The mechanism responsible for spermatid translocation in the mammalian seminiferous epithelium was proposed to be the microtubule-based transport of specialized junction plaques (ectoplasmic specializations) that occur in Sertoli cell regions attached to spermatid heads. These plaques each consist of a cistern of endoplasmic reticulum, a layer of actin filaments and the adjacent plasma membrane. It is predicted that motor proteins function to move the junction plaques, and hence the attached spermatids, first towards the base and then back to the apex of the epithelium, along microtubules. If this hypothesis is true, motor proteins should be associated with the cytoplasmic face of the endoplasmic reticulum component of ectoplasmic specializations. In addition, isolated junction plaques should support microtubule movement both in the plus and minus directions to account for the bidirectional translocation of spermatids in vivo. To determine if cytoplasmic dynein is localized to the endoplasmic reticulum of the plaques, perfusion-fixed rat testes were immunologically probed, at the ultrastructural level, for the intermediate chain of cytoplasmic dynein (IC74). In addition, testicular fractions enriched for spermatid/junction complexes were incubated with and without gelsolin, centrifuged and the supernatants compared, by western blot analysis, for Glucose Regulated Protein 94 (a marker for endoplasmic reticulum) and IC74. At the ultrastructural level, the probe for IC74 clearly labelled material associated with the cytoplasmic face of the endoplasmic reticulum component of the junction plaques. In the gelsolin experiments, both probes reacted more strongly with appropriate bands from the gelsolin-treated supernatants than with corresponding bands from controls. To determine if the junction plaques support microtubule transport in both directions, polarity-labelled microtubules were bound to isolated spermatid/junction complexes and then assessed for motility in the presence of ATP and testicular cytosol (2 mg/ml). Of 25 recorded motility events, 17 were in a direction consistent with a plus-end directed motor being present, and 8 were in the minus-end direction. The results are consistent with the conclusion that the junction plaques have the potential for moving along microtubules in both the plus and minus directions and that both a kinesin-type and a dynein-type motor may be associated with the junction plaques. The data also indicate that cytoplasmic dynein is localized to the cytoplasmic face of the endoplasmic reticulum component of the plaques.

As developing sperm cells (spermatids) acquire an elongate shape during spermatogenesis, they become situated in invaginations in the apices of Sertoli cells. Initially, these crypts are shallow and spermatids are positioned in apical regions of the epithelium. As spermatogenesis continues, the crypts deepen. This deepening results in the repositioning of spermatid heads closer to the base of the epithelium. Still later, the crypts again become shallow and the spermatids are moved to the apex of the epithelium where they are eventually released. Although the biological significance of this ‘down and up’ translocation is not known, the mechanism responsible for the movement was proposed to be the microtubule-based transport of unique intercellular adhesion junctions (ectoplasmic specializations) that develop in Sertoli cell regions associated with the crypts (Vogl, 1988; Redenbach and Vogl, 1991).

Ectoplasmic specializations, which develop in association with apical crypts, consist of the Sertoli cell plasma membrane in regions attached to spermatid heads, a layer of actin filaments and an attached cistern of endoplasmic reticulum (see Fig. 1A). The three components of the junction plaques are structurally very robust and together remain firmly adherent to spermatids when the latter cells are mechanically dissociated from the epithelium (Romrell and Ross, 1979; Vogl, 1996). A significant consequence of these features is that mechanical dissociation of spermatids from the epithelium can be used as a means of isolating intact junctions for use in experimental systems, and that the spermatid head can be used as a morphological marker for locating the junction at the light microscopic level.

Fig. 1.

Basic architecture of ectoplasmic specializations, and the spermatid translocation model. (A) A cross section through an elongate spermatid head and related Sertoli cell regions in the rat. The ectoplasmic specialization consists of the plasma membrane of the Sertoli cell in regions of attachment to the spermatid head, a layer of actin filaments and an attached cistern of endoplasmic reticulum (ER). (B) The spermatid translocation model. It is proposed that microtubule-dependent motor proteins are anchored to the endoplasmic reticulum of ectoplasmic specializations, and move the junction plaques and hence the attached spermatids along adjacent microtubule tracts. (C) The close relationship between ectoplasmic specializations and microtubules (mts) is shown in the electron micrograph. Bars, 0.25 μm.

Fig. 1.

Basic architecture of ectoplasmic specializations, and the spermatid translocation model. (A) A cross section through an elongate spermatid head and related Sertoli cell regions in the rat. The ectoplasmic specialization consists of the plasma membrane of the Sertoli cell in regions of attachment to the spermatid head, a layer of actin filaments and an attached cistern of endoplasmic reticulum (ER). (B) The spermatid translocation model. It is proposed that microtubule-dependent motor proteins are anchored to the endoplasmic reticulum of ectoplasmic specializations, and move the junction plaques and hence the attached spermatids along adjacent microtubule tracts. (C) The close relationship between ectoplasmic specializations and microtubules (mts) is shown in the electron micrograph. Bars, 0.25 μm.

Our general working hypothesis of spermatid translocation is that motor proteins move the junction plaques, and hence the attached spermatids, along adjacent microtubule tracts (see Fig. 1B). A number of observations are consistent with this model. First, microtubules are abundant in apical regions of Sertoli cells (Christensen, 1965; Fawcett, 1975), oriented parallel to the movement of spermatids (Vogl, 1988; Amlani and Vogl, 1988; Vogl et al, 1995), and are closely related to ectoplasmic specializations attached to the adjacent spermatid heads (Russell, 1977) (see Fig. 1C). Interestingly, the majority of these microtubules have their plus ends positioned near the nucleus, situated basally, and their minus ends at the apex of the cell (Redenbach and Vogl, 1991). A second observation is that more microtubules bind to isolated junction plaques in the absence of ATP than when nucleotide is present (Vogl, 1996); a result consistent with the possibility that microtubules cycle off of motors on the plaque in the presence of nucleotide. Moreover, the microtubules attach to the endoplasmic reticulum component of the plaques in these binding assays (Vogl, 1996). An important, and recent, observation is that isolated spermatid/junction complexes support microtubule transport in the presence of ATP (Beach and Vogl, 1999). Also significant is the observation that antibodies to the intermediate chain of cytoplasmic dynein (IC74) react with Sertoli cell regions surrounding apical crypts (Miller et al., 1999).

If the microtubule-based spermatid translocation model is correct, then there are two major predictions that can be made about the system. First, microtubule dependent motor proteins should be anchored to the cytoplasmic face of the endoplasmic reticulum. Second, microtubule transport of the junction plaques should occur bidirectionally to account for the ‘down and up’ movement that actually occurs in the epithelium; in other words, there should be both plus-end and minus-end directed motors associated with the junction plaques. In this study, we use two approaches to determine if cytoplasmic dynein, one of the motors predicted to be involved with the system, is associated with the endoplasmic reticulum component of the junction plaques. The first approach is to use immunoelectron microscopy to determine the ultrastructural localization of IC74 at the junction plaques. The second approach is to enzymatically disassemble the junction plaques attached to isolated spermatids, remove the spermatids by centrifugation, and probe the supernatants for the intermediate chain of cytoplasmic dynein (IC74) and Glucose Regulated Protein 94 (GRP94, a marker for ER). If our hypothesis is correct, then the levels of both these proteins should be greater in supernatants collected from gelsolin-treated material than in supernatants collected from controls. To verify the prediction that isolated junction plaques support microtubule transport both in the plus- and in the minus-end directions, we have developed and used an in vitro polarity-marked microtubule motility assay. Collectively, the data are consistent with the conclusion that both a dynein and a plus-end motor, likely a kinesin, are associated with the junction plaque.

Chemicals and reagents

Unless otherwise indicated, most reagents used in the study were from Sigma Chemical Co. (St Louis, MO, USA). Rhodamine-tubulin and unlabelled tubulin were obtained from Cytoskeleton (Denver, CO, USA), as were the motility chambers. The fluorescent phallotoxins and DiOC6 were obtained from Molecular Probes (Eugene, OR, USA). Paraformaldehyde was obtained from Fisher Scientific (Suwanee, GA, USA) and the gluteraldehyde, sodium cacodylate, OsO4, Epon 812, Lowicryl K4M and Unicryl were obtained from J.B. EM Services (Dorval, PQ, USA).

Antibodies were obtained from the following sources. The monoclonal antibody to IC74 was obtained both commercially from Covance (Richmond, CA, USA) and as a gift from Dr Pfister (University of Virginia Health Sciences Center, Charlottesville, USA) (Dillman and Pfister, 1994). The monoclonal antibody to glucose-regulated protein (GRP94) (endoplasmin) was obtained from Stressgen (Vancouver, BC, Canada). Secondary antibodies conjugated either to 10 nm gold, for ultrastructural localization, or to horseradish peroxidase, for immunoblots, were obtained from Jackson Immunoresearch Laboratories Inc. (West Grove, PA, USA) and Santa Cruz Biotechnology Inc. (Santa Cruz, CA, USA), respectively.

Animals

All animals used in the study were reproductively active Sprague-Dawley rats. They were obtained from a colony in the Animal Care Facility at the University of British Columbia and were maintained and used in accordance with guidelines established by the Canadian Council on Animal Care.

Immunoelectron microscopy (IC74)

To localize IC74 at the ultrastructural level, perfusion-fixed tissue was labelled with antibody either before or after being embedded in resin.

Pre-embedded labelling

For pre-embedded labelling, testes were perfused briefly with phosphate buffered saline (PBS: 150 mM NaCl, 3.2 mM Na2HPO4, 0.8 mM KH2PO4, 5.0 mM KCl, pH 7.3) to clear the organs of blood, and then for 30 minutes with fixative (PBS containing 3.0% paraformaldehyde at 33°C). Following fixation, the testes were washed for 30 minutes by perfusion with PBS. Each testis was removed from the canula and cut into small pieces with razor blades. To mechanically fragment the material, the pieces were aspirated through an 18-gauge needle. The fragments were allowed to settle through approximately 10 ml of PBS for 5-10 minutes and the upper half of the sedimented material was collected. In addition to other components of the testis, this material contained elongate spermatids with attached ectoplasmic specializations.

The fragments were pelleted by centrifugation (setting 6 for 4 minutes on an Eppendorf table top centrifuge) and then resuspended in blocking solution (PBS, 0.1% bovine serum albumin (BSA), 5% normal goat serum (NGS), 0.05% Tween-20). After 10 minutes, the samples were incubated, either for 2 hours at room temperature or overnight at 4°C, in primary antibody, washed with PBS containing 0.1% BSA and 0.05% Tween-20, and then incubated for 2 hours at room temperature with gold-conjugated secondary antibody. The samples were washed with PBS, pelleted by centrifugation, and the pellets fixed with PBS containing 3% paraformaldehyde and 0.5% glutaraldehyde (pH 7.3). The pellets were washed with 0.1 M sodium cacodylate (pH 7.3), postfixed on ice for 1 hour in 1% OsO4 in 0.1 M sodium cacodylate (pH 7.3), washed with dH2O, and then stained en bloc in 1% uranyl acetate in dH2O. Following a final series of washes in dH2O, the samples were dehydrated through ethanol and embedded in EPON 812. Controls included replacing the primary antibody with a similar protein concentration of normal mouse IgG, replacing the primary antibody with buffer alone, and replacing both the primary and secondary antibody with buffer alone.

Post-embedded labelling

Testes to be used for post-embedded labelling were perfused first with PBS (2 minutes) and then with a fixative containing 3% paraformaldehyde and 20 mM ethylacetimidate in PBS (pH 7.3) for 30 minutes followed by perfusion with a fixative containing 3% paraformaldehyde and 0.1% glutaraldehyde in PBS (protocol modified after Geiger et al., 1981). The testes again were perfused with PBS for 30 minutes and then cut into small pieces. The pieces then were treated with 50 mM NH4Cl in PBS for 30 minutes, washed with PBS, dehydrated through methanol, and then embedded either in Lowicryl or Unicryl resin. After the blocks were polymerized at −20°C and using ultraviolet light, sections were cut on an ultramicrotome and then mounted on formvar/carbon-coated nickel grids for immunolabelling.

Sections on grids were treated first with a blocking buffer (PBS/0.01 M glycine, 0.1% BSA, 5% NGS, 0.05% Tween 20) for 10 minutes, and then with the primary antibody to IC74 followed by a secondary gold (10 nm)-conjugated antibody. Grids were washed with PBS, treated with 2.0% glutaraldehyde in PBS, washed with dH2O, and then stained with 1% uranyl acetate in dH2O. After a series of washes with dH2O the grids were air dried and then photographed using a Philips 300 electron microscope operated at 60 kV.

Controls were similar to those used with pre-embedded material.

Quantification of labelling

To quantify labelling in pre-embedded material, gold particles were counted within 0.18 μm of the cytoplasmic face of the endoplasmic reticulum of 11 spermatids. Particles also were counted in similar areas of 11 cells treated with normal IgG instead of primary antibody. A dependent sample t-test was used to determine if the difference between the two groups was statistically significant. P<0.05 was taken as statistically significant.

For post embedded material, gold particles were counted within a rectangular area (0.12 μm wide × 1.27 μm long) placed over the following four regions: (1) Sertoli cell regions bordering the cytoplasmic face of the endoplasmic reticulum of the junction plaque, (2) the related actin zone of the plaque, (3) the spermatid head immediately adjacent to the junction plaque and (4) in a random area of cytoplasm in the perinuclear zone of Sertoli cells. Similar areas were counted in sections treated with normal IgG instead of the primary antibody. A single set of areas for each of 11 cells both from experimental and from control groups were counted. Analysis of variance and Neuman-Keuls Post Hoc analysis were performed to test for significant differences amongst groups.

Gelsolin digestion of spermatid/junction complexes

If motor proteins are anchored to the endoplasmic reticulum of the junction plaque, then preparations enriched for this component of the junction plaque should contain more motors than preparations that are not enriched for plaque ER. To verify this prediction we used gelsolin to detach the ER from junction plaques attached to spermatid heads and then removed the latter cells from suspension by centrifugation. Equivalent supernatants from gelsolin-treated and control-treated samples were compared by western blot analysis for reaction with immunological probes for IC74 and GRP94 (endoplasmin – a molecular chaperone found in the endoplasmic reticulum; Sargan et al., 1986; Mazzarella et al., 1987; Csermely et al., 1995; Muresan and Arvan, 1997). To determine the status of actin filaments associated with spermatids in gelsolin-treated and control material, samples were stained with fluorescent phallotoxin and evaluated microscopically.

Isolation of epithelia, enrichment for spermatid/junction complexes and gelsolin treatment

Protocols used to obtain testicular fractions enriched for spermatid/ junction complexes and to digest actin filaments of the junction complexes are described in detail elsewhere (Miller et al., 1999). Testes from a single animal were used for each individual experiment. Each experiment was repeated at least twice and in some cases five times. In brief summary, testes were decapsulated and the seminiferous tubule masses cut into small pieces. Using a microscope with darkfield optics, epithelia were carefully isolated and collected over a period of 30-60 minutes from individual tubules. The epithelia were mechanically fragmented by aspiration through a small bore pipette and the material was loaded onto step sucrose gradients. Following centrifugation, fractions enriched for spermatid/junction complexes were collected and divided into two equal volumes of material. One sample was incubated for 90 minutes at room temperature with gelsolin and the other with an equivalent volume of buffer under the same conditions. Following incubation, a 5 μl portion was removed from each tube and labelled with fluorescent phallotoxin, for actin filaments. The remaining samples were centrifuged to remove spermatids and the supernatants collected for analysis by western blotting.

Electrophoresis and western blotting of supernatants

Equivalent volumes of supernatant from gelsolin-treated and control samples were loaded onto either mini (10×8.5 cm) or standard (16×20 cm) gels (6.7% acrylamide running gel) and separated by electrophoresis (Laemmli, 1970). The resolved proteins were then transferred to PVDF membranes. The membranes were washed twice in TTBS (100 mM Tris-Cl, pH 7.5, 0.9% NaCl, 0.1% Tween 20), and then incubated for 2 hours at room temperature in blocking buffer (TTBS, 1:5000 NGS, 4% BSA, 1:5000 fetal bovine serum (FBS), 2% non-fat dry milk). Blots then were probed for 2 hours at room temperature with antibodies to IC74 [1:10,000] and GRP94 [1:500] in TTBS. Replicate blots were probed under the same conditions with normal IgG or serum. Following incubation with primary antibodies, they were washed twice with each of TTBS and TBS, and then incubated for 2 hours with goat anti-mouse horse-radish peroxidase-conjugated antibody-diluted 1:5,000 in TTBS. The blots were washed, treated with ECL solution (Amersham, Buckinghamshire, UK), and antibody-reactive bands exposed on Kodak X-OMAT film.

The motility assay

General design

The basic design of the motility assay is described elsewhere (Beach and Vogl, 1999). In summary, spermatids with attached junction complexes were loaded into commercially available chambers and assayed for their ability to support the transport of taxol stabilized and fluorescently labelled microtubules. Assays were done in the presence of 2 mg/ml testicular cytosol at room temperature. Cytosol was included in experiments because, in other systems, microtubule binding is increased in the presence of cytosol (Blocker et al., 1996). Moreover, recombinant kinesin (35 of 37 recorded microtubules moved with the minus-end of the microtubule leading the movement; that is, movement generated by a plus-end directed motor).

The buffer systems (PEM, PEM/250, motility buffer) used in the assay are described elsewhere (Beach and Vogl, 1999), in some systems, motility has been reported only in the presence of as are the methods used to prepare testicular cytosol and to obtain elongate spermatids with attached junction plaques.

Running the assay

To increase the number of polarity-labelled microtubules bound to junction plaques, microtubules were added to testicular fractions enriched for elongate spermatids prior to adding the cells to the motility chamber. Approximately 10 μl of buffer containing polar microtubules was combined with 5 μl of spermatids with attached ectoplasmic specializations in a small 1.5 ml tube. The tube was placed at a 45° angle on a small rotating platform and allowed to gently rotate, around its longitudinal axis, for 10 minutes. Following this, the mixture was added very slowly to a motility chamber and the cells allowed to attach to the glass. After 10 minutes, the chamber was washed with 10 μl of PEM/250 containing 2 μM taxol and 1 mg/ml casein and placed on the microscope stage. A spermatid was located using phase microscopy and data collection initiated. After 10-15 control frames, 10 μl of motility buffer (containing 5 mM ATP) was placed on the slide and drawn through the chamber with a piece of filter paper. Data collection continued until 80 frames were captured. The direction (polarity) of any microtubule movement was determined by animating the captured images.

A microtubule was considered polarity-labelled only if there was a seed flanked by one long and one short end. In a few cases, a microtubule was also considered labelled if there was only one long end detectable, but only if the end was visibly longer than the seed. If there was any doubt about the polarity, the microtubule was considered nonpolar and was not used in the final data set.

Although the majority of experiments were conducted at room temperature (20-25°C), a limited number were conducted at 33°C (testicular temperature). cytosol (Schroer et al., 1988; Muresan et al., 1996). Data were recorded as described elsewhere (Beach and Vogl, 1999). Image sets consisted of 80× 3-second exposures separated by approximately 3-second intervals. Motility buffer containing 5 mM ATP was added after 10-15 images had been recorded. These images served as controls for nucleotide. Following the assay, cells were stained with coumarin phallacidin (100 U/ml PEM250) and DiOC6 (0.5 μg/ml PEM250) to verify the presence of the actin and endoplasmic reticulum components, respectively, of ectoplasmic specializations. In addition, an image was collected in the plane of the glass to verify that any microtubules that had moved were on ectoplasmic specializations and not on the glass.

Unlike in our previously reported assays (Beach and Vogl, 1999), those reported here involved the use of polarity-marked microtubules. Although we tried a number of methods to generate polar microtubules, we obtained good quality microtubules of the appropriate length and at an appropriate concentration to use in our assay system only by modifying a protocol originally described by Howard and Hymann (1993) in which GMPCPP (guanylyl (alpha,beta) methylenediphosphonate) is used to construct stable seeds onto which are polymerized more dimly labelled ends. We controlled the polymerization time to obtain microtubules of 5-8.5 μm in length, and removed unpolymerized tubulin by centrifuging the polar microtubules into a layer of PEM/250. We confirmed that the microtubules moved in the appropriate direction on bovine brain or

Movement of polar microtubules on testicular cytosol

To determine what direction microtubules move on cytosol, 5 mg/ml of testicular cytosol was added to the motility chamber and allowed to stand for 10 minutes at room temperature. Following this, buffer containing 2 mg/ml casein was added to the chamber for 10 minutes in order to block any non-specific sites on the glass. Microtubules were added to the chamber and allowed bind for 10 minutes. The chamber was washed once and then mounted on the microscope. Assays were done both with and without 2 mg/ml testicular cytosol present in the motility buffer.

Cytosol in the motility buffer

To determine if testicular cytosol in the motility buffer alone could support microtubule movement in the assay system, a 10 μl portion of polar microtubules was added to 5 μl of PEM/250 and gently rotated, as described above. This mixture was added to the motility chamber for 10 minutes incubation and then washed with 10 μl of PEM/250 containing 2 μM taxol and 1 mg/ml casein. As was done in the motility assays when ectoplasmic specializations on spermatids were used, motility buffer (containing 2 mg/ml testicular cytosol) was added to the motility chamber after approximately 15 frames had elapsed.

Antibodies to IC74 react with the endoplasmic reticulum of ectoplasmic specializations

Both in pre- and post-embedded material, gold particles were positioned near the cytoplasmic face of the endoplasmic reticulum of ectoplasmic specializations. Interestingly, the gold particles were not restricted to these sites, but also occurred in Sertoli cell cytoplasm surrounding the plaques.

In fragmented material that was labelled prior to embedding, structural elements of spermatids and junction plaques were well defined. In particular, the endoplasmic reticulum of the junction plaques was clearly visible.

Gold particles were present in Sertoli cell regions adjacent to the junction plaque (Fig. 2A,B) and were associated with the cytoplasmic face of endoplasmic reticulum component of the plaques (Fig. 2C-G). The number of gold particles in this region was significantly different from the number present in a similar region from normal IgG-treated controls (t(23)=2.55, P<0.02, n=11) (Table 1). Although associated with the endoplasmic reticulum, gold particles were also present deeper in the Sertoli cell cytoplasm in relation to elements other than those directly associated with junction plaques. Few gold particles were observed in sections of material in which the primary antibody was replaced with an equivalent concentration of normal mouse IgG or with buffer alone (data not shown). As expected, no gold was present when both primary and secondary antibodies were replaced with buffer alone (data not shown).

Table 1.

Quantification of gold associated with Sertoli cell junction plaques (ectoplasmic specializations) incubated with an immunological probe for IC74 before or after embedding

Quantification of gold associated with Sertoli cell junction plaques (ectoplasmic specializations) incubated with an immunological probe for IC74 before or after embedding
Quantification of gold associated with Sertoli cell junction plaques (ectoplasmic specializations) incubated with an immunological probe for IC74 before or after embedding
Fig. 2.

Pre-embedding immunogold labelling for IC74 in perfusion-fixed and mechanically fragmented rat testis. The mechanically dissociated material was labelled with a primary antibody to IC74 then with second antibody conjugated to colloidal gold. The material then was further processed using standard techniques for electron microscopy. Gold particles are associated with the cytoplasmic face of the endoplasmic reticulum (A-H). Also notice that in some of the images (B,H) gold occurs somewhat distant from the endoplasmic reticulum, in association with other structures.

Fig. 2.

Pre-embedding immunogold labelling for IC74 in perfusion-fixed and mechanically fragmented rat testis. The mechanically dissociated material was labelled with a primary antibody to IC74 then with second antibody conjugated to colloidal gold. The material then was further processed using standard techniques for electron microscopy. Gold particles are associated with the cytoplasmic face of the endoplasmic reticulum (A-H). Also notice that in some of the images (B,H) gold occurs somewhat distant from the endoplasmic reticulum, in association with other structures.

In testis material that was labelled after embedding, the morphology of the junction plaques was not nearly as good as in material labelled prior to embedding. Nevertheless, junction plaques were easily located by their association with spermatid heads, and the junction-related cisternae of endoplasmic reticulum appeared as a clear zone between the actin layer and other cytoplasm deeper in the Sertoli cell. Gold particles were concentrated in regions adjacent to the cytoplasmic face of the endoplasmic reticulum (Fig. 3A-E). Quantitatively, analysis of variance revealed a significant difference in the number of gold particles amongst the areas that were measured (F(7,80)=20.26, P<0.0001, n=11). Neuman-Keuls Post Hoc analysis demonstrated that the number of gold particles in cytoplasm adjacent to the cytoplasmic face of the endoplasmic reticulum at the junction plaques was significantly different from the numbers present in actin regions of the plaques, in related head regions of the attached spermatids, in perinuclear regions of Sertoli cell cytoplasm, and in IgG controls (Table 1) (P<0.0005, n=11).

Fig. 3.

IC74 staining of sections of fixed rat testis embedded in Unicryl. (A-E) Sertoli cell regions adjacent to spermatid heads. The actin layer and endoplasmic reticulum component (ER) of ectoplasmic specializations are indicated. In each case, the gold particles occur in Sertoli cell regions associated with the cytoplasmic face of the endoplasmic reticulum. Bars, 0.25 μm.

Fig. 3.

IC74 staining of sections of fixed rat testis embedded in Unicryl. (A-E) Sertoli cell regions adjacent to spermatid heads. The actin layer and endoplasmic reticulum component (ER) of ectoplasmic specializations are indicated. In each case, the gold particles occur in Sertoli cell regions associated with the cytoplasmic face of the endoplasmic reticulum. Bars, 0.25 μm.

IC74 and GRP94 are enriched in supernatants collected from gelsolin-treated spermatid/junction plaques

Probes for cytoplasmic dynein were more reactive with supernatants from gelsolin-treated spermatid/ junction complexes than with supernatants from control material. Spermatids treated with gelsolin and stained with fluorescent phallotoxin had little or no staining relative to control cells (Fig. 4). Moreover, supernatants from gelsolin-treated cells reacted strongly with the marker for endoplasmic reticulum (GRP94) (Fig. 4), whereas supernatants from control cells reacted weakly or not at all. Significantly, the probe for IC74 reacted more strongly on blots of supernatants from gelsolin-treated than of control-treated testicular fractions enriched for spermatid/junction complexes.

Fig. 4.

Immunoblot analysis of supernatants collected from spermatid/junction complexes treated with or without gelsolin. Treatment with gelsolin resulted in a loss of filamentous actin, and presumably of the endoplasmic reticulum, from the junction plaques attached to spermatids. This is shown in the top four images (phase and fluorescence) in which control (−Gelsolin) and gelsolin-treated (+Gelsolin) cells have been stained for filamentous actin. Probes for GRP94 (top set of blots) and IC74 (lower set of blots) reacted more intensely with supernatants from gelsolin-treated spermatid/junction complexes than with controls. The positions of molecular mass markers (kDa) are shown.

Fig. 4.

Immunoblot analysis of supernatants collected from spermatid/junction complexes treated with or without gelsolin. Treatment with gelsolin resulted in a loss of filamentous actin, and presumably of the endoplasmic reticulum, from the junction plaques attached to spermatids. This is shown in the top four images (phase and fluorescence) in which control (−Gelsolin) and gelsolin-treated (+Gelsolin) cells have been stained for filamentous actin. Probes for GRP94 (top set of blots) and IC74 (lower set of blots) reacted more intensely with supernatants from gelsolin-treated spermatid/junction complexes than with controls. The positions of molecular mass markers (kDa) are shown.

Testicular cytosol supports only plus-end type motility

When 5 mg/ml testicular cytosol was used in the motility chamber instead of kinesin, all recorded microtubules moved leading by the minus end (n=158), that is, in a direction generated by a plus-end directed motor. This was true both in the presence and absence of 2 mg/ml cytosol in the motility buffer. Importantly, no motility was detected in assays where 2 mg/ml cytosol was included in the motility buffer and polar microtubules were added to the motility chamber without previously adhering kinesin, testicular cytosol, or spermatid/junction complexes to the glass.

Ectoplasmic specializations support both plus-end and minus-end motility

The polar microtubules used in the motility assays were of a length and a concentration that resulted both in sufficient binding to the junction plaques (Fig. 5) and in minimal breakage during the assay. A total of 25 polar microtubule movements were recorded on spermatid/junction complexes. Of these, 8 microtubules moved with their plus ends leading (Fig. 6) and 17 moved with their minus ends leading (Fig. 7). Interestingly, the polar microtubules that moved in the assay system were not always oriented, relative to the spermatid heads, as are microtubules in Sertoli cells; that is, with their plus ends positioned towards the tips of the spermatid heads. In approximately 50% of the cases, the polar microtubules were oriented in the opposite orientation to that found in Sertoli cells in vivo. Once the assay was established, motility was recorded from approximately one in every four cells tested. At no time did we ever observe any single microtubule to move in both the plus and the minus direction. Microtubules moved either smoothly or in a saltatory fashion along the plaque, and then either stopped moving or cycled off of the plaque.

Fig. 5.

Polarity-marked GMPCPP microtubules bound to the ectoplasmic specializations. A phase image followed by images of actin (to show the presence of the ectoplasmic specialization) and the GMPCPP polarity-marked microtubules attached to the junction plaque as well as an overlay of the polar microtubules on the ectoplasmic specialization. Actin = red, polar microtubules = green. Bar, 2.5 μm.

Fig. 5.

Polarity-marked GMPCPP microtubules bound to the ectoplasmic specializations. A phase image followed by images of actin (to show the presence of the ectoplasmic specialization) and the GMPCPP polarity-marked microtubules attached to the junction plaque as well as an overlay of the polar microtubules on the ectoplasmic specialization. Actin = red, polar microtubules = green. Bar, 2.5 μm.

Fig. 6.

Phase image of a spermatid (A) followed by fluorescence images of a microtubule (large arrow) moving with the plus-end leading (direction of movement indicated by the small arrow) (D). This type of movement is consistent with the conclusion that a minus-end motor is associated with the junction plaque. The microtubule is oriented in the same orientation as that found in vivo (long end is marked by +; short end is marked by −). The microtubule shown here moved along the junction plaque and up onto the glass (E) where it stopped moving. Actin and DiOC6 (C) (endoplasmic reticulum) images were taken once the run was complete. Bar, 2.5 μm.

Fig. 6.

Phase image of a spermatid (A) followed by fluorescence images of a microtubule (large arrow) moving with the plus-end leading (direction of movement indicated by the small arrow) (D). This type of movement is consistent with the conclusion that a minus-end motor is associated with the junction plaque. The microtubule is oriented in the same orientation as that found in vivo (long end is marked by +; short end is marked by −). The microtubule shown here moved along the junction plaque and up onto the glass (E) where it stopped moving. Actin and DiOC6 (C) (endoplasmic reticulum) images were taken once the run was complete. Bar, 2.5 μm.

Fig. 7.

Images demonstrating transport, on an ectoplasmic specialization, of a polarity-marked microtubule with the minus-end leading (small arrow indicates the direction of movement). This direction of motion is consistent with the conclusion that a plus-end type motor is anchored to the junction plaque. The large arrow points to the seed of the polar microtubule in the second panel and the asterisks indicate the position of the seed throughout the rest of the sequence. The time (minutes:seconds) is indicated in each panel. Bar, 2.5 μm.

Fig. 7.

Images demonstrating transport, on an ectoplasmic specialization, of a polarity-marked microtubule with the minus-end leading (small arrow indicates the direction of movement). This direction of motion is consistent with the conclusion that a plus-end type motor is anchored to the junction plaque. The large arrow points to the seed of the polar microtubule in the second panel and the asterisks indicate the position of the seed throughout the rest of the sequence. The time (minutes:seconds) is indicated in each panel. Bar, 2.5 μm.

Rates could not always be calculated because of difficulties encountered in establishing the exact frame that movement either began or ended, or because of the saltatory nature of the movement. Also, in some cases, the microtubule clearly was positioned at an oblique angle on the cell and to the plane of focus of the lens. Nevertheless, rates calculated when microtubules moved leading and trailing by the minus-end were 0.062±0.066 μm/second (n=12) and 0.085±0.031 (n=3), respectively, in the presence of 2 mg/ml cytosol.

In this study, we provide evidence that microtubule-dependent motor proteins are associated with specialized junction plaques (ectoplasmic specializations) that occur at sites of contact between Sertoli cells and elongate spermatids in the mammalian testis. Data from immunoelectron micrographs, western blots and motility assays support the conclusion that a cytoplasmic dynein is associated with the endoplasmic reticulum of the junction plaques. In addition, data from in vitro motility assays indicate that a plus-end directed microtubule-dependent motor, likely a kinesin, is also present. Collectively, the data are consistent with the hypothesis that changes in the position of elongate spermatids within the mammalian seminiferous epithelium are due to the microtubule-based transport of the junction plaques that occur in Sertoli cells at sites of attachment to spermatid heads.

Ectoplasmic specializations are a class of actin-related intercellular adhesion junctions that occur at sites of intercellular contact between Sertoli cells and adjacent cells. The structures are unique in a number of ways. The actin filaments within the plaques are hexagonally packed and form a distinct layer adjacent to the Sertoli cell plasma membrane (reviewed by Vogl et al., 1993). In addition, a cistern of endoplasmic reticulum is attached to the cytoplasmic side of the filament layer and is a structural component of the plaque. Moreover, at least one of the intercellular adhesion molecules present at the sites appears to be an integrin (α6β1) (Palombi et al., 1992; Salanova et al., 1995). Finally, the isotype of espin, an actin-binding and -bundling protein found in ectoplasmic specializations, is much larger than the form found associated with bundles of parallel actin filaments in other cell types and contains an additional actin-binding site (Chen et al., 1999).

During spermatogenesis, developing spermatids dramatically change position in the semininiferous epithelium.

At one point, they become situated very deep in the epithelium, in some cases reaching a position adjacent to the tubule wall. Subsequently, the spermatids are moved to the apices of the Sertoli cells where they are eventually released as spermatozoa. The functional significance of this translocation is not known, although it has been suggested that it may provide structural support for elongate spermatids or may increase the area of contact between Sertoli cells and spermatids for ‘communication’ between the two cell types (Beach and Vogl, 1999). Spermatid translocation, to some extent, occurs in most vertebrates that have an elongate spermatid shape and in which the spermatids occur within apical crypts of Sertoli cells. The velocity of translocation, in vivo, has not been determined.

There are a number of possible mechanisms by which the position of developing spermatids could be altered in the epithelium. One mechanism would be by the use of actin-based motor proteins and involve either contraction of the actin filament bundles of the plaque or of the movement of the actin filaments relative to adjacent structures. This may be the mechanism of translocating spermatids to a deep position in the epithelium of some non-mammalian vertebrates (Stanley and Lambert, 1985); however, it is likely not the case in mammals because conventional myosin is not present (Vogl and Soucy, 1985), the actin bundles are close-packed into hexagonal arrays, and the actin bundles are restricted to the plaque and do not extend deeper into the cytoplasm or to the base of the cell. Although myosin VIIa is present within mammalian ectoplasmic specializations (Hasson et al., 1997; Wolfrum et al., 1998), there is no evidence that this motor participates in moving the junction plaque relative to the rest of the Sertoli cell.

A second possible mechanism is that translocation occurs by changes in the polymerisation state of associated microtubules. In this mechanism, basally directed polymerization of microtubules from nucleating sites in apical regions might ‘push’ the junctions, and hence the attached spermatids, deep into the epithelium. Depolymerization would be coupled in some way to the return of spermatids to the apices of the Sertoli cells. If this hypothesis were true, one would predict that, at all stages of spermatogenesis, microtubules should be concentrated apical to the positions of spermatid heads. This is not observed (Vogl et al., 1995).

A third possibility, and our working hypothesis, is that microtubule-dependent motor proteins are associated with ectoplasmic specializations and function to move the junction plaques, with attached spermatids, along adjacent microtubules. This hypothesis is consistent with the known morphology of the system, and with the observations that isolated junction complexes bind and transport exogenous microtubules in a nucleotide dependent fashion (Vogl, 1996; Beach and Vogl, 1999). In addition, the intermediate chain of cytoplasmic dynein (IC74) has been localized, by immunofluorescence, to regions of the Sertoli cell in which translocation occurs (Miller et al., 1999).

Based on the structure of ectoplasmic specializations and on the translocation model presented above, the predicted position of any microtubule dependent motor proteins associated with the junction plaque is on the cytoplasmic face of the endoplasmic reticulum. Our finding that immunological probes for IC74 react, at the ultrastructural level, with material on or immediately adjacent to this region of the plaque is consistent with the conclusion that a cytoplasmic dynein is present at the predicted location. Although certainly not conclusive, the finding that probes for IC74 and a marker for endoplasmic reticulum (GRP94) are enriched in immunoblots of supernatants collected from gelsolin-treated spermatid/junction complexes relative to supernatants from control-treated complexes supports this conclusion, as does our observation that approximately one third of the recorded motility events on spermatid/junction complexes in vitro were in a direction consistent with a minus-end motor being present on the plaque. The concept that the endoplasmic reticulum in general is closely associated with and transported along microtubules is not new (Franke, 1971; Terasaki et al., 1986; Dabora and Sheetz, 1988; Lee et al., 1989; Vale and Hotani, 1989; Klopfenstein et al., 1990; Cole and Lippincott-Schwartz, 1995; Lane and Allan, 1999). In fact, motor proteins (Hollenbeck, 1989; Schmitz et al., 1994; Allan, 1995) and at least one of their putative receptors (Toyoshima et al., 1992) have been localized to this organelle. The endoplasmic reticulum appears pre-adapted for coupling transport machinery to an intercellular junction plaque.

The observation that immunological probes for IC74 also react with cytoplasmic regions not directly associated with the plaque indicates that cytoplasmic dynein is likely involved with transport processes other than and in addition to spermatid translocation within the Sertoli cell. In some species there are dramatic changes in the distribution of organelles during spermatogenesis (Vogl et al., 1983). Another possibility is that motor proteins in these areas may indirectly contribute to positioning of the junction plaques by altering the position of junction-related microtubules relative to microtubules more distant from the plaques.

Data obtained from the motility assays confirm our earlier results (Beach and Vogl, 1999), demonstrating that isolated spermatid/junction complexes support microtubule transport. Importantly, results presented here indicate that transport on the junction plaques occurs both in the minus- and in the plus-end direction. The observation that virtually all microtubules on testicular cytosol and on pure kinesin move in a direction generated by plus-end directed motors strengthens the argument that the bidirectional nature of movement associated with ectoplasmic specializations is specific to the plaque. The unidirectional nature of microtubule transport on testicular cytosol is consistent with the report that plus-end type movement predominates when both dynein and kinesin are artificially attached to the same cargo (Muresan et al., 1996). Although it is impossible for us to rule out conclusively the possibility that some ambiguity may exist in the data due to microtubule breakage, the fact that approximately one third of the polar microtubules moved in a direction that would be generated by a minus-end type motor on the spermatid/junction complexes while none did so on cytosol suggests that the results are real and not artifactual. The localization of dynein to regions of the Sertoli cell in which translocation occurs (Miller et al., 1999) and to the endoplasmic reticulum of the junction plaque itself (present study) support this conclusion. The differences between results with cytosol, which presumably contains both kinesin and dynein, and the spermatid/junction complexes may be related to the relative numbers of motors present in each preparation, to differences in arrangement or distribution of the motors on glass versus endoplasmic reticulum, and/or to differences in motor activities in the two preparations. It is pertinent to note that bidirectional movement of Xenopus egg-derived ER, in the presence of somatic Xenopus tissue culture cell extracts, has been reported (Lane and Allan, 1999). Our motility data constitute the first evidence that a plus-end directed motor, most likely a kinesin, is associated with ectoplasmic specializations. This system is one of the few in which there is evidence both for plus- and minus-end directed motor proteins on the same organelle.

Data presented here indicate that isolated spermatid/junction complexes support bidirectional microtubule transport in vitro. They also support the argument that a cytoplasmic dynein is associated with the cytoplasmic face of the endoplasmic reticulum component of ectoplasmic specializations, and that at least one form of kinesin also is likely present. Based on the orientation of microtubules in Sertoli cells, and if our hypothesis is correct, a plus-end directed motor would move spermatids towards the base of the epithelium and a minus-end directed motor would move the cells back to the apex. Although the data presented here are consistent with our working hypothesis of spermatid translocation in the mammalian seminiferous epithelium, it remains to be determined if microtubule-dependent motors associated with the junction plaque actually do translocate spermatids within the seminiferous epithelium in vivo, or if perhaps they function in some other fashion as has been suggested by others (Hall et al., 1992). Also, our results do not rule out the possibility that a minus-end directed kinesin may be associated with the junction plaque in addition to dynein.

This work was funded by Operating Grant number MT-13389 awarded to A.W.V. by the Medical Research Council of Canada. We would also like to thank Dr Pfister for the IC74 antibody, Dr Hancock for the GMPCPP and Candace Hofmann for assistance with the statistical analysis.

Allan
,
V.
(
1995
).
Protein phosphatase 1 regulates the cytoplasmic dynein-driven formation of endoplasmic reticulum networks in vitro
.
J. Cell Biol
.
128
,
879
891
.
Amlani
,
S.
and
Vogl
,
A. W.
(
1988
).
Changes in the distribution of microtubules and intermediate filaments in mammalian Sertoli cells during spermatogenesis
.
Anat. Rec
.
220
,
143
160
.
Beach
,
S. F.
and
Vogl
,
A. W.
(
1999
).
Spermatid translocation in the rat seminiferous epithelium: Coupling membrane trafficking machinery to a junction plaque
.
Biol. Reprod
.
60
,
1036
1046
.
Blocker
,
A.
,
Deverin
,
F. F.
,
Habermann
,
A.
,
Hyman
,
A. A.
,
Griffiths
,
G.
and
Burchardt
,
J. K.
(
1996
).
Microtubule-associated protein-dependent binding of phagosomes to microtubules
.
J. Biol. Chem
.
271
,
3803
3811
.
Csermely
,
P.
,
Miyata
,
Y.
and
Schnaider
,
T.
(
1995
).
Autophosphorylation of grp 94 (edoplasmin)
.
J. Biol. Chem
.
270
,
6381
6388
.
Chen
,
B.
,
Li
,
A.
,
Wang
,
D.
,
Wang
,
M.
,
Zheng
,
L.
and
Barles
,
J. R.
(
1999
).
Espin contains an additional actin-binding site in its N terminus and is a major actin-bundling protein of the Sertoli cell-spermatid ectoplasmic specialization junction plaque
.
Mol. Biol. Cell
10
,
4327
4339
.
Christensen
,
A. K.
(
1965
).
Microtubules in Sertoli cells of the guinea pig testis
.
Anat. Rec
.
151
,
335a
.
Cole
,
N. B.
and
Lippincott-Schwartz
,
J.
(
1995
).
Organization of organelles and membrane traffic by microtubules
.
Curr. Opin. Cell Biol
.
7
,
55
64
.
Criswell
,
P. S.
and
Asai
,
D. J.
(
1998
).
Evidence for cytoplasmic dynein heavy chain isoforms in rat testis
.
Mol. Biol. Cell
9
,
237
247
.
Dabora
,
S. L.
and
Sheetz
,
M. P.
(
1988
).
The microtubule-dependent formation of a tubulovesicular network with characteristics of the ER from cultured cell extracts
.
Cell
54
,
27
35
.
Dillman
,
J. F.
and
Pfister
K. K.
(
1994
).
Differential phosphorylation in vivo of cytoplasmic dynein with anterogradely moving organelles
.
J. Cell Biol
.
127
,
1671
1681
.
Fawcett
,
D. W.
(
1975
).
Ultrastructure and function of the Sertoli cell
.
In Handbook of Physiology; Section 7, Endocrinology; Vol. 5, Male Reproductive System
. (ed.
R. O.
Greep
), pp.
21
55
.
Baltimore
:
Williams & Wilkins
.
Franke
,
W. W.
(
1971
).
Cytoplasmic microtubules linked to endoplasmic reticulum with cross-bridges
.
Exp. Cell Res
.
66
,
486
489
.
Geiger
,
B.
,
Dutton
,
A. H.
,
Tokuyasu
,
K. T.
and
Singer
,
S. J.
(
1981
).
Immunoelectron microscope studies of membrane-microfilament interactions: Distributions of alpha-actinin, tropomyosin, and vinculin in intestinal epithelial brush border and chicken gizzard smooth muscle cells
.
J. Cell Biol
.
91
,
614
628
.
Hall
,
E. S.
,
Eveleth
,
J.
,
Jiang
,
C.
,
Redenbach
,
D. M.
and
Boekelheide
,
K.
(
1992
).
Distribution of the microtubule-dependent motors cytoplasmic dynein and kinesin in rat testis
.
Biol. Reprod
.
46
,
817
828
.
Hasson
,
T.
,
Walsh
,
J.
,
Cable
,
J.
,
Mooseker
,
M. S.
,
Brown
,
S. D. M.
and
Steel
,
K. P.
(
1997
).
Effects of shaker-1 mutations on myosin-VIIa protein and mRNA expression
.
Cell Motil. Cytoskel
.
37
,
127
138
.
Hollenbeck
,
P. J.
(
1989
).
The distribution, abundance and subcellular localization of kinesin
.
J. Cell Biol
.
108
,
2335
2342
.
Howard
,
J.
and
Hymann
,
A. A.
(
1993
).
Preparation of marked microtubules for the assay of the polarity of microtubule-based motors by fluorescence microscopy
.
Meth. Cell Biol
.
39
,
105
113
.
Klopfenstein
,
D. R. C.
,
Kappeler
,
F.
and
Hauri
,
H-P.
(
1998
).
A novel direct interaction of endoplasmic reticulum with microtrubules
.
EMBO J
.
17
,
6168
6177
.
Laemmli
,
U. K.
(
1970
).
Cleavage of structural proteins during the assembly of the head of bacteriophage T4
.
Nature
227
,
680
685
.
Lane
,
J. D.
and
Allan
,
V. J.
(
1999
).
Microtubule-based endoplasmic reticulum motility in Xenopus laevis: Activation of membrane-associated kinesin during development
.
Mol. Biol. Cell
,
10
,
1909
1922
.
Lee
,
C.
,
Ferguson
,
M.
and
Chen
,
L. B.
(
1989
).
Construction of the endoplasmic reticulum
.
J. Cell Biol
.
109
,
2045
2055
.
Mazzarella
,
R. A.
and
Green
,
M.
(
1987
).
Erp99, an abundant, conserved glycoprotein of the endoplasmic reticulum, is homologous to the 90-kDa heat shock protein (hsp90) and the 94-kDa glucose regulated protein (GRP94)
.
J. Biol. Chem
.
262
,
8875
8883
.
Miller
,
M. G.
,
Mulholland
,
D. J.
and
Vogl
,
A. W.
(
1999
).
Rat testis motor proteins associated with spermatid translocation (dynein) and spermatid flagella (kinesin-II)
.
Biol. Reprod
.
60
,
1047
1056
.
Muresan
,
V.
,
Godek
,
C. P.
,
Reese
,
T. S.
and
Schnapp
,
B. J.
(
1996
).
Plus-end motors override minus-end motors during transport of squid axon vesicles on microtubules
.
J. Cell Biol
.
135
,
383
397
.
Muresan
,
Z.
and
Arvan
,
P.
(
1997
).
Thyroglobulin transport along the secretory pathway; Investigation of the role of molecular chaperone, GRP94, in protein export from the endoplasmic reticulum
.
J. Biol. Chem
.
272
,
26095
26102
.
Palombi
,
F.
,
Salanova
,
M.
,
Tarone
,
G.
,
Farini
D.
and
Stefanini
,
M.
(
1992
).
Distribution of β1 integrin subunit in rat seminiferous epithelium
.
Biol Reprod
.
47
,
1173
1182
.
Redenbach
,
D. M.
and
Vogl
,
A. W.
(
1991
).
Microtubule polarity in Sertoli cells: a model for microtubule-based spermatid transport
.
Eur. J. Cell Biol
.
54
,
277
290
.
Romrell
,
L. J.
and
Ross
,
M. H.
(
1979
).
Characterization of Sertoli cell-germ cell junctional specializations in dissociated testicular cells
.
Anat. Rec
.
193
,
23
42
.
Russell
,
L.
(
1977
).
Observations on rat Sertoli ectoplasmic (‘junctional’) specializations in their association with germ cells of the rat testis
.
Tissue Cell
,
9
,
475
498
.
Salanova
,
M.
,
Stefanini
,
M.
,
De Curtis
,
I.
and
Palombi
,
F.
(
1995
).
Integrin receptor α6β1 is localized at specific sites of cell-to-cell contact in rat seminiferous epithelium
.
Biol. Reprod
.
52
,
79
87
.
Sargan
,
D. R.
,
Tsai
,
M. J.
and
O’Malley
,
B. W.
(
1986
).
hsp 108…a novel heat shock inducible protein of chicken
.
Biochemistry
25
,
6252
6258
.
Schmitz
,
F.
,
Wallis
,
K. T.
,
Rho
,
M.
,
Drenckhahn
,
D.
and
Murphy
,
D.
(
1994
).
Intracellular distribution of kinesin in chromaffin cells
.
Eur. J. Cell Biol
.
63
,
77
83
.
Schroer
,
T. A.
,
Schnapp
,
B. J.
,
Reese
,
T. S.
and
Sheetz
,
M. P.
(
1988
).
The role of kinesin and other soluble factors in organelle movement along microtubules
.
J. Cell Biol
.
107
,
1785
1792
.
Stanley
,
H. P.
and
Lambert
,
C. C.
(
1985
).
The role of a Sertoli cell actin-myosin system in sperm bundle formation in the ratfish, Hydrolagus colliei (Chondrichthyes, Holocephali)
.
J. Morphol
.
186
,
223
236
.
Teresaki
,
M
,.
Chen
,
L. B.
and
Fujiwara
,
K.
(
1986
).
Microtubules and the endoplasmic reticulum are highly interdependent structures
.
J. Cell Biol
.
103
,
1557
1568
.
Toyoshima
,
I.
,
Yu
,
H.
,
Steuer
,
E. R.
and
Sheetz
,
M. P.
(
1992
).
Kinectin, a major kinesin-binding protein on ER
.
J. Cell Biol
.
118
,
1121
1131
.
Vale
,
R. D.
and
Hotani
,
H.
(
1988
).
Formation of membrane networks in vitro by kinesin-driven microtubule movement
.
J. Cell Biol
.
107
,
2233
2241
.
Vogl
,
A. W.
(
1988
).
Changes in the distribution of microtubules in rat Sertoli cells during spermatogenesis
.
Anat. Rec
.
222
,
34
41
.
Vogl
,
A. W.
(
1996
).
Spatially dynamic intercellular adhesion junction is coupled to a microtubule-based motility system: Evidence from an in vitro binding assay
.
Cell Motil. Cytoskel
.
34
,
1
12
.
Vogl
,
A. W.
and
Soucy
,
L. J.
(
1985
).
Arrangement and possible function of actin filament bundles in ectoplasmic specializations of ground squirrel Sertoli cells
.
J. Cell Biol
.
100
,
814
825
.
Vogl
,
A. W.
,
Lin
,
Y. C.
,
Dym
,
M.
and
Fawcett
,
D. W.
(
1983
).
Sertoli cells of the golden-mantled ground squirrel (Spermophilus lateralis): a model system for the study of shape change
.
Am. J. Anat
.
168
,
83
98
.
Vogl
,
A. W.
,
Pfeiffer
,
D. C.
,
Redenbach
,
D. M.
and
Grove
,
B. D.
(
1993
).
Sertoli cell cytoskeleton
.
In The Sertoli Cell
(ed.
L. D.
Russell
and
M. D.
Griswold
), pp.
39
86
.
Clearwater
:
Cache River Press
.
Vogl
,
A. W.
,
Weis
,
M.
and
Pfeiffer
,
D. C.
(
1995
).
The perinuclear centriole-containing centrosome is not the major microtubule organizing center in Sertoli cells
.
Eur. J. Cell Biol
.
66
,
165
179
.
Wolfrum
,
U.
,
Liu
,
X. R.
,
Schmitt
,
A.
,
Udovichenko
,
I. P.
and
Williams
,
D. S.
(
1998
).
Myosin VIIa as a common component of cilia and microvilli
.
Cell Motil. Cytoskel
.
40
,
261
271
.