|Home | About | Journals | Submit | Contact Us | Français|
All four CATSPER channel pore-forming subunits (CATSPER1–4) are localized in the sperm principal piece. They form an alkalization-activated Ca2+-permeable channel and are required for sperm-hyperactivated motility, egg coat penetration, and male fertility. Unlike many other ion channels, the composition of the CATSPER protein complex is poorly defined. Herein, we describe the novel protein CATSPERG associated with the CATSPER complex. CATSPERG is predicted to be a single transmembrane-spanning protein with a large extracellular domain and a short intracellular tail. Like all the CATSPERs and the previously identified CATSPER-associated protein CATSPERB, CATSPERG is only expressed in testis and is localized in the sperm principal piece. In CATSPER1-deficient sperm, the CATSPERG protein (but not the K+ channel protein KCNU1) is also lost. Together with previous findings, our data suggest that the CATSPER protein complex contains pore-forming proteins and two additional proteins (CATSPERB and CATSPERG) and that the trafficking and/or assembly of these proteins depends on CATSPER1.
Sperm cells in animals ranging from sea urchins to humans use Ca2+ as a messenger to control cellular processes such as capacitation, motility, and the acrosome reaction . Proteins of many Ca2+-permeable ion channels, including the highly selective voltage-gated Ca2+ channels (CaVs), the less selective transient receptor potential channels, and the cyclic nucleotide-gated channels, have been detected in testis and sperm [2–4]. In addition, these proteins are found in other cell types such as neurons and muscle cells. The in vivo significance of these proteins to sperm physiology and fertilization remains unclear, as targeted disruption of these channel genes in mice resulted in no obvious male fertility defect [5–10]. Sperm also have a family of sperm-specific ion channels, CATSPER [11, 12]. CATSPER channels have characteristics of other classes of cation channels. Like voltage-gated K+ channels (KVs), the CATSPER pore-forming subunits are six transmembrane (TM)-spanning proteins, but CATSPER selectivity filter sequences are similar to those of CaV channels (24 TM) . Mammals have four CATSPER genes (CATSPER1–4), each of which is required for sperm-hyperactivated motility and male fertility, as shown by targeted gene disruption in mice [11, 12, 14]. In addition, mutation in the CATSPER2 gene is implicated in male infertility in humans . CATSPER channels are required for the Ca2+ influxes activated by an alkaline depolarizing medium  and cyclic nucleotides . Calcium ions entering the channels in the principal piece trigger intracellular Ca2+ concentration increases in the midpiece and head through unknown mechanisms . The CATSPER proteins presumably form an alkalization-activated Ca2+-permeable channel in sperm, as disruption of the Catsper genes in mice also led to the elimination of such a channel current [11, 14, 18]. Attempts to express functional CATSPER channels in heterologous systems have been unsuccessful.
In general, a complete ion channel complex is composed of pore-forming proteins and one or more auxiliary subunits . For example, CaV channels consist of a pore-forming α1 subunit that determines the ion selectivity, a β subunit (an intracellular protein), a single TM-spanning α2δ subunit, and a multiple membrane-spanning γ subunit . Similarly, voltage-gated Na+ (NaV) channels are composed of the pore-forming α subunit and the single TM-spanning β subunits. The auxiliary subunits have fundamental roles in the formation and localization of the channels, as their presence influences the biophysical properties of the channels reconstituted in heterologous expression systems . These subunits are also essential for the channel function in vivo, as mutations in CaV β  or α2δ subunits  lead to severe disorders or lethality.
Unlike the composition of NaVs and CaVs, the subunits of CATSPER channels are not well studied. Because all CATSPER pore-forming proteins (CATSPER1–4) are required for the functional alkalization-activated current, the channel pore is thought to be a tetramer of the four CATSPERs . In addition, the channel complex contains the multiple TM-spanning protein CATSPERB (previously called CATSPERβ) . Herein, we identify the novel single-TM protein CATSPERG associated with the CATSPER complex.
All procedures described herein were reviewed and approved by the University of Pennsylvania Institutional Animal Care and Use Committee and were performed in accord with the Guiding Principles in the Care and Use of Laboratory Animals by the National Institutes of Health. The Catsper1−/− mouse was previously described . Tg(Catsper1/EGFP)1Dren transgenic mice carry a transgene encoding a fusion protein between an HA-tagged green fluorescence protein (eGFP) and CATSPER1 (HA.EGFP.CATSPER1) in the Catsper1-null background . Unlike the Catsper1-null mice, the male transgenic mice are fertile, suggesting that the fusion protein functionally replaced the wild-type CATSPER1.
The CATSPER1 protein complex was purified from mouse testis as previously described . Purified proteins were separated on SDS-PAGE gel, and specific protein bands were excised and subjected to protein identification using mass spectrometry. Five previously nonanalyzed CATSPERG peptides were identified in the band also containing CATSPERB (predicted to be 126 kDa, close to the 131 kDa of CATSPERG).
Novel peptide sequences from mass spectrometry were used to search databases to assemble putative open reading frames (ORFs). The whole CATSPERG ORF (GenBank accession number GQ225581) was amplified from mouse testis first-strand cDNA by PCR. To ensure that no mutations were incorporated in the PCR products, sequences were compared with genomic and cDNA expression sequence tag sequences; clones without unwanted mutations were used to assemble the full length into pcDNA3-based vector. The start of the ORF was determined by the presence of an in-frame stop codon in the 5′ untranslated region. Sequences of human, sea urchin, and Ciona intestinalis CATSPERG homologs were obtained by searching National Center for Biotechnology Information databases and the C. intestinalis cDNA databases .
To identify the overall expression pattern, CATSPERG RT-PCR was performed using a multiple-tissue mouse cDNA panel (Clontech, Palo Alto, CA). The forward and reverse primers had the following sequences: 5′-AGT CGA GTG GCT GTG CTT GGA GAA C-3′ and 5′-CTA TGC TGT CTT AGC TTG AAC ATT GTC C-3′, respectively. The RT-PCR amplification included 35 cycles of 20 sec at 94°C, 20 sec at 58°C, and 30 sec at 72°C. The mouse G3PDH was amplified as an input control for the cDNA panel kit, and the PCR samples in Figure 2A were drawn from the same reactions used in the study by Liu et al. . To determine the specific localization of CATSPERG in the testis, a 1.4-kilobase single-strand digoxigenin-labeled RNA probe was synthesized with a T7 primer (starting at base 2068) and was used in in situ hybridization to testis sections (10 μm thick) as previously described .
Two polyclonal antibodies were developed by immunizing rabbits with keyhole limpit hemocyanin-conjugated peptides with the following sequences derived from mouse CATSPERG: QDHEPIDTVAKVFQC (for anti-CATSPERGNT [N-terminal amino acids 67–80]) and CRAGSKEDNVQAKTA (for anti-CATSPERGCT [C-terminal amino acids 1132–1145]). Antisera were affinity purified against the peptides linked to SulfoLink agarose (Pierce, Rockford, IL). Antibodies against CATSPER1 and UNC-80 (a neuronal ion channel-associated protein) were previously described [13, 26]. The following antibodies were purchased from commercial suppliers: anti-KCNU1 (mSLO3; NeuroMab, Davis, CA), anti-FLAG (Sigma, St. Louis, MO), anti-HA (Santa Cruz Biochemicals, Santa Cruz, CA), horseradish peroxidase (HRP)-conjugated anti-rabbit IgG secondary antibody (Pierce), and anti-GFP and Alexa 568-conjugated anti-rabbit IgG (Invitrogen, San Diego, CA).
Testis membrane protein was prepared as previously described . About 500 μg of protein was solubilized in radio-immunoprecipitation assay (RIPA) buffer (50 mM Tris-Cl [pH 7.4], 150 mM NaCl, 1% NP-40, 0.5% [w/v] deoxycholate, and 0.1% [w/v] SDS) supplemented with Complete Protease Inhibitor Cocktail (Roche, Indianapolis, IN) at 4°C for 1 h. After a 30-min spin at ~21000 × g, the supernatant was mixed with anti-HA, anti-GFP, or anti-FLAG (negative control) to precipitate HA.EGFP.CATSPER1 (or with anti-CATSPERGCT [anti-UNC-80 and anti-FLAG as negative controls] to precipitate CATSPERG). Following incubation at 4°C for 2 h, samples were mixed with 50 μl of RIPA buffer-equilibrated protein A agarose at 4°C for 2 h or overnight. After three washes (10 min each) with RIPA buffer, bound protein was eluted in 50 μl (for Fig. 3B) or 60 μl (for Fig. 3C) of lithium dodecyl sulfate (LDS) sample buffer, and 15 μl (for Fig. 3B) or 3 μl (for Fig. 3C) of the elution was analyzed using Western blot.
Testis membrane protein (~1 mg) was solubilized with 1% Triton X-100 in 250 μl of binding buffer (50 mM Na2HPO4, 300 mM NaCl, and 1× protease inhibitor cocktail [pH 8.0]) at 4°C for 1 h. Soluble protein was added to 100 μl of buffer-equilibrated cobalt resin (Clontech) supplemented with 20 mM imidazole (pH 7.5) in 250 μl of binding buffer and mixed at 4°C for 2 h. Following a quick spinning of the protein-resin mixture, the supernatant (flow through) was collected and concentrated with a Microcon-50 ultrafiltration column (Millipore, Bedford, MA), and the cobalt resin was washed twice with binding buffer. Bound protein was eluted with 300 mM imidazole in binding buffer. Flow through and eluate were mixed with LDS loading buffer and heated for 15 min at 70°C before electrophoresis.
SDS-PAGE was performed using the 4%–12% Bis-Tris gradient gels in MOPS-SDS running buffer (NuPAGE gels; Invitrogen) supplemented with NuPAGE antioxidant according to the manufacturer's instructions. Resolved proteins were transferred onto polyvinylidene difluoride membrane (Immuno-Blot; Bio-Rad, Cambridge, MA). The membrane was blocked with 5% nonfat dry milk in PBS with 0.1% Tween-20 and incubated with primary antibody at 4°C overnight. Anti-CATSPERGNT and anti-CATSPERGCT were used at ~0.3 μg/ml and ~0.65 μg/ml, respectively. Following incubation with HRP-labeled secondary antibody for 1 h at room temperature, the membrane was developed with SuperSignal ECL (Pierce).
Because CATSPERG is only detected in testis, the anti-CATSPERGCT antibody was preabsorbed against cyanogen bromide-activated-Sepharose beads (Sigma-Aldrich, St. Louis, MO) conjugated with total protein from mouse brain, heart, and kidney to minimize nonspecificity. Sperm cells were spotted onto coverslips and fixed in cold acetone at −20°C for 2 min. After permeabilization with 0.2% Triton X-100 in PBS for 15 min, cells were blocked with 3% bovine serum albumin (BSA) in PBS for 30 min and incubated with 0.65 μg/ml preabsorbed anti-CATSPERGCT in PBS containing 3% BSA at 4°C overnight. After three washes with 0.05% Triton X-100 in PBS, coverslips were incubated with Alexa 568-conjugated donkey anti-rabbit IgG secondary antibody (2.5 μg/ml) in PBS containing 10% normal goat sera at room temperature for 90 min, washed three times for 10 min each with 0.05% Triton X-100 in PBS, and mounted in antifade reagent (ProLong Gold; Invitrogen) for visualization with a Nikon (Natick, MA) fluorescence microscope with a 60× oil immersion lens. Wild-type and Catsper1−/− sperm were processed identically.
We previously purified the CATSPER1-containing protein complex from mouse testis using transgenic mice expressing an HA- and eGFP-tagged CATSPER1 protein (HA.EGFP.CATSPER1) in the Catsper1-null background . When separated on SDS-PAGE gel, the purified protein complex has two specific bands in addition to the CATSPER1-containing one. One of the bands was identified as HSPA1B, a testis-specific heat shock protein. Twelve peptides from the other band at ~130 kDa were identified as CATSPERB (predicted molecular weight, 126 kDa) . Further protein purification preparation and mass spectrometry analysis of the ~130-kDa band revealed five additional peptides that do not match CATSPERB. Using these five peptide sequences, we cloned a novel protein, CATSPERG. The mouse CATSPERG is predicted to be 1145 amino acids, with a predicted molecular weight of 131 kDa and a isoelectric point (pI) of 6.7 (Fig. 1A). The CATSPERG sequence is not similar to that of any other protein with known function. A hydrophilicity plot predicts one signal peptide, a large extracellular domain (~1000 amino acids), a TM-spanning domain, and a short intracellular tail (58 amino acids) (Fig. 1, B and C). The predicted TM topology is similar to that of the α2δ subunit of CaV channels and the β subunits of NaVs .
CATSPERG homologs are present in the genomes of all animals that have CATSPER channels, including humans, mice, C. intestinalis, and sea urchins, but are absent from the genomes of animals without CATSPER channels, including Drosophila melanogaster and Caenorhabditis elegans. Like other CATSPER complex proteins, CATSPERG has a low sequence identity (55%) between human and mouse.
We used RT-PCR to determine when and where CATSPERG is expressed. CATSPERG mRNA was only detected in testis (Fig. 2A). The expression pattern is highly similar to that of CATSPER1–4 and CATSPERB [11, 13, 24, 27, 28]. To determine the expression within testis, we performed in situ hybridization with labeled RNAs derived from CATSPERG cDNA. Sense probe as a specificity control did not detect significant signal; antisense probe detected expression in the seminiferous tubules but not in the interstitial cells. Within the seminiferous tubules, CATSPERG was detected in spermatocytes and spermatids (Fig. 2B). Further experiments are needed to compare the expression time windows of CATSPER1–4, CATSPERB, and CATSPERG.
We developed two polyclonal antibodies against the amino-termini and carboxy-termini of CATSPERG (α-CATSPERGNT and α-CATSPERGCT, respectively). Both antibodies specifically recognized CATSPERG protein in CATSPERG cDNA-transfected HEK293T cells (Fig. 3A). Proteins immunoprecipitated from the HA.EGFP.CATSPER1 transgenic testis with anti-HA or anti-GFP antibodies were recognized by the antibody against CATSPERG (Fig. 3B). Conversely, immunoprecipitation with an anti-CATSPERG antibody also precipitated CATSPER1 as detected by an anti-CATSPER1 antibody (Fig. 3C). These data confirmed the association between CATSPER1 and CATSPERG in testis.
CATSPER1 protein binds to a cobalt column, presumably because of the protein's histidine-rich N-terminus (Fig. 3D, upper panel) . CATSPERG, which has no histidine-rich region, could be detected from proteins eluted from cobalt columns incubated with wild-type mouse testis protein but not from those incubated with Catsper1−/− testis protein (Fig. 3D), suggesting a binding of CATSPERG to the cobalt column via its association with CATSPER1. In the cobalt column flow through, CATSPER1 protein was largely undetectable (Fig. 3D, upper panel, lane 3), whereas CATSPERG could be detected (Fig. 3D, lower panel, lane 3), suggesting that not all CATSPERG protein is associated with CATSPER1 in testis.
In wild-type mouse sperm, both of the CATSPERG antibodies recognized a protein that migrated at the same position as that from CATSPERG cDNA-transfected HEK293T cells (Fig. 4, A and B). In the Catsper1−/− sperm, however, CATSPERG was undetectable; CATSPERG protein was restored in the Catsper1−/− sperm rescued with the transgene Tg(Catsper1/EGFP)1Dren (Fig. 4). Unlike in sperm, CATSPERG protein levels in testis were similar between wild type, Catsper1−/−, and transgenic (Fig. 4, A and B). These data suggest that the presence of CATSPERG protein in sperm, but not in testis, depends on the presence of CATSPER1 protein.
The loss of CATSPERG protein (Fig. 4, A and B) (as well as another CATSPER channel pore-forming protein, CATSPER2 , and the CATSPER-associated protein CATSPERB ) in Catsper1−/− sperm raised the suspicion that the disruption of the Catsper1 gene in the knockout leads to nonspecific degradation or instability of sperm proteins. This effect, however, appears to be specific for CATSPER-associated proteins, as KCNU1 (a testis-specific protein  that presumably forms the alkalization-activated potassium channel KSper ) was not disrupted in the Catsper1−/− sperm (Fig. 4C). The intact KCNU1 protein in Catsper1−/− sperm is consistent with the detection of the KSper channel current in the Catsper1−/− mutant sperm . These data suggest that only the CATSPER-related pore-forming subunits and auxiliary proteins are CATSPER1 dependent.
To determine the subcellular localization of CATSPERG, we performed immunofluorescence staining with wild-type and Catsper1−/− sperm. A CATSPERG antibody recognized signals localized in the principal piece (Fig. 5). Consistent with the absence of protein in the Catsper1−/− sperm, the antibody detected minimal signals in the mutant sperm.
Herein, we describe CATSPERG, a novel single TM-spanning protein associated with CATSPER. We provide three lines of evidence supporting CATSPERG as a component of the CATSPER complex: 1) CATSPERG copurified with CATSPER1, 2) it colocalized with CATSPER proteins, and 3) it was undetectable in the Catsper1−/− sperm and was restored with an HA.EGFP.CATSPER1 protein in the transgenic mice. In addition, CATSPERG shares many common features with other CATSPER-related proteins. For example, all four CATSPERs (CATSPER1–4), CATSPERB, and CATSPERG are expressed exclusively in testis; the proteins are strikingly localized to the sperm principal piece [11, 13, 24, 28] (see also the study by Jin et al. ), consistent with the detection of functional ionic currents  and Ca2+ influx  mediated by the CATSPER channels in the same sperm subcellular region. Furthermore, they are all found in the same unique set of animals but not in D. melanogaster and C. elegans . Like many other fertilization-specific proteins, the CATSPER-related ones between human and mouse have a much lower sequence identity (~55%) than those most other proteins [32, 33].
The CATSPER protein complex consists at least three TM proteins: the pore-forming six-TM CATSPER1–4, a predicted two-TM CATSPERB , and the single-TM CATSPERG described herein. This structure is comparable to that of other ion channels such as CaVs, NaVs, and KVs. In several CaVs, the auxiliary subunits control the trafficking of channel complex and are required for a robust functional reconstitution of proper ion channel currents in heterologous systems . Further functional reconstitution studies will be needed to test whether CATSPERG functions as an auxiliary subunit of the CATSPER channel. In preliminary experiments, we have coexpressed CATSPERs with CATSPERB and CATSPERG but have not detected functional channel currents (Ren et al., unpublished results). The inability to reconstitute a functional CATSPER channel remains unclear. In the protein purification, we did not detect other major proteins in the CATSPER1-containing complex . Other proteins might have a role in the complex, but they may have a weak association with CATSPER1 that does not survive the purification with detergents . Alternatively, the necessary subunits may interact with CATSPER only in sperm and not in testis cells, which are the major source of the purified CATSPER protein complex.
CATSPER channels can be directly activated by intracellular alkalization . However, the channels might also be activated by other biological stimuli. Before fertilization, sperm cells interact with many other cell types within the seminiferous tubules, the male and female reproduction tracts, the cumulus cells around the egg, and finally the egg coat and membrane. In NaV channels, the extracellular domains of the single TM-spanning β subunits modulate the NaV's biophysical properties and function as adhesion molecules in the interaction between neurons and glia . It will be relevant to determine whether the large extracellular domain of CATSPERG (~1000 amino acids) has similar roles in mediating the interaction between sperm and other cells and/or in coupling of environmental stimuli to channel opening.
We thank Dr. Chao-Xing Yuan at the Proteomic Core Facility at the University of Pennsylvania (supported by the Genomics Institute, Abramson Cancer Center, and School of Medicine) for help in protein identification, Urs Berger for in situ hybridization, and members of the Ren lab for discussion.
1Supported by NIH grants 1R01HD047578 and 1R03HD045290.