BIOLOGY OF REPRODUCTION 63, 89–98 (2000)
Vesicular Traffic and Golgi Apparatus Dynamics During Mammalian
Spermatogenesis: Implications for Acrosome Architecture1
Ricardo D. Moreno,3,7 João Ramalho-Santos,3,8 Peter Sutovsky,4 Edward K.L. Chan,6 and
Gerald Schatten2,4,5
Oregon Regional Primate Research Center,4 Beaverton, Oregon 97006
Department of Obstetrics and Gynecology, and Cell and Developmental Biology,5 Oregon Health Sciences University,
Beaverton, Oregon 97006
WM Keck Autoimmune Disease Center,6 The Scripps Research Institute, La Jolla, California 92037
Reproduction and Developmental Biology Unit,7 Faculty of Biological Sciences, Pontifical Catholic University of Chile,
Santiago, Chile
Center for Neuroscience and Cell Biology of Coimbra,8 Department of Zoology, University of Coimbra,
Coimbra, Portugal
ABSTRACT
matocyte stage and continues throughout the round and
elongating spermatid stages [4–6]. In mammals, acrosome
biogenesis begins with the fusion of proacrosomal granules
synthesized by the Golgi apparatus in pachytene spermatocytes, similar to the formation of secretory granules in
many other cell types [1, 7–9]. During the ‘‘Golgi phase,’’
the acrosomal vesicle attaches to the nuclear envelope and
grows as a result of the constant arrival and fusion of Golgiderived vesicles [1, 8, 10]. At later stages (cap and acrosome phase), the acrosomal vesicle flattens and spreads
over the nucleus, covering up to two-thirds of its surface.
During this process, acrosomal proteins condense and are
packed in a paracrystalline structure called the acrosomal
granule, which eventually forms the acrosomal matrix in
mature sperm [4, 5, 11, 12].
During the early stages of spermiogenesis (Golgi and cap
phase), there is a close relationship between the forming
acrosomal vesicle and the Golgi apparatus. The Golgi is
actively engaged in the formation of the acrosomal vesicle,
producing and delivering the proteins and membranes needed for its enlargement and differentiation [3, 12, 13]. There
are also a number of small coated vesicles (40–50 nm diameter) that may correspond to the coatomer-coated vesicles (COP), as described in somatic cells [14–16]. One of
the components, b-COP, has been localized both in the Golgi apparatus and in the acrosomal membrane of rat spermatids, suggesting a role for COP-I vesicles in membrane
trafficking between these two organelles [17]. At later stages (acrosome and maturation phase), the Golgi apparatus
migrates toward the cell pole opposite the nascent acrosome, and the acrosomal vesicle lies underneath the plasma
membrane [8, 18, 19]. The molecular mechanisms involved
in this cytoplasmic polarization are still unknown. Eventually, the Golgi apparatus remains in the cytoplasmic lobe
and is discarded in the cytoplasmic droplet along with most
of the other organelles [1, 20, 21].
The acrosome has a characteristic shape and size depending of the species. The shaping of this organelle does
not seem to be related to changes in the cytoskeleton or
cell-cell interactions. We have previously defined some of
the molecular components present in the Golgi apparatus
of rhesus monkey spermatids [19]. We hypothesized that
either presence or distribution of the molecular components
involved in the vesicular traffic machinery may account for
differences in the shaping and kinetics of acrosome biogenesis between mammals. Our data indicate that rhesus
monkey and bovine spermatids share most of the Golgi
Vesicular membrane trafficking during acrosome biogenesis
in bull and rhesus monkey spermatogenesis differs from the somatic cell paradigm as imaged dynamically using the Golgi apparatus probes b-COP, giantin, Golgin-97, and Golgin-95/
GM130. In particular, sorting and delivery of proteins seemed
less precise during spermatogenesis. In early stages of spermiogenesis, many Golgi resident proteins and specific acrosomal
markers were present in the acrosome. Trafficking in both round
and elongating spermatids was similar to what has been described for somatic cells, as judged by the kinetics of Golgi protein incorporation into endoplasmic reticulum-like structures after brefeldin A treatment. These Golgi components were retrieved from the acrosome at later stages of differentiation and
were completely devoid of immature spermatozoa. Our data
suggest that active anterograde and retrograde vesicular transport trafficking pathways, involving both b-COP- and clathrincoated vesicles, are involved in retrieving Golgi proteins missorted to the acrosome and in controlling the growth and shape
of this organelle.
gametogenesis, sperm, spermatid, spermatogenesis, testes
INTRODUCTION
The acrosome is a secretory vesicle containing a number
of hydrolytic enzymes that help the sperm penetrate the
egg’s coats, and it is assembled only after the formation of
the haploid spermatid [1–3]. However, the synthesis of
many proteins involved in acrosome biogenesis, such as the
proteolytic enzyme acrosin, starts at the pachytene sperThe Mellon Foundation sponsored R.M. J.R.S. is a recipient of a Praxis
XXI postdoctoral fellowship from Fundação para a Ciência e Tecnologia
(FCT, Portugal). This research was sponsored by research grants from the
NIH (NICHD, NCRR) to G.S and by NICHD/NIH through cooperative
agreement [U54 18185] as part of the Specialized Cooperative Centers
Program in Reproduction Research. The ORPRC infrastructure is sponsored as an NCRR Regional Primate Research Center.
2
Correspondence: Gerald Schatten, Oregon Regional Primate Research
Center, 505 NW 185th Ave., Beaverton, OR 97006. FAX: 503 614 3725;
e-mail: [email protected]
3
Current address: Oregon Regional Primate Research Center, 505 NW
185th Ave., Beaverton, OR 97006.
1
Received: 29 December 1999.
First decision: 21 January 2000.
Accepted: 15 February 2000.
Q 2000 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
89
90
MORENO ET AL.
components used by somatic cells. These components have
similar spatial and temporal distribution in both species.
Thus, acrosome shaping may involve other unidentified
components or unique interactions between the cytoskeleton and the vesicular traffic machinery.
MATERIALS AND METHODS
Chemicals
Unless otherwise stated, all chemicals were purchased
from Sigma Chemical Company (St. Louis, MO). Brefeldin
A was obtained from Epicentre Technologies (Madison,
WI) and was stored as a stock solution of 1 mg/ml at
2208C. BODIPY FL C5-ceramide and NBD C6-ceramide
were obtained from Molecular Probes (Eugene, OR) and
were stored as a stock solution of 5 mM at 2208C.
Antibodies
Antibodies against Golgin-97, Golgin-160, Golgin95/
GM130, and giantin were prepared as described previously
[22, 23]. Anti-Golgin and anti-giantin antibodies were used
at dilutions of 1:200 and 1:500, respectively. The polyclonal antibody against perinuclear theca (pAb 427, dilution 1:
200) was a gift from Dr. Richard Oko (Department of Anatomy and Cell Biology, Queens University, Kingston, ON,
Canada). Monoclonal antibodies against human acrosin
were a gift from Dr. Claudio Barros (Reproduction and Development Unit, Pontifical Catholic University of Chile,
Santiago, Chile) and used in a dilution of 1:200. The antibodies against trans-Golgi network (TGN38), clathrin, and
b-COP were purchased from Affinity Bioreagents (Golden,
CO) and used according to the manufacturer’s instructions.
Goat antibody against mouse or rabbit IgG, conjugated with
either tetramethylrhodamine B isothiocyante (TRITC) or
fluorescein isothiocyante (FITC), were obtained from Zymed Laboratories (San Francisco, CA).
Isolation and Culture of Spermatogenic Cells
Rhesus monkey testes were obtained from males undergoing necropsy for reasons unrelated to fertility. Bovine
testes were obtained from a local slaughterhouse. Testes
were dissected and transferred to a Petri dish filled with
TALP-Hepes medium ([24]; modified Tyrode-lactate medium with pyruvate and albumin: 114 mM NaCl, 3.2 mM
KCl, 2 mM CaCl2, 0.5 mM MgCl2, 25 mM NaHCO3, 0.4
mM NaH2PO4, 10 mM sodium lactate, 6.5 IU penicillin,
25 mg/ml gentamicin, 3 mg/ml fatty acid-free BSA, 0.2 mM
pyruvate, buffered with 10 mM Hepes at pH 7.4) and
minced with fine forceps as previously described [25]. The
final cell suspension contained all the spermatogenic stages
and was used for vital labeling, immunofluorescence, and
colloidal gold labeling.
Cell Culture
Rhesus monkey 6308 cells were propagated in Dulbecco
modified Eagle medium (D-MEM) supplemented with 10%
fetal calf serum, 2 mM glutamine, and 0.3% gentamycin at
378C in 5% CO2. When needed for immunofluorescence or
vital labeling, the cells were grown onto glass coverslips.
Vital Labeling and Imaging of Spermatids and Cell Lines
The procedure for labeling the Golgi apparatus with
BODIPY FL C5-ceramide or NBD C6-ceramide was essen-
tially the same as described by Pagano et al. [26]. Spermatids attached to poly-L-lysine-coated coverslips or 6308
cells grown onto glass coverslips were incubated with
BODIPY FL C5-ceramide at final concentration of 5 mM
in D-MEM containing Hepes buffer instead of bicarbonate
at 48C for 30 min. The coverslips were then washed with
TALP-HEPES and prepared for live epifluorescence microscopy. Labeling of cells with NBD C6-Ceramide was
carried out as described except that the cells were incubated
at 48C for 2 h.
Brefeldin A Treatment
Rhesus spermatids attached to poly-L-lysine-coated coverslips or 6308 cells grown onto glass coverslips were incubated in D-MEM containing brefeldin A for 30 min at
378C with 5% CO2. The coverslips were then washed with
TALP-HEPES and fixed for immunofluorescence. For drug
recovery experiments, the coverslips were washed and incubated in D-MEM without brefeldin A for 30 min at 378C
and then fixed for immunofluorescence.
Immunofluorescence and Cell Imaging
Bull or rhesus monkey spermatogenic cells were attached to poly-L-lysine-coated coverslips in 1 ml of KMT
medium (100 mM KCl, 2 mM MgCl2, 10 mM Tris-HCl,
pH 7.0) and fixed for 1 h with 2% formaldehyde in 0.1 M
PBS. The cells were permeabilized for 1 h in 1% Triton X100 in PBS. Nonspecific antibody cross-reaction was
blocked by a 1-h preincubation in 0.1 M PBS containing
2% BSA and 130 mM glycine. The coverslips were incubated with the primary antibody for 2 h and then for 1 h
with either TRITC- or FITC-conjugated appropriate secondary antibodies. After the final wash with PBS containing 0.5% Tween 20, the DNA was labeled with 5 mg/ml
of 49-69-diamino-2-phenylindole (Molecular Probes) for 10
min. The coverslips were rinsed three times and mounted
in a drop of Vectashield mounting medium (Vector Laboratories, Burlingame, CA). Coverslips were examined using
either a Zeiss Axiophot or a Nikon Eclipse E1000 epifluorescence microscope and photographed using a chilled
CCD camera (Princeton Instruments Inc., Trenton, NJ) operated by Metamorph software.
Cell lines grown onto coverslips were processed as described for spermatogenic cells.
Colloidal Gold Labeling and Electron Microscopy
Testicular cells were isolated, fixed in formaldehyde,
permabilized with 0.1% Triton-X-100, and processed with
anti-b-COP and anti-Golgin-95/GM130 antibodies as described for immunofluorescence, except that the fluorescently conjugated secondary antibodies were replaced by a
10-nm gold-conjugated goat anti-rabbit IgG (British
BioCell, Cardiff, UK). Instead of being attached to microscopy coverslips, the cells were handled in suspension and
washed by centrifugation in PBS with 1% BSA and 0.1%
Triton X-100 between individual processing steps. The labeled cells were pelleted by centrifugation, fixed for transmission electron microscopy, and embedded in Epon 812
as described previously [25]. Ultrathin sections were cut on
a Sorvall MT-5000 ultramicrotome (Ivan Sorvall, Norwalk,
CT), stained by uranyl acetate and lead citrate, and examined in a Phillips EM 300 electron microscope.
GOLGI APPARATUS IN MAMMALIAN SPERMATOGENESIS
91
FIG. 1. Distribution of Golgi proteins in a rhesus monkey cell line. Rhesus monkey 6308 cells were grown onto glass coverslips, fixed when they
reached about 50% of confluence, and stained with antibodies against giantin (A), b-COP (B), Golgin-97 (C), Golgin-95 (D), and Golgin 160 (E). All
the antibodies displayed a tubulelike appearance in the perinuclear region. This distribution pattern disappeared after brefeldin A treatment, as shown
for giantin (F) and Golgin-97 (G).
RESULTS
Intracellular Traffic in Mammalian Spermatids
Imaging the Golgi Apparatus in a Rhesus Monkey
Cell Line
In pachytene spermatocytes, giantin was present in a
rounded structure formed by at least two crescent-shaped
structures (Fig. 2A). These structures faced each other and
surrounded additional central material with a dotted pattern.
b-COP also showed a rounded shape formed by two crescent-shaped structures, which were probably the same as
labeled by giantin. However, b-COP did not seem to label
the central material as strongly as did giantin.
In round spermatids, both giantin and b-COP antibodies
labeled a typical horseshoe-shaped Golgi apparatus, with
the convex side facing the acrosomal vesicle (Fig. 2, B and
B9). Giantin and b-COP appeared to colocalize in the spermatid Golgi apparatus, although whether they were present
in the identical cisternae remains unresolved. As the acrosomal vesicle spread over the nucleus, the structure, position, and orientation of the Golgi apparatus was the same
as in the earlier stage (Fig. 2, C and C9). The Golgi apparatus was located very close to the acrosomal vesicle, and
at this resolution, it seemed that giantin and b-COP were
present on the acrosomal vesicle. In early elongating spermatids, the Golgi apparatus migrated to the cell pole opposite the acrosome and was included in the cytoplasmic
lobule (Fig. 2, D and D9).
To confirm the specificity of the labeling, rhesus monkey
Giantin is a membrane protein that appeared to colocalize with b-COP-containing vesicles [27]. This protein displayed a tubulelike pattern around the nucleus in a nonhuman primate cell line derived from the mammary gland
(Fig. 1A). Likewise, b-COP showed a perinuclear distribution and partially colocalized with giantin, although giantin had a wider distribution pattern (Fig. 1B). Golgin-97,
Golgin-160, and Golgin-95/GM130 stained a compact perinuclear tubulelike structure very similar to the label obtained with giantin (Fig. 1, C, D, and E, respectively). However, the giantin-containing tubules appeared to protrude
towards the plasma membrane and did not wrap around
each other, as in the case of the Golgin proteins. Evidence
supporting the localization of these antigens in the Golgi
apparatus came from cells treated with the fungal metabolite brefeldin A. Our results confirm previous findings
showing a rapid redistribution of Golgin-97 to the endoplasmic reticulum (ER) throughout the cell cytoplasm (Fig.
1G). Giantin undergoes a similar redistribution but with an
increase in signal over the nucleus that may represent a
preferential localization of the protein in the ER cisternae
forming the nuclear envelope (Fig. 1F).
92
MORENO ET AL.
FIG. 2. Distribution of Giantin and b-COP in rhesus monkey spermatocytes and spermatids. Rhesus monkey spermatogenic cells were attached to
poly-L-lysine-coated coverslips and fixed in 2% formaldehyde for 1 h. Following permeabilization with 1% Triton X-100, the Golgi apparatus was
visualized by immunofluorescence using antibodies against giantin (green; A–D) or the vesicular protein b-COP (red; A9–D9). The Golgi apparatus in
secondary spermatocytes is composed of two half-circles facing each other through their concave sides (A and A9). This arrangement produces a rounded
shape with some material inside, clearly seen in those cells stained with anti-giantin antibody but not with anti-b-COP (A9). In spermatids, both giantin
and b-COP display a horseshoelike appearance, with the concave side facing the acrosome (B and B9). At this stage, it seems that both proteins are
also located on the acrosome (B and B9). This distribution is maintained during stages 6–7, when the acrosomal vesicle has already started to flatten
and spread over the nucleus (C and C9). At later stages of differentiation, the Golgi apparatus migrates towards the opposite pole of the cell, and giantin
and b-COP are no longer detected on the acrosome (D and D9). Bar 5 5 mm.
spermatids were treated with 7.5 mM brefeldin A. The effect
of the drug was tested at two differentiation stages: Golgi
phase (Fig. 3, A and C) and cap phase (Fig. 3, B and D)
spermatids. Both cell types showed a rapid redistribution of
giantin throughout the cytoplasm after brefeldin A treatment.
Restoration of Golgi and vesicular staining was observed 90
min after removal of brefeldin A (data not shown). The effects of brefeldin A on spermatids were therefore in accordance with what has been described for somatic cells.
In the bovine round spermatids, the Golgi had a crescentlike shape as visualized with anti-Golgin-97 and anti-
Golgin-95/GM130 antibodies (Fig. 4, A and B). In elongating spermatids, besides staining the Golgi itself, both
antibodies gave a strong signal extending along the acrosomal vesicle (Fig. 4, C and D). This acrosomal localization
is similar to the distribution pattern of the proteins forming
the perinuclear theca, a cytoskeletal structure implicated in
acrosomal biogenesis (Fig. 4E) [28]. Testicular spermatozoa
did not show any labeling with either Golgi-97 or Golgin95/GM130 (Fig. 4, F and G).
Immunogold electron microscopy of bull spermatids
showed that Golgin-95/GM130 localized in a vesicle-rich
GOLGI APPARATUS IN MAMMALIAN SPERMATOGENESIS
93
FIG. 3. Effect of brefeldin A in rhesus monkey spermatids. Rhesus monkey spermatids were attached to poly-L-lysine-coated glass coverslips and
incubated with brefeldin A (5 mg/ml) in D-MEM for 30 min at 378C. The cells were then fixed, permeabilized, and stained for giantin. Two different
stages of differentiation are depicted. At stages 3–4 (A and C) and stages 6–7 (B and D), giantin is distributed throughout the cell (C and D), with no
typical horseshoelike pattern near the acrosomal vesicle (A and B; compare with Fig. 3, B and C). Because brefeldin A disrupts ER-to-Golgi traffic, these
results are interpreted as giantin being incorporated in the spermatid ER. Bar 5 5 mm.
FIG. 4. Imaging the Golgi apparatus and acrosome in bull spermatids. Bull spermatids were mounted onto poly-L-lysine-coated coverslips, fixed, and
stained with antibodies against acrosin (red) and Golgin-95 (A and C) or Golgin-97 (B and D). At the round spermatid stage (A and B), both Golgin95/GM130 and Golgin-97 display a similar distribution in the Golgi apparatus. In elongating spermatids (C and D), both antibodies label both the
acrosome vesicle and the Golgi apparatus, localized on the other side of the cell. This label is on the acrosome because it shows the same pattern as
an antibody against the perinuclear theca (green, E). In both round and elongating spermatids, the Golgi shows a horseshoe-shaped structure that does
not change in elongating spermatids (A and C). Neither Golgin-95 nor Golgin-97 label a fully differentiated testicular sperm (F and G). Bar 5 5 mm.
94
MORENO ET AL.
FIG. 5. Localization of Golgin-95 and b-COP in bull spermatids by immunogold electron microscopy. Golgin-95 (A–C) and b-COP (D–H) were
immunolocalized by electron microscopy using 5-mm colloidal gold particles. Golgin-95 shows a strong label over the Golgi apparatus (A) and a
weaker label on the acrosomal membrane (B and C). b-COP shows a weak label on the acrosomal membrane (D and E) and a stronger label in
cytoplasmic vesicles near the acrosomal vesicle (G). b-COP is also localized at the rim of Golgi stacks (E, arrowhead). The Golgi apparatus in round
spermatids appears associated with the axoneme (A and E, arrow). Neither b-COP (F) nor Golgin-95 (B) labeled the acrosomal granule (asterisk) or any
other structure in testicular sperm (H).
area near the nascent sperm tail (Fig. 5A). Golgin-95/
GM130 localized at the surface of the acrosomal vesicle in
round spermatids (Fig. 5, B and C). However, b-COP was
also found over the acrosomal vesicle in round spermatids
(Fig. 5, D and F). In addition, cytoplasmic vesiclelike profiles near the acrosome and the rims of the Golgi stacks
were highly enriched in b-COP label (Fig. 5, E and G).
Neither b-COP nor Golgin-95/GM130 was detected in tes-
ticular sperm (Fig. 5H). The label was specific because no
gold particles were found in other spermatid organelles,
such as the nucleus or mitochondria.
Distribution of Post-Golgi and Sorting Compartments
in Rhesus Monkey Spermatids
In Golgi stage round spermatids, TGN38 was concentrated at the apical pole cell over the acrosomal vesicle (Fig.
GOLGI APPARATUS IN MAMMALIAN SPERMATOGENESIS
95
sphyngolipids. Fluorescent lipid probes such as BODIPY
FL C5-ceramide or NBD C6-ceramide can easily enter into
the cell and then be metabolized in the Golgi apparatus and
incorporated into its membrane system. These probes gave
a pattern very similar to the one described for giantin and
Golgin-97 in live 6308 cells (Fig. 7, A and B). To visualize
the Golgi apparatus in live rhesus monkey spermatids, we
used the lipid probe BODIPY FL C5-ceramide. This probe
gave a clear and bright signal in round spermatids, labeling
a horseshoe-shaped Golgi apparatus with the convex side
facing the acrosomal vesicle (Fig. 7C). In cap phase spermatids, BODIPY FL C5-ceramide-labeled Golgi faced the
acrosomal vesicle, which at this stage attached to the nucleus and started to expand (Fig. 7D). In early elongating
spermatids (cap phase), the Golgi apparatus was at the opposite pole of the acrosome (Fig. 7D). In contrast to b-COP
and giantin, the fluorescent lipid probes never labeled the
acrosomal vesicle. The inserts show the phase contrast image of each cell.
FIG. 6. Localization of post-Golgi compartments in rhesus monkey spermatids. Rhesus monkey round (Golgi phase) and early elongating (acrosome phase) spermatids are immunostained with antibodies against
TGN38 and b-COP (A and C) or clathrin and acrosin (B and D). In round
spermatids, b-COP (green) appears as a horseshoe-shaped structure facing
the acrosomal vesicle (A). TGN38 (red) is partially colocalized with bCOP at this stage but with a broader distribution pattern (A). In elongating
spermatids, the distribution of both proteins does not overlap (C). Clathrin
(red) has a wide distribution pattern in Golgi phase spermatids and overlies the acrosome as evidenced by acrosin staining (green). In elongating
spermatids (D), clathrin strongly labels the whole acrosomal vesicle. This
protein also shows a fuzzy cytoplasmic distribution in the area over the
acrosome and is present around the spermatid tail. Bar 5 5 mm.
6A). At this stage, TGN38 was in close association with
the Golgi apparatus, as shown by a partial colocalization
with b-COP. However, TGN38 had a wider distribution pattern than did b-COP vesicles (Fig. 6A). In elongating spermatids, the compartments labeled by TGN38 and b-COP
appeared completely distinct, without any visible overlap
(Fig. 6C). At this stage, the acrosomal vesicle did not contain b-COP or TGN38. We never detected surface localization of TGN38 in round or elongating spermatids.
Clathrin is another protein involved in post-Golgi trafficking and recycling of proteins between the TGN and
plasma membrane. In round spermatids, clathrin was present in a region overlying the acrosome with great intensity
and was found as scattered dots around the nucleus and
near the cell surface (Fig. 6B). However, clathrin was not
in the region occupied by the Golgi apparatus at this stage.
In elongating spermatids, the label of clathrin followed the
shape of the acrosomal vesicle, avoiding the region of the
acrosomal granule (Fig. 6D). In addition, the protein was
also found in a dotted pattern over the acrosomal vesicle
and in large aggregates of vesicles around the spermatid
tail (Fig. 6D) but could not be detected in testicular sperm
(data not shown).
Live Imaging of the Golgi Apparatus
During Spermatogenesis
The Golgi apparatus is an organelle that contains a particularly high concentration of phosphoinositidiios and
DISCUSSION
Spermatogenesis is a complex process that involves nuclear remodeling, cytoplasmic movements and the formation of new and unique organelles [1, 10]. The haploid
round spermatid becomes polarized in the first stages of
differentiation, as evidenced by the asymmetric distribution
of its organelles [10, 29, 30]. Plasma membrane domains
that will eventually perform specialized functions during
fertilization are also generated at this stage [31]. However,
little is known of the protein sorting and mechanisms involved in the formation of the acrosome and the different
sperm head domains.
In this work, we have shown that four Golgi proteins,
giantin, b-COP, Golgin-97, and Golgin-95/GM130, are present in primary pachytene spermatocytes and round and
elongating spermatids. In round spermatids, these proteins
are localized in a horseshoe-shaped Golgi apparatus and in
membranes surrounding the acrosomal vesicle. In elongating spermatids, giantin, b-COP, Golgin-97, and Golgin-95/
GM130 are no longer found in the acrosome, suggesting
that they may have been retrieved into the Golgi. At the
final stages of differentiation, these proteins are shed in the
discarded cytoplasmic droplet [19]. In somatic cells, Golgin-95/GM130 interacts with GRASP65 to be correctly targeted to the Golgi apparatus [32]. GRASP65 is a membrane-associated protein involved in the reassembly of Golgi stacks in a cell-free system [32, 33]. In mammalian spermatids, both GRASP65 and Golgin-95/GM130 may keep
this interaction in order to bind to the acrosomal membrane.
The localization of this complex in the acrosome may allow
the proper targeting of b-COP vesicles to this organelle.
The shape of the Golgi apparatus varies among different
cell types and is cell cycle dependent [16, 34]. Biogenesis
and maintenance of this organelle is dependent on protein
transport through the cisternae [16]. Using the fungal metabolite brefeldin A, we showed here that there is an active
ER-to-Golgi traffic in round and elongating spermatids. The
kinetics of giantin redistribution to the ER in the presence
of the brefeldin A and the dose required to achieve this
effect are comparable to those found in somatic cells [35–
37]. Thus, it seems that the formation of b-COP-coated
vesicles and anterograde/retrograde trafficking pathways
are similar in both systems.
Formation of the mammalian acrosome is a very slow
process when compared with the biogenesis of other secretory vesicles, such as insulin granules [9]. Spermiogenesis,
96
MORENO ET AL.
FIG. 7. Live imaging of the Golgi apparatus in rhesus monkey spermatids. NBD C6-ceramide (A) and BIODIPY FL-C5-ceramide (B) are fluorescent
lipids that specifically stain the Golgi apparatus in rhesus 6308 cells. Live rhesus monkey spermatids are labeled with the fluorescent lipid BIODIPY
FL-C5-ceramide. The plate depicts early spermatids in stages 3–4 (C), 5–7 (D), and 9–10 (E). Fluorescence is observed in a horseshoe-shaped Golgi
apparatus (C and D) near the acrosomal vesicle (E, arrow). The Golgi apparatus faces the acrosomal vesicle at a later stage of differentiation and then
migrates towards the opposite pole of the cell, as already described using other Golgi markers (C–E). Insert shows the phase contrast image. Bar 5 5
mm.
FIG. 8. Role of vesicular trafficking in acrosome biogenesis. In early
round spermatids (stages 1–4; A), there is an anterograde Golgi-to-acrosome vesicular traffic. Incoming vesicles bind preferentially to the border
of the acrosomal vesicle, thus increasing the surface area attached to the
nucleus as the acrosome expands and flattens. At this stage, few clathrin
(red) and b-COP (green) vesicles are budding from the acrosomal membrane. Clathrin coats are also found in vesicles budding from the TGN,
and b-COP is also detected at the rims of the Golgi cisternae. At later
stages (stages 5–8; B), the budding rate of clathrin- and b-COP-coated
vesicles from the acrosome increases. At this time, the anterograde vesicle
flow begins to decrease, and the Golgi apparatus starts to migrate towards
the opposite pole of the cell.
the process of spermatid differentiation, may take from 14
days in mice up to 23 days in humans [38]. Acrosome
formation accounts for about 50% of the total time required, which may be due to either slow vesicular traffic
or slow interaction kinetics between the different docking/
fusion machinery components. It is not known when or
how acrosome components are sorted out from proteins
with other destinations. During spermiogenesis, there are a
number of lysosomal proteins detected in the acrosomal
vesicle that are retrieved subsequently throughout differentiation [39, 40]. Clathrin- and probably also b-COP1coated vesicles may participate in the retrieval of missorted
proteins from the acrosome. The acrosome of mature sperm
contains both unique acrosomal enzymes and common enzymes of lysosomal origin [41]. Many of these enzymes
may have no functional role in fertilization, and their presence in the acrosome may merely reflect the low efficiency
of the spermatid sorting machinery. This process may be
similar to the processing and maturation of prosecretory
granules in endocrine cells [42, 43]. Besides having a role
in the retrieval of misrouted proteins from the acrosomal
vesicle, vesicular trafficking may also have a profound influence in the modeling and shaping of the acrosome. During stages 4–7 of spermatogenesis, the acrosomal vesicle
flattens and spreads around the nucleus, finally covering up
GOLGI APPARATUS IN MAMMALIAN SPERMATOGENESIS
to two-thirds of its surface at stages 10–12 in the mouse or
the equivalent stages in other mammals [1, 30, 44]. In addition, the mechanism by which the species-specific final
shape of the acrosome is reached is not known [1, 30, 44].
We propose that acrosome flattening and spreading over the
nucleus may be the result of vesicle fusion at the edges of
the acrosome coupled with membrane retrieval at the center. This appears to be the case in early rhesus monkey
round spermatids. According to this model (Fig. 8), there
is a net Golgi-to-acrosome membrane flux during the first
stages of differentiation (Fig. 8A) that is responsible for the
initial growth of the organelle. Over time, Golgi-to-acrosome traffic begins to decline, and there is a concomitant
increase in the opposite acrosome-to-Golgi pathway. The
spermatid constantly adjusts both trafficking routes until the
acrosome has flattened and the Golgi apparatus has migrated towards the opposite pole of the cell (Fig. 8B).
18.
19.
20.
21.
22.
23.
ACKNOWLEDGMENT
We thank Michael Webb for the preparation of ultrathin sections.
24.
REFERENCES
1. Burgos M, Fawcett D. Studies on the fine structure of the mammalian
testis. I. Differentiation of the spermatid in the cat (Felis domestica).
J Biophys Biochem 1955; 1:287–313.
2. de Kretser D, Kerr J. The cytology of the testis. In Knobil E, Neil J
(eds.), The Physiology of Reproduction. New York: Raven Press;
1988: 837–919.
3. Burgos MH, Gutierrez LS. The Golgi complex of the early spermatid
in guinea pig. Anat Rec 1986; 216:139–145.
4. Anakwe OO, Gerton GL. Acrosome biogenesis begins during meiosis:
evidence from the synthesis and distribution of an acrosomal glycoprotein, acrogranin, during guinea pig spermatogenesis. Biol Reprod
1990; 42:317–328.
5. Escalier D, Gallo JM, Albert M, Meduri G, Bermudez D, David G,
Schrevel J. Human acrosome biogenesis: immunodetection of proacrosin in primary spermatocytes and of its partitioning pattern during
meiosis. Development 1991; 113:779–788.
6. Kashiwabara S, Arai Y, Kodaira K, Baba T. Acrosin biosynthesis in
meiotic and postmeiotic spermatogenic cells. Biochem Biophys Res
Commun 1990; 173:240–245.
7. Klumperman J, Kuliawat R, Griffith JM, Geuze HJ, Arvan P. Mannose
6-phosphate receptors are sorted from immature secretory granules via
adaptor protein AP-1, clathrin, and syntaxin 6-positive vesicles. J Cell
Biol 1998; 141:359–371.
8. Susi FR, Leblond CP, Clermont Y. Changes in the Golgi apparatus
during spermiogenesis in the rat. Am J Anat 1971; 130:251–267.
9. Tooze SA. Biogenesis of secretory granules in the trans-Golgi network
of neuroendocrine and endocrine cells. Biochim Biophys Acta 1998;
1404:231–244.
10. Russell L, Ettlin R, Hikim A, Clegg E. Histological and histopathological evaluation of the testis, 2nd ed. Clearwater, FL: Cache River
Press; 1990.
11. Kallajoki M, Parvinen M, Suominen JJ. Expression of acrosin during
mouse spermatogenesis: a biochemical and immunocytochemical
analysis by a monoclonal antibody C 11 H. Biol Reprod 1986; 35:
157–165.
12. Westbrook-Case VA, Winfrey VP, Olson GE. Sorting of the domainspecific acrosomal matrix protein AM50 during spermiogenesis in the
guinea pig. Dev Biol 1995; 167:338–349.
13. Tang XM, Lalli MF, Clermont Y. A cytochemical study of the Golgi
apparatus of the spermatid during spermiogenesis in the rat. Am J
Anat 1982; 163:283–294.
14. Lowe M, Kreis TE. Regulation of membrane traffic in animal cells
by COPI. Biochim Biophys Acta 1998; 1404:53–66.
15. Scales SJ, Pepperkok R, Kreis TE. Visualization of ER-to-Golgi transport in living cells reveals a sequential mode of action for COPII and
COPI. Cell 1997; 90:1137–1148.
16. Lippincott-Schwartz J, Cole NB, Donaldson JG. Building a secretory
apparatus: role of ARF1/COPI in Golgi biogenesis and maintenance.
Histochem Cell Biol 1998; 109:449–462.
17. Martinez MJ, Geuze HJ, Ballesta J. Identification of two types of beta-
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
97
COP vesicles in the Golgi complex of rat spermatids. Eur J Cell Biol
1996; 71:137–143.
Moreno RD, Ramalho-Santos J, Chan EKL, Schatten G. Molecular
membrane marker dynamics during rhesus spermatogenesis: acrosome
biogenesis, Golgi apparatus and mitochondria differentiation. Mol
Biol Cell 1998; 9:66a.
Moreno RD, Ramalho-Santos J, Chan EKL, Wessel G, Schatten G.
The Golgi apparatus segregates from the lysosomal/acrosomal vesicle
during rhesus spermiogenesis: structural alterations. Dev Biol 2000;
219:334–349.
Dietert S. Fine structure of the formation and fate of the residual body
of mouse spermatozoa with evidence for the participation of lysosomes. J Morphol 1922; 120:317–346.
Sakai Y, Yamashina S. Mechanism for the removal of residual cytoplasm from spermatids during mouse spermiogenesis. Anat Rec 1989;
223:43–48.
Griffith KJ, Chan EK, Lung CC, Hamel JC, Guo X, Miyachi K, Fritzler MJ. Molecular cloning of a novel 97-kd Golgi complex autoantigen associated with Sjogren’s syndrome. Arthritis Rheum 1997; 40:
1693–1702.
Fritzler MJ, Hamel JC, Ochs RL, Chan EK. Molecular characterization of two human autoantigens: unique cDNAs encoding 95- and
160-kD proteins of a putative family in the Golgi complex. J Exp
Med 1993; 178:49–62.
Bavister BD, Boatman DE, Leibfried L, Loose M, Vernon MW. Fertilization and cleavage of rhesus monkey oocytes in vitro. Biol Reprod
1983; 28:983–999.
Sutovsky P, Ramalho-Santos J, Moreno RD, Oko R, Hewitson L,
Schatten G. On-stage selection of single round spermatids using a
vital, mitochondrion-specific fluorescent probe MitoTracker(TM) and
high resolution differential interference contrast microscopy. Hum Reprod 1999; 14:2301–2312.
Pagano RE, Martin OC, Kang HC, Haugland RP. A novel fluorescent
ceramide analogue for studying membrane traffic in animal cells:
accumulation at the Golgi apparatus results in altered spectral properties of the sphingolipid precursor. J Cell Biol 1991; 113:1267–
1279.
Sonnichsen B, Lowe M, Levine T, Jamsa E, Dirac SB, Warren G. A
role for giantin in docking COPI vesicles to Golgi membranes. J Cell
Biol 1998; 140:1013–1021.
Oko R, Morales CR. A novel testicular protein, with sequence similarities to a family of lipid binding proteins, is a major component of
the rat sperm perinuclear theca. Dev Biol 1994; 166:235–245.
Hermo L, Rambourg A, Clermont Y. Three-dimensional architecture
of the cortical region of the Golgi apparatus in rat spermatids. Am J
Anat 1980; 157:357–373.
Sinowatz F, Wrobel KH. Development of the bovine acrosome. An
ultrastructural and cytochemical study. Cell Tissue Res 1981; 219:
511–524.
Pelletier RM, Friend DS. Development of membrane differentiations
in the guinea pig spermatid during spermiogenesis. Am J Anat 1983;
167:119–141.
Barr FA, Nakamura N, Warren G. Mapping the interaction between
GRASP65 and GM130, components of a protein complex involved in
the stacking of Golgi cisternae. EMBO J 1998; 17:3258–3268.
Shorter J, Warren G. A role for the vesicle tethering protein, p115, in
the post-mitotic stacking of reassembling Golgi cisternae in a cell-free
system. J Cell Biol 1999; 146:57–70.
Warren G. Membrane partitioning during cell division. Annu Rev
Biochem 1993; 62:323–348.
Lippincott-Schwartz J, Yuan LC, Bonifacino JS, Klausner RD. Rapid
redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell
1989; 56:801–813.
Lippincott-Schwartz J, Cole N, Presley J. Unravelling Golgi membrane traffic with green fluorescent protein chimeras. Trends Cell Biol
1998; 8:16–20.
Torii S, Banno T, Watanabe T, Ikehara Y, Murakami K, Nakayama K.
Cytotoxicity of brefeldin A correlates with its inhibitory effect on
membrane binding of COP coat proteins. J Biol Chem 1995; 270:
11574–11580.
Courot M, Hochereau-de Reviers MT, Ortavant R. Spermatogenesis.
In: Johnson AD, Gomes WR, Vandemark NL (eds.), The Testis. New
York: Academic Press; 1970: 339–432.
Griffiths G, Warren G, Stuhlfauth I, Jockusch BM. The role of clathrin-coated vesicles in acrosome formation. Eur J Cell Biol 1981; 26:
52–60.
98
MORENO ET AL.
40. Jin QS, Kamata M, Garcia del Saz E, Seguchi H. Ultracytochemical
study of trimetaphosphatase activity during acrosomal formation in
the mouse testis. Histol Histopathol 1995; 10:681–689.
41. Tulsiani DR, Abou-Haila A, Loeser CR, Pereira BM. The biological
and functional significance of the sperm acrosome and acrosomal
enzymes in mammalian fertilization. Exp Cell Res 1998; 240:151–
164.
42. Kuliawat R, Arvan P. Distinct molecular mechanisms for protein sort-
ing within immature secretory granules of pancreatic beta-cells. J Cell
Biol 1994; 126:77–86.
43. Kuliawat R, Klumperman J, Ludwig T, Arvan P. Differential sorting
of lysosomal enzymes out of the regulated secretory pathway in pancreatic beta-cells. J Cell Biol 1997; 137:595–608.
44. Berndston WE, Desjardins C. The cycle of the seminiferous epithelium and spermatogenesis in the bovine testis. Am J Anat 1974; 140:
167–179.