Introduction
Mammalian spermatogenesis is a complex multistage process in which immature germ cells undergo division, meiosis and differentiation to form round spermatids. Finally, after a series of biochemical and morphological changes, spermatids elongate and then functional spermatozoa are generated in the testis (Senoo et al., Reference Senoo, Hoshino, Mochida, Matsumura and Habu2002; McLean et al., Reference McLean, Friel, Johnston and Griswold2003; Dunleavy et al., Reference Dunleavy, O’Bryan, Stanton and O’Donnell2019)
Spermatogenesis is a well coordinated developmental programme in which all the various steps are defined by cell type- and stage-specific induction or repression of the expression of specific genes (Zhao and Garbers, Reference Zhao and Garbers2002; Chu and Shakes, Reference Chu and Shakes2013; Pariante et al., Reference Pariante, Dotolo, Venditti, Ferrara, Donizetti, Aniello and Minucci2016a; Chemek et al., Reference Chemek, Venditti, Boughamoura, Mimouna, Messaoudi and Minucci2018; Venditti and Minucci Reference Venditti and Minucci2017, Reference Venditti and Minucci2019).
The process of spermatogenesis is under a genetic and molecular programme (Sassone-Corsi, Reference Sassone-Corsi2002; de Mateo and Sassone-Corsi, Reference de Mateo and Sassone-Corsi2014) and requires precise sequential expression of many genes that are unique to spermatogenesis (Escalier, Reference Escalier2001; Ergoli et al., Reference Ergoli, Venditti, Picillo, Minucci and Politano2020). In addition, accurate transcriptional and post-transcriptional regulation is necessary to support the highly coordinated expression of specific genes for each step of spermatogenesis (Zhou et al., Reference Zhou, Chen, Jiang and He2019).
Given this background, the identification and characterization of spermatogenic-related genes is an important contribution to understanding the mechanisms of both spermatogenesis and human reproduction (Guo et al., Reference Guo, Shen, Xia, Zhang, Zhang, Zhao, Xing, Chen, Chen and Lin2010; Santillo et al., Reference Santillo, Venditti, Minucci, Chieffi Baccari, Falvo, Rosati and Di Fiore2019; Venditti et al., Reference Venditti, Fasano, Minucci, Serino, Sinisi, Dale and Di Matteo2020). In this study, an attempt was made to isolate genes involved in rat spermatogenesis and to use them as new markers for true testicular activity. In this context, we isolated, for the first time in rat testis, a partial cDNA clone for EH domain binding protein 1-like 1 (Ehbp1l1; previously named Tangerin) that had a single calponin homology domain (CH) and two transmembrane domains (Chauhan et al., Reference Chauhan, Reed, Zhang, Duncan, Kilimann and Cvekl2002; Friedberg, Reference Friedberg2010). These proteins have been described to link vesicular transport to the actin cytoskeleton (Guilherme et al., Reference Guilherme, Soriano, Bose, Holik, Bose, Pomerleau, Furcinitti, Leszyk, Corvera and Czech2004; Shi et al., Reference Shi, Chen, Banerjee, Glodowski, Audhya, Rongo and Grant2010) as an effector of the small GTPase Rab8 (Nakajo et al., Reference Nakajo, Yoshimura, Togawa, Kunii, Iwano, Izumi, Noguchi, Watanabe, Goto, Sato and Harada2016; Eguchi et al., Reference Eguchi, Kuwahara, Sakurai, Komori, Fujimoto, Ito, Yoshimura, Harada, Fukuda, Koike and Iwatsubo2018).
Calponin is an actin filament-associated regulatory protein expressed in smooth muscle and many types of non-muscle cells (Liu and Jin, Reference Liu and Jin2016). In vitro studies have demonstrated that calponin binds actin (Takahashi et al., Reference Takahashi, Abe, Hiwada and Kokubu1988a; Singh et al., Reference Singh, Bandi, Winder and Mallela2014) and cross-links microfilaments (Leinweber et al., Reference Leinweber, Parissenti, Gallant, Gangopadhya, Kirwan-Rhude, Leavis and Morgan2000). In addition, calponin interaction with many other cytoskeleton and related proteins have been described to date (Takahashi et al., Reference Takahashi, Hiwada and Kokubu1988b; Childs et al., Reference Childs, Watson, Novy, Lin and Mak1992; Fujii and Koizumi, Reference Fujii and Koizumi1999; Fujii et al., Reference Fujii, Takagi, Arimoto, Ootani and Ueeda2000; Szymanski, Reference Szymanski2004; Wu and Jin, Reference Wu and Jin2008).
Interestingly, the CH domain is a specific structural feature that has been related to actin binding; it is composed of about 120 residues located on the N-terminal side of calponin (Rozenblum and Gimona, Reference Rozenblum and Gimona2008; Singh et al., Reference Singh, Bandi, Winder and Mallela2014). The CH domain has been found in both signalling and cytoskeleton-related proteins (Leinweber et al., Reference Leinweber, Leavis, Grabarek, Wang and Morgan1999; Orlova et al., Reference Orlova, Rybakova, Prochniewicz, Thomas, Ervasti and Egelman2001; Ishisaki et al., Reference Ishisaki, Takaishi, Furuta and Huh2001; Sjöblom et al., Reference Sjöblom, Ylänne and Djinović-Carugo2008). Moreover, in most of these proteins the specific actin-binding region consists of two CH domains in tandem (Galkin et al., Reference Galkin, Orlova, VanLoock and Egelman2003; Reference Galkin, Orlova, Salmazo, Djinović-Carugo and Egelman2010), however the actin-binding capacity of the single CH domain is still controversial (Gimona and Mital, Reference Gimona and Mital1998; Gimona and Winder, Reference Gimona and Winder1998; Stradal et al., Reference Stradal, Kranewitter, Winder and Gimona1998). In this paper, to expand current knowledge on the molecular mechanisms underlying the differentiation processes of germ cells into mature spermatozoa, we characterized Ehbp1l1 expression and localization in adult rat testis and also during testicular development.
Materials and methods
Animals and tissue collection
Male Wistar rats (Rattus norvegicus) were housed under defined conditions (12D:12L) and were fed with standard food and provided with water ad libitum. Animals were sacrificed at several stages of the development: infant–prepuberal period (14 and 30 days old), adult (3 and 6 months old) and 12 months old (old; three animal/each time). All the animals were killed by decapitation under ketamine anaesthesia (100 mg/kg i.p.) in accordance with local and national guidelines covering experimental animals. Testis were collected from rat of all stages, several tissues (lung, spleen, kidney, liver, small intestine, seminal vesicles, and muscle and one testes) were dissected from 3-month-old rats, tissues were frozen quickly by immersion in liquid nitrogen, and stored at −80°C until RNA extraction. In addition, the other 3-month-old testis was fixed in Bouin’s fluid for histological analysis. Finally, to isolate germinal cells by centrifugal elutriation, an additional two 3-month-old rats were sacrificed and the testes were used (see below).
Cloning of Ehbp1l1 cDNA by RT-PCR
Total RNA from tissues and testis collected at different stages of development was isolated using TRIzol reagent (Sigma-Aldrich, St. Louis, MO, USA). The quantity (ng/ml) and purity (260/280 and 260/230 ratios) of total RNAs were assessed using a NanoDrop 2000 spectrophotometer (Thermo, Waltham, MA, USA). To remove potential contamination of genomic DNA, RNA aliquots (10 μg) were treated with 2U DNase I (Amersham Bioscience) according to the manufacturer’s recommendations (Falvo et al., Reference Falvo, Chieffi Baccari, Spaziano, Venditti, Rosati, Di Fiore and Santillo2018). First-strand cDNA was synthesized using 5 μg of total pooled RNA, the reverse transcription reaction was conducted in 20 µl of total volume reaction according to the manufacturer’s recommendations (Venditti et al., Reference Venditti, Fasano, Santillo, Aniello and Minucci2018a, Reference Venditti, Aniello, Santillo and Minucci2019). As a negative control, the same reaction was carried out on a pool of RNA without using the reverse transcriptase enzyme (RT−).
Based on the published mRNA sequence of Rattus norvegicus Ehbp1l1 variant 2 (NCBI data bank accession NM_001129997.1), polymerase chain reaction (PCR) was performed using the following primers: Ehbp1l1 For: 5′-GCAACAGAAAGCAGAGAGGG-3′ and Ehbp1l1 Rev: 5´-ATGTAAAATAGCACAGAAGGCCA-3´. An appropriate region of Rattus norvegicus β-actin mRNA (Act; NCBI GenBank accession no. NM_031144.3) was amplified with specific oligonucleotide primers (Act For: 5´-CTCTTCCAGCCTTCCTTCCT-3´; Act Rev: 5´-CTGCTTGCTGATCCACATC-3´) and used as a control. PCR amplification was carried out for 35 cycles with denaturing at 94ºC for 45 s, annealing at 58ºC for 45 s and extension at 72ºC for 45 s, followed by a final extension at 72ºC for 5 min. Amplification products were electrophoresed on 1.2% agarose gel and visualized with ethidium bromide staining and ultraviolet (UV) light. The expected DNA amplicons for variant 2 were 382 bp for Ehbp1l1, and 300 bp for β-actin. The amplicons were purified using the QIAGEN gel extraction kit (QIAGEN, Hilden, Germany) and were cloned into a pGEM-T Easy Vector and sequenced on both strands to confirm specificity, in accordance with the manufacturer’s instruction. Nucleotide sequences were compared with the NCBI GenBank database (www.ncbi.nlm.nih.gov) to confirm the specificity of PCR products.
In situ hybridization
For in situ hybridization, randomly chosen testis sections from 3-month-old animals (7 sections/animal) were treated following the same conditions as previously described (Pariante et al., Reference Pariante, Dotolo, Venditti, Ferrara, Donizetti, Aniello and Minucci2016b; Venditti et al., Reference Venditti, Donizetti, Fiengo, Fasano, Santillo, Aniello and Minucci2018b) using a plasmid containing Ehbp1l1 cDNA linearized with either XhoI or BamHI enzymes to produce a template for an antisense or sense probe, using T7 or T3 RNA polymerases, respectively. The sense (control) and antisense cRNA probes were prepared by in vitro transcription with DIG-uridine triphosphate (UTP) (Roche Diagnostics) as recommended by the manufacturer.
Isolation of germinal cells by centrifugal elutriation
For isolation of germinal cells, we followed the protocol of Quesada et al. (Reference Quesada, Atorino, Cardone, Ciarcia and Farina1996). Testes from two 3-month-old rats were decapsulated, resuspended in 10 ml of Dulbecco’s minimal essential medium (DMEM) and seminiferous tubules, free of interstitial tissues, were obtained using collagenase treatment (0.25 mg/ml). Seminiferous tubules were then incubated at 37°C for 60 min in DMEM containing 0.25 mg/ml collagenase, 0.075 mg/ml DNase I, and 0.5% bovine serum albumin (BSA). After incubation, the cell suspension was centrifuged for 10 min at 1200 g. Aliquots of the pellet were complexed with propidium iodide and subjected to cytofluorimetry analysis in a Becton–Dickinson cytofluorimeter. Total germinal cells were resuspended in DMEM, in the presence of 0.1 mg/ml DNase I and 0.5% BSA, and separated into fractions enriched in various cell types using centrifugal elutriation, as described by Meistrich (Reference Meistrich1977). A cell suspension of 10 ml (180–220 × 106 cells) was loaded into a JE-6 Beckman elutriator rotor and separation was performed at 3000–2000 rpm and flow rates of 13–40 ml/min. The buffer employed was PBS containing 0.5% BSA; several 50 ml fractions were collected. Aliquots of the pellet of single fractions were complexed with propidium iodide and subjected to cytofluorimetry analysis in a Becton–Dickinson cytofluorimeter.
Results and Discussion
Correct male gamete formation requires the precise coordination of expression of many genes that harmonize the delicate cell cycle events as well as the differentiation mechanisms underlying spermatozoa production (Lin et al., Reference Lin, Ke, Wang, Chen, Chen, Ku, Chiang and Yeh2017). As most of these genes remains to be identified, an attempt to characterize those specific to spermatogenesis in the rat testis was made. Here we isolated a partial cDNA clone encoding for a fragment of about 120 amino acids of EH domain binding protein 1-like 1 (Ehbp1l1, previously named Tangerin). This protein contained a CH domain (CH) and two transmembrane domains (Chauhan et al., Reference Chauhan, Reed, Zhang, Duncan, Kilimann and Cvekl2002; Friedberg, Reference Friedberg2010). The first identified protein of this family was calponin (Takahashi et al., Reference Takahashi, Abe, Hiwada and Kokubu1988a), which is involved directly in the regulation of actomyosin interactions in several tissues, most prominently in smooth muscle contraction/relaxation cycles and in neuronal outgrowth (Burgstaller and Gimona, Reference Burgstaller and Gimona2004).
Analysis of the NCBI database revealed that the rat Ehbp1l1 gene includes different transcriptional isoforms. The longest transcript variant (NM_001129997.1) is characterized by a long exon of 2689 bp in length, also found in the corresponding human transcript (NM_001099409.3) and encodes for a protein that contains three well known domains: the N-terminal C2-like domain (NT-C2), which binds phosphatidylserine and phosphatidylethanolamine (Lemmon, Reference Lemmon2008), the CH domain and the C-terminal bivalent Mical/EHBP Rab binding (bMERB) domain (Fig. 1). The evolutionary conservation of these three domains is evidenced by the per cent identity obtained when comparing the corresponding EHBP1L1 sequences of rat and human (Fig. 1). In particular, the per cent identity of the three domains was higher compared with the per cent identity of the overall corresponding sequences (Fig. 1). Rat and human EHBP1L1 proteins also shared many proline-rich (PxxP) motifs (PR), known to be involved in SH3 protein binding (Kaneko et al., Reference Kaneko, Li and Li2008), such as Bin1 and amphiphysin1 and dynamin1 (Nakajo et al., Reference Nakajo, Yoshimura, Togawa, Kunii, Iwano, Izumi, Noguchi, Watanabe, Goto, Sato and Harada2016). As evidenced in Fig. 1, these motifs were clustered mainly in two regions, one region was just downstream of the NT-C2 domain, and the second region was included between the CH and bMERB domains (Fig. 1). In addition, we also evaluated the presence of the C-terminal prenylation motif (CaaX-box) in the mammalian proteins; this box is probably needed for EHBP1l1 prenylation and delivery to the plasma membrane (Gao et al., Reference Gao, Liao and Yang2009). As shown in Fig. 1, both rat and human EHBP1L1 share this motif at their C-terminal ends. Interestingly, rat and human EHBP1L1 proteins bear a central region that is responsible for their sequence diversity. This amino acid sequence is encoded in the vast majority by the amino acid sequence of the longest exon. In fact, the per cent identify corresponding to this amino acid region for the rat and human proteins was 52% (Fig.1).
Schematic representation of domain architecture of rat and human EHBP1L1. The single domains or the entire amino acid sequence were aligned using Clustal Omega software with the default parameters to find the per cent identity. Black brackets indicate the overall per cent identity between rat and human proteins. Green brackets identify the amino acid region encoded by the long exon of rat and human transcript variants (NM_001129997.1 and NM_001099409.3, respectively). Red arrows indicate the proline-rich (PxxP) motifs. The black arrow indicates the C-terminal prenylation motif (CaaX-box). NT-C2, N-terminal C2 domain; CH, calponin homology domain, bMERB, Mical/EHBP Rab binding domain.
As we first isolated the Ehbp1l1 clone from rat testis, to verify if Ehbp1l1 mRNA is distributed ubiquitously, its presence was assessed using PCR analysis on 3-month-old rat tissues (Fig. 2). Preliminarily, the primers were tested for their sequence specificity using an amplification reaction carried out on cDNA from 14-day-old Wistar testis rats. The PCR was analysed by gel electrophoresis and the resulting amplicon was purified, subcloned, and sequenced to confirm reaction specificity (data not shown). The analysis confirmed that Ehbp1l1 was expressed in all analyzed tissues, with a highest levels in the liver, small intestine, kidney and testis (Fig. 2A). As our interest was specifically for gonads, we performed an expression pattern analysis using a RT-PCR assay at different time points of testis development. As reported in the Fig. 3(A), Ehbp1l1 mRNA was revealed at each stage of testis maturation, showing a higher level at the end of the analyzed period (12 months) compared with the earlier stages. These data indicated that Ehbp1l1 is required for all the rat reproductive period. In both experiments, quality control of cDNA was performed using specific primers for β-actin mRNA (Figs 2B and 3B), while, as expected, no amplification bands were detected in the negative controls (RT−).
Expression of Ehbp1l1 mRNA in adult rat tissues. Agarose gel electrophoresis of RT-PCR products performed on total extracts of liver, intestine, spleen, muscles, lung, kidney, seminal vesicles and testis. Ehbp1l1 is expressed in all analyzed tissues. Quality of the cDNA samples was checked by amplifying a fragment of housekeeping gene rat β-actin mRNA. M: molecular weight marker (GeneRuler Express DNA Ladder). nc: negative, no-cDNA control. PRT−: pool of negative control of the RT reaction, performed by omitting reverse transcriptase. TRT−: testis-specific negative control of the RT reaction, performed as for PRT−.
Expression of Ehbp1l1 mRNA in rat testis development. Agarose gel electrophoresis of RT-PCR products performed on total extracts from 12 week, and 1-, 3-, 6- and 12-month-old rat testis. Ehbp1l1 was expressed in all analyzed tissues. Quality of cDNA samples was checked by amplifying a fragment of the housekeeping rat β-Actin mRNA. M: molecular weight marker (GeneRuler Express DNA Ladder). nc: negative, no-cDNA control. All the RT− controls were negative.
To determine which testis cell types expressed the Ehbp1l1 gene, we performed in situ hybridization experiments. In particular, we used a DIG-labelled antisense RNA probe to mark Ehbp1l1-expressing cells (Fig. 4A, B) and the corresponding DIG-labelled sense RNA probe as the experimental control to evaluate the specificity of the revealed hybridization signals (Fig. 4C, D). An in situ hybridization assay was carried out on 3-month-old testes, the youngest but completely sexually mature of the previously analyzed stages. Interestingly, as showed in Fig. 4A, B, a strong signal was detected only in the primary and secondary spermatocytes (SPC, arrows) while it was completely absent in spermatogonia, spermatids (SPT), spermatozoa, as well as in somatic cells (i.e. Sertoli and Leydig cells).
Localization of Ehbp1l1 mRNA in the testis of adult rat and expression analysis on isolated germ cells. (A–D) Sections were treated with antisense probe (A); high magnification in (B); or sense probe as a negative control (C, D). Blue staining indicates positive cells, which include meiotic types I and II SPC. All the other cell types were negative. Scale bars represent 20 µm in (A, C), 5 µm in (B, D). (E, F) Agarose gel electrophoresis of RT-PCR products performed on total extracts from type I SPC, type II SPC and SPT isolated from adult rat testis. Ehbp1l1 is expressed in types I and II SPC and not in SPC. Quality of the cDNA samples was checked by amplifying a fragment of the housekeeping gene rat β-actin mRNA. M: molecular weight marker (GeneRuler Express DNA Ladder). nc: negative, no-cDNA control. All samples for the RT− controls were negative.
As well known, SPCs (and their female counterpart) are the only cells that undergo meiosis division to form type I SPC and type II SPC. Interestingly, proteins containing CH domains have been associated with cell division, and in particular to binding the actin filaments that are required for the assembly and dynamics of the contractile actomyosin ring (Shannon and Li, Reference Shannon and Li1999) and, in some cases, the domains were specific for meiotic division (Mintz et al., Reference Mintz, Galperin, Pasmanik-Chor, Tulzinsky, Bromberg, Kozak, Joyner, Fein and Horowitz1999; Connolly et al., Reference Connolly, Osterberg, Christensen, Price, Lu, Chicas-Cruz, Lockery, Mains and Bowerman2014). Therefore Ehbp1l1, thanks to its CH domain, could participate in the cytoskeletal remodelling, necessary for a proper cell division.
Moreover, it is possible to hypothesize also an additional role for Ehbp1l1 in these phases: via the proline-rich domain, the relative protein is able to interact with Rab8, Bin1/amphiphysin II, and dynamin to form a multiprotein complex, which regulates the transport of protein cargos from the trans-Golgi network (TGN) to the apical plasma membrane (Nakajo et al., Reference Nakajo, Yoshimura, Togawa, Kunii, Iwano, Izumi, Noguchi, Watanabe, Goto, Sato and Harada2016; Ang et al., Reference Ang, Taguchi, Francis, Fölsch, Murrells, Pypaert, Warren and Mellman2004). During SPT differentiation, one of the key aspects is acrosome formation, a membrane-bound organelle that releases hydrolytic enzymes to facilitate sperm penetration through the zona pellucida of the oocyte (O’Donnell, Reference O’Donnell2015). Its biosynthesis begins in round SPT soon after meiosis, with coated vesicles growing from the TGN sorted gradually to the nuclear membrane (Ventela et al., Reference Ventela, Mulari, Okabe, Tanaka, Nishimune, Toppari and Parvinen2000; Kierszenbaum et al., Reference Kierszenbaum, Rivkin and Tres2011). It has to be highlighted that, in mammals, during spermiogenesis a progressive transcriptional inactivation occurs (Kierszenbaum et al., Reference Kierszenbaum, Rivkin and Tres2011) and, immediately before this, two significant bursts of RNA synthesis happen in midpachytene SPC and in round SPT, respectively, therefore supporting the translation of proteins incorporated in the components of differentiating SPT (Monesi, Reference Monesi1965; Monesi et al., Reference Monesi, Geremia, D’Agostino and Boitani1978). As we already characterized the prothymosin-alpha mRNA/protein system, which showed a comparable temporal expression pattern (Ferrara et al., Reference Ferrara, Izzo, Pariante, Donizetti, d’Istria, Aniello and Minucci2010), we hypothesized that expression of the Ehbp1l1 transcript starts in types I and II SPC, and that the expression of the protein would probably take place in differentiating SPT, to sustain the vesicle trafficking from TGN (Ventela et al., Reference Ventela, Mulari, Okabe, Tanaka, Nishimune, Toppari and Parvinen2000; Kierszenbaum et al., Reference Kierszenbaum, Rivkin and Tres2011).
To further validate these results, we performed an elutriation experiment on testes of two adult rats (3 months old) to separate the different classes of testis cells. We obtained enriched fractions of three different cell types from which total RNA was extracted and used for RT-PCR. As expected, we observed a specific Ehbp1l1 amplification band of the expected size only in types I and II SPC fractions, but not in the SPT fraction (Fig.4E). As mentioned previously, quality control of cDNA was performed using specific primers for β-actin mRNA (Fig. 4F) and no amplification bands were detected in the negative controls (RT−).
In conclusion, from our data, we could hypothesize a role for Ehbp1l1 during rat spermatogenesis. Given the presence of specific domains in the protein sequence, which allow EHBP1l1 to participate both in cytoskeletal remodelling and regulation of vesicle sorting from the TGN to the apical plasma membrane, and given the specific expression of this mRNA in the rat meiotic cells, its possible role during the differentiation phases of spermiogenesis could be hypothesized.
Acknowledgements
In memory of our friend, Prof. Anna Cardone.
Financial support
The authors received financial support for research, authorship, and publication of this article from the Department of Experimental Medicine, University of Campania ‘Luigi Vanvitelli’ (2018), Italy.
Conflict of interest
None
Ethical standards
The experimental protocol and animal housing conditions were in accordance with the Italian guidelines (D. Lvo 116/92) and authorized by the local Animal Care Committee (Servizio veterinario ASL 44, Prot. Vet. 22/95).
