BIOLOGY OF REPRODUCTION 58, 361-370 (1998)
Acrosome Biogenesis in the Hamster: Ultrastructurally Distinct Matrix Regions Are
Assembled from a Common Precursor Polypeptide'
Gary E. Olson,2 Virginia P. Winfrey, and Subir K. NagDas
Department of Cell Biology, Vanderbilt University, Nashville, Tennessee 37232
ABSTRACT
The hamster sperm acrosome contains a stable matrix complex that binds specific hydrolases and appears to regulate their
release during the acrosome reaction. This complex comprises
two contiguous but ultrastructurally distinct regions that are
segregated to specific sites within the acrosome. In this study,
we define the temporal expression, processing, and localization
of major matrix proteins of 29 kDa (AM29) and 22 kDa (AM22)
during spermiogenesis and post-testicular sperm maturation in
the epididymis. Peptide mapping, N-terminal microsequence
analysis, immunoblotting, and immunocytochemistry were used
to demonstrate that AM29 and AM22 of mature spermatozoa
are structurally related and appear to arise from a common 40kDa precursor protein expressed in round spermatids. A monoclonal antibody that recognized only the mature forms of the
matrix proteins and a polyclonal antibody that recognized both
the precursor and fully processed matrix proteins were prepared
and used to demonstrate that the precursor protein is present
in the acrosome of round spermatids and that it undergoes size
processing during the terminal stages of spermiogenesis so that
the mature matrix polypeptides are evident in epididymal spermatozoa. Finally, using light and electron microscopic immunocytochemistry, we demonstrated that the matrix polypeptides
are excluded from the equatorial segment and are localized to
both structurally distinct matrix domains of the mature acrosome. These data show that processing of the major proteins of
the acrosomal matrix occurs in a temporally regulated fashion
after their transport to the acrosome and that the processed
products can assemble into ultrastructurally distinct matrix elements.
INTRODUCTION
The mammalian sperm acrosome is a membrane-bounded organelle containing several hydrolases that facilitate
penetration of the egg investments [1-4]. These enzymes
are released to the extracellular environment during the acrosome reaction, a receptor-regulated exocytotic process involving fusion between the periacrosomal plasma membrane and the outer acrosomal membrane [5, 6]. Cell fractionation experiments have demonstrated a difference in the
solubility properties of specific acrosomal proteins, and
during the acrosome reaction specific hydrolases exhibit
different temporal release patterns [7-13]. For example, in
guinea pig spermatozoa, dipeptidyl peptidase is rapidly released during the acrosome reaction, and this enzyme is
sequestered to a specific domain along the dorsal surface
of the acrosomal apical segment [9]. In contrast, a considerable amount of proacrosin is associated with a particulate
sperm fraction, and it is released more slowly during the
acrosome reaction [7, 8, 11, 13, 14]. This particulate nature
and temporal release of different hydrolases may reflect a
sequential requirement for specific hydrolases as sperm encounter different elements of the egg vestments during fertilization. The differences in solubility properties and release rates between acrosomal proteins appear to reflect
their compartmentalization within the acrosome and their
association with, or assembly into, stable acrosomal matrix
assemblies [9-11, 14, 15].
In several mammalian species, the acrosomal contents
are segregated into spatially distinct domains of differing
ultrastructural appearance [1, 16, 17]. In the guinea pig,
specific acrosomal matrix polypeptides localized to restricted domains of the apical segment have been identified and
characterized [15, 17-20]. This suggests that regionally localized matrix assemblies may partition the acrosomal interior into distinct functional domains. The acrosomal contents of hamster spermatozoa also exhibit regional differences in structural appearance [21-23], and specific elements of the acrosomal matrix remain intact and associated
with the hybrid membrane complex after the acrosome reaction [5, 24]. These acrosomal matrix elements also resist
solubilization by various extraction regimens, emphasizing
their structural stability [10, 22, 23, 25]. We previously purified a stable acrosomal assembly from cauda epididymal
hamster spermatozoa that comprises two ultrastructurally
distinct acrosomal matrix domains and the detergent-insoluble membrane skeleton of the outer acrosomal membrane,
and we demonstrated its ability to bind both proacrosin and
N-acetylglucosaminidase [10, 13]. This acrosomal structure
was termed the acrosomal lamina-matrix (ALM) complex.
The two matrix components of the ALM adhere to different
regions of the outer acrosomal membrane and are spatially
restricted within the acrosome [10, 22]. Thus the ALM represents a stable infrastructure that may segregate hydrolases
within the acrosome and regulate their release during the
acrosome reaction; moreover, it may also represent a cytoskeleton-like framework that affects the vesiculation of
the outer acrosomal membrane during the acrosome reaction and that maintains the integrity of the shed acrosomal
cap. The molecular mechanisms that regulate the assembly
and define the precise spatial distribution of distinct matrix
elements within the acrosome are poorly understood. In this
study, we identify the temporal expression, the molecular
relationships, and the spatial distribution of the major acrosomal matrix polypeptides in hamster spermatids and
spermatozoa.
MATERIALS AND METHODS
Animals
Care and use of animals conformed to NIH guidelines
for humane animal care and use in research. All protocols
were approved by the institutional Animal Care Committee,
and animals were housed under the supervision of University Veterinarians in an AAALAC-approved Central Animal Care Facility. Hamsters and mice were killed by CO 2
Accepted September 15, 1997.
Received July 21, 1997.
'Supported by NIH grant HD20419; support for Morphology and Protein Sequencing Core Laboratories provided by Center Grant HD05797.
2
Correspondence. FAX: (615) 343-4539;
e-mail: [email protected]
361
362
OLSON ET AL.
asphyxiation, and guinea pigs were killed with methoxyflurane.
Preparation of Sperm Suspensions
The caput and cauda epididymides were separated and
minced in Hank's saline or Tyrode's solution (Sigma Chemical Co., St. Louis, MO) at 37°C. Cauda sperm were pelleted by centrifugation for 10 min at 500 x g. Caput sperm
were purified by centrifugation for 5 min at 650 x g on a
step gradient of 40%, 60%, and 80% Percoll (Pharmacia
Biotech, Piscataway, NJ); the spermatozoa at the 60%-80%
interface were diluted with Tyrode's solution and pelleted
g for 10 min.
by centrifugation at 500
Isolation of Acrosomal Fraction
The ALM complex was isolated from cauda epididymal
spermatozoa as described previously [10, 13]. Sperm from
the cauda epididymides of 4 animals were resuspended into
10 volumes of ice-cold extraction solution (T-TNI) composed of 0.1% Triton X-100 in TNI (TNI = 150 mM NaCl,
25 mM Tris-HCl [pH 7.0], 2 mM benzamidine, 1 pxg/ml
leupeptin, and 1 xg/ml pepstatin) and immediately centrifuged at 500
g for 10 min. The pellet was extracted for
30 min at 4C with T-TNI and homogenized with 10-20
strokes of a glass silicone-coated homogenizer to detach the
ALM complex from the sperm heads. Twenty milliliters of
the sperm suspension was mixed with 100 ml of 45% Percoll, 0.25 M sucrose, 0.1% Triton X-100, and 25 mM Trisg for 35
HC1 (pH 7.0), and then centrifuged at 60 000
min in a Beckman 60Ti rotor (Beckman Instruments, Palo
Alto, CA). The ALM fraction formed a band at the top of
the gradient which was collected, diluted with TNI, and
pelleted at 100 000 X g for 20 min in a Beckman TL55
rotor.
Spermatogenic Cell Isolation
Spermatogenic cells were purified by unit gravity sedimentation [26, 27]. Testes were minced and rinsed twice in
Krebs-Ringer-bicarbonate (KRB) medium containing both
nonessential and essential amino acids. The tissue was then
incubated for 30 min at 37C in KRB containing 1 mg/ml
collagenase (Sigma Chemical Co.) with gentle oscillation
and then incubated for 10 min in KRB containing 1 mg/ml
trypsin and 0.1 U/ml micrococcal nuclease (Sigma Chemical Co.). After 3 rinses with KRB, the seminiferous tubule
fragments were resuspended in KRB containing 0.5% BSA,
gently pipetted, and filtered through a 40-iM mesh sieve
(Falcon 2340; Falcon Plastics, Los Angeles, CA) to obtain
a suspension of single cells. Cells were counted and assessed for viability by Trypan blue dye exclusion, and samples exhibiting > 90% viability were used for unit gravity
sedimentation. Cells were separated for 2 h at 40C on a 24% gradient of BSA in KRB medium using a Celsep apparatus (Brinkman Instruments, Westbury, NY). Purity of
gradient fractions was determined by phase contrast and/or
differential interference contrast (DIC) microscopy. Fractions containing > 80% round spermatids were pooled, and
the cells were pelleted by centrifugation for 10 min at 500
X g.
Antibody Preparation
To prepare monoclonal antibodies, BALB/c mice were
immunized with 25-50 g of the purified ALM fraction
emulsified in Freund's adjuvant. Animals received two
booster injections at 2-wk intervals followed by a final tailvein injection of 5 g ALM suspended in PBS. After four
days, animals were killed, and their splenocytes were fused
with X63-Ag 8.653 myeloma cells (ATCC, Rockville, MD)
using polyethylene glycol [28]. Hybrid cells were selected
by culture in RPMI 1640 HAT (hypoxanthine, aminopterin,
thymidine) medium (Sigma Chemical Co.), and positive
wells were identified by ELISA. Positive cells were cloned
twice by limiting dilution and used to prepare ascites fluid
[28].
Polyclonal antiserum against the 22-kDa ALM polypeptide (AM22) was prepared in guinea pigs. The ALM fraction was subjected to preparative SDS-PAGE [29], polypeptides were visualized by staining with CuC12 [30], and
the AM22 band was excised. Guinea pigs received a primary injection and two booster injections, at 2-wk intervals, of AM22 emulsified in Freund's adjuvant. Either
whole serum or an IgG fraction, purified by protein A Sepharose affinity chromatography [31], was used for immunocytochemistry or immunoblotting.
SDS-PAGE, Western blotting, Peptide Mapping, and NTerminus Sequencing
SDS-PAGE [29] was performed on gradient or continuous gels as specified in the text. For peptide mapping,
polypeptide bands were visualized by CuC12 staining [30],
excised, and equilibrated with 0.1% SDS, 1 mM EDTA, 2
mg/ml dithiothreitol (DTT), 20% glycerol, and 25 mM TrisHC1, pH 6.8. The polypeptides were then fractionated on
15% SDS-polyacrylamide gels in the presence of V8 protease [32]. Peptide fragments were electrophoretically
transferred to polyvinylidene difluoride (PVDF) membranes
[33] and stained with Coomassie blue dye for N-terminus
sequencing with an Applied Biosystems (Foster City, PA)
475S sequencer [34]. Peptide sequences were analyzed with
the National Center for Biotechnology Information (NCBI)
BLAST program using the blastp database [35].
Immunoblot Staining
Protein patterns on Western blots [33] were visualized
using Coomassie blue dye or colloidal gold [36]. Immunoblots were incubated in 5% goat serum, 2.5% BSA, 0.1%
Tween 20 in PBS (150 mM NaCl, 20 mM sodium phosphate, pH 7.6) to block nonspecific binding sites. After
three rinses in PBS containing 0.05% Tween 20 and 1%
goat serum (PBS-Tw-GS), blots were incubated in primary
antibody diluted in PBS-Tw-GS; controls received equivalent dilutions of nonimmune serum or normal ascites fluid.
After three washes in PBS-Tw-GS, the blots were incubated
in an affinity-purified horseradish peroxidase-conjugated
secondary antibody (KPL Inc., Gaithersburg, MD) diluted
in PBS-Tw-GS. Immunoreactive bands were visualized using diaminobenzidine for color development.
Immunofluorescence
Cells were fixed for 30 min on ice with 4% formaldehyde in 0.1 M sodium phosphate, pH 7.4, and then permeabilized for 30 min by dilution into an equal volume of
fixative containing 0.2% Triton X-100. In some experiments, spermatozoa were permeabilized either by incubation in 0.1% Triton X-100 in TNI for 30 min at 4C or by
nitrogen cavitation at 400 psi for 10 min [10] before fixation. After adhering to poly-L-lysine-coated coverslips, cells
were rinsed in PBS and blocked with PBS containing 5%
ACROSOMAL MATRIX ASSEMBLY
363
FIG. 1. A) Electron micrograph showing
cross section through the head region of a
cauda epididymal spermatozoon. The acrosomal cap overlies the dorsal surface of
the nucleus (n). Note that the acrosomal
contents located dorsally appears less
electron-dense than the acrosomal contents located ventrally (*). B) Electron micrograph of cross section of Triton X-1 00extracted sperm head showing an acrosomal lamina-matrix complex separating
from the underlying nucleus (n). Each
complex consists of two contiguous acrosomal matrix domains of differing electron
density (ml and m2) and detergent-insoluble elements of the outer acrosomal membrane, termed the acrosomal lamina (al),
which remains adherent to the ml matrix.
C) Electron micrograph of purified ALM
fraction; note the absence of other contaminating sperm structures. A and B,
x48 000; C, x18 000 (reproduced at
69%). al, Acrosomal lamina; oa, outer acrosomal membrane; ia, inner acrosomal
membrane; es, equatorial segment; pm,
plasma membrane; ps, postacrosomal segment.
FIG. 2. Coomassie blue dye-stained PVDF blot of the ALM fraction separated by SDS-PAGE on a 7.5%-15% gradient gel demonstrates that polypeptides of 29 kDa (AM29) and 22 kDa (AM22) are major polypeptides
of the acrosomal fraction. Molecular weight markers (x 10- 3 ) are at left.
goat serum and 2.5% BSA. Coverslips were then rinsed
three times in PBS containing 1% goat serum (PBS-GS)
and incubated in primary antibody diluted in PBS-GS; controls received equivalent dilutions of preimmune serum or
normal ascites fluid. After three washes in PBS-GS, the
coverglasses were incubated in PBS-GS containing affinity-
FIG. 3. Immunoblot analysis of ALM fraction separated on 7.5%-15%
gradient gel; 5 Rg of protein was loaded per lane. Lane 1, immunostained
with polyclonal anti-AM22, reveals staining of both AM29 and AM22 and
an adjacent set of minor polypeptide bands. Lane 2, stained with preimmune serum, shows no immunoreactive bands. Lane 3, immunostained
with monoclonal anti-AM29/22, shows the same set of immunoreactive
bands as noted in lane 1. Lane 4, immunostained with normal ascites
fluid, shows no immunoreactive bands. Molecular weight markers (x
10-3 ) are at left.
364
OLSON ET AL.
FIG. 5. Immunoblot of 12% SDS-PAGE minigel showing total caput
sperm lysate (lane 1), total cauda sperm lysate (lane 2), Triton X-100soluble cauda sperm proteins (lane 3), and Triton X-100-insoluble cauda
sperm proteins (lane 4) stained with polyclonal anti-AM22. Each lane was
loaded with extracts representing equivalent numbers of sperm. Note the
similar patterns of immunoreactive polypeptides in lysates of caput and
cauda spermatozoa. Also note that all immunoreactive cauda sperm polypeptides are present in the pellet fraction. Molecular weight markers (x
10 3) are at left.
purified secondary antibodies of fluorescein isothiocyanate
(FITC)-conjugated anti-guinea pig IgG or Cy3-conjugated
anti-mouse IgG (KPL Inc. and Jackson Immuno Research,
West Grove, PA, respectively). Coverslips were then rinsed
several times in PBS, and cells were examined by phase
contrast and epifluorescence microscopy. For double-antibody immunostaining, cells were incubated with a mixture
of monoclonal and polyclonal primary antibodies and then
in a mixture of anti-guinea pig and anti-mouse secondary
antibodies conjugated with different fluorochromes. As a
control, each secondary antibody was tested individually
against both primary antibodies, and no species cross-reactivity was noted.
Immunoelectron Microscopy
FIG. 4. A) Peptide maps of AM29 and AM22 using V8 protease. SDSPAGE-purified AM 29 (lanes 1 and 2) and AM22 (lanes 3 and 4) were
incubated in the absence (lanes 1 and 3) or presence (lanes 2 and 4)
of V8 protease and then fractioned by SDS-PAGE on 15% acrylamide
gels, transferred to PVDF membranes, and stained with Coomassie
blue dye. Both AM29 and AM22 exhibit an identical pattern of V8
protease-generated polypeptides of 20, 16, 13, and 12 kDa (lanes 2
and 4). Molecular weight markers (x 10 3) are at left. B) N-terminal
sequences of intact AM22 and V8-digest polypeptides of both AM29
and AM22. Apparent molecular mass of V8 protease generated polypeptides are given in parentheses. Single letter codes are given, "X"
denotes that no definitive sequence was obtained. *Deduced sequence of mouse SP-10 residues 170-182 [37]. The GenBank accession number for the mSP-10 cDNA sequence is U31992. Single underlined letters represent conservative amino acid substitution between mSP-10 and hamster AM29 and/or AM22; double underlined
residue represents difference between mSP-10 and hamster AM29 and/or AM22.
For post-embedding immunolabeling, cells were fixed on
ice with 4% formaldehyde, 0.25% glutaraldehyde in 0.1 M
sodium phosphate buffer, pH 7.4. The samples were then
rinsed with phosphate buffer, dehydrated through an ethanol
series, and embedded in LR White Resin (Ted Pella Inc.,
Burlingame, CA). Thin sections were mounted on nickel
grids and immunostained as described for immunofluorescence except that gold-conjugated secondary antibodies
were used (Amersham Inc., Arlington Heights, IL). After
immunolabeling, the sections were washed in PBS, fixed
with 1% glutaraldehyde, rinsed with water, and stained with
uranyl acetate and lead citrate.
For pre-embedding immunolabeling, spermatozoa were
permeabilized and immunolabeled as for immunofluorescence except that gold-conjugated secondary antibodies
were used. After labeling, sperm were fixed with 4% glutaraldehyde in 0.1 M sodium cacodylate (pH 7.4), postfixed
with 1% osmium tetroxide in sodium cacodylate buffer, dehydrated in an ethanol series, equilibrated with propylene
oxide, and embedded in epoxy resin. Thin sections were
stained with uranyl acetate and lead citrate.
365
ACROSOMAL MATRIX ASSEMBLY
FIG. 6. Matched phase contrast and fluorescence photographs of immunostained
cauda epididymal spermatozoa. A, B)
Formaldehyde-fixed Triton X-1 00-permeabilized sperm stained with polyclonal antiAM22 showing intense fluorescence of the
acrosomal cap and absence of fluorescence of equatorial segment. C, D) Triton
X-100-pretreated spermatozoa immunostained with polyclonal anti-AM22. Note
the intense fluorescence of the partially
detached acrosomal caps while the underlying anterior head and equatorial segment
are negative. E,F)Triton X-1 00-pretreated
sperm immunostained with preimmune serum. No fluorescence is noted. G, H)
Spermatozoa permeabilized by nitrogen
cavitation and immunostained with polyclonal anti-AM22 show intense fluorescence of the partially detached acrosomal
caps, whereas the underlying sperm head
is negative. I, ) Triton X-1 00-pretreated
spermatozoa immunostained with monoclonal anti-AM29/22 show fluorescence of
the acrosomal cap, whereas all other
sperm structures are negative. ac, Acrosomal cap; es, equatorial segment; ah, anterior head; ps, postacrosomal segment; mp,
midpiece. Bar = 10 jIm.
RESULTS
Isolation of Acrosomal Fraction
The acrosomal matrix of intact spermatozoa exhibited
subtle regional variations in electron density; the matrix
positioned over the dorsal surface of the sperm head appeared less electron-dense than the matrix located ventrally
within the acrosome (Fig. 1A). Specific structural elements
of the acrosomal cap are released as an integrated complex
after extraction of spermatozoa with Triton X-100 [10, 13].
This assembly, termed the ALM complex, comprised three
acrosomal structures (Fig. B): the first a less electrondense matrix domain (ml) derived from the dorsal surface
of the acrosome, the second a more electron-dense matrix
domain (m2) located ventrally in the intact acrosome; and
the third the acrosomal lamina, which appeared to include
the detergent-insoluble membrane-skeleton of the outer acrosomal membrane and remained adherent to the dorsal surface of the ml matrix domain. Fractionation of Triton X100-extracted spermatozoa on Percoll density gradients
yielded a highly purified ALM fraction; each complex retained the ml and m2 matrix and the acrosomal lamina
(Fig. 1C).
Structural Comparison of AM29 and AM22
The polypeptide composition of the ALM fraction has
been characterized previously [10]. It contains major lowmolecular-mass bands of 29 kDa and 22 kDa (Fig. 2). A
monoclonal antibody prepared against the ALM fraction,
termed anti-AM29/22, recognized both the 29-kDa (AM29)
and 22-kDa (AM22) bands and a set of adjacent minor
bands by immunoblot analysis (Fig. 3, lane 1); the other
polypeptides of the ALM fraction did not react with antiAM29/22. Parallel blots stained with normal ascites fluid
exhibited no immunostained bands (Fig. 3, lane 2). To explore the antigenic relationships of these proteins, AM22
was purified by preparative SDS-PAGE and used to prepare
a polyclonal antiserum. By immunoblot analysis, polyclonal anti-AM22 stained the same set of polypeptides recog-
366
OLSON ET AL.
FIG. 7. Post-embedding immunoelectron
microscopy of epididymal spermatozoa using polyclonal anti-AM22. A) Cross section
through anterior region of sperm head.
Note the specific localization of gold particles to the regions of the acrosome corresponding to the location of the ml and
m2 matrix elements. No specific labeling
of the nucleus (n) or perinuclear material
(p) is noted. B) Cross section through anterior region of sperm head again showing
specific intra-acrosomal labeling at the
sites occupied by ml and m2 matrix assemblies. C) Section through sperm head
at level of the equatorial segment. Labeling is apparent over regions occupied by
the ml and m2 matrix, but the equatorial
segment (es) is not labeled. A and B, x78
000; C, x60 000 (reproduced at 69%). n,
Nucleus; p, perinuclear material; ia, inner
acrosomal membrane; ps, postacrosomal
segment.
nized by monoclonal anti-AM29/22 (Fig. 3, lane 3); preimmune serum reacted with no polypeptides in the ALM fraction (Fig. 3, lane 4). These results demonstrate that AM29,
AM22, and the set of adjacent polypeptides are an antigenically related family.
To identify structural relationships of AM29 and AM22,
the polypeptides were purified by SDS-PAGE (Fig. 4A,
lanes 1 and 3) and compared by peptide mapping. V8 digestion of AM29 and AM22 generated an identical set of
four polypeptide bands of 20 kDa, 16 kDa, 13 kDa, and 12
kDa (Fig. 4A, lanes 2 and 4), demonstrating a structural
similarity of AM29 and AM22.
The structural relationship of AM29 and AM22 was also
examined by N-terminal amino acid sequencing (Fig. 4B).
Intact AM29 was N-blocked; however, intact AM22 was
sequenced through 17 residues. The V8 protease-generated
polypeptide fragments of 20 kDa, 16 kDa, and 12 kDa for
both AM29 and AM22 also yielded N-terminal sequence
data. The N-terminal sequence of each V8 protease-generated polypeptide fragment was distinct from intact AM22,
indicating that internal sequences were obtained. The 20kDa and 12-kDa polypeptides from both AM29 and AM22
exhibited identical N-terminal sequences. The 16-kDa
bands of AM29 and AM22 were also identical to one another in N-terminal sequence but differed from the 20-kDa
and 12-kDa fragments. These data demonstrate that AM29
and AM22 share sequence homology and may originate
from a common parent molecule.
The N-terminus sequence of AM22 and of the 16-kDa
V8 protease fragment exhibited no homologies to sequenc-
es in the blastp database (NCBI BLAST program). However, the N-terminal sequence of the 12-kDa and 20-kDa
peptides exhibited sequence homology to the mouse acrosomal protein mSP-10, the murine homologue of human
SP-10 [37]. Over their first 10 amino acids, a 80% residue
identity, or 90% similarity including conservative amino
acid substitutions, was noted with amino acids 170-179 of
mSP-10 (Fig. 4B).
Distribution of AM29 and AM22 in Epididymal
Spermatozoa
The distribution of AM22 and AM29 between soluble
and particulate sperm fractions was assessed by immunoblotting. Total lysates of caput and cauda spermatozoa
stained with polyclonal anti-AM22 exhibited the same pattern of immunoreactive polypeptides (Fig. 5, lanes 1 and
2). Triton X-100 lysates of cauda sperm were centrifuged
in a microcentrifuge at 12 000
g for 2 min, and the supernatant and pellet fractions were adjusted to the same
volume. Immunoblot analyses revealed that immunoreactive polypeptides were absent in the supernatant fluid (Fig.
5, lane 3) but that they localized to the sperm pellet (Fig.
5, lane 4). This result demonstrates that the total pool of
AM22 and AM29 is associated with a particulate sperm
fraction.
To define the intracellular distribution of AM22 and
AM29, cauda epididymal spermatozoa were immunostained
with polyclonal anti-AM22 (Fig. 6, A-H) or with monoclonal anti-AM29/22 (Fig. 6, I and J). The two antibodies
ACROSOMAL MATRIX ASSEMBLY
gave identical staining patterns. Sperm exhibited intense
fluorescence of the acrosomal cap and no other structures
under all conditions of permeabilization and fixation employed (Fig. 6). In formalin-fixed Triton X-100-permeabilized spermatozoa, staining was restricted to the apical and
principal segments of the acrosome, and no staining of the
equatorial segment was noted (Fig. 6, A and B). Spermatozoa that were permeabilized by Triton X-100 extraction
(Fig. 6, C, D, I, and J) or by nitrogen cavitation (Fig. 6, G
and H) before formaldehyde fixation exhibited intense fluorescence of both partially or fully detached acrosomal
caps, while the exposed principal segment and equatorial
segment of the sperm head were negative. Control specimens immunostained with identical dilutions of preimmune
serum or normal ascites fluids exhibited no fluorescence
(Fig. 6, E and F). These data demonstrate that AM22 and
AM29 are restricted to the apical and principal segments
of the acrosome and are primarily associated with the acrosomal cap and not the inner acrosomal membrane.
Immunoelectron microscopy was employed to define
precisely the intra-acrosomal distribution of AM22 and
AM29. Post-embedding immunolabeled cauda epididymal
spermatozoa exhibited specific labeling throughout the ml
and m2 matrix regions of the acrosome (Fig. 7, A-C), and
no labeling was noted in the equatorial segment (Fig. 7C).
Other structures of the head, including the perinuclear material, the nucleus, and the postacrosomal sheath, as well
as all flagellar organelles, exhibited no specific labeling.
For pre-embedding immunolabeling, spermatozoa were
permeabilized using nitrogen cavitation, before treatment
with antibody probes. Specific labeling of both the ml and
m2 matrix elements of the acrosomal cap was noted (Fig.
8, A and B), but neither the outer nor the inner acrosomal
membranes were specifically labeled (Fig. 8, A and B).
Gold particles were located at the surface of both the ml
and m2 acrosomal matrix with the pre-embedding labeling
protocol, while no gold particles were noted within the matrix; this probably reflects the inability of the primary and/
or secondary antibodies to penetrate the interior of the intact matrix (Fig. 8, A and B).
AM22/AM29 Expression during Spermatogenesis
Light and electron microscopic immunolabeling was employed to define the temporal expression and localization
of the acrosomal matrix polypeptides in spermatogenic
cells. When polyclonal anti-AM22 was used, protein expression was noted in the acrosome at all stages of spermatid development (Fig. 9, A, B, D, and E). In late maturation-phase spermatids, the staining with anti-AM22 became restricted to the apical and principal segments of the
acrosome whereas the equatorial segment was negative
(long arrows on Fig. 9, D and E). When monoclonal antiAM29/22 was used, the acrosome of late maturation-stage
spermatids [38] exhibited intense fluorescence staining
(long arrows Fig. 9, C and F); however, the acrosome of
earlier spermatids was negative (Fig. 9, C and F). Ultrastructural immunolabeling using anti-AM22 demonstrated
labeling throughout the matrix of the acrosomal vesicle of
round spermatids (Fig. 10). These data indicate that AM29
and/or AM22, or alternatively a precursor protein, is expressed in early spermatids. Moreover, the data suggest that
an intra-acrosomal posttranslational protein-processing
event, occurring in late maturation-phase spermatids, is required for antigen recognition by monoclonal anti-AM29/
22.
367
FIG. 8. Pre-embedding immunoelectron microscopy of epididymal spermatozoa permeabilized by nitrogen cavitation and then labeled with antiAM29/22. A) Portion of released acrosomal cap showing specific labeling
of ml and m2 matrix elements. Gold particles are confined to surface of
the matrix assemblies and appear unable to penetrate its interior. Note
the absence of gold particles on the outer acrosomal membrane (oa),
which remains associated with the acrosomal matrix. B) Cross section
through anterior tip of sperm head showing the inner acrosomal membrane (ia), the perinuclear material (p), the outer acrosomal membrane
(oa), and m2 matrix. Only the m2 matrix exhibits specific labeling. A,
x46 000; B, x75 000 (reproduced at 91%).
Immunoblotting was performed on Triton X-100-soluble
and -insoluble fractions of round spermatids using both
polyclonal anti-AM22 and monoclonal anti-AM29/22 (Fig.
11). The Triton X-100-soluble fraction possessed a polypeptide of 40 kDa that reacted with polyclonal anti-AM22
(Fig. 11, lane 2) and no immunoreactive polypeptide was
found in the pellet fraction (Fig. 11, lane 3). In addition,
no AM22 or AM29 were noted in these fractions. Parallel
immunoblots stained with monoclonal anti-AM29/22 exhibited no immunoreactive bands in either the supernatant
or pellet fractions (Fig. 11, lanes 5 and 6). These data suggest that the 40-kDa polypeptide of round spermatids represents a precursor that is processed during late spermiogenesis into mature AM29 and AM22. Since the monoclonal anti-AM29/22 interacts only with the fully mature acrosomal matrix polypeptides and not the precursor form,
this suggests that other posttranslational modifications, in
368
OLSON ET AL.
FIG. 9. Matched phase contrast (A, D)
and fluorescence (B, C, E, F) images of
spermatogenic cells that were doublestained with polyclonal anti-AM22 and
monoclonal anti-AM29/22. B and E used
filter sets to visualize staining with polyclonal anti-AM22 while C and F used filters to visualize staining with monoclonal
anti-AM29/22. Note that round spermatids
(rs) show fluorescent acrosomes (a) when
immunostained with polyclonal anti-AM22
(B, E) but not when stained with monoclonal anti-AM29/22 (C, F). Late maturationphase spermatids (long arrows) show intense acrosomal fluorescence with both
polyclonal (B, E)and monoclonal (C, F)
antibody, whereas early maturation-phase
spermatids (short arrow) stain only with
polyclonal antibody (E,F). Bar = 10 I.m.
addition to size modification, may occur during processing
of the acrosomal matrix precursor polypeptide.
DISCUSSION
The outer acrosomal membrane of hamster spermatozoa
adheres to a stable acrosomal matrix assembly composed
of two contiguous but structurally distinct elements [10,
13, 22]. This matrix complex binds proacrosin and N-acetylglucosaminidase [13], and possibly other hydrolases, and
it remains intact and associated with the hybrid membrane
complex after the acrosome reaction [5, 24]. In the present
study, we addressed the composition and assembly of these
FIG. 10. Post-embedding immunolabeling of round spermatid with polyclonal anti-AM22 shows that the acrosomal matrix (am) is specifically
labeled. x 30 000 (reproduced at 74%). n, Nucleus.
structurally distinct, outer acrosomal membrane-associated
matrix elements.
Using both biochemical and immunological approaches,
we demonstrated that the major polypeptides of the ALM
complex, AM22 and AM29, are structurally related proteins
and are processed from a round spermatid precursor protein
of 40 kDa. Internal sequence data demonstrated a homology of AM22 and AM29 to mouse sperm acrosomal pro-
FIG. 11. Immunoblot of 12% SDS-PAGE minigel showing cauda sperm
ALM fraction (lanes 1 and 4) and Triton X-100 supernatant (lanes 2 and
5) and pellet (lanes 3 and 6) fractions of round spermatids (RS) immunostained with polyclonal anti-AM22 (lanes 1-3) and monoclonal antiAM29/22. Note that the polyclonal antibody recognizes a 40-kDa polypeptide in the supernatant (lane 2) but not the pellet fraction (lane 3) of
round spermatids, and the monoclonal antibody (lanes 5 and 6) does not
react with any round spermatid protein. Molecular weight markers (x
10 ) are at left.
-
3
ACROSOMAL MATRIX ASSEMBLY
teins mSP10 [37], which were originally termed the HS-63
antigens [39]. SP10-related acrosomal proteins are present
in several species including the human, baboon, fox,
mouse, and bovine, but their function has not been resolved
[40-44]. The human SP10 proteins have been demonstrated
to associate with the outer and inner acrosomal membranes
of the anterior acrosome and are also present within the
equatorial segment [45]. In the mouse, the SP10-related
polypeptides localize to the acrosome by immunofluorescence [39], but their incorporation into a stable acrosomal
matrix assembly was not demonstrated. Our data on hamster spermatozoa demonstrate a more restricted intra-acrosomal distribution and membrane association of AM22 and
AM29. These acrosomal proteins are not present in the
equatorial segment, and they are associated with the outer
acrosomal membrane of the apical and principal segments
but not with the inner acrosomal membrane. We have noted
an identical distribution pattern of related proteins in bovine
spermatozoa [46], suggesting that this may represent the
general acrosomal distribution pattern of SP10-related proteins in nonhuman spermatozoa.
A surprising finding was that the major structural proteins of the ml and m2 domains represent a related polypeptide family. Since both the monoclonal antibody and the
polyclonal antibody used in the present study recognized
both AM22 and AM29, it remains unknown whether the
individual polypeptides are restricted to separate matrix domains or whether they co-localize to both the ml and m2
domains. In the apical segment of guinea pig spermatozoa,
homopolymeric complexes of nonrelated structural proteins
form the basis for the distinct matrix assemblies [15, 1720]. Whether AM29 and AM22 assemble into higher-order
complexes remains to be determined, but it is possible that
homopolymeric complexes of AM22 and of AM29 represent the basis for the distinct structural appearance of ml
and m2 matrix domains in the hamster acrosome.
Although the mature AM22 and AM29 polypeptides are
assembled into a Triton X-100-insoluble acrosomal matrix
assembly in epididymal spermatozoa, the 40-kDa putative
precursor protein of the round spermatids is readily soluble
in Triton X-100. It is possible that size processing of the
precursor polypeptide is required for the assembly of the
insoluble matrix elements, and this is under current investigation. It is also possible that processing of the precursor
protein is required for the binding of the matrix to the outer
acrosomal membrane; this possibility, too, is under investigation. In other species, including the baboon, human, and
mouse, the testicular forms of the SP10 or SP10-related
polypeptides are high-molecular-weight precursors of the
mature polypeptides found in epididymal spermatozoa [37,
39, 42, 47]. Although several high-molecular-weight precursor forms of mSP10 have been identified in the mouse
testis [37], we found only a single immunologically reactive
band of 40 kDa in hamster round spermatids. This may
reflect species differences, but we may identify intermediate forms in the hamster as more mature spermatids are
examined.
The protease(s) responsible for the intra-acrosomal size
processing of the 40-kDa polypeptide in the hamster or of
the SP10-related proteins in other species are unidentified.
In the hamster, the size processing results in two major
polypeptides, AM22 and AM29, as well as a set of minor
related bands; however, in other species such as the human,
baboon, mouse, and bovine, the processing results in a family of several polypeptides but does not appear to result in
the production of two major forms [40-42, 44, 46]. The
369
functional significance of these multiple polypeptide forms
is unclear. As suggested above, in the hamster they may
assemble into distinct matrix regions, and it is also possible
that they may have different hydrolase-binding properties.
We have previously demonstrated by blot overlay analyses
that AM29, but not AM22, exhibits proacrosin-binding activity [13], and whether AM22 binds other hydrolases has
not been established. Possibly the multiple polypeptides in
other species also have different hydrolase-binding specificity or affinity, but this has not been addressed. Our data
show that the monoclonal anti-AM29/22 does not recognize
the precursor 40-kDa polypeptide. This raises the possibil-
ity that in addition to size processing, the polypeptides may
undergo further posttranslational modifications within the
acrosome that are required for recognition by the monoclonal but not the polyclonal antibody. Immunoblotting and
immunofluorescence studies have also identified antigenic
changes in acrosomal proteins of guinea pig and mouse late
spermatids and spermatozoa [48, 49] consistent with intraacrosomal protein processing. We have been unable to demonstrate any carbohydrates associated with AM22 or AM29
by lectin staining of Western blots (unpublished data), so
the specific nature of potential posttranslational modifications occurring during the late stages of spermiogenesis
needs identification.
AM22 and AM29 assemble into stable matrix elements
that are specifically associated with the fusigenic domains
of the outer acrosomal membrane. The mechanisms that
establish this specific interaction are not known, but it probably reflects specific protein-protein interactions between
matrix polypeptides and proteins localized to the luminal
surface of the outer acrosomal membrane. Different elements of the acrosomal matrix display specific fates during
acrosomal exocytosis. The matrix assembly associated with
the outer acrosomal membrane appears particularly stable
and remains associated with the hybrid membranes after the
acrosome reaction [5, 24, 46, 50, 51]. Thus it is possible
that the role of these matrix elements is not limited to regulating hydrolase distribution and release during the acrosome reaction but they could also function in the membrane
fusion events or in sperm-zona interactions. These potential
functions as well as identification of specific interactions
between the acrosomal matrix and the outer acrosomal
membrane will be addressed in future studies.
REFERENCES
1. Fawcett DW. The mammalian spermatozoon. Dev Biol 1975; 44:394436.
2. Meizel S. The importance of hydrolytic enzymes to an exocytotic
event, the mammalian sperm acrosome reaction. Biol Rev 1984; 59:
125-157.
3. Eddy EM, O'Brien DA. The spermatozoon. In: Knobil E, Neill JD
(eds.), The Physiology of Reproduction. New York: Raven Press;
1993: 29-77.
4. Yanagimachi R. Mammalian fertilization. In: Knobil E, Neill JD
(eds.), The Physiology of Reproduction. New York: Raven Press;
1993: 189-317.
5. Barros C, Bedford JM, Franklin LE, Austin CR. Membrane vesiculation as a feature of the mammalian acrosome reaction. J Cell Biol
1967; 34:CI-C5.
6. Kopf GS, Gerton GL. The mammalian sperm acrosome and the acrosome reaction. In: Wassarman PM (ed.), Elements of Mammalian
Fertilization. Boca Raton: CRC Press; 1991: 153-203.
7. Srivastava PN, Munnell JF, Yang CH, Foley CW. Sequential release
of acrosomal membranes and acrosomal enzymes of ram spermatozoa.
J Reprod Fertil 1974; 36:363-372.
8. Green DPL. The activation of proteolysis in the acrosome reaction of
guinea-pig sperm. J Cell Sci 1978; 32:153-164.
9. DiCarlantonio G, Talbot P. Evidence for sequential deployment of se-
370
OLSON ET AL.
cretory enzymes during the normal acrosome reaction of guinea pig
sperm in vitro. Gamete Res 1988; 21:425-438.
10. Olson GE, Winfrey VP, Davenport GR. Characterization of matrix
domains of the hamster acrosome. Biol Reprod 1988; 39:1145-1158.
11. Hardy DM, Oda MN, Friend DS, Huang TTE A mechanism for differential release of acrosomal enzymes during the acrosome reaction.
Biochem J 1991; 275:759-766.
12. Nuzzo NA, Anderson RA, Zaneveld LJD. Proacrosin activation and
acrosin release during the guinea pig acrosome reaction. Mol Reprod
Dev 1990; 25:52-60.
13. NagDas SK, Winfrey VP, Olson GE. Hydrolase binding activities of
the acrosomal matrix of hamster spermatozoa. Biol Reprod 1996; 55:
1405-1414.
14. NagDas SK, Winfrey VP, Olson GE. Proacrosin-acrosomal matrix
binding interactions in ejaculated bovine spermatozoa. Biol Reprod
1996; 54:111-121.
15. Foster JA, Friday BB, Maulit MT, Blobel C, Winfrey VP, Olson GE,
Kim K, Gerton GL. AM67, a secretory component of the guinea pig
sperm acrosomal matrix, is related to mouse sperm protein sp56 and
the complement component 4-binding proteins. J Biol Chem 1997;
272:12714-12722.
16. Phillips DM. Substructure of the mammalian acrosome. J Ultrastruct
Res 1972; 38:591-604.
17. Westbrook-Case VA, Winfrey VP, Olson GE. A domain-specific 50kilodalton structural protein of the acrosomal matrix is processed and
released during the acrosome reaction in guinea pig. Biol Reprod
1994; 51:1-13.
18. Noland TD, Friday BB, Maulit MT, Gerton GL. The sperm acrosomal
matrix contains a novel member of the pentaxin family of calciumdependent binding proteins. J Biol Chem 1994; 269:32607-32614.
19. Reid MS, Blobel CP. Apexin, an acrosomal pentaxin. J Biol Chem
1994; 269:32615-32620.
20. 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.
21. Yanagimachi R, Noda YD. Fine structure of the hamster sperm head.
Am J Anat 1970; 128:367-388.
22. Olson GE, Winfrey VP. Substructure of a cytoskeletal complex associated with the hamster sperm acrosome. J Ultrastruct Res 1985; 92:
167-179.
23. Longo FJ, Cook S, Baillie R. Characterization of an acrosomal matrix
protein in hamster and bovine spermatids and spermatozoa. Biol Reprod 1990; 42:553-562.
24. Yanagimachi R, Phillips DM. The status of acrosomal caps immediately before fertilization in vivo. Gamete Res 1984; 9:1-19.
25. Hyatt H, Gwatkin RBL. Characterization of isolated acrosomal matrices from hamster spermatozoa. J Reprod Fertil 1988; 83:419-429.
26. Romrell LJ, Bellve AR, Fawcett DW. Separation of mouse spermatogenic cells by sedimentation velocity. A morphological characterization. Dev Biol 1976; 49:119-131.
27. Bellve AR, Millette CE Bhatnagar YM, O'Brien DA. Dissociation of
the mouse testis and characterization of isolated spermatogenic cells.
J Histochem Cytochem 1977; 25:480-494.
28. Galfre G, Milstein C. Preparation of monoclonal antibodies: strategies
and procedures. Methods Enzymol 1981; 73:1-46.
29. Laemmli UK. Cleavage of structural proteins during the assembly of
the head of bacteriophage T4. Nature 1970; 227:680-685.
30. Lee C, Levin A, Branton D. Copper staining: a five-minute protein
stain for sodium dodecyl sulfate-polyacrylamide gels. Anal Biochem
1987; 166:308-312.
31. Harlow E, Lane D. Antibodies. A Laboratory Manual. Cold Spring
Harbor, NY: Cold Spring Harbor Laboratory; 1988.
32. Cleveland DW, Fischer SG, Kirschner MW, Laemmli UK. Peptide
mapping by limited proteolysis in sodium dodecyl sulfate and analysis
by gel electrophoresis. J Biol Chem 1977; 252:1102-1106.
33. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins
from polyacrylamide gels to nitrocellulose sheets: procedure and some
applications. Proc Natl Acad Sci USA 1979; 76:4350-4354.
34. Matsudaira P. Sequence from picomole quantities of proteins electroblotted onto polyvinylidene difluoride membranes. J Biol Chem 1987;
262:10035-10038.
35. Altschul SE Gish W, Miller W, Myers EW, Lipman DJ. Basic local
alignment search tool. J Mol Biol 1990; 215:403-410.
36. Yamaguchi K, Asakwa H. Preparation of colloidal gold for staining
proteins electrotransferred onto nitrocellulose membranes. Anal Biochem 1988; 172:104-107.
37. Reddi PP, Naaby-Hansen S, Aguolnik I, Tsai J, Silver LM, Flickinger
CJ, Herr JC. Complementary deoxyribonucleic acid cloning and characterization of mSP-10: the mouse homologue of human acrosomal
protein SP-10. Biol Reprod 1995; 53:873-881.
38. Clermont Y. Cycle de l'epithelium seminal et mode de renouvellement
des spermatogonies chez le hamster. Rev Can Biol 1954; 3:208-245.
39. Liu MS, Aebersold R, Fann CH, Lee CG. Molecular and developmental studies of a sperm acrosome antigen recognized by HS-63
monoclonal antibody. Biol Reprod 1992; 46:937-948.
40. Herr JC, Flickinger CJ, Homyk M, Klotz K, John E. Biochemical and
morphological characterization of the intra-acrosomal antigen sp-10
from human sperm. Biol Reprod 1990; 42:181-193.
41. Herr JC, Wright RM, John E, Foster J, Kays T, Flickinger CJ. Identification of human acrosomal antigen SP-10 in primates and pigs.
Biol Reprod 1990; 42:377-382.
42. Freemerman AJ, Wright RM, Flickinger CJ, Herr JC. Cloning and
sequencing of baboon and cynomologus monkey intraacrosomal protein SP-10: homology with human SP-10 and a mouse sperm antigen
(MSA-63). Mol Reprod Dev 1993; 34:140-148.
43. Beaton S, Have JT, Bradley MP Cloning and partial characterization
of the cDNA encoding the fox sperm protein FSA-Acr.l with similarities to the SP-10 antigen. Mol Reprod Dev 1995; 40:242-252.
44. Coonrod SA, Herr JC, Westhusin ME. Inhibition of bovine fertilization in vitro by antibodies to SP-10. J Reprod Fertil 1996; 107:287297.
45. Foster JA, Klotz KL, Flickinger CJ, Thomas TS, Wright RM, Castillo
JR, Herr JC. Human SP-10: acrosomal distribution, processing, and
fate after the acrosome reaction. Biol Reprod 1994; 51:1222-1231.
46. Olson GE, Winfrey VP, Neff JC, Lukas TJ, NagDas SK. An antigenically related polypeptide family is a major structural constituent of a
stable acrosomal matrix assembly in bovine spermatozoa. Biol Reprod
1997; 57:325-334.
47. Herr JC, Klotz K, Shannon J, Wright RM, Flickinger CJ. Purification
and microsequencing of the intra-acrosomal protein SP-10. Evidence
that SP-10 heterogeneity results from endoproteolytic processes. Biol
Reprod 1992; 47:11-20.
48. O'Brien DA, Gerton GL, Eddy EM. Acrosomal constituents identified
with a monoclonal antibody are modified during late spermiogenesis
in the mouse. Biol Reprod 1988; 38:955-967.
49. Gerton GL, O'Brien DA, Eddy EM. Antigens recognized by monoclonal antibody to mouse acrosomal components differ in guinea pig
spermatogenic cells and sperm. Biol Reprod 1988; 39:431-441.
50. Russell L, Peterson R, Freund M. Direct evidence for formation of
hybrid vesicles by fusion of plasma and outer acrosomal membranes
during the acrosome reaction in boar spermatozoa. J Exp Zool 1979;
208:41-56.
51. VandeVoort CA, Yudin Al, Overstreet JW. Interaction of acrosomereacted macaque sperm with the macaque zona pellucida. Biol Reprod
1997; 56:1307-1316.