhibited no change in phenotype [29]. Although the reason for the discrepancy between the two mice without HSF2 remains unclear, it may depend on the technical methods for the generation of the knockout mice or the genetic background of the mice [14, 29].
Interestingly, the expression level of Hsf2 is not correlated with the expression level of Hsp, suggesting that HSF2 has functions other than the transcription of Hsp [42, 43]. There are several reports that suggest the functions of HSF2 other than the regulation of HSPs. For example, it has been reported that HSF2 can interact with the subunit of protein phosphatase 2A, PR65, and modulate the activity of protein phosphatase 2 [44, 45]. Alastalo et al. showed, using an electron micrograph, that HSF2 could be involved in the mRNA of protein transport, as certain phases of gametogenesis are known to be dependent on stored mRNA species and probably also on presynthesized stored proteins [16].
Although HSFY and LW-1 are similar in presumed DNA-binding domains, other regions of the two HSF-like factors are considerably different. It is interesting to investigate the expression profiles of the two HSF-like factors in testes.
HSF family members have isoforms that are generated with alternative splicing and that have reverse functions for the transcription of HSPs [46, 47]. A recent study has shown that HSFY has three alternative transcripts, although it remains unclear whether HSFY transcript 2 and 3, which are not analyzed in this study, are translated and functional [30]. For LW-1 and mHSFYL, so far, it is not known whether alternative transcripts exist.
Evolution of HSFY
Most recently, Tessari et al. reported that HSFY genomic blocks on the Y chromosome originate from chromosome 22, where four of seven exons of HSFY exist [30]. This information seems to be important in understanding the origin of HSFY, although it remains unclear whether the HSFY-like sequence on chromosome 22 is transcribed and translated.
In conclusion, we show some characteristics of a novel HSF-like factor, HSFY, which is located on the human Y chromosome and is deleted in some azoospermic males. This HSF-like factor is considerably different from conventional HSFs, HSF1, HSF2, and HSF4, in its structure and expression. HSFY is presumed to have important roles in spermatogenesis and may be related to human spermatogenic failures, such as azoospermia and oligospermia.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Correspondence: Yutaka Nakahori, Department of Human Genetics and Public Health, School of Medicine, The University of Tokushima, 3-18-15 Kuramoto-cho, Tokushima-City, 770-8503 Japan. FAX: 81 88 633 7453; nakahori{at}basic.med.tokushima-u.ac.jp
Received: 29 September 2003.
First decision: 29 October 2003.
Accepted: 26 February 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Hu and N. F. Mivechi Association and Regulation of Heat Shock Transcription Factor 4b with both Extracellular Signal-Regulated Kinase Mitogen-Activated Protein Kinase and Dual-Specificity Tyrosine Phosphatase DUSP26 Mol. Cell. Biol., April 15, 2006; 26(8): 3282 - 3294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Vinci, F. Raicu, L. Popa, O. Popa, R. Cocos, and K. McElreavey A deletion of a novel heat shock gene on the Y chromosome associated with azoospermia Mol. Hum. Reprod., April 1, 2005; 11(4): 295 - 298. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
