SEARCH

SEARCH BY CITATION

Keywords:

  • acroplaxome;
  • head–tail coupling apparatus;
  • chromatoid body;
  • manchette;
  • sperm tail;
  • Rnf19a;
  • Psmc3

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We report the cDNA cloning of rat testis Rnf19a, a ubiquitin protein ligase, and show 98% and 93% protein sequence identity of testicular mouse and human Rnf19a, respectively. Rnf19a interacts with Psmc3, a protein component of the 19S regulatory cap of the 26S proteasome. During spermatid development, Rnf19a and Psmc3 are initially found in Golgi-derived proacrosomal vesicles. Later on, Rnf19a, Psmc3, and ubiquitin are seen along the cytosolic side of the acrosomal membranes and the acroplaxome, a cytoskeletal plate linking the acrosome to the spermatid nuclear envelope. Rnf19a and Psmc3 accumulate at the acroplaxome marginal ring–manchette perinuclear ring region during spermatid head shaping and in the developing sperm head–tail coupling apparatus and tail. Rnf19a and Psmc3 may interact directly or indirectly with each other, presumably pointing to the participation of the ubiquitin–proteasome system in acrosome biogenesis, spermatid head shaping, and development of the head-tail coupling apparatus and tail. Developmental Dynamics 238:1851–1861, 2009. © 2009 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Spermiogenesis involves a complex series of molecular and structural events consisting in the assembly of permanent structures and disassembly of transient ones during the construction of the male gamete. Consequently, the spermatid intracellular environment contains a mixture of structural and functional proteins targeted to the assembling components. In addition, the correct folding of newly synthesized proteins requires monitoring before they become constituents of the developing spermatid. Stable components of the spermatid are the highly polarized acrosome–acroplaxome complex and the head–tail coupling apparatus (HTCA), each assembling at opposite poles of the spermatid nucleus. The HTCA consists of a pair of centrioles enclosed by pericentriolar dense material with specific morphological characteristics differing from the centrosome of somatic cells. The microtubular manchette, developing at the caudal site of the acrosome–acroplaxome complex, is a transitory structure. Increasing data indicate that the manchette is involved in the transport of cargoes between the nucleus and the cytoplasm and toward the HTCA and the tail (Kierszenbaum,2002). The spermatid nucleus houses initially a transcriptional active genome consisting of DNA and bound somatic histones gradually replaced first by transient proteins and later by stable protamines (reviewed in Meistrich,1993) in correlation with progressive transcriptional silencing. Ran GTPase, housed in the manchette, has been proposed to participate in spermatid nucleocytoplasmic trafficking and contribute to the complex structure of the HTCA (Kierszenbaum et al.,2002).

Structural assembly/disassembly and nucleocytoplasmic trafficking during spermiogenesis suggest transport, turnover, and storage of disposed proteins. The ubiquitin–proteasome system (UPS) participates in protein degradation by two distinct steps: the covalent attachment of multiple ubiquitin molecules to the protein substrate and the degradation of the polyubiquitinated protein by the 26S proteasome. Ubiquitination of protein substrates involves the sequential activation of three enzymes, ubiquitin-activating enzyme (E1), ubiquitin-conjugating enzymes (E2), and ubiquitin protein ligases (E3). Ubiquitin protein ligases have been classified into two major distinct families, the HECT (for Homologous to E6-AP Carboxyl Terminus; E6-AP being the founder of the family) and the RING (for Really Interesting New Gene)-finger families (reviewed in Weissman,2001). The Siah1a (for Seven in absentia homolog 1) male mouse mutant, displaying sterility due to a block in the completion of meiosis I (Dickins et al.,2002), highlights the significance of ubiquitin protein ligases of the RING-finger family in spermatogenesis. In addition, chromatin protein turnover during spermatogenesis is accurately regulated by ubiquitination (reviewed in Baarends et al.,1999). Yet, little is known about the expression and distribution of ubiquitin protein ligases during spermatid development.

We have previously reported the cDNA cloning, expression, and localization of the protein Psmc3 (previously called TBP-1) in rat spermatids (Rivkin et al.,1997). Psmc3 is one of the 20 subunits of the 19S cap complex of the 26S proteasome (DeMartino et al.,1996). The 19S cap acts as a chaperone that recognizes polyubiquitinated substrates and enables their subsequent translocation into the 20S proteolytic barrel for degradation (reviewed in Ciechanover,2005). Six of the 20 subunits of the 19S cap, including Psmc3, are ATPases members of the ATPases-associated-with-different-cellular-activities (AAA) family (Hoyle et al.,1997). The previously observed association of Psmc3 with the microtubular manchette, chromatoid bodies, and outer dense fibers of the developing spermatid tail suggested a role of the 26S proteasome in protein turnover (Rivkin et al.,1997). We further reported the fractionation by chromatography of the 26S proteasome from rat testis and sperm tail and its visualization by whole-mount electron microscopy using negative staining (Mochida et al.,2000). An outstanding issue was which ubiquitin protein ligases exist in spermatids to account for the proteasome degradation of ubiquitinated protein. Our attention was directed to Rnf19A, a ubiquitin protein ligase initially cloned from mouse fetal ovary and testis (López-Alañon and del Mazo,1995). XYbp (for XY body protein), with 94% identity to Rnf19a, was reported to be associated with the XY chromosomal pair in mouse spermatocytes and the centrosome of somatic and germinal cells (Párraga and del Mazo,2000). A protein with the characteristics of Rnf19a was later cloned from human spinal cord, designated dorfin (for double ring-finger protein), and shown to have ubiquitin ligase activity (Niwa et al.,2001). Rnf19a consists of two double RING-finger motifs and an IBR (for In Between RINGS) motif with binding affinity to ubiquitin conjugating enzymes. Partial deletion of the RING-finger/IBR domains prevented interaction with ubiquitin conjugating enzymes and protein ubiquitination (Niwa et al.,2001). Overexpression of Rnf19a reduced aggregates of superoxide dismutase-1 (SOD1) presumably by targeting SOD1 proteins for proteasome degradation (Niwa et al.,2002). A potential association of Rnf19a with the spermatid centrosome was of interest for three main reasons. First, little is known about the regulatory mechanism that precludes centrosome duplication in spermatids that may produce sperm with multiple tails. Second, molecular details of centrosomal reduction or degradation during spermiogenesis in some mammalian species (Manandhar et al.,2000) are partially known. Third, no substantial data are available on the mechanism responsible for sperm decapitation in azh (for abnormal sperm head shape) mutant mice (Mochida et al.,1999).

We report here the cDNA cloning of rat testis Rnf19a and the localization of the encoded protein during spermiogenesis. We show that Rnf19a coimmunoprecipitates with Psmc3 and that Rnf19a and Psmc3 coexist with ubiquitin at the cytosolic side of outer and inner membranes of the acrosome. Rnf19a also associates with the acroplaxome, a cytoskeletal plate linking the acrosome to the spermatid nuclear envelope (Kierszenbaum et al.,2003a). Upon manchette development, Rnf19a and Psmc3 appear to translocate by intramanchette transport to the developing HTCA and the spermatid tail.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Amino Acid Sequence of Rat Testis Rnf19a and Its Primary Structure

A comparison of the protein sequence of rat, mouse, and human Rnf19a reveals approximately 98% and 93% identity to mouse and human, respectively (Fig. 1A). Rat testis Rnf19a has a molecular mass of approximately 90 kDa and contains 840 amino acids. The N-terminus consists of two cysteine- and histidine-rich RING-zinc finger domains (amino acids 132–179 and 301–332) separated by an IBR domain (amino acids 219–264). The two RING-finger domains and the IBR domain (Fig. 1B) are the site of ubiquitin protein ligase activity and interaction with ubiquitin conjugating enzyme (Niwa et al.,2001). The ubiquitin conjugating–ubiquitin protein ligase activity region is identical in rat, mouse and human Rnf19a, an indication of its evolutionary conservation. Two amino acids (801TS802) present in the C-terminal region of rat and mouse Rnf19a are not present in human Rnf19a. As a result, rat and mouse Rnf19a consists of 840 residues when compared with the 838 amino acids of human Rnf19a. Rnf19a has three nuclear localization sequence (NLS) motifs (amino acids 61–75, 357–360, and 396–412). In addition, Rnf19a is 94% identical to mouse XYbp protein, a component of the XY chromosomal pair in mouse spermatocytes nuclei (Párraga and del Mazo,2000). This sex chromosomal localization, together with the presence of three NLS domains, indicates that Rnf19a may translocate to the nucleus to carry out presently unknown functions. Figure 1C shows that Rnf19a transcripts are present in rat pachytene spermatocytes, round spermatocytes, testis, and sperm.

thumbnail image

Figure 1. A: Comparison of Rnf19a protein sequence in rat, mouse, and human. Dissimilar amino acids are shown for mouse and human sequences. Two amino acids missing in human Rnf19a are indicated by a black box. The position of the RING finger domains and IBR domain is indicated by dashed brackets. NLS (nuclear localization sequence) motifs are indicated by gray boxes. The dashed box indicates the synthetic peptide sequence used as antigen for the production of anti-Rnf19a serum. Note the 100% identity of the RING finger and IBR domains in rat, mouse and human. B: Diagrammatic representation (not to scale) of Rnf19a protein indicating the various domains and motifs. C: Reverse transcriptase-polymerase chain reaction detection of Rnf19a transcripts in rat pachytene spermatocyte and round spermatid cDNA expression libraries, adult testis, and epididymal sperm.

Download figure to PowerPoint

Coimmunoprecipitation of Rnf19a With Psmc3, a Component of the 19S Cap of the 26S Proteasome

A ubiquitin protein ligase activity site in Rnf19a suggested the possible involvement of the 26S proteasome. As a first step to explore this possibility, we conducted immunoblotting experiments to ascertain the specificity of the anti-Rnf19a polyclonal antibody. Figure 2A shows that anti-Rnf19a recognizes a distinct 90-kDa protein band (lane 1). Absorption of the antibody with the corresponding antigenic peptide completely suppresses recognition of Rnf19a (lane 2). Preimmune serum was negative (not shown). A coimmunoprecipitation approach was used to determine whether Rnf19a and Psmc3 interacted with each other in testicular lysates. Figure 2B, lane 1, shows Rnf19a in the starting sample. Rnf19a immunoreactive protein band is identified following immunoprecipitation with anti-Psmc3 serum (lane 2). Preimmune serum and preabsorbed antiserum with the antigenic peptide yielded negative results (data not shown). These results showed that the Rnf19a antibody was specific and that Psmc3 protein interacted either directly or indirectly with Rnf19a.

thumbnail image

Figure 2. A: Immunoblotting analysis of Rnf19a antiserum specificity and coimmunoprecipitation of the Rnf19a- Psmc3 complex. Lane 1 shows a 90-kDa protein band seen in rat testis lysates. Anti-Rnf19a serum, pre-incubated with the immunogenic peptide, does not detect an Rnf19a (lane 2). B: Coimmunoprecipitation of Rnf19a using anti-Psmc3 serum. Lane 1 illustrates a 90-kDa protein band in the starting sample stained with anti-Rnf19a serum. Lane 2 shows Rnf19a immunoprecipitated with anti-Psmc3 and stained with anti-Rnf19 serum.

Download figure to PowerPoint

Rnf19a Localization Follows a Golgi/Proacrosomal Vesicle-Acrosome/Acroplaxome-Manchette-HTCA-Tail Sequence During Spermiogenesis

An indirect immunofluorescence approach was used to recapitulate the distribution of Rnf19a during rat spermiogenesis. A composite of phase-contrast microscopy and immunofluorescence images included in Figure 3 highlights details of the distribution of Rnf19a in spermatids collected from spermatogenic stage-specific seminiferous tubules. Cells were fixed under conditions to enable visualization of the Golgi-acrosome/acroplaxome-manchette relationship in the absence of the plasma membrane removed by a hypotonic effect of the fixative (Kierszenbaum et al,2003a). Panels 1–3 illustrate spermiogenesis step (S) 1–2 (S1–S2) spermatids double-stained with anti-Rnf19a and anti–γ-tubulin (to identify the centriolar pair). The Rnf19a-immunoreactive Golgi and centrioles can be seen in panel 2; the γ-tubulin immunoreactive centrioles are shown in panel 3. Panels 2 and 3 can be correlated with the corresponding phase-contrast microscopy shown in panel 1. The accompanying diagrams (S1/S2 spermatids) depict the relocation of the centrioles and chromatoid body from a Golgi proximal site to the opposite nuclear pole. Panels 4–6 show S3 spermatids double-stained with Rnf19a and myosin Va antibodies. The Rnf19a labeling of the Golgi (panel 5) correlates with myosin Va staining (panel 6). Myosin Va was used to trace Golgi-derived proacrosomal vesicles (Kierszenbaum et al.,2003b). Panels 7–9 illustrate the Rnf19a staining of the Golgi and adjacent developing acrosome in a S4 spermatid. The chromatoid body, stained red with propidium iodide, does not display Rnf19a immunoreactivity, an indication that protein degradation may not be in progress. Panels 10,11 (S5 spermatid), panels 12,13 (S6 spermatid), and panels 14,15 (S7 spermatid) illustrate the progressive enlargement of the acrosome whose boundaries are Rnf19a immunoreactive. Rnf19a staining of the developing HTCA is depicted in panel 15. The emerging tail from the HTCA, seen in panel 14, is not stained with Rnf19a. The development of the manchette (panels 16–18, S8 spermatid), detected with anti–α-tubulin (panel 18), indicates that Rnf19a immunoreactivity is spreading to the manchette (panel 17). The corresponding phase-contrast microscopy (panel 16) shows the proximity of the acrosome–acroplaxome complex and the manchette. Rnf19a is visualized in the acrosome–acroplaxome and the manchette of a S9 spermatid (panels 19–21). Panels 22,23 illustrate in a S10 spermatid Rnf19a in the acrosome–acroplaxome region, in the manchette and the chromatoid body. Panels 24,25 show the presence of Rnf19a in the acrosome–acroplaxome unit and manchette and, to a lesser extent, in the relocating Golgi. Upon completion of spermiogenesis (panels 26–29, S19 spermatid), the acrosome–acroplaxome unit, the HTCA, and connected spermatid tail display strong Rnf19a immunoreactivity. The tubulobulbar complex region, corresponding to cytoplasmic processes extending from the concave site of the spermatid head into the adjacent Sertoli cell cytoplasm, shows the presence of Rnf19a (panel 27). A detailed view of the principal segment of the spermatid tail (panel 29) indicates a regular banded distribution of Rnf19a, suggesting a role of the Rnf19a-Psmc3 complex in the construction of the tail. Epididymal sperm (panel 30) display Rnf19a in the sperm head, the HTCA, and the tail. In summary, an analysis of Rnf19a localization during rat spermiogenesis demonstrates a pathway sequence with a starting point in the Golgi-derived proacrosomal vesicles followed by the outer acrosome membrane-inner acrosome membrane-acroplaxome unit. An Rnf19a particulate, vesicle-like pattern is later seen at the acroplaxome–manchette boundary, intramanchette, adjacent to chromatoid bodies and the HTCA.

thumbnail image

Figure 3. Distribution of Rnf19a during rat spermiogenesis. Phase contrast microscopy images are in gray (panels 1, 4, 7, 12, 14, 16, 19, 22, 24, and 26). The accompanying diagrams indicate in green the localization of Rnf19a in the Golgi, centrioles/head–tail coupling apparatus (HTCA), acrosome (Acr), acrosome–acroplaxome complex (Acr/Apx), acroplaxome (Apx), acroplaxome marginal ring (ApxMR), manchette (M), chromatoid body (ChrB), tubulobulbar complex region (TBCr), and tail. The propidium iodide red staining corresponds to the nucleus (Nu) and nucleic acid-containing ChrB. A similar notation identifies similar structures in the immunofluorescence microscopy images. Step 1–2 (S1–2) spermatids; panels 1–3: Rnf19a is initially detected in the Golgi (panel 2) and adjacent centrosome (panel 3, anti–γ-tubulin; arrows). The S2 diagram indicates the displacement of the ChrB and centrioles to the nuclear pole opposite to the Golgi. S3; panels 4–6: Rnf19a decorates the Golgi (panel 5) and derived proacrosomal vesicles stained with anti-myosin Va (panel 6) to identify the motor protein involved in vesicular transport. S4; panels 7–9: The Golgi and adjacent acrosome (arrowhead) are Rnf19a immunoreactive (panel 8). Panel 9 identifies in red the nucleus and associated ChrB that is not stained with Rnf19a. S5; panels 10,11: As the acrosome–Apx enlarge, Rnf19a staining identifies the boundaries of the two closely apposed structures visualized at different inclination angles in several spermatids with respect to the propidium iodide (PI) -stained nuclei (panel 10). The ChrB, identified by red PI staining, is stained with Rnf19a (panel 11). S6; panels 12,13: The profile of the acrosome is well defined by the Rnf19a linear staining. Golgi-specific staining gradually disappears. The arrowhead points to the overlapping inner acrosome membrane–Apx immunoreactive site. S7; panels 14,15: A side view of the Acr/Apx complex displays strong Rnf19a immunoreactivity along the outer acrosome region and a diffuse staining throughout the overlapping complex. The arrow points to the Rnf19a-stained HTCA (panel 15) whose derived tail is seen in panel 14. S8; Panels 16–18: Rnf19a is localized in the Acr/Apx complex (panel 17) and in the developing manchette stained with anti–α-tubulin (panel 18). The folded ApxMR (indicated by a dashed contour line in panel 16) marks the boundary between the Acr/Apx complex and the manchette. S9; panels 19–21: As spermatids start to elongate, Rnf19a is detected in the Acr/Apx complex, the manchette and the HTCA (arrows in panels 19–21). S10; panels 22,23: Rnf19a staining is seen in the Acr/Apx complex, the migrated Golgi and a ChrB adjacent to the red-stained nucleus (panel 23). A punctuated Rnf19a immunoreactive pattern is seen in the manchette. S11; panels 24,25: the Acr/Apx and migrated Golgi are Rnf19a immunoreactive. A diffuse punctuated immunoreactive pattern is seen in the manchette (panel 25). S19; panels 26–29: Spermatids completing their differentiation display Rnf19a in the Acr/Apx complex, the TBCr (panel 27), the HTCA (panels 27 and 28) and along the entire tail (panels 27 and 29). A regular immunoreactive banding pattern is seen along the tail (panel 29). Panel 30: Strong Rnf19a immunoreactivity is seen in the concave side of Apx, the HTCA and less intense along the tail of epididymal sperm. Scale bar: 5 μm.

Download figure to PowerPoint

Rnf19a Colocalizes With Psmc3 and Ubiquitin at the Outer Acrosome Membrane, Inner Acrosome Membrane Unit, and the HTCA

Immunogold electron microscopy was used to define at high resolution the distribution of Rnf19a and establish a correlation with Psmc3, a 26S proteasome marker, and with ubiquitin, targeting cooperatively proteins for 26S proteasome degradation. Figure 4A–C shows that Rnf19a is predominantly associated with the cytosolic side of Golgi-derived proacrosomal vesicles and the outer and inner acrosomal membrane (panel A, S7 spermatid), vesicles adjacent to the chromatoid body (panel B), and the HTCA and proximal tail segment (panel C, S10 spermatid). We previously reported the immunolocalization of Psmc3 in the manchette, chromatoid bodies, and the spermatid tail (Rivkin et al.,1997). We extend here these observations by showing that Psmc3 is also present in the acrosome membrane–acroplaxome complex and the HTCA (Fig. 4D–J). It is noteworthy the staining pattern of the acrosome membrane–acroplaxome–manchette complex. Figure 4D,E illustrates strong Psmc3 immunofluorescence along the marginal ring of the acroplaxome and the adjacent perinuclear ring of the manchette of a S8 spermatid. A moderate punctuate immunoreactive pattern is seen in the microtubular region of the manchette. A Psmc3 particle-like string pattern is recognized along the region where the marginal ring of the acroplaxome faces the perinuclear ring of the manchette. An analogous particulate immunofluorescent pattern can also be appreciated when the acrosome–acroplaxome–manchette complex is observed from the top (S8, Fig. 4F,G). Similar to Rnf19a localization is the distribution of Psmc3 along the outer acrosome membrane, the inner acrosome membrane–acroplaxome unit (Fig. 4H, S12 spermatid) and the HTCA (Fig. 4I; S13 spermatid). Epididymal sperm display strong Psmc3 immunoreactivity in the head, predominantly along the concave side, and in the implantation fossa region of the HTCA (Fig. 4J). Ubiquitin is also visualized at the outer and inner acrosome membrane and the adjacent acroplaxome (Fig. 4K, S10 spermatid) and the HTCA (Fig. 4L, S18 spermatid). Details of the peripheral and central localization of Rnf19a in the spermatid tail are shown in Figure 5A,B. Immunofluorescent (Fig. 5A) and gold-labeled (Fig. 5B) Rnf19a can be seen at the periphery and core of the middle, principal, and terminal piece of the tail. In summary, Rnf19a localization in the spermatid tail, together with the previous fractionation of intact 26S proteasomes from sperm tail (Mochida et al.,2000), lend support to a protein degradation mechanism during spermatid development, sperm maturation, and most likely during sperm capacitation.

thumbnail image

Figure 4. Localization of Rnf19a, Psmc3, and ubiquitin in the acrosome–acroplaxome complex and head–tail coupling apparatus (HTCA). A–C: Immunogold electron microscopy localization of Rnf19a. A: Rnf19a is associated with the surface of Golgi-derived vesicles and the outer (OAM) and inner (IAM) acrosome membranes (gold particles are denoted by circles) of a S7 spermatid (compare with Figure 3, panel 13). The bracket indicates the close apposition of the inner acrosome membrane with the acroplaxome. B: Localization of Rnf19a in the core of the chromatoid body (ChrB) of a S9 spermatid. The circle indicates an adjacent vesicle with gold particles on its surface. C: Distribution of Rnf19a in the HTCA of a S10 spermatid. The various components of the illustrated HTCA are: 1–2: The closely associated implantation fossa and basal plate. 3: Capitulum. 4: Striated collar. 5: Distal centriole. Gold particles are seen in 1–2, 4, and 5 and in the associated axoneme (Ax) segment. The circles indicate dense material labeled with gold particles. D–G: Indirect immunofluorescent localization of Pcm3, a 26S proteasome marker. D: Phase-contrast microscopy of a S8 spermatid. The acrosome–acroplaxome (Acr/Apx) complex and the caudally located manchette (M) are shown. The bracket identifies the close apposition of the marginal ring of the acroplaxome (black arrow) and the perinuclear ring of the manchette (white arrow). E: Psmc3 particulate immunoreactivity signal is predominantly located at the acroplaxome marginal ring (black arrow)–manchette perinuclear ring (white arrow) interface. F,G: The top view of the Acr/Apx complex of a S6 spermatid. The white arrows indicate the presence of Psmc3 at the acroplaxome–manchette contact side. H: Immunogold electron microscope localization of Psmc3 along the outer acrosome membrane (OAM, gold particles within circles) and the inner acrosome membrane–acroplaxome complex (IAM/Apx, gold particles within dashed circles), all seen in a tangential view. F-actin bundles of the Sertoli cell ectoplasmic region are closely associated to Sertoli cell–spermatid plasma membranes and the adjacent OAM, all of them poorly defined in this off-plane orientation. I: Immunogold electron microscopy distribution of Psmc3 in the implantation fossa region of the HTCA of a S13 spermatid. The numbers identify the various components of the HTCA as indicated in C, with 1 corresponding to the implantation fossa and 2 to the basal plate. Gold particles are seen in similar sites as Rnf19a (see C). The circles identify dense material with immunoreactive Psmc3. The position of the nonimmunoreactive annulus is indicated. Gold particles are associated with the developing tail (ax). J: Indirect immunofluorescence of epididymal sperm showing Psmc3 in the acroplaxome, predominantly on its concave region (similar to Rnf19a; see Figure 3, panel 30) and in the implantation fossa region of the HTCA (arrows). The arrowheads indicate the presence of specific immunoreactivity along the middle piece of the tail (see Rivkin et al.,1997, for additional details). K,L: immunogold electron microscopy localization of ubiquitin in the outer acrosome membrane (OAM, circle) and the inner acrosome membrane/acroplaxome (IAM/Apx; dashed circles) complex, both seen in a tangential view S10 spermatid. L: Immunogold electron microscope localization of ubiquitin in the HTCA of a S18 spermatid.

Download figure to PowerPoint

thumbnail image

Figure 5. A: Indirect immunofluorescence localization of Rnf19a in the luminal region of a rat seminiferous tubule. The arrows indicate Rnf19a localization in the acrosome–acroplaxome complex of elongated spermatids. The open arrowheads identify the propidium iodide (red)-stained nuclei. Rnf19a immunoreactive spermatid tails, seen in longitudinal and cross-section views, extend into the lumen. The inset is a high magnification of cross-sectioned spermatid tails (area denoted by the rectangular box). The arrow points to one of the tails to indicate the peripheral circumferential distribution of Rnf19a as well as in the core. B: Immunogold electron microscopy of cross-sectioned spermatid tails. The peripheral distribution of gold particles, representing the localization of Rnf19a, is indicated dashed circles. Specific immunoreactive sites are also seen in the center of the tail segments (rectangular box). 1: middle piece. 2: principal piece: 3: end piece. Mit, mitochondria. Scale bars = 5 μm in A.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

We report here rat testis Rnf19a protein structure, sequence identity in mouse and human orthologs and localization of transcripts and protein in developing spermatids. We show that the ubiquitin protein ligase Rnf19a colocalizes with Psmc3, a component of the 26S proteasome, and with ubiquitin. Rnf19a consists of two RING-finger domains separated by an inter-RING domain, IBR. Ubiquitin conjugating enzymes and ubiquitin protein ligases, together with a ubiquitin-activating enzyme, are members of a cascade of enzymatic reactions leading to the ubiquitination of lysine residues of a protein substrate. Results from coimmunoprecipitation experiments and immunogold electron microscopy show a relationship between Rnf19a and Psmc3, the latter a member of the 19S regulatory cap of the 26S proteasome. Future studies should determine whether antisera to protein components of the 20S proteasome core can also coimmunoprecipitate or colocalize with Rnf19a.

Molecular details of the interaction of a ubiquitin protein ligase with Psmc3, one of the six ATPase subunits of the 19S regulatory cap, were reported (Corn et al.,2003). The biological significance of Psmc3 is stressed by preimplantation lethality due to defective blastocyst development in proteasome Psmc3-deficient mice (Sakao et al.,2000). Rnf19a-Psmc3 interaction may be indicative of the feeding of ubiquitinated proteins into the central 20S catalytic chamber of the 26S proteasome for proteolysis. Each of the six ATPases of the 19S cap is a specific chaperone driving substrates to be ubiquitinated by additional ubiquitin protein ligases (Corn et al.,2003). Essentially, Rnf19a may be one of several ubiquitin protein ligases determining specificity and timing of substrate ubiquitination leading to protein degradation by the 26S proteasome during spermiogenesis.

Rnf19a-Psmc3 colocalization is first detected in Golgi-derived proacrosomal vesicles transported to the acrosome membranes. Proacrosomal vesicle fusion with the outer acrosome membrane results in the recruitment of Rnf19a and Psmc3 to the inner acrosome membrane and eventually to the acroplaxome, a cytoskeletal plate positioned strategically between the acrosome and the spermatid nuclear envelope. It needs to be determined whether Rnf19a and Psmc3 interact directly or indirectly with each other and whether the two proteins interact individually or cooperatively during acrosome biogenesis.

The colocalization of the Rnf19a–Psmc3 complex in proacrosomal vesicles and outer and inner membranes of the acrosome deserves further discussion. There are several precedents for the association of structural and functional proteins with the cytosolic side of proacrosomal vesicles and acrosome membranes (for example, Hrb [Kang-Decker et al.,2001], the myosin Va-Rab27a/Rab27b proacrosomal vesicle transport complex [Kierszenbaum et al.,2004], truncated nonreceptor tyrosine kinases FerT [Kierszenbaum et al.,2008], and truncated nonreceptor tyrosine kinase Fyn [Kierszenbaum et al.,2009]). Of particular interest, and in agreement with our data, is the presence of ubiquitinated proteins on the surface of proacrosomal vesicles and acrosome membranes in rat spermatids (Haraguchi et al.,2004). The presence of Rnf19a and Psmc3 (this study) and ubiquitinated proteins (Haraguchi et al.,2004) at the cytosolic side of proacrosomal vesicles and acrosome membranes is reminiscent of the endoplasmic reticulum-associated protein degradation (ERAD) system involved in the disposal of misfolded or unassembled proteins from the endoplasmic reticulum (reviewed in Vembar and Brodsky,2008). Further work should determine whether components of an ERAD-like mechanism are lodged in the outer and inner acrosomal membranes during acrosome biogenesis.

A noteworthy finding is the recruitment of Rnf19a and Psmc3 to the acroplaxome and its marginal ring through the inner acrosome membrane pathway. The marginal ring of the acroplaxome is a desmosome-like complex that fastens the descending recess of the acrosome to a shallow recess in the spermatid nuclear envelope–dense nuclear lamina (Kierszenbaum et al.,2003a,b). The marginal ring of the acroplaxome is closely associated to the perinuclear ring of the manchette. The nucleation of microtubules by the perinuclear ring of the manchette takes place after the marginal ring of the acroplaxome has assembled (Kierszenbaum et al.,2004). We show here that the Rnf19a–Psmc3 complex accumulates along the closely associated acroplaxome marginal ring–manchette perinuclear ring. A defective or absent acroplaxome marginal ring in the Hrb mutant mouse correlates with the ectopic assembly site of the manchette and the development of spermatids with spherical nuclei (Kierszenbaum et al.,2004). It may be possible that, in wild-type spermatids, the accumulation of the Rnf19a-Psmc3 complex at the acroplaxome–manchette boundary represents catalytic activities leading to the adjustment in diameter of the stacked acroplaxome–manchette rings. A gradual dual-ring diameter reduction may steer the apical–caudal axial elongation of the spermatid nucleus elicited by exogenous clutching forces of Sertoli cell origin (reviewed in Kierszenbaum and Tres,2004). In fact, perinuclear rings of various diameters have been visualized in fractionated manchette preparations (Mochida et al.,1998).

Three additional aspects reported here are the presence of Rnf19a in chromatoid bodies, the HTCA, and the developing spermatid tail. A chromatoid body is an aggregate of dense material in the cytoplasm of spermatocytes and spermatids, proximal or direct contact with the nuclear envelope. We have shown that Psmc3 accumulates in the chromatoid bodies of pachytene spermatocytes and spermatids (Rivkin et al.,1997). We report here the presence Rnf19a-bound vesicles adjacent to spermatid chromatoid bodies and suggest that Rnf19a and Psmc3 may be involved in the turnover of molecules accumulated in the chromatoid bodies. It has been proposed that spermatid chromatoid bodies are a site for the organization and control of RNA processing (reviewed in Kotaja and Sassone-Corsi,2007). Whether Rnf19a and Psmc3 have a role in the disposal of molecules stored in the chromatoid bodies needs to be determined.

The presence of Rnf19a–Psmc3 in the centrosome-derived HTCA is not surprising. Proteasome activity in the centrosome of somatic cells has been known for quite some time (Wigley et al.,1999). Components of the multisubunit SCF (for Skp1-cullin-F-box) RING-containing ubiquitin protein ligase are observed in the centrosome of somatic cells and they can regulate centrosome duplication (Freed et al.,2008). Centrosome duplication may have adverse consequences in haploid spermatids because of the possibility of generating sperm with more than one tail. Whether the UPS complex prevents the activity of factors involved in centrosome duplication in spermatids is not known. Neither is known the molecular mechanism that stabilizes the attachment of the sperm head to the HTCA to endure vigorous sperm forward motility avoiding head detachment (decapitation). The colocalization of Rnf19a–Psmc3–ubiquitin in the HTCA points to a probable role of the UPS in the assembly and maintenance of the functional integrity of the HTCA. In some species (for example, mouse and monkey sperm), the centriolar component of the HTCA undergoes reduction or degeneration (Manandhar et al.,2000) by an unknown mechanism. In the azh mutant, an abnormally developed HTCA may account for sperm decapitation (Mochida et al.,1999), one of the causes of male subfertility or infertility in humans. Finally, the assembly of outer dense fibers and a fibrous sheath around the axoneme of the developing spermatid tail may require the participation of Rnf19a, together with the 26S proteasome, to ensure correct protein assembly. We reported the fractionation of the 26S proteasome from rat testis and sperm tail (Mochida et al.,2000). Further studies, should define molecular details of the Rnf19a–Psmc3 interaction and the contribution of the UPS to acrosome biogenesis, spermatid head shaping, and HTCA and tail assembly.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

Cloning and Distribution of Rnf19a Transcripts

The nucleotide sequence of rat, mouse, and human Rnf19a sequences are from the NCBI database (accession nos. NM_001130560, NM_013923, and NM_183419, respectively). Parts of the rat sequence, corresponding to amino acids 1–626 and 651–840, were verified by direct sequencing of rat PCR products. A primer (forward, TTCTGAATTCTTACATCCCTCTGG; reverse, CCTGGCTTTTGCTGGTAACTTA) corresponding to the sequence of rat Rnf19a was used for reverse transcriptase-polymerase chain reaction (RT-PCR) analysis of transcripts of adult rat testis, epididymal sperm and rat pachytene spermatocytes, and round spermatids cDNA expression libraries. Sequencing confirmed that the products corresponded to the nucleotide sequence of rat Rnf19a with 93% identity to human Rnf19a. Technical details of the construction of rat pachytene and round spermatid cDNA expression libraries were reported (Rivkin et al.,1997). The procedure for extraction of total RNA of adult rat testis and epididymal sperm and the conditions for RT-PCR were described (Kierszenbaum et al.,2008).

Antibody Production, Immunoblotting, and Immunoprecipitation

An immunogenic peptide from human Rnf19a was selected using a protein analysis program (Protean, DNASTAR, Madison, WI). A peptide with the sequence 570KERYSLSGESGTVSLGTV587 was synthesized, the amino acid sequence verified by mass spectrometry and used for rabbit immunization according to a standard immunization protocol (Sigma Genosys, The Woodlands, TX). The specificity of anti-Rnf19a serum was determined by immunoblotting using lysates of rat testis (adult, Sprague-Dawley), absorption of the antibody with the antigenic peptide, and immunoprecipitation using preimmune and immune serum. Immunoblotting was carried out as previously described (Kierszenbaum et al.,2003a). The Rnf19a–Psmc3 association was determined by coimmunoprecipitation of Rnf19a with anti-Psmc3 serum using rat testis lysates. The specificity and characteristics of the anti-Psmc3 polyclonal serum were reported (Rivkin et al.,1997). For coimmunoprecipitation, anti-Psmc3 serum (55 μl) in RSBT buffer (10 mM Tris, pH 7.4. 0.1 M NaCl, 2.5 mM MgCl2, 0.5% Triton X-100) was added to 80 μl of a suspension of Protein A–Sepharose CL-B beads in phosphate-buffered saline and rotated at 4°C for 1 hr. The antibody-coated Sepharose beads were washed 4 times with RSBT buffer and combined with 60 μl/5 μg of rat testis lysate in 0.5 ml of RSBT and rotated for 20 min at 4°C. The beads with the antibody–antigen complex were washed 4 times in RSBT, 1 time with RSBT without Triton X-100, resuspended in 4× sample buffer (8% sodium dodecyl sulfate [SDS], 40% glycerol, 20% β-mercaptoethanol, 40 mM Tris, 0.02% bromophenol blue) and boiled to separate bound proteins. Proteins in the sample were fractionated on a 12% SDS-polyacrylamide gel electrophoresis gel, transferred to an Immobilon P membrane (Millipore, Bedford MA), and incubated with anti-Rnf19a serum (1:2,500) according to standard procedures.

Immunolocalization of Rnf19a, Psmc3, and Ubiquitin by Indirect Immunofluorescence and Immunogold Electron Microscopy

Spermatogenic cells from adult rats were collected as reported (Kierszenbaum et al.,2003a) from seminiferous tubular fragments. Spermatogenic stages were identified using a dissecting stereomicroscope to identify specific transillumination patterns of the tubular fragments. Dispersed spermatogenic cells were fixed for 15 min by adding a drop of 3.7% paraformaldehyde (electron microscopy grade) in 0.1 M sucrose in phosphate buffer, pH 7.4, on microscope slides coated with Vectabond (Vector Laboratories, Burlingame, CA). This fixation procedure yields plasma membrane-free preparations in which the spermatid nucleus with associated Golgi, acrosome, acroplaxome, manchette, chromatoid body, and HTCA can be readily visualized. Antibodies to α-tubulin (1:100; Sigma, St. Louis, MO), anti–γ-tubulin (1:50; catalog number T6557; Sigma), and myosin Va (1:50; catalog no. AB5887P; Chemicon International, Temecula, CA), were used to localize the manchette, centrioles, and Golgi-derived proacrosomal vesicles, respectively. Cells were immunoreacted with anti-Rnf19a rabbit serum (working dilution, 1:100); anti-Psmc3 rabbit serum (working dilution, 1:100); and anti-ubiquitin monoclonal antibody that recognizes conjugated and unconjugated ubiquitin (working dilution, 1:50; catalog no. 13-160, clone Ubi-1; Zymed Laboratories, San Francisco, CA). Preimmune serum and preabsorbed antiserum with antigenic peptide were used as controls (data not shown). Second antibodies were Alexa Fluor 488-conjugated goat anti-rabbit and anti-mouse IgG (working dilution, 1:200; Molecular Probes, Eugene, OR). Specimens were mounted with Vectashield (Vector Laboratories) without or with propidium iodide (to detect nucleic acids by a red emission color). Specimens were observed using a fluorescence microscope equipped with episcopic illumination. Images were recorded using a Magnafire digital camera (Optronics, Goleta, CA).

For immunogold electron microscopy, testes from adult rats were fixed in a mixture of 1.5% glutaraldehyde and 3.4% paraformaldehyde (electron microscope grade) in 0.1 M phosphate buffer, pH 7.2, embedded in Lowicryl K4M (Polysciences, Warrington, PA), and processed for immunogold microscopy as described previously (Rivkin et al.,1997).

Anti-Rnf19a, anti-Psmc3, and anti-ubiquitin were used at working dilutions of 1:100 in phosphate-buffered saline containing 0.1% Tween 20, 1% bovine serum albumin, and 1% goat serum. Bound antibody was detected by incubating the sample overnight at 4°C with goat anti-rabbit IgG conjugated with 10-nm gold particles (Amersham Biosciences, Piscataway, NJ) by using a 1:25 working dilution. Thin sections were stained for 5 min with 5% uranyl acetate in deionized water, and specimens were examined using a JEM-100CX transmission electron microscope operated at an accelerating voltage of 60 kV. Thick sections (1–2 μm thick) of testis embedded in Lowicryl were stained with anti-Rnf19a serum as reported (Kierszenbaum et al.,2003b) and examined using a fluorescence microscope.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES

L.L.T. and A.L.K. were funded by the National Institutes of Health.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. Acknowledgements
  8. REFERENCES