| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
1 Isoforms of the Catalytic Subunit of Cyclic 3',5'-Adenosine Monophosphate-Dependent Protein Kinase Testicular Cells1
a Department of Cell Biology, University of Massachusetts Medical School, Worcester, Massachusetts 01655
ABSTRACT
The amino terminus of the sperm cAMP-dependent protein kinase catalytic subunit (termed Cs) differs from that of the C
1 isoform expressed in most tissues due to the use of alternative transcripts of the C gene. Both C1 and Cs transcripts are present in testis; Cs is expressed specifically in spermatogenic cells and is the only C isoform detected in mature sperm. Immunohistochemistry of mouse testis using antibodies specific for Cs and C1 now shows that C1 is present in somatic testicular cells, spermatogonia, and preleptotene spermatocytes but not in cells that are in later stages of spermatogenesis. In contrast, Cs is expressed only in midpachytene and later stage spermatocytes and in spermatids. Therefore, Cs and C1 expression do not overlap. Immunofluorescence microscopic localization of Cs in murine and ovine sperm reveals that Cs is located primarily in sperm tail components, including the midpiece mitochondria and the axoneme. Quantitative analysis of Western blots indicates that individual ovine sperm contain 
4 x 105 molecules of Cs, a seemingly large number for a protein that acts catalytically.
cyclic adenosine monophosphate, kinases, sperm, spermatogenesis, testis
INTRODUCTION
Cyclic AMP-dependent protein kinase (PKA) is essential for development of the capacity to be motile during mammalian sperm maturation [1–4] and for the maintenance of motility in mature sperm [5, 6]. The sperm PKA catalytic subunit (Cs) differs from the C1 isoform expressed in most tissues in that the amino-terminal myristate and first 14 amino acids of C1 are replaced by an amino-terminal acetate and 6 different amino acids in Cs [7–10]. This difference is due to the use of alternative transcripts of the C1 gene [8, 9]. Both C1 and Cs transcripts are expressed in testis [8, 9], Cs is expressed specifically in spermatogenic cells [8], and only Cs is present in mature sperm [7]. This raises the important questions of which testicular cells express C1, and whether C1 and Cs are both expressed in spermatogenic cells. If so, then Cs must be targeted specifically to the developing sperm. On the other hand, if only Cs is present in spermatogenic cells, then it must carry out any and all cAMP-dependent protein phosphorylation in these cells.
To address these questions, we have used immunohistochemistry with antibodies specific for Cs and C1 to examine the distribution of these two isoforms in mouse testis. In contrast to Cs, which first appears in midpachytene [8], C1 is detected only in somatic cells, in spermatogonia, and in preleptotene spermatocytes. Neither isoform is detected in cells in the leptotene or zygotene stages of the first meiotic division. Therefore, Cs and C1 expression are mutually exclusive, and the two isoforms do not overlap in spermatogenic cells. As a result, it is not necessary to postulate a specific transport mechanism for moving Cs but not C1 into the elongating sperm tail. Germ cells that completely lack either isoform apparently do not use PKA-mediated protein phosphorylation for signal transduction. In later stages of spermatogenesis, all PKA-mediated protein phosphorylation must be carried out by Cs.
We also have used immunofluorescence confocal microscopy to examine the distribution of Cs in mature ovine and murine sperm. Consistent with previous biochemical results [7], Cs was detected predominantly in the sperm tail. Staining was brightest in the midpiece, suggesting that Cs is associated with the midpiece mitochondria. Cs also was detected in the remainder of the tail, where it appeared to be associated with the axoneme.
Finally, we have quantitated western blots probed with the anti-Cs antibody to estimate the number of Cs molecules per ovine sperm. The results indicate that each sperm contains 4 x 105 Cs subunits. For comparison, it is estimated that each sperm contains only 16 000 outer arm dyneins. The relatively large number of Cs molecules is consistent with the subunit being located in multiple structures within the sperm. The abundance of Cs molecules also may be related to the subunit's cAMP-insensitive anchorage in the sperm tail [7, 10], which may result in each Cs having access to only a small number of substrate molecules.
MATERIALS AND METHODS
Biochemicals
Anti-rabbit IgG-biotin conjugate, streptavidin-alkaline phosphatase conjugate, 5-bromo-4-chloro-3-indoxyl phosphate/nitro blue tetrazolium chloride/iodonitrotetrazolium violet (BCIP/NBT/INT), and basic fuchsin substrate systems were purchased from DAKO (Carpinteria, CA). Ovine Cs (oCs) and ovine C1 (oC1) were purified as previously described [7]. Recombinant mouse C1 (rC1) was a gift from Prof. Susan Taylor (University of California San Diego, La Jolla, CA). Aprotinin, digitonin, leupeptin, pepstatin, N-p-tosyl-L-lysine chloromethyl ketone (TLCK), N-tosyl-L-phenylalanine chloromethyl ketone (TPCK), polyethyleneimine, and formaldehyde were purchased from Sigma (St. Louis, MO); 4-aminobenzamidine was from Aldrich (Milwaukee, WI). All other reagents were of the highest purity available.
Antibodies
The rabbit anti-mouse Cs and rabbit anti-ovine Cs polyclonal antibodies were raised against the peptides acetyl-ASSSNDVK and acetyl-ASNPNDVK, respectively, and affinity-purified as described earlier [8]. Affinity-purified rabbit polyclonal antibody raised against a peptide (KKGSEQESVKEFLAKC) containing consensus residues 7–21 of C1 was purchased from Upstate Biotechnology (Lake Placid, NY; anti-PKA, [NT], #06–386) (C1 residues are here numbered beginning with the amino-terminal glycine of the native protein). Although the anti-C1 antibody did not cross-react with Cs in Western blots (see Fig. 1C), the peptide used to generate this antibody contained C1 residues 15–21 that are shared with Cs. Therefore, to neutralize any antibodies that had the potential to cross-react with Cs, the diluted antibody preparation used for immunohistochemistry (see below) was incubated overnight at 4°C with 10 nmol/ml of the peptide ESVKEFLAKC (Research Genetics, Huntsville, AL) before use [11].
|
Western Blotting and Quantitation of Cs
Mouse brain and testis extracts were prepared as before [8] except that the homogenization buffers were modified to reduce proteolysis. The testis homogenization buffer was 0.08% digitonin in PBS, pH 7.4, containing the following protease inhibitors: 2 mM EDTA, 10 mM 4-aminobenzamidine, 50 µM leupeptin, 15 µM pepstatin, 200 µM TPCK, 1 mM PMSF, 1 µg/ml aprotinin, 100 µM TLCK. Brain homogenization buffer had the same components except that 4-aminobenzamidine was at 2.5 mM. Ovine and murine sperm were collected and washed as previously described [12, 13]. The sperm concentrations were determined using a hemacytometer. Protein concentrations were determined using the Coomassie protein assay reagent (Pierce, Rockford, IL) with BSA as protein standard. Coomassie blue dye binding is proportional to the number of positive charges on the protein [14]; because the basic amino acid compositions of BSA, oC1, and oCs are very similar, all three proteins are expected to bind similar amounts of dye per unit mass. Sperm, tissue extracts, and protein samples were run in 10% SDS-polyacrylamide gels, transferred to polyvinylidene difluoride (PVDF) membranes, and the PVDF blots probed for cross-reacting proteins as described before [8].
Quantitation of Cs in ovine sperm was carried out using the Western blot in Figure 2. The blot was exposed to film for various times (5–30 sec). The developed film was photographed (Kodak DC120 Zoom Digital Camera; Eastman Kodak, Rochester, NY) and the images saved as tif files using Adobe Photoshop 5.5 (Adobe Systems, San Jose, CA). The bands in the blot were then quantitated by summing the entire signal over the area of the bands using Image Quant software (Molecular Dynamics/Amersham Pharmacia Biotech, Piscataway, NJ). The exposure that produced a linear correlation between pixel volume (the integrated intensity of all the pixels in a band after background correction) and amount of purified oCs was used to determine the amount of Cs present in ovine sperm.
|
Immunohistochemistry
Fixation of mouse testes, antigen retrieval in testis sections, labeling with affinity-purified antibody, and detection of label with a biotin-streptavidin-alkaline phosphatase amplification system were as described [8]. The anti-murine Cs (mCs) antibody (0.83 mg/ml stock) and the anti-C1 antibody (1 mg/ml stock) were both used at 1:2000 dilution. BCIP/NBT/INT was used to visualize mCs, and basic fuchsin was used to detect murine C1 (mC1). Sections were counterstained with Harris hematoxylin (5 min) and mounted on glass slides with an aqueous-based mountant (Glycergel; DAKO).
Immunofluorescence Microscopy
Ovine sperm Ejaculated ovine sperm were collected and washed as previously described [12]. The loose pellet of sperm was carefully overlayed with 4 ml of wash buffer, and the sperm were allowed to swim up for 1–2 h. The motile sperm were then collected, transferred to 1.5-ml tubes, centrifuged (6000 rpm, 5 min in an IEC microfuge), and resuspended in PBS. The sperm concentration was adjusted to 1.2–2.4 x 107/ml, after which an equal volume of 2x fixing solution (0.1 M sodium phosphate, pH 7.4, 4% formaldehyde) was added. Fixation was for 2 h at 4°C. Intact, fixed sperm were permeabilized for 30 min by adding an equal volume of the fixer that contained 0.2% Triton X-100 (Amersham Pharmacia Biotech, Piscataway, NJ). About 50 µl of the suspension of fixed and permeabilized sperm (3–6 x 106/ml) were applied to each coverslip that previously had been coated with 0.1% polyethyleneimine. The coverslips were placed in individual humidors, and the sperm were allowed to settle for 10 min, after which the remaining fluid was blotted off the coverslips. The coverslips next were rinsed three times with 250 µl PBS and then incubated with 250 µl blocking buffer (PBS, 5% BSA, fatty acid free [Calbiochem, San Diego, CA], 1% cold water fish gelatin [Sigma], 10% normal goat serum [Sigma]) for 30 min at room temperature. This was followed by an overnight incubation at 4°C with anti-oCs antibody diluted 1:10 000 with 1/5 blocking buffer [PBS, 1% BSA, 0.2% cold water fish gelatin, 2% normal goat serum]. After three washes with 250 µl 1/5 blocking buffer, the cover slips were incubated for 1 h at room temperature with 250 µl of a 1:500 dilution of the secondary antibody (goat anti-rabbit IgG conjugated to Oregon Green 500 [Molecular Probes, Eugene, OR]). The coverslips were washed twice with 250 µl of 1/5 blocking buffer, then with 250 µl PBS. Excess PBS was blotted off, and the coverslips were mounted in 20 µl of antifade reagent (Prolong Antifade Kit; Molecular Probes).
Murine sperm Mice were killed by cervical dislocation, the cauda epididymis rapidly removed, and the sperm gently extruded from the epididymis into PBS. Sperm that had been stripped of their mitochondria, and sperm from which the axonemal microtubules and associated outer dense fibers had been partially extruded from the fibrous sheath, were prepared according to Si and Okuno [4]. The sperm were fixed as described above, except that the permeabilization step was carried out only with the intact sperm. About 100 µl of fixed murine sperm (5–10 x 106/ml) were applied to each polyethyleneimine-coated coverslip in a humidor and allowed to settle for 10 min. Excess solution was blotted off, and the coverslips were washed three times with 250 µl PBS. This was followed by a 30-min incubation in blocking solution (10 mM sodium phosphate, pH 7.4, 150 mM NaCl, 5% fatty acid-free BSA, 1% cold water fish gelatin, 10% nonimmune swine serum [DAKO]). The blocking solution was drained and replaced with 250 µl of anti-mCs antibody diluted 1:5000 in 1/5 blocking solution, and the coverslips incubated overnight at 4°C. As a control, the primary antibody was preincubated with the peptide antigen (acetyl-ASSSNDVK, 10 nmol peptide/ml diluted antibody). The coverslips were then brought to room temperature, the primary antibody was drained, and the coverslips washed three times with 250 µl TBST (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 0.1% Tween-20). This was followed by a 30-min incubation with 250 µl linker (biotin-conjugated swine anti-rabbit IgG [DAKO]) diluted 1:300 with linker diluent (50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1% fatty acid-free BSA, 10% nonimmune mouse serum). The coverslips were washed three times with TBST and then incubated for 30 min with 250 µl Alexa Fluor 488-conjugated streptavidin (1 mg/ml stock, Molecular Probes) diluted 1:300 with TBS, 0.5% fatty acid-free BSA. After washing a final time with TBST, the coverslips were mounted in 20 µl of Prolong antifade reagent.
Confocal microscopy Images were captured using a Leica DM IRBE confocal microscope (Heidelberg, Germany) and were digitized using the TCS software (Lasertechnik GmbH, Heidelberg, Germany).
RESULTS
Production and Specificity of Anti-C Antibodies
Anti-mCs antibody
We previously reported that an antibody prepared against a peptide based on the unique amino terminus of mCs reacted specifically with mCs, did not cross-react with mC1, and recognized a single band in murine testis and epididymal sperm [8] (see Fig. 1A). This antibody is ideal for immunohistochemistry of mouse testis, and immunofluorescence microscopy of mouse sperm (see below), but the antibody does not cross-react with oCs (Fig. 1A, lanes 4 and 5). This lack of cross-reactivity is not surprising, because the unique amino terminus of mCs (Ac-ASSSND) differs from that of oCs (Ac-ASNPND) in two out of six residues. Because the sequence targeted by the antibody is short and possesses minimal secondary structure, the difference between mCs and oCs apparently is sufficient to alter the epitopes recognized by the antibody.
Anti-oCs antibody
It was desirable to have an anti-oCs antibody for quantitation of the amount of Cs in sperm and also for use in immunofluorescence microscopic studies on the location of Cs in ovine sperm. Therefore, a peptide (Ac-ASNPNDVK) based on the oCs amino terminus was synthesized and used to generate a rabbit anti-peptide antibody. The affinity-purified antibody reacted strongly with oCs isolated from ram sperm flagella (Fig. 1B, lane 2) and with a single 40-kDa protein in whole ram sperm (Fig. 1B, lane 1). The antibody was highly specific for Cs and did not cross-react with oC1 (Fig. 1B, lane 3). The antibody also did not cross-react with any protein in murine sperm (Fig. 1B, lane 4), indicating that it, like the anti-mCs antibody, is species specific.
Anti-C1 antibody
The commercially obtained C1 antibody was raised against a peptide based on the consensus residues 7–21 (KKGSEQESVKEFLAK) of mammalian C1 [7, 15–18]. This antibody recognized recombinant mouse C1 (Fig. 1C, lane 5) and oC1 purified from skeletal muscle (Fig. 1C, lane 4). It reacted very strongly with a 40-kDa band in mouse brain extract (Fig. 1C, lane 7). This band is likely to be composed of C1 and Cß, both of which are present in brain and would be expected to react with this antibody. In mouse testis extract, the antibody recognized two proteins, with apparent masses of 40 and 30 kDa (Fig. 1C, lane 6). The 40-kDa protein is undoubtedly intact C1; the 30-kDa band may represent a natural product of the degradation of C1 in spermatogenic cells (see Discussion). It is unlikely to be an artifact arising from proteolysis during the extraction, because protease inhibitors were included in the extraction buffer, and Cs apparently did not undergo any proteolysis even though it is present in the same extract (cf. Fig. 1A, lane 3). The antibody did not cross-react with purified oCs (Fig. 1C, lane 3), or with any protein in mouse sperm or ram sperm (Fig. 1C, lanes 1 and 2, respectively) that contain Cs but not C1 [7]. Therefore, the antibody is highly specific for C1 and does not recognize Cs.
Amount of Cs in Sperm
The availability of the anti-oCs antibody together with highly purified oCs [7] provided us the opportunity to use quantitative Western blotting to estimate the amount of Cs in ovine sperm. A Western blot containing various amounts of purified oCs was probed with the antibody under conditions where the intensities of the resulting bands were linearly proportional to the amount of oCs in each band (Fig. 2, A and B). Intact ram sperm proteins on the same Western blot were probed identically; the number of sperm loaded was adjusted so that the intensity of the resulting band fell within the linear range established for the standards. The results for the experiment shown indicated that 0.5 x 106 sperm contained 13.6 ng of Cs, corresponding to 4.14 x 105 molecules of Cs per sperm. Other experiments using a different preparation of sperm or different concentrations of sperm from the same preparation gave similar results. The average value for five determinations using sperm from three different preparations was 4.15 (±0.48 SD) x 105 molecules of Cs per sperm.
Distribution of Cs in Testicular Cells
We reported earlier that the mCs protein was detected only in spermatogenic cells, specifically in midpachytene spermatocytes, in round and elongate spermatids, and in sperm that have been released to the lumen of the seminiferous tubules [8]. To obtain a more definitive picture of mCs expression in the testis, all stages of spermatogenesis were analyzed for immunostaining by the anti-mCs antibody. Figure 3 shows micrographs of sections of mouse seminiferous tubules at spermatogenic stages I–XII [19] after labeling with anti-mCs antibody. Figure 6 diagrammatically depicts the meiotic and spermiogenic cells present in each of the stages. Comparison of stages V and VI in Figure 3 shows that Cs was first detectable as a faintly staining cytosolic component of midpachytene spermatocytes at stage VI. Staining was still faint in pachytene spermatocytes in stage VII, but by stage VIII, Cs staining in the cytosol of pachytene spermatocytes had increased considerably. By stage X, Cs staining in the cytosol of late pachytene spermatocytes appeared to be maximal. Cs remained in the cytosol of diplotene spermatocytes and secondary spermatocytes as they passed through the first and second meiotic divisions, respectively (m, stage XII). After the second meiotic division, the newly formed round spermatids (spermiogenesis step 1, stage I) retained Cs in their cytoplasm. As the round spermatids elongated and developed flagellar structures (spermiogenesis steps 9–16), Cs was recruited from the cytoplasm to the flagella (see, e.g., stage V, inset, arrowheads).
|
|
|
As previously observed [8], the anti-mCs antibody did not stain somatic cells, including interstitial cells, peritubular cells, or Sertoli cells, portions of which were clearly evident as radial streaks of unstained cytoplasm standing out in contrast to the more darkly stained spermatogenic cells. The antibody also did not stain relatively undifferentiated germ cells, including spermatogonia and preleptotene, leptotene, zygotene, or stage I–IV pachytene spermatocytes.
The staining pattern obtained with the anti-Cs antibody made it much easier to distinguish spermatogenic cells from each other and from Sertoli cells. With the exception of stages VIII and IX, the spermatogenic cell generation exhibiting highest intensity staining (see circled generation above each micrograph in Fig. 3) occupied a region of the seminiferous tubule flanked by two other generations of spermatogenic cells that were either lightly stained (midpachytene spermatocytes and elongate spermatids) or not stained at all. In stage VIII, pachytene spermatocytes and round spermatids were stained with about the same degree of intensity, but the former are readily recognized by the larger size of their nuclei. Similarly in stage IX, the elongating nuclei of the spermatids easily set them apart from the equally stained pachytene spermatocytes. As a result, staining with the anti-Cs antibody greatly facilitates staging of seminiferous tubules.
Distribution of C1 in Testicular Cells
The pattern of staining obtained with the anti-C1 antibody was almost completely complementary to that obtained with the anti-Cs antibody. In contrast to the anti-Cs antibody, the anti-C1 antibody stained somatic testicular cells, especially Sertoli cells, and undifferentiated germ cells, specifically spermatogonia and preleptotene spermatocytes. However, the antibody did not stain cells in leptotene or later stages of spermatogenesis.
At low magnification, the antibody labeling resulted in a pattern of punctate foci around the periphery of the seminiferous tubule, as well as spoke-like regions extending from the periphery of the tubule in toward the lumen (see tubules with white stars in Fig. 4A). Control sections in which the primary antibody was omitted did not show any staining (Fig. 4B).
|
At higher magnification, it was apparent that the darkly stained spoke-like regions were Sertoli cells (S, insets of stages VI, VII, and VIII, Fig. 5). Regions of intensely stained cytoplasm could be seen surrounding the Sertoli cell nuclei (insets of stages VI, VIII, Fig. 5). Interstitial cells were very faintly stained, but peritubular cells (white asterisk, insets, stages VIII and XII, Fig. 5) did not appear to be stained at all. The punctate labeling seen at the periphery of the seminiferous tubule was due mainly to C1 staining of the cytosol of the spermatogonia that line the base of the seminiferous epithelium (B, In, and Sp, insets of stages II–III, VI and XII, Fig. 5). C1 labeling was still present in preleptotene spermatocytes (Pl, inset of stage VII, Fig. 5) but diminished in late preleptotene spermatocytes (Pl, inset of stage VIII, Fig. 5). By the time the spermatocytes had entered leptotene (L, inset of stage IX, Fig. 5), C1 labeling was no longer detectable. Zygotene (Z, stages XI and XII, Fig. 5), pachytene (P, stages II–III, VI, VII, VIII and IX, Fig. 5), diplotene, and secondary spermatocytes (Di and II, respectively, stages XI and XII, Fig. 5) were not labeled. Similarly, no staining was seen in round spermatids (spermiogenesis steps 2–8), elongate spermatids (spermiogenesis steps 9–16), and testicular sperm occupying the lumen of the seminiferous tubule (spermiogenesis step 16). Thus, the C1 protein is expressed only in somatic cells, spermatogonia, and preleptotene spermatocytes. The results of the Cs and C1 immunolabeling experiments in mouse testis are summarized in Table 1 and Figure 6.
|
|
Location of Cs in Mature Sperm
To determine the location of Cs in the mature sperm, ejaculated ovine and epididymal murine sperm were examined by immunofluorescence confocal microscopy following labeling with anti-oCs or anti-mCs antibodies, respectively (Fig. 7). In both cases, staining was almost exclusively in the tail (Fig. 7, A and B); little if any Cs was detected in the sperm heads. The most intense immunofluorescence occurred in the midpiece of the tail. The remainder of the tail also was labeled, with the intensity of immunofluorescence increasing from the proximal to the distal region of the principal piece, and increasing still more in the end piece. When the primary antibody was preincubated with the peptide antigen, no immunofluorescence staining was observed (Fig. 7B, 5).
|
The very bright staining of the midpiece suggested that some of the Cs is associated with the midpiece mitochondria. To test this, the midpiece mitochondria of murine sperm were removed by treatment with Triton X-100 and dithiothreitol [4]. The extraction eliminated much of the midpiece immunofluorescence (Fig. 7C), consistent with at least some of the midpiece Cs being associated with the mitochondria. However, the residual immunofluorescence in the midpiece was still brighter than that in the adjoining part of the principal piece and about equal to that of the end piece. Following the extraction procedure, slight immunofluorescence also was detected in the head.
The end piece of the sperm flagellum contains the 9+2 axoneme but no accessory structures. Therefore, the strong immunofluorescence of the end piece, even after demembranation (Fig. 7C, 7), indicates that Cs also is associated with the axoneme.
The immunofluorescence of the principal piece also is likely to have been due to axonemal Cs, access to which was partially restricted by the fibrous sheath. The intensity of immunofluorescence in the principal piece increased with increasing distance from the annulus, in inverse relationship to the size of the fibrous sheath. Similarly, the residual immunofluorescence of the midpiece after mitochondrial removal may have been due to axonemal Cs; inasmuch as the fibrous sheath does not extend into the midpiece, access of the antibody to Cs in this region would not have been restricted.
To investigate these possibilities, demembranated murine sperm were treated with ATP to induce sliding of the outer doublet microtubules (and their associated outer dense fibers) out of the fibrous sheath [4]. When the partially disintegrated sperm were examined by immunofluorescence microscopy, the extruded doublet microtubules exhibited a relatively uniform fluorescence that was brighter than that of the adjacent portion of the principal piece (Fig. 7D, 11). In some sperm, multiple fibers (presumably representing individual doublets or small groups of axonemal microtubules) could be detected by both differential interference contrast and immunofluorescence microscopy (Fig. 7D, 13 and 14). These experiments provide strong evidence that Cs is associated with the outer doublet microtubules throughout the length of the tail, but that access of the antibody to the axonemal Cs was partially blocked by the fibrous sheath in the principal piece.
DISCUSSION
Expression of Cs and C1 Are Mutually Exclusive
In this paper we confirm our earlier observation that Cs is expressed only in meiotic spermatogenic cells, initially appearing in midpachytene spermatocytes [8]. The protein is readily detected in the cytoplasm of spermatocytes throughout the first and second meiotic divisions, eventually becoming localized to the flagellum of the elongating spermatid. In contrast, C1 is expressed only in somatic cells, spermatogonia, and preleptotene spermatocytes. A major finding of the current work is that C1 disappears from the spermatogenic cells coincident with the onset of meiosis. Therefore, there is no overlap in the expression of Cs and C1 in testicular cells. One consequence of this mutually exclusive expression is that the mechanism by which C is targeted to the nascent flagellum need not distinguish between the Cs and C1 isoforms. It also is apparent that it must have been Cs mRNA and protein that was being detected in previous studies reporting C mRNA [20–23] and C protein [23] in pachytene spermatocytes.
A feature shared in common by those testicular cells that express C1 is that they are in direct contact with the interstitial fluid and presumably require a functional cAMP-signaling pathway for proper response to extracellular stimuli and agonists in the blood. The PKA-mediated responses in these cells clearly must be mediated by C1. However, following differentiation of the spermatogonia into preleptotene spermatocytes, the latter move into the adluminal compartment of the seminiferous tubule, where they are protected by the blood-testis barrier provided by the Sertoli cells. During this movement, or soon thereafter, C1 disappears from the cytoplasm of the spermatocytes, probably through a process of programmed degradation. Indeed, the 30-kDa protein detected by the anti-C1 antibody in Western blots of murine testis extract is likely to be a natural product of this degradation process. A relatively long period of meiotic prophase ensues as the diploid primary spermatocytes develop from preleptotene to pachytene, and in this period no C isoform is present in the germ cells. Cs is first expressed in midpachytene cells, long after C1 has disappeared from the precursor preleptotene spermatocytes. Apparently a functional PKA signaling pathway is not needed by cells undergoing early meiotic prophase, either because PKA-dependent processes do not contribute to the development of the cells during this period, or because these processes are mediated by the Sertoli cells as part of the housekeeping chores that they assume.
The absence of C1 in spermatogenic cells may be related to an unusual feature of cAMP-responsive element modulator (CREM)-mediated transcriptional activation in male germ cells. CREM is a widely expressed cAMP-responsive transcription factor that binds to and activates transcription of genes containing cAMP-responsive elements (CREs) [24]. A number of these genes encode sperm structural proteins and are transcribed postmeiotically in the testis [25, 26]. CREM is essential for the expression of these genes, and disruption of the CREM gene in the mouse blocks spermiogenesis at an early step [27, 28]. Transcriptional activation by CREM requires that CREM also bind to a protein cofactor, an interaction that in somatic cells occurs only after CREM has been phosphorylated by PKA. However, CREM in the testis is not phosphorylated, so that CREM-mediated transcriptional activation in the testis is PKA independent [24]. Recently, it was shown that, in the testis, CREM is activated by a testis-specific coactivator termed ACT (for activator of CREM in testis) that can bind to the nonphosphorylated form of CREM [29]. The development of this mechanism for bypassing the requirement for PKA activation of CREM may be an evolutionary adaptation to the absence of C1 in spermatogenic cells, or conversely, may have allowed C1 to be dispensed within these cells.
Cs Is Present in the Midpiece Mitochondria and Axoneme of Mature Sperm
Immunofluorescence microscopy of ovine and murine sperm indicated that a significant amount of the Cs in the sperm flagellum is in the midpiece, apparently associated with the midpiece mitochondria. This finding is in partial agreement with the recent immunofluorescence microscopic results of Reinton et al. [10] that Cs in human sperm is in the midpiece region. However, these authors concluded that Cs was restricted to the midpiece, a conclusion at variance with our results that Cs is located throughout the tail (see below). Because the immunofluorescence of the midpiece is so intense, it is possible that Reinton et al. overlooked less intense staining in other parts of the sperm tail. Localization of Cs to the midpiece also is consistent with a previous immunofluorescence microscopic study showing strong staining of the midpiece by antibodies to C [30], and with immunoelectron microscopic localization of C to the outer membrane of the midpiece mitochondria in rat sperm [31] and to the space between the mitochondria and the dense fibers in ram sperm [21]. Similarly, both immunoflourescence microscopic [30, 32] and immunoelectron microscopic [21, 31, 33] studies have detected RI and RII in the midpiece. Recently, two A-kinase-anchoring proteins, hAKAP220 and S-AKAP84, have been localized to the midpiece of mature sperm [32, 34]. Thus, the sperm midpiece appears to have all of the components needed for a functional PKA signaling system. PKA in the midpiece may have a role in the regulation of respiratory function [35, 36], in the control of mitochondrial uptake of intracellular Ca2+ [37], and/or in the direct control of sperm motility.
In the current study, the anti-Cs antibodies also labeled the principal piece, where the intensity of immunofluorescence increased from proximal to distal, and the end piece, where the intensity was greater still. Inasmuch as the end piece contains only the axoneme surrounded by the membrane, staining of this region both before and after demembranation indicates that some Cs must be associated with the axoneme. The pattern of staining of the principal piece also may be due to axoneme-associated Cs, access to which was partially blocked by the fibrous sheath (see below). This possibility is strongly supported by our finding that individual outer doublet microtubules (or small groups of doublets) that were extruded from the anterior region of the principal piece fluoresced at least as intensely as the principal piece fluoresced before extrusion of the doublets. Although the outer dense fibers are likely to have been extruded along with the outer doublet microtubules [38], there is no evidence that the former structures were labeled by the antibody. The intensity of immunofluorescence along the principal piece was inversely proportional to the size of the outer dense fibers, whereas the opposite would have been expected if the outer dense fibers had contributed to the immunofluorescence pattern.
Previous localization studies have not been in agreement on whether PKA is located in the axoneme. As discussed above, Cs was not detected in the principal or end pieces of human sperm by Reinton et al. [10]. On the other hand, Vijayaraghavan et al. [39], using an anti-RII antibody, reported an immunofluorescence staining of the principal and end pieces of bovine sperm that was similar to that which we observed; these authors concluded that RII was located predominantly in the axoneme. One immunoelectron microscopic study reported a noticeable labeling of the ram sperm axoneme with both anti-C and anti-RII antibodies [21], whereas another reported only sparse labeling of the rat sperm axoneme using an anti-RII antibody [31]. In view of the current findings, it is likely that this sparse labeling was indeed due to the presence of PKA in the axoneme.
The axonemal substructure with which Cs is associated has not yet been determined. However, in Chlamydomonas, the radial spoke protein RSP3 is an AKAP [40]. Moreover, we have recently determined that one of the substrates phosphorylated in a cAMP-dependent manner concomitant with the development of motility in ram sperm is related to Chlamydomonas radial spoke proteins RSP4 and RSP6 (unpublished results) that are located in the radial spoke head [41]. The radial spokes are involved in the control of flagellar movement [42, 43], so association of Cs with the radial spokes would put it in an excellent position to regulate axonemal motility.
It should be noted that the pattern of immunofluorescence observed with our anti-Cs antibodies did not provide evidence for the location of Cs in the fibrous sheath. In contrast, immunoelectron microscopic studies have consistently detected C and RII in the fibrous sheath [21, 31, 33]. Moreover, several AKAPs have been shown to be associated with the fibrous sheath [44–50], although it has not been demonstrated that these AKAPs bind PKA in vivo. If Cs is present in the fibrous sheath, it is possible that our antibody simply did not have access to it due to the physical structure of the fibrous sheath. This presumably would not have been a problem in the immunoelectron microscopic studies, where Cs inside the fibrous sheath would have been exposed by thin sectioning.
In the intact mouse and ram sperm, little if any labeling of the head by the anti-Cs antibodies was observed, consistent with our previous biochemical data that the vast majority of ram sperm Cs in present in the tail [7]. Following demembranation of the murine sperm, a low level of immunofluorescence was observed in the head. This could have been due to unmasking of a relatively small number of epitopes in the head by the demembranation procedure or to artifactual redistribution of Cs to the head following removal of the midpiece mitochondria.
Cs Is Highly Abundant in Sperm
An intriguing finding of this study is that each ram sperm contains 4 x 105 molecules of Cs. In an earlier study, Horowitz et al. [51] measured 0.3 pmol of [3H]cAMP bound to 106 rat sperm; from this value it can be calculated that each rat sperm contains at least 1.8 x 105 R subunits. These values, obtained by completely independent methods, are in reasonably good agreement. The value that we obtained clearly indicates that Cs is much more abundant in sperm than are major axonemal structures such as the dynein arms and radial spokes. The outer dynein arms repeat at 24-nm intervals [52] on eight of the nine outer doublet microtubules [53]; thus, a 50-µm long ram sperm flagellum contains 16 000 outer dynein arms. The radial spokes occur in triplets that repeat every 96 nm along each doublet microtubule [52], so a 50-µm long flagellum contains 13 500 radial spokes. Thus, the number of Cs molecules in the sperm far exceeds that of the motility-generating and motility-regulating proteins that are likely to be important targets of the PKA signal transduction pathway. Much of the Cs in sperm may be associated with the midpiece mitochondria or the fibrous sheath (see above). Nevertheless, the amount of Cs present in sperm seems high for a protein expected to act catalytically. One possible explanation is that because activated Cs appears to be tethered to sperm tail structures in a cAMP-insensitive manner [6, 7, 10], it cannot diffuse freely within the sperm tail following its release from R. Similarly, the Cs substrates are likely to be structural components of the tail [6] and therefore unable to diffuse to the tethered Cs. As a result, each Cs may have access to a relatively small number of substrate molecules, and need to be present in a relatively high stoichiometry (perhaps approaching 1:1) compared to its substrates.
ACKNOWLEDGMENTS
The authors thank Dr. Jeffrey Nickerson for his assistance with the confocal microscopy, Kevin Hall for valuable instruction and help in the preparation of testis sections, and Dr. Lindsay Gillan for assistance in the collection of mouse epididymal sperm. We also are grateful to Dr. Susan Taylor for her gift of recombinant mC1.
FOOTNOTES
First decision: 4 December 2000.
1 Supported by National Institutes of Health grant HD23858 and by the Robert W. Booth Fund at the Greater Worcester Community Foundation. 
2 Correspondence: George Witman, Department of Cell Biology, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655. FAX: 508 856 5612; george.witman{at}umassmed.edu
Accepted: February 16, 2001.
Received: October 10, 2000.
REFERENCES
mRNA expressed specifically in spermatogenic cells. Mol Biol Cell 2000; 11:3031-3044-s, localizes to sperm midpiece. Biol Reprod 2000; 63:607-11. Nucleic Acids Res 1988; 16:8189-8190 catalytic subunit of the cAMP-dependent protein kinase: cDNA sequence and evidence that it is the only isoform expressed in myoblasts. Biochim Biophys Acta 1991; 1089:254-256[Medline] catalytic subunit of the bovine cAMP-dependent protein kinase. Biochim Biophys Acta 1992; 1171:93-96[Medline] regulatory subunit delocalizes protein kinase A in mouse sperm without affecting motility or fertilization. J Biol Chem 1999; 274:24131-24136 of cAMP-dependent protein kinase in human sperm: biochemical and cytochemical study. Mol Reprod Dev 1994; 39:415-422[CrossRef][Medline]This article has been cited by other articles:
|
|
 |
|
  K. A. Burton and G. S. McKnight PKA, Germ Cells, and Fertility Physiology, February 1, 2007; 22(1): 40 - 46. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Gur and H. Breitbart Mammalian sperm translate nuclear-encoded proteins by mitochondrial-type ribosomes Genes & Dev., February 15, 2006; 20(4): 411 - 416. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kotaja, B. Macho, and P. Sassone-Corsi Microtubule-independent and Protein Kinase A-mediated Function of Kinesin KIF17b Controls the Intracellular Transport of Activator of CREM in Testis (ACT) J. Biol. Chem., September 9, 2005; 280(36): 31739 - 31745. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Rajkovic, R. Middendorff, M. G. Wetzel, D. Frkovic, S. Damerow, H. J. Seitz, and J. M. Weitzel Germ Cell Nuclear Factor Relieves cAMP-response Element Modulator {tau}-mediated Activation of the Testis-specific Promoter of Human Mitochondrial Glycerol-3-phosphate Dehydrogenase J. Biol. Chem., December 10, 2004; 279(50): 52493 - 52499. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. A. Nolan, D. F. Babcock, G. Wennemuth, W. Brown, K. A. Burton, and G. S. McKnight Sperm-specific protein kinase A catalytic subunit C{alpha}2 orchestrates cAMP signaling for male fertility PNAS, September 14, 2004; 101(37): 13483 - 13488. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Jurczyk, A. Gromley, S. Redick, J. S. Agustin, G. Witman, G. J. Pazour, D. J.M. Peters, and S. Doxsey Pericentrin forms a complex with intraflagellar transport proteins and polycystin-2 and is required for primary cilia assembly J. Cell Biol., August 30, 2004; 166(5): 637 - 643. [Abstract] [Full Text] [PDF] |
