| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
research-article |
Departments of Biochemistry3 and Structural Biology,4 National Institute for Research in Reproductive Health (NIRRH), Indian Council of Medical Research, Parel 400 012, Mumbai, India
ABSTRACT
The sperm from the testis acquires complete fertilizing ability and forward progressive motility following its transit through the epididymis. Acquisition of these characteristics results from the modification of the sperm proteome following interactions with epididymal secretions. In our attempts to identify epididymis-specific sperm plasma membrane proteins, a partial 2.83-kb clone was identified by immunoscreening a monkey epididymal cDNA library with an agglutinating monoclonal antibody raised against washed human spermatozoa. The sequence of the 2.83-kb clone exhibited homology to the region between 1 and 1097 bp of the homeobox gene, Hoxb2. This sequence was found to be species conserved, as revealed by RT-PCR analysis. To obtain a full-length clone of the sequence, 5' RACE-PCR (rapid amplification of cDNA ends PCR) was carried out using rat epididymal RNA as the template. It resulted in a full-length 1.657-kb cDNA encoding a 32.9-kDa putative protein. The protein designated HOXBES2 exhibited homology to the conserved 61-amino acid homeodomain region of the HOXB2 homeoprotein. However, characteristic differences were noted in its amino and carboxyl termini compared with HOXB2. A putative 30-kDa protein was detected in the tissue extracts from adult rat epididymis and caudal spermatozoa, and a 37-kDa protein was detected in the rat embryo when probed with a polyclonal antibody against HOXB2 protein. Multiple tissue Western blot and immunohistochemical analysis further indicated its expression in the cytoplasm of the principal and basal epithelial cells, with maximal expression in the distal epididymal segments. Northern blot analysis detected a single ~2.5-kb transcript from the adult epididymis. Indirect immunofluorescence localized the protein to the acrosome, midpiece, and equatorial segments of rat caudal and ejaculated human and monkey spermatozoa, respectively. In conclusion, we have identified and characterized a novel epididymal homeoprotein different from HOXB2 protein and hereafter referred to as HOXBES2, (HOXB2 homeodomain containing epididymis-specific sperm protein) with a probable role in fertilization.
epididymis, gene regulation, sperm, sperm maturation, testosterone
Homeoproteins are DNA binding transcription factors encoded by the homeobox-containing genes that are known to regulate embryonic development [1]. The homeobox originally identified in Drosophila is a 183-bp DNA sequence that encodes a 61-amino acid conserved homeodomain (HD) [2]. The HD located either at the terminal or subterminal position of the corresponding homeoprotein is responsible for recognizing and binding sequence-specific DNA motifs. The specific binding allows homeoproteins to activate or repress the expression of a battery of downstream effector target genes [3]. Various groups of homeobox gene families are classified through sequence similarities within the HDs and flanking regions. They include the Drosophila Antennapedia (Antp) HD defined class I HD (Hox genes) [4], bicoid (bcd), caudal (cad), engrailed (en), even-skipped (eve), muscle segment homeobox genes (MSX), paired (PAX), Pit-Oct-Unc (POU), empty spiracles (EMX), and orthodenticle (OTX) [5]. The role of homeobox-containing genes in embryo development has been investigated extensively since their discovery. These genes in the homeobox cluster are expressed in a colinear fashion from 5' to the 3' direction, particularly in the nervous system during the establishment of the rostra-caudal axis of the embryo [6] and the formation of eye [7]. Neural expression of homeoproteins is strong enough even to be prolonged throughout adulthood [8]. Furthermore, the expression of HD transcription factors in adult tissues, including the liver, kidney, and intestine, signifies the involvement of these gene families in regulating crucial biological processes of adult eukaryotic cells, such as cellular morphogenesis, cell growth and differentiation, and cell-cell and cell-extracellular matrix interactions [9].
In addition to their expression in the embryonic and adult somatic tissues, several homeobox genes have been extensively characterized as downstream targets of male and female sex hormones during prenatal and postnatal life. The increased expression of 5' hox genes Hoxa10 and Hoxa11 in response to circulating estrogen and progesterone levels indicates their roles in endometrial maturation, implantation, and maintenance of pregnancy [10]. The target disruption of Hoxa10 gene in mice generates uterine factor infertility [11]. The expression of Hoxd10 and Hoxd13 genes has been reported in the male reproductive tract of mammalian embryos [12]. Homeotic genes expressed in the testis regulate events related to development during spermatogenesis [13–15]. Recently, MacLean and coworkers [16] have reported a cluster of 12 homeobox genes expressed both in male and female mouse reproductive tissues and named them as Rhox genes. However, the expression and function of individual Rhox genes, including Rhox5, Rhox6, Rhox9, and Rhox8 (previously known as Pem, Psx1, Psx2, and Tox, respectively), had been described already prior to their identification as members of the Rhox cluster [17–20]. Among the Rhox genes identified so far, Rhox2, Rhox5, and Rhox9 were found to be expressed at higher levels in the mouse epididymis than in the testis and placenta [16]. More recently, 20 additional Rhox genes have been identified as paralogs to the 
-subcluster in the mouse X chromosome [21]. Wang and Zhang [22] have reported an additional 18 Rhox genes and 3 pseudogenes in mice, 16 Rhox-related genes from rat, and 1 new Rhox gene (PEPP3) in the human. Nevertheless, until recently less was known about the expression pattern of homeobox genes discovered in the adult epididymis. Hoxa10 and Hoxa11 knockout male mice were found to be sterile, and their epididymis showed a homeotic transformation [23, 24]. Pax2, Rhox5, Etv4, and Hoxc8 [13, 17, 25, 26] are the other homeoproteins expressed in the epididymis. The DNA-binding Meis 1 protein has been identified as a cofactor of Hox11A protein in the adult mouse epididymis [27].
In the present study, we report on the identification and characterization of a novel homeoprotein from epididymis, until now designated HOXBES2 (HOXB2 homeodomain containing epididymis-specific sperm protein). The homeobox gene Hoxb2 belongs to the Proboscipedia subfamily of the Antennapedia homeobox gene family called class I homeobox genes, or Hox genes [6]. It is located on the human chromosome 17, and its orthologs in the mouse and rat genome are localized to chromosomes 11 and 10, respectively. Hoxb2 gene is maximally expressed in the rhombomere 4 of the developing embryo and functions together with Hoxb1 in the specification of the motor component of the VII nerve, and with Hoxb4 in the closure of the ventral thoracic body wall. Nearly 75% of the Hoxb2 homozygous mutant mice died within 24 h of birth, owing to split sternum [28]. Hoxb2 gene is also expressed in the VII/VIII ganglia of hindbrain and hind limb of the developing embryo and the k562 nuclear extracts [29], and its expression at transcriptional and translational levels has been reported in the human hematopoietic and pulmonary systems [30, 31].
The present study also attempts to investigate the relevance of HOXBES2 expression in the epididymis. The highly specialized epididymis is a multifunctional, androgen-regulated male accessory organ [32]. The epididymis provides a unique and conducive microenvironment for the transport, maturation, and, finally, the storage of mature spermatozoa [33, 34]. Its segmental nature together with its region-specific gene expression enacts various morphological, biochemical, and physiological modifications on the luminal spermatozoa by synthesizing and secreting a constellation of epididymal-specific proteins. Our focus for the last few years has been to identify, characterize, and understand the functional role of novel sperm membrane proteins acquired specifically from the adult epididymis. In addition to the above, the current paper describes our endeavor to identify and characterize a novel homeoprotein, HOXBES2, from adult epididymis, and it proposes its probable functional role in fertilization.
For the studies described, 180-day-old adult male rats of the Holtzmann strain weighing ~400 g were used as the animal model. The rats were housed under conditions of 12L:12D with access to rat chow and water ad libitum. The monkey spermatozoa were collected by electro-ejaculation from proven fertile bonnet monkeys (Macaca radiata) bred in the animal colony of our institute (National Institute for Research in Reproductive Health [ICMR], Mumbai, India). All experimental procedures were conducted in accordance with the guidelines of the NIRRH's Animal Ethics Committee for the use and care of animals for biomedical research. The human spermatozoa used in our study were collected by our lab from fertile, healthy human volunteers. Informed consent was obtained from the patients, and the investigations were approved by the ICMR's Human Research Committee.
Experimental rats were anesthetized with ether, and the somatic (brain, heart, liver, lung, kidney, skeletal muscle, large and small intestine, and spleen) and the reproductive (testis, prostate, seminal vesicle, vas deferens, and epididymis) tissues were dissected out. The epididymis was divided into initial segment, proximal and distal caput, proximal and distal corpus, and proximal and distal cauda [35], was processed under sterile conditions for RNA extraction, and was used for RT-PCR, 5'-RACE (rapid amplification of cDNA ends), and Northern blot analysis.
Complementary DNA Library Screening
Using a monoclonal antibody raised against washed human spermatozoa, a monkey epididymal 
ZAP cDNA library was screened to identify clones encoding sperm membrane proteins [36, 37]. A positive clone identified with an insert size of ~2.8 kb was PCR amplified using T3 and T7 vector-specific primers (Bangalore Genei, Bangalore, India) and was sequenced commercially at Bangalore Genei. The sequence showed homology to the homeobox gene Hoxb2 and revealed the presence of a poly-A tail at the 3' end and the absence of an initiation site at the 5' end, suggesting that it was a partial clone.
Rat was selected as the animal model to characterize the Hoxbes2 gene. The upstream sequence was extended by 5' RACE-PCR using the SMART RACE cDNA amplification kit (Clontech). Five micrograms of total RNA from rat epididymis was reverse transcribed using a set of gene-specific primers (5'-AGATAACCGAGTGCCCAA-3') and a universal primer mix (UPM). The resultant RACE-PCR product was diluted 50-fold and subjected to nested PCR using Hoxbes2-specific primers (forward primer: 5'-GTGTTCACAGGAACCAA-3'; reverse primer: 5'-TCGTATTATAAAGAACA-3') and the advantage GC-Polymerase PCR kit (Clontech). RACE-PCR was carried out with an initial denaturation at 94°C for 3 min, followed by 35 cycles, each of denaturation at 94°C for 15 sec, annealing and extension at 56.2°C for 3 min, and finally an extension at 68°C for 10 min. The PCR products then were purified on a 1.2% agarose gel, subcloned into a TA cloning vector (Invitrogen), and sequenced at the ICMR's central sequencing facility using the Applied Biosystems DNA sequencer version 3.7.
The cDNA and amino acid sequences were subjected to a homology search using the standard nucleotide-nucleotide BLAST (blastn) and protein-protein BLAST (blastp) tools (http://www.ncbi.nlm.nih.gov/BLAST). These sequences were subjected to multiple sequence alignment (ClustalW: EMBL, EBI) with human (NM_002145 and NP_002136), rat (XM_220894 and XP_220894), and mouse (NM_134032 and NP_598793) Hoxb2 (HOXB2 for human) mRNA and protein sequences. The full-length cDNA and its putative protein were analyzed in silico using various computational bioinformatics tools, such as the ORF finder (http://www.ncbi.nlm.nih.gov/ORF), Phylogram, PROSITE, PFAM, HMM, and NCBI Conserved Domain database (CDD; http://www.expasy.ch), SWISS PROT, SignalP (http://www.cbs.dt.dk), SubLoc V 1.0, and iPSORT, DAS (http://www.sbs.su.sc) and HMMTOP 1.1 (http://www.enzim.hu, http://www.ncbi.nlm.nih.gov/GenomeBLAST), and Lasergene, DNA Star package (DNA Star Inc.).
RNA Isolation and Northern Blot Analysis
Total RNA was extracted from rat tissues using the single-step acid guanidium thiocyanate-phenol-chloroform extraction method described by Chomozynski and Sacchi [38], quantitated, and purity determined by absorbance at 260 and 280 nm (Smartspec; BioRad). Northern blot analysis was carried out according to the modified protocols of Chirgwin [39] and Sambrook et al. [37]. The total RNA (20 µg) separated on 1% agarose formaldehyde gel was denatured and transferred onto nylon membrane (Amersham Pharmacia Biotech, Piscataway, NJ), then subjected to Northern blot analysis together with a commercially available rat multiple tissue Northern blot (Clontech). The blots were hybridized with a digoxigenin-labeled, 581-bp cDNA fragment (bp 561-1134, including the conserved homeodomain region specific to Hoxb2; 20 ng/ml) overnight at 55°C, and hybrids were detected using the chemiluminescent substrate CSPD (Roche, Germany). The blot was stripped and reprobed with ß-actin, the housekeeping gene to confirm equal loading of the template RNA.
RT-PCR with Rat, Monkey, and Human Epididymal RNA
Five micrograms each of total RNA from rat, monkey, and human epididymis was reverse transcribed using BD PowerScript Reverse Transcriptase (BD Biosciences, Clontech, CA) and oligo-dT primer. Hoxbes2 was amplified using gene-specific primers: forward primer 5'-GTGTTCACAGGAACCAA-3' (50–66 bp) and reverse primer 5'-TCGTATTATAAAGAACA-3' (1638–1622 bp). The primers were designed so as to not share any identity with Hoxb2 gene and were Hoxbes2 specific. The PCR conditions used were as follows: initial denaturation at 94°C for 3 min, followed by 35 cycles of denaturation at 94°C for 30 sec, annealing and extension at 56.2°C for 3 min, and a final extension at 68°C for 10 min. Amplification of a 580-bp ß-actin housekeeping gene served as the internal control.
Protein Extraction from Rat Epididymis, Caudal Sperm, and Embryo
The rat epididymal segments and somatic tissues were dissected, minced, and incubated in 10 ml of 0.1 M PBS (pH 7.4) containing 0.1 M phenylmethylsulfonyl fluoride (PMSF) for 30 min at room temperature. The supernatant containing sperm was aspirated and centrifuged at 700 x g for 10 min, and the sperm pellet was washed thrice in 0.1 M PBS, resuspended in minimum volume of PBS containing 0.1M PMSF and 1% SDS, and incubated at 4°C overnight. The sperm suspension then was sonicated on ice for 5 min and centrifuged at 7000 x g for 10 min at 4°C. The clear supernatant was mixed with 50 µl protease inhibitor cocktail (Sigma-Aldrich, St. Louis, MO) and incubated at 37°C for 1 h, followed by extraction of tissue proteins as described by Hu et al. [40]. Protein extraction also was carried out from the hindbrain and hind limb regions of the 7.5-days postcoitus rat embryo, in which the HOXB2 transcription factor is known to be expressed during embryo development [41, 42], and the protein concentration of the extracts was estimated by the method of Lowry et al. [43].
To identify the cognate protein expression in the epididymis, an aliquot of 40 µg protein per lane from each tissue extract was analyzed on 10% SDS-PAGE [44], followed by Western blot analysis [45]. To identify the immunoreactive protein, the blot was incubated with an affinity-purified, monospecific goat polyclonal primary antibody (HOXB2 [P-20]; catalog no. SC-17165; Santa Cruz-Biotechnology) raised by epitope mapping within an internal region of HOXB2 protein of human origin. This was followed by 1:1000 diluted horseradish peroxidase-conjugated rabbit anti-goat secondary antibody. The peroxidase activity was detected using enhanced chemiluminescence (ECL; Amersham Life Science Inc.). The blot was stripped and reprobed with a ß-tubulin monoclonal antibody as an internal control to confirm equal loading of protein samples in the gel.
Paraffin sections (5-µm thick) of Bouins fixed and processed epididymal tissue were immunostained with a goat polyclonal HOXB2 antibody diluted 1:25 in blocking buffer for 1 h at room temperature. Controls were incubated with nonimmune goat serum at a dilution similar to that used for the primary antibody. All other protocols were followed according to the manufacturer's instructions (Santa Cruz Biotechnology). Finally, the sections were dehydrated through an alcohol gradation (10%–100%), cleaned in xylene, mounted in DPX Permount (Qualigens), and observed under the microscope (Leitz). Immunoreactivity to the HOXB2 antibody was analyzed using the Biovis Image Plus software (REV 2.0 Expert Vision Lab) based on the intensity of color reaction observed in the epididymal tissue sections. The intensity of the color reaction was expressed as the ratio between either the integrated optical densities (IOD) and/or percentage of area (% area) stained and the total area analyzed. The mean ratios of immunoreactivity in various segments versus the total epididymis were depicted graphically with IOD on the y-axis, and epididymal segments were on the x-axis.
Air-dried smears of approximately 1.5 x 104 rat caudal spermatozoa and swim up samples from proven fertile monkey and human ejaculates were fixed with 4% paraformaldehyde for 5 min at 4°C, washed with 0.1 M PBS six times for 10 min each, and blocked in 20% normal rabbit serum (blocking serum) containing 5% BSA in PBS for 2 h in a humidified chamber. The slides then were incubated overnight with primary HOXB2 antibody/nonimmune goat serum diluted 1:10 in blocking serum-PBS at 4°C. Following overnight incubation, the slides were washed six times for 10 min each with 0.1 M PBS, followed by incubation for 2 h with secondary antibody fluorescein isothiocyanate (FITC)-labeled rabbit anti-goat (Bangalore Genei) diluted 1:100 in PBS containing 0.5% BSA and then washed six times for 10 min each with PBS. Finally, the slides were mounted using paraphenylene diamine and observed under a fluorescent microscope (Leitz).
Identification and Characterization of the Positive cDNA Clone
A monkey epididymal ZAP cDNA library (titer 2 x 107 pfu) was screened using a monoclonal antibody raised against washed human spermatozoa [36] to identify clones encoding sperm membrane proteins. The immunoscreening resulted in the identification of a 2.83-kb positive cDNA clone, which on sequence analysis revealed significant homology to the transcription factor encoding homeobox gene, Hoxb2. However, the presence of a poly-A tail in the 3' region and absence of an initiation site at the 5' end indicated the truncated nature of the clone. The upstream region of the sequence was extended by 5' RACE-PCR using rat epididymal RNA as the template, and a 1.657-kb full-length cDNA obtained (Fig. 1A). The full-length clone was found to encode a putative protein of 305 amino acids named HOXBES2. The nucleotide sequence of Hoxbes2 was deposited to the NCBI GenBank and ascribed with the accession number DQ399532, and it also was submitted to the rat genome database (Hoxbes2/RGD_ID: 1309853). This full-length Hoxbes2 cDNA with an extension of 560 bp in its 5' region showed 100% identity to the previously identified partial 2.83-kb monkey cDNA clone (NCBI GenBank accession number AF 255949). Hoxbes2 cDNA consists of an open reading frame (ORF) of 918 bp, with an initiation codon (ATG) at positions 218–220 bp and a stop codon (TAG) at positions 1133–1135. The ORF is flanked by a 217-bp 5' untranslated region and a 521-bp 3' untranslated region, a putative polyadenylation signal 5'-AAATAAA-3' at positions 1615 to 1621 that ends with a poly-A tail (Fig. 1B). The criteria determining the full-length cDNA were: a Kozak sequence at the site of the initiation Met [46]; presence of a signal peptide cleavage site between amino acids 47 and 48; and the presence of a stop codon in the cDNA sequence immediately 5' to the predicted initiator methionine.
|
In Silico Analysis of Hoxbes2 cDNA (DQ399532) and its Putative Protein (ABD73307)
The homology search against the NCBI database revealed that the full-length of Hoxbes2 cDNA shares maximum similarity between bp 590 and 1635 with human (97% identity), rat (94% identity), and mouse (91% identity) Hoxb2 (HOXB2 for human) cDNA, with significant differences in the 5' (bp 1–568) and the 3' (bp 1636–1657) regions. Similarly, the putative translated protein (ABD73307) of 305 amino acids exhibited maximum similarity with human (89% identity) followed by rat (69% identity) and mouse (68% identity) HOXB2 proteins. The protein identity is 100% in the conserved homeodomain region (amino acids 125–185), whereas it differs considerably in its amino (amino acids 1–150) and carboxyl termini (amino acids 272–305). HOXBES2 protein has a calculated molecular mass of 32.9 kDa and an isoelectric point (pI) of 10.1. Analysis using NetOGly 3.1 and Netphos 2.0 tools unraveled a potential glycosyl-phosphatidyl inositol (GPI) modification site at position 287 serine, and as many as 16 phosphorylation sites. PROSITE and the conserved domain database (CDD) search indicated the presence of a 61-amino acid conserved motif of HOXB2 homeodomain ranging between amino acids 125 and 185 (Fig. 2). Multiple sequence alignment and phylogenetic comparison analysis of HOXBES2 HD reiterated the above observation that it displays the highest sequence identity and closer evolutionary proximity (100%) to human, rat, and mouse HOXB2 homeodomain (Fig. 3, A and B). SOSUI signal indicated its soluble nature, and the DAS and HMMTOP 1.1 predicted two short (16–19, 77–79) helices and a 20-amino acid-long (2–21) transmembrane helix in the HOXBES2 protein. SubLoc V 1.0 and iPSORT predicted the existence of a nuclear localization signal (NLS; 175–185; N'-QNRRMKHKRQ-C') and a 30-amino acid-long (amino acids 1–30) mitochondrial targeting peptide, respectively. Lasergene program (DNA Star package; DNA Star Inc.) predicted a 20-amino acid (171–190; N'-QNRRMKHKRQTQHREPPDGE-C') highly antigenic and most hydrophilic region.
|
|
Chromosomal Localization of Hoxbes2 Gene
The gene cognate to the Hoxbes2 cDNA was mapped to the chromosomes 17, 10, and 11 of the human, rat, and mouse genome, respectively, by the NCBI Genome BLAST (http://www.ncbi.nlm.nih.gov/GenomeBLAST) search. Genome search further revealed five blast hits (Hoxb1, Hoxb2, Hoxb3, and Hoxb4 genes and myeloid ecotropic viral integration site 1 protein) for Hoxbes2 in the mouse chromosome 11 and one each in rat and human chromosomes 10 and 17, respectively, where Hoxb2 gene was identified originally [6]. Pairwise alignment and CDD search indicated that Hoxbes2 exhibits 55%, 68%, 45%, and 51% identity to Hoxb1, Hoxb2, Hoxb3, and Hoxb4 genes, respectively, and no significant homology to the myeloid protein, while sharing maximum 70%, 100%, 70%, and 67% identity with the HD of the respective Hox proteins.
Identification of the Cognate Epididymal Protein
Based on the sequence homology to HOXB2 protein, Western blot analysis was carried out using an affinity-purified monospecific goat polyclonal primary antibody (HOXB2 [P-20]: catalog no. SC-17165; Santa Cruz Biotechnology) raised by epitope mapping within an internal region of HOXB2 protein of human origin to identify the expression of cognate protein, if any, in the rat epididymis. The results indicated the presence of a single band of ~30 kDa in the epididymal tissue and rat caudal sperm extract and a protein of ~37 kDa, as expected, in the rat embryonic extract (Fig. 4). No bands were detected in any of the lanes when probed using nonimmune goat serum (NGS) or with secondary antibody alone (negative control, not shown). A single band of 55 kDa was observed uniformly in all lanes when the blot was reprobed using a monoclonal antibody to ß-tubulin, which served as the internal control.
|
Tissue-Restricted Expression of Hoxbes2
To determine the transcript size and tissue-specific expression, Northern blot analysis performed using a digoxigenin-labeled, 581-bp cDNA probe identified a single 2.5-kb transcript exclusively in caput, corpus, and cauda epididymal RNA samples (Fig. 5A). Detection of equivalently intense 651-bp product uniformly in all the lanes of the blot reprobed with ß-actin gene confirmed the equal loading and integrity of the RNA template. Similarly, multiple-tissue Western blot analysis was carried out for proteins extracted from brain, heart, liver, lung, kidney, skeletal muscle, large and small intestine, spleen, testis, prostate, seminal vesicle, vas deferens, and the epididymis probed using the HOXB2 antibody. A single band of ~30 kDa protein was detected only in the epididymis (Fig. 5B). No reactivity was evident in any of the other tissues analyzed either using antibody to HOXB2 or nonimmune goat serum or secondary antibody alone (negative control). A single band of 55 kDa observed uniformly in all the lanes of the blot reprobed using a monoclonal antibody to ß-tubulin served as the internal control. Thus, the results of multiple-tissue Northern and Western blot analysis were complementary in suggesting the tissue-restricted expression of Hoxbes2 in the epididymis.
|
Hoxbes2 Is Conserved Across the Species
RT-PCR analysis carried out using gene-specific primers for Hoxbes2 revealed a single product of ~1.6 kb from the adult rat, monkey, and human epididymis (Fig. 6A). No band was observed in the negative control, and a 580-bp ß-actin amplified product of uniform intensity in all the lanes was indicative of equal loading.
|
Domain-Specific Localization of Sperm HOXBES2 Protein
Indirect immunofluorescence to study whether and where the sperm acquires the HOXBES2 protein during its transit through the epididymis revealed a fluorescent signal on the midpiece and acrosome of rat caudal, monkey, and human spermatozoa and on the equatorial segment of monkey sperm (Fig. 6B). Further, the absence of a fluorescent signal on the sperm traversing from the testis until it reached the distal caput in rat epididymis indicated that the transfer of HOXBES2 protein to spermatozoa occurs from the distal caput region onwards (Fig. 6C). The specificity of immunofluorescence was demonstrated by the absence of fluorescent signal on spermatozoa incubated with NGS or with secondary antibody alone (negative control).
Immunohistochemical Analysis of HOXBES2 Protein
When subjected to immunohistochemical analysis using HOXB2 antibody, the sections of rat epididymal segments and somatic tissues revealed the presence of the HOXBES2 protein in the principal and basal cells of the caput, corpus, and caudal epididymal epithelium (Fig. 7A). Whereas the staining was visible in all three regions of the epididymis, the maximum expression was observed in the corpus and caudal regions, as confirmed by relative quantitation of immunoreacted products using image analysis software (Fig. 7B). At higher magnification, the immunoreactivity to the protein could be visualized in the apical region of the principal cells, absent in the nucleus, and observed on the luminal spermatozoa and on the stereocilia layering the epithelium (Fig. 7C). The protein was not seen when the tissues were probed with NGS or with secondary antibody alone (negative control). The absence of fluorescent signal for the HOXBES2 protein on the testicular spermatozoa (Fig. 7D) and absence of immunolocalization in the testicular tissue sections or on the testicular germ cells (Fig. 7E) confirmed the epididymal origin of the HOXBES2 protein.
|
The study reports the identification and characterization of HOXBES2, a novel sperm membrane protein expressed and secreted specifically from the adult epididymis and exhibiting homology to the HOXB2 homeoprotein. The 1.657-kb full-length Hoxbes2 cDNA (DQ399532) encodes a putative protein (ABD73307) of 305 amino acids from an open reading frame of 918 bp. The nonidentity between Hoxbes2 and the recently discovered Rhox genes in mouse, rat, and human indicates its phylogenetic divergence and origin from an ancestor (Hox) different from that of the Rhox genes that are more closely related to the paired and prd-like families [5]. Also, Hoxbes2 gene expression contrasts with that of the 5' hox genes such as Hoxa9, Hoxa10, and Hoxa11 and their Hoxd paralogs in mouse epididymis with 63% HD identity. As described in Results, multiple sequence alignment and genome database search indicated that Hoxbes2 is more closely related to Hoxb2 of Hox gene cluster, for which the Drosophila protein antennapedia represents the prototype. Mouse and rat genes are 94% identical in their peptide and nucleotide sequences [47], and similar identity also was seen between rat and mouse Hoxb2 genes. Hoxbes2 displays its closest proximity to human and shares almost identical similarity (68% and 69%) with both mouse and rat HOXB2 proteins. Therefore, it could be a new paralog of the single blast hit embryonic hypothetical rat HOXB2 protein (XP_220894). However, at this time point we are unable to conclude whether Hoxbes2 is a novel ortholog of human (HOXB2) and mouse Hoxb2 genes, owing to the difference in its temporal and spatial expression pattern. Observations similar to this have been reported for the Rhox4 genes in rat [48]. Recent discovery of paralogous Rhox genes in the mouse and rat genome implicate the participation of gene duplication events [21, 22]. Therefore, the identification of multiple blast hits for Hoxbes2 in mouse but not in rat suggests that the duplication events in the mouse Hox cluster took place at or soon after the time of divergence of the two species. This is further substantiated by the existence of almost 39 functional Hox members in the mouse and human Hox clusters.
The epididymis-specific expression of Hoxbes2 demonstrated by the presence of its single 2.5-kb transcript in the adult epididymis differs considerably in terms of transcript size and tissue specificity from the previously reported 1.7- and 3.2-kb Hoxb2 transcripts of the developing embryo [6] and the k562 nuclear extracts [29]. Nevertheless, these observations are not surprising, owing to the alternative splicing pattern and polyadenylation signals and use of multiple promoters that are the hallmarks of homeobox gene expression. A study recently carried out by Makova and Li [49] on the relationship between gene duplication and gene expression changes supports the divergence in the transcription profiles of a large number of paralogous genes. It could be surmised that the presence of epididymis-specific Hoxbes2 transcript could be due to either subfunctionalization or neofunctionalization processes following the gene duplication events. According to the concept of neofunctionalization [50, 51], Hoxbes2 should have evolved either through changes in the amino acid sequence and/or through changes in the gene expression pattern, resulting in the expression of Hoxb2 in a new tissue, the epididymis, where it was not expressed previously. It is likely that the Hoxbes2 expression in the epididymis could be regulated by factors (such as androgens [data not shown]) other than those responsible for Hoxb2 (known to be retinoic acid) in the embryo.
In addition, Western blot analysis revealed a 30-kDa protein (HOXBES2) in adult rat epididymal tissue and caudal sperm extract, and a 37-kDa protein representing the HOXB2 transcription factor in the 7.5-dpc rat embryo. This finding suggests the sharing of a common epitope within the HOXB2 homeodomain by the cognate epididymal and embryonic proteins. These observations also reiterate the fact that there exists only a single protein with HOXB2 homeodomain in the epididymis and caudal sperm. The absence of a 37-kDa embryonic HOXB2 protein or the 87-kDa protein of k562 nuclear extracts [52] corresponding to the Hoxb2 locus in the adult rat epididymis could be explained by the absence of their corresponding transcripts or negative regulation of their expression. Since no protein was recognized in the somatic tissues either by the antibody or when the blot was probed using either normal goat serum or secondary antibody alone, this indicated the specificity of the primary antibody. These data also confirm that the Hoxbes2 gene is both transcribed and translated in the epididymis, which lends credence to the possible functional significance of the protein. On the other hand, since the developmental pattern in the embryo is similar to the well-differentiated architecture of adult epididymis, the preferential expression of paralogous Hoxbes2 gene in the adult epididymis is similar to the Rhox4.5 gene [48] in the rat testis. The results from multiple-tissue Western blot analysis complement the observations made by Northern blot analysis in indicating the tissue-restricted/specific expression of Hoxbes2 in the epididymis. These observations augur well for Hoxbes2, as the tissue-specific genes are more likely to belong to large gene families, as Hoxb2 itself has evolved from the large cluster of homeobox gene family [51]. Further, support from the predicted molecular mass of 32.9 kDa of the ex vivo HOXBES2 protein is almost similar to and in agreement with the molecular size of the cognate epididymal protein (30 kDa). Thus, the difference in the molecular size of the predicted and cognate epididymal proteins from the embryonic protein, detection of a single band of 30 kDa in the Western blot, sequence divergence in the cDNA and at the protein level, difference in the transcript size, and variations at the temporal and spatial level expression suggest the novelty of Hoxbes2 while ruling out the possibility of being a posttranslationally modified form of embryonic HOXB2 protein. Therefore, these observations led us to believe that the Hoxbes2 expression in the adult epididymis may have additional functions to perform, as suggested for the moonlighting protein [53].
The cellular localization of HOXBES2 protein in the epididymis differs from that of HOXB2 transactivator protein and displays cell type-specific expression in the principal and basal epididymal epithelial cells. At higher magnification, intense immunostaining was observed in the supranuclear cytoplasm rich in organelles, such as mitochondria, rough and smooth endoplasmic reticulum, free ribosomes, numerous vacuoles and vesicles, multivesicular bodies, a large Golgi apparatus, membrane-delimited granules, caveolae, and lysosomes, indicating the synthesis and possible secretion of the protein. This observation was further strengthened by the staining observed on the stereocilia and on the luminal spermatozoa. Being a homeoprotein, localization would be expected to be present in the nucleus; surprisingly, HOXBES2 expression was restricted to the cytoplasm, although in silico analysis indicated the presence of an NLS (N'-QNRRMKHKRQ-C') at amino acids 170–180. Studies on homeoproteins indicate that they could reside exclusively in the cytoplasm, although their functional relevance has not yet been clearly understood, as in the case of Csx/Nkx 2.5 homeodomain protein in the cranial skeletal muscles in vertebrates [54] and 5' Hox genes [27], Growth arrest-specific (Gas) protein 8, testicular proteins like Arc (activity regulated, cytoskeleton associated) and Haprin (haploid germ cell-specific RBCC protein) [55–57]. Meanwhile, an unexpected link between secretion and passage of certain homeoproteins through the nucleus has been established by its NLS, as described for the scaffold protein Ste5 in yeast [58] and for several other proteins as reviewed by Prochiantz and Joliot [59]. Thus, a possible explanation for the cytoplasmic accumulation of HOXBES2 protein with NLS could be that its passage through the nucleus renders the protein competent for secretion. Therefore, the presence of NLS in HOXBES2 may be a prerequisite for its intercellular transfer.
Does Hoxbes2 have a role in sperm maturation? Maximum expression of Hoxbes2 in the distal regions of the epididymis that are associated with acquisition of sperm fertilizing ability and forward progressive motility correlates with the expression of other epididymal genes, like Hoxa11 [27] and Rhox5 [17]. Proteins P26h, P25b, and P34H synthesized in these regions with or without signal peptide, were carried on to the sperm by epididymosomes, and later attached to the sperm surface through a GPI anchor [60]. The presence of a potential GPI modification site at position 287-serine of the putative HOXBES2 protein might be involved in its association with such membrane-bound vesicles. Joliot and coworkers [61] have reported the association of engrailed homeoproteins with microdomains or caveolaelike particles essential for their secretion. However, further studies are required to confirm such an association of HOXBES2. Homeodomain-containing transcription factors Hoxa5, Hoxc8, Fushi-Tarazu, Engrailed, Knotted-1, Antennapedia Penetratin (Antp), herpes simplex virus (HSV) type 1 protein vp22, and human immunodeficiency virus (HIV-1) transactivator TAT protein are some of the secretory transactivators reported to be transferred between the cells [62–69]. Penetratin is a 16-amino acid (43–58) secretory peptide (RQIKIWFQNRRMKWKK) derived from the third helix of the conserved 61-amino acid homeodomain of antennapedia homeoprotein [65–67], which is used as vector for cellular internalization of hydrophilic molecules and considered to be a member of the "Trojan peptides," or cell-permeable peptides or protein translocation domains. The characteristic ability of penetratin to induce cell internalization [66] and micelle formation [70] lies in its basic nature (four arginine residues) and the 48th tryptophan residue of the antennapedia homeodomain. The 75% identity between the amino acids 168–183 of HOXBES2 with 43–58 residues of penetratin, including the tryptophan and the basic residues, suggests that the 16-amino acid "homologous peptide" might help in the secretion of HOXBES2. We propose that the secretion and internalization of the HOXBES2 protein by the principal cells is mediated mainly by the electrostatic interaction between the basic residues of the protein (predicted pI 10.1) and the negatively charged lipids in the plasma membrane. The observations by Prochiantz and Theodore [71] on homeoproteins are supportive of the notion that the HOXBES2 protein can be traded between cells and secreted in spite of the absence of a signal peptide. The presence of a fluorescent signal for the HOXBES2 protein on live unfixed spermatozoa and its disappearance following exposure to various detergents (data not shown) confirm the predicted soluble and surface-oriented nature of the protein. The differential localization of HOXBES2 protein on the spermatozoa suggests that the antibody recognizes a species-specific but identical epitope of HOXBES2 in which the protein may or may not have an identical role to play in different species. Similar observations have been reported for a testis-specific sperm autoantigen, TSA70, identified using a postvasectomy monoclonal antibody [72]. Absence of immunostaining for HOXBES2 in the testis and on the testicular germ cells and its presence on the traversing sperm from the distal caput onward suggests that the protein is synthesized and released as a secretory protein by the epididymis and is later acquired by the transiting luminal spermatozoa. Therefore, the domain-specific localization and a ~1.6-kb RT-PCR product obtained using Hoxbes2-specific primers from the epididymis of the respective mammalian species correlate with its epididymal origin and species conservation. The amino acids 1-30 of the putative HOXEBES2 protein sequence have been identified as a targeting peptide (intracellular signal peptide/intracellular localization signal sequence) by in silico analysis, which may help in the possible association of HOXBES2 protein to the mitochondria. This observation was substantiated by the immunofluorescence data showing the localization of the protein on the sperm midpiece where mitochondria lie [73].
Although the functional relevance of this intercellular transfer is not yet known, recent data on homeoproteins indicate that, following transfer, the transcription factors can regulate transcription and translation in the recipient cell as "messenger proteins" [74]. While attempts are being made to understand the role of HOXBES2 in sperm function, its localization on the acrosome of rat caudal, monkey, and human spermatozoa suggest its possible role in acrosome reaction and sperm-egg interaction. On the other hand, its presence on the sperm midpiece may suggest its participation in the energy-generating process for sperm motility. HOXBES2 localization in the equatorial segment of the monkey spermatozoa is similar to rat epididymal protein DE (also known as cysteine-rich secretory protein, or CRISP-1), which has been proposed as being involved in sperm-egg fusion and as a probable molecule in the process of fertilization [75]. Finally, the burning question is how significant is the epididymal HOXBES2 protein? To date, neither descriptive nor functional data are available on Hoxb2 expression in male reproductive tissues, particularly the epididymis; Earlier reports indicated HOXB2 to be only a transcription factor at transcription and translational levels during embryogenesis and in human haematopoietic and pulmonary systems [30, 31]. This is the first-ever study to reveal the expression of a protein homologous to HOXB2 in the adult epididymis; it also describes the secretory nature of homeoproteins for HOXBES2 that are being released from epididymis and, finally, the identity and domain specificity of a homeodomain-containing protein on the spermatozoa of various mammalian species.
Thus, it is proposed that HOXBES2 is a novel sperm homeoprotein specific to the epididymis that differs in its molecular size, subcellular localization, and functionality from the HOXB2 transactivator with a probable role in sperm physiology.
ACKNOWLEDGMENTS
The authors gratefully acknowledge Dr. C. P. Puri, Director, National Institute for Research in Reproductive Health (ICMR), for his constant support and encouragement. The authors express sincere thanks to Mrs. Shanti Ganeshan, Library and Information Officer, and Dr Geetanjali Sachdeva, Senior Research Officer, for assisting in editing the manuscript, and Mr. Ravi B. Kadam, Mr. Hemant C. Karekar, Madhavi Pusalkar, and Dr A. Maitra of the NIRHH's Central DNA Sequencing Facility, and the animal house staff for their technical assistance.
FOOTNOTES
1This work (NIRRH/MS/7/2006) was supported by grants from the Indian Council of Medical Research (ICMR). E.P. is a recipient of the ICMR Senior Research Fellowship. 
Correspondence: 2Vijaya P. Raghavan, Department of Biochemistry, National Institute for Research in Reproductive Health (ICMR), J.M. Street, Parel 400 012, Mumbai, India. FAX: 91 22 2413 9412; e-mail: vijuraghavan{at}hotmail.com
Received: 2 May 2006.
First decision: 23 June 2006.
Accepted: 17 October 2006.
REFERENCES
-globin enhancer–binding protein is a product of the Hox-B2 (HOX2H) locus. Blood 1994; 83:1420–1427-globin gene regulatory sequences. Proc Natl Acad Sci U S A 1991; 88:7318–7322 locus on mouse chromosome X. Mammalian Genome 2006; 17:178–187[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
