(Received for publication, November 18, 1996, and in revised form, February 7, 1997)
From the The guinea pig sperm acrosomal matrix is the
dense core of the acrosome and is likely to be important in acrosome
biogenesis and fertilization. Isolated acrosomal matrices are composed
of a limited number of major bands when analyzed by SDS-polyacrylamide gel electrophoresis, among which is a Mr 67,000 protein that we have termed AM67. Indirect immunofluorescence
demonstrated that AM67 is localized to the apical segment of the cauda
epididymal sperm acrosome. Immunoelectron microscopy further refined
the localization of AM67 to the M1 (dorsal bulge) domain within the acrosome. Using a polymerase chain reaction product based upon tryptic
peptide sequences from AM67, a The sperm acrosome reaction has the characteristic hallmarks of
regulated secretion: 1) secretory products are concentrated and
condensed; 2) secretory granules are stored for long periods of time;
and 3) secretion is coupled to an extracellular stimulus (in this case,
the zona pellucida) (1). Regulated secretion from sperm
(i.e. the acrosome reaction) is obligatory for fertilization (2). It is presumed that the release of specific secretory components
from the acrosome is involved in assisting the sperm in the penetration
of the investments surrounding the egg. Furthermore, although sperm
contain only one secretory vesicle, different enzymatic activities are
released from sperm at various times following the acrosome reaction
(3-8). The differential temporal release of secretory components
occurs in other regulated secretory tissues such as the pancreas, but
since cells of such tissues contain multiple secretory granules, it has
been proposed that this "non-parallel secretion" is due to
exocytosis from heterogeneous sources within the tissue or cell of
study (9). In guinea pig sperm, the acrosome reaction results in the
differential release of acrosomal components possibly due to their
compartmentalization within soluble or insoluble phases of the acrosome
(5, 8). Following the exocytotic acrosome reaction, certain acrosomal
proteins, such as proacrosin, remain associated with the matrix that
establishes the "dense core" of the acrosome and are released at a
slower rate than components, such as dipeptidyl peptidase, that are
found in a more readily solubilized compartment.
The results presented here and in recent papers (10-14) demonstrate
that two of the components of the dense core (matrix) of the acrosome,
AM50 and AM67, represent monomers of large homopolymeric proteins. AM50
has been shown to be a novel, testis-specific member of the pentraxin
family of proteins (11, 13). Preliminary tryptic peptide sequencing
indicated that AM67 is related to members of the complement-binding
protein family (11). In this paper, we show that AM67 is, indeed, a
member of this class of proteins. Furthermore, the amino acid sequence
of AM67 is most closely related to that of the mouse sperm protein
sp56. This protein is a candidate for the cell-surface receptor of ZP3,
a glycoprotein of the egg's extracellular matrix, the zona pellucida
(15). A standard post-embedding immunoelectron microscopic technique
demonstrated that AM67 is restricted to a specific dorsal compartment
(M1 domain of the apical segment) within the acrosome of guinea pig
sperm, while mouse sp56 was detected only within the
acrosome and not on the plasma membrane over the acrosome as was
previously reported using an unconventional colloidal gold surface
replica immunoelectron microscopic procedure (15).
Guinea pigs (male retired breeders) were
purchased from Charles River Laboratories (Wilmington, MA). Reagents
used for electrophoresis were obtained from Bio-Rad. Percoll,
leupeptin, pepstatin A, and benzamidine HCl were from the Sigma.
Nitrocellulose and Immobilon-P membranes were from Schleicher & Schuell
and Millipore Corp. (Bedford, MA), respectively. An enhanced
chemiluminescence kit (ECL, Amersham Corp.) was used for immunoblot
analyses. Digoxigenin labeling reagents, detection kits, and Genius
buffers were from Boehringer Mannheim. Plasmid pGEM-11Zf(+) was from
Promega (Madison, WI). Taq DyeDeoxyTM Terminator Cycle
sequencing kits were from Applied Biosystems, Inc. (Foster City, CA).
Fluorescein isothiocyanate-conjugated goat anti-rabbit IgG was
purchased from Zymed Laboratories, Inc. (South San Francisco, CA), and
colloidal gold-labeled goat anti-mouse IgG antibody and Texas
Red-conjugated horse anti-mouse IgG were from Jackson ImmunoResearch
Laboratories, Inc. (West Grove, PA). LR Gold was obtained from
Polysciences (Fort Washington, PA). Monoclonal antibody 7C5, specific
to mouse sperm sp56, was purchased from QED Biologicals (La Jolla,
CA).
Acrosomal matrices were
prepared by a modification of the procedure of Hardy et al.
(5). Sperm were extruded from the caudae epididymides and vasa
deferentia by retrograde perfusion with 20 mM sodium
acetate, pH 5.2, containing 0.15 M NaCl, 5% sucrose, and
0.5 mM p-aminobenzamidine. Following a wash in
the perfusion buffer, the sperm were treated with 20 mM
sodium acetate, pH 5.2, containing 0.15 M NaCl, 0.625%
Triton X-100, 5% sucrose, and 0.5 mM
p-aminobenzamidine. The acrosomal matrices were dislodged
from the remainder of the detergent-extracted sperm by extrusion
through a 26-gauge needle two times. The resulting sperm/acrosomal
matrix suspension was passed over a glass bead column, and the
acrosomal matrices eluting from the column were collected by
centrifugation. After washing once with 20 mM sodium
acetate, pH 5.2, containing 0.15 M NaCl and 0.5 mM p-aminobenzamidine, the pelleted acrosomal matrices were then extracted with 1% acetic acid containing 5 mM benzamidine.
Protein
concentrations were determined by the Pierce BCA method according to
the manufacturer's specifications and employing bovine serum albumin
as a standard (16). SDS-polyacrylamide gel electrophoresis
(PAGE)1 was performed according to Laemmli
(17). Following electrophoresis, gels were stained with silver by the
method of Gerton and Millette (18).
For the partial amino acid sequencing of AM67, an acid extract of
guinea pig sperm (consisting mostly of solubilized acrosomal components) was separated on several gels by non-equilibrium pH gradient electrophoresis followed by SDS-PAGE, and the gels were transferred to Immobilon-P (19, 20). The transfer membranes were
stained with Ponceau S, and the spots corresponding to AM67 (as
determined by probing a stained replicate blot with anti-AM67) were
excised. Tryptic peptides were generated from the transfer membranes,
purified by high-performance liquid chromatography, and sequenced by
the Edman degradation procedure by Dr. John Leszyk (Worcester
Foundation for Experimental Biology).
AM67 was also affinity-purified using the monoclonal antibody PH1,
which binds the guinea pig sperm protein fertilin (21) and apparently
cross-reacts with AM67.2 Affinity-purified
AM67 was reduced and alkylated, fractionated by SDS-PAGE, and
electroblotted onto nitrocellulose membranes. The Ponceau S-stained
Mr 67,000 band was excised and processed for
internal amino acid analysis by Drs. P. Tempst, H. Erdjument-Bromage, and S. Geronamos (Sloan-Kettering Institute Microchemistry Core Facility) as described (22). The immobilized protein samples were
subjected to trypsinization, and tryptic fragments were separated by
reverse-phase HPLC. HPLC peak fractions (over trypsin background) were
analyzed by automated Edman degradation using an Applied Biosystems
Model 477A Sequencer, with instrument and procedure optimized for
femtomole level analysis as described (23).
Acrosomal matrices were
separated by preparative SDS-PAGE and electrophoretically transferred
to nitrocellulose paper by the method of Towbin et al. (24).
The protein band corresponding to AM67 was excised, and antisera were
produced commercially from the nitrocellulose strips (Cocalico,
Reamstown, PA). For immunoblot analysis, proteins were separated by
SDS-PAGE, transferred to nitrocellulose membranes, and processed as
described previously using ECL procedures (20).
Indirect immunofluorescence of guinea pig sperm attached to
polylysine-coated coverslips was performed as described previously (25). Mouse sperm obtained from caudae epididymides were washed twice
by centrifugation for 5 min at 300 × g, resuspended in
PBS, and placed onto polylysine-coated coverslips. Sperm were then treated either with PBS containing 4% paraformaldehyde or with PBS
alone for 15 min at room temperature. For immunostaining, coverslips
were incubated in PBS containing 10% normal goat serum (blocking
buffer) for 30 min at room temperature followed by primary antibody (40 µg/ml) diluted in blocking buffer for 1 h at 37 °C. Following
a washing step (3 × 5 min in PBS), coverslips were incubated with
Texas Red-conjugated horse anti-mouse IgG diluted 1:40 in blocking
buffer for 1 h at 37 °C, washed again, and mounted using Fluoromount G. Slides were examined using a Zeiss Photomicroscope III
equipped with epifluorescence and photographed with Kodak T-Max
P3200.
Immunoelectron microscopy of
guinea pig sperm from caudae epididymides was performed on LR
White-embedded sections using 10-nm gold particles as described
previously (12). Mouse sperm were obtained from the caudae epididymides
and were washed twice by centrifugation for 5 min at 300 × g and resuspension in PBS. For the post-embedding
immunostaining procedure, sperm were fixed in 0.2 M
cacodylate buffer, pH 7.4, containing 2% paraformaldehyde for 1 h
at 4 °C; dehydrated in an increasing series of ethanol; and embedded
in LR Gold. Grids were incubated in blocking buffer (Tris-buffered
saline (20 mM Tris, pH 7.5, 0.5 M NaCl)
containing 10% normal goat serum) for 30 min at room temperature and
were subsequently incubated in blocking buffer containing either
anti-sp56 monoclonal antibody 7C5 (40 µg/ml) or normal mouse IgG (40 µg/ml) overnight (~16 h) at 4 °C. Following three washes in
Tris-buffered saline, the grids were incubated in blocking buffer
containing 12-nm colloidal gold conjugated to a goat anti-mouse IgG
antibody diluted 1:40. The same procedure was used for AM67
immunostaining, except that the grids were treated with either an
anti-AM67 rabbit antiserum or a preimmune control diluted 1:100
followed by an 18-nm colloidal gold-conjugated goat anti-rabbit
antibody diluted 1:50. After washing three times in Tris-buffered
saline, all grids were fixed and counterstained with 1% osmium
tetroxide followed by aqueous uranyl acetate. Specimens were observed
and photographed using a Phillips 201 transmission electron
microscope.
To screen for cDNA clones
encoding AM67, a probe was prepared by PCR of a guinea pig testis
After labeling by the random priming method with digoxigenin-labeled
dUTP as one of the nucleotides, the probe was used to screen the
library for AM67 according to standard procedures (26, 27). Briefly,
plates of Y1090 cells infected with phage particles were grown at
42 °C for 5-7 h. Phage plaques were lifted onto Hybond-N (Amersham
Corp.) filters, and phage DNA was denatured by sequentially placing
filters onto filter paper saturated with 0.5 M NaOH, 1.5 M NaCl; 0.5 M Tris-HCl, pH 7.5; 1.5 M NaCl; and finally, 2 × SSC. Filters were dried and
fixed with UV light. Following prehybridization in modified Church
buffer (0.25 M sodium phosphate, pH 7.4, 1 mM
EDTA, 7% SDS, and 1% bovine serum albumin), filters were hybridized
in modified Church buffer containing 10 ng/ml digoxigenin-labeled AM67
PCR product overnight at 62 °C. Filters were washed in Genius Buffer
1 (150 mM NaCl, 100 mM Tris-HCl, pH 7.5) for 10 min at room temperature and then again at 65 °C. Filters were
blocked in Genius Buffer 2 (100 mM NaCl, 100 mM
Tris-HCl, pH 7.5, and 2% blocking reagent) for 30 min, followed by
incubation for 1 h in anti-digoxigenin antibody conjugated to
peroxidase (1:1000 in Genius Buffer 2) at room temperature. Filters
were washed in Genius Buffer 1 and detected using ECL reagents.
Positive clones were plaque-purified, and the cDNA inserts were
subcloned into pGEM-11Zf(+) and sequenced using the Taq
DyeDeoxyTM Terminator Cycle sequencing kit chemistry and the
appropriate primers. The products were separated by electrophoresis and
analyzed with an Applied Biosystems Model 373A Automated Sequencer.
Both strands of each insert were sequenced.
The deduced amino acid sequence of AM67 was analyzed with the
MacVectorTM molecular biology program (Kodak Scientific Imaging Systems) or the MacPattern program using the Prosite and Blocks data
bases (28). Homology searches of GenBankTM and other sequence data
bases were performed using the BLAST program of the National Center for
Biotechnology Information (29). Alignments were created using the
Clustal V program (30) and were displayed using SeqVu Version
1.0.1.3 The N-terminal amino acid of mature
AM67 was deduced with the AnalyzeSignalase2.03 program, which predicts
signal peptidase cleavages based upon the method of von Heijne
(31).
Total RNAs (20 µg) from various
guinea pig tissues were denatured by glyoxal, separated on 1% agarose
gels, and transferred onto Hybond-N membranes. The blots were probed
with an AM67 PCR probe 32P-labeled by the random priming
method, and the hybridized bands were detected by autoradiography as
described previously (20).
The protein
composition of acrosomal matrices from guinea pig sperm was relatively
simple (Fig. 1). The protein profile was dominated by a
major protein band of Mr ~50,000. As
determined by non-equilibrium pH gradient electrophoresis/SDS-PAGE
followed by immunoblotting, this band actually comprised two proteins
of similar molecular weights, proacrosin and AM50 (also called p50 and
apexin in previous publications (11, 13)). A less prominent protein was
AM67, represented by a protein band of Mr
67,000. Both AM50 and AM67 form high molecular weight homomeric
complexes in the absence of reducing agents (11-14). A more
variable constituent of these preparations was a
Mr ~32,000 protein, believed to be proacrosin-binding protein (5, 32).
An antiserum to AM67 was produced by immunization of
rabbits with nitrocellulose blot-purified AM67. The antiserum was
specific to AM67 as shown by immunoblot analysis (Fig.
2). As demonstrated previously by the electrophoresis of
AM67 under nonreducing conditions followed by separation after
reduction, AM67 formed a Mr
Sperm prepared in various ways were examined to determine whether all
of the AM67 remained associated with the acrosomal matrix during
fractionation of sperm and to ascertain whether there were any gross
alterations in the size of AM67 as a result of germ cell maturation. As
can be seen in Fig. 3, AM67 can be detected in the
Triton X-100-soluble fraction of sperm (solubilized during the course
of acrosomal matrix preparation) as well as in the acrosomal matrix
(pellet). This suggested that there was a subpopulation of AM67 that
was not tightly bound to acrosomal matrix and that could be released
with gentle non-ionic detergents. This result was in contrast to AM50,
which remains firmly associated with the matrix during isolation (11).
In addition, the size of AM67 acid-extracted from testicular germ cells
approximated that from epididymal sperm, indicating that there were no
gross structural alterations in the AM67 monomer as a result of sperm
maturation. However, if sperm were extracted with Laemmli sample buffer
without reducing agents, AM67 was partially degraded, presumably by
proteases, such as acrosin, that remain active in SDS in the absence of
reducing agent (25, 33).
The biochemical studies presented
above indicate that AM67 is part of the acrosomal matrix. However, the
morphology of the acrosome is somewhat complex and can be
differentiated into three segments: the apical segment (the region that
extends beyond the anterior tip of the sperm nucleus), the principal
segment (the region that encases the anterior half of the nucleus), and
the equatorial segment (the region that forms a narrow band around the
posterior region of the acrosome and where the outer acrosomal membrane
makes the transition to the inner acrosomal membrane). Furthermore,
individual acrosomal proteins have been localized to one or more
segments of the guinea pig sperm acrosome (5, 12, 34). To determine
whether AM67 was restricted to a particular segment of the acrosome,
sperm were examined by indirect immunofluorescence. AM67 protein was
found to be restricted to the apical segment of the acrosome (Fig.
4B). Staining was also not observed in sperm treated with preimmune serum instead of anti-AM67 (Fig. 4, C
and D) or in cells that had not been permeabilized prior to
antibody treatment (data not shown).
Ultrastructurally, the apical segment can be morphologically
differentiated into three domains: M1 ("dorsal bulge"), M2
(intermediate zone), and M3 (ventral zone). Immunoelectron microscopic
localization using thin sections of guinea pig sperm provided the
resolution demonstrating that AM67 is restricted to the M1 domain of
the apical segment (Fig. 5). No staining was observed in
the principal and equatorial segments of the acrosome. Few gold
particles were seen in the M2 and M3 domains of the acrosome, and none
were detected on the sperm plasma membrane. Within the M1 domain, the
gold particles were generally excluded from apparently spherical,
denser regions. Thus, AM50 and AM67, which are both high molecular
weight homomeric complexes of the acrosomal matrix and are found in the
M3 and M1 domains, respectively, do not overlap in their
subcellular localization (10-14).
As reported earlier, two tryptic peptides of AM67
(MALEVYK and LSLEIEQLEKEKY) were sequenced and were found to be highly
homologous to contiguous regions in human C4BP
Using the tryptic peptide sequencing information, degenerate primers
were designed for PCR using DNA from the The cDNA sequence of AM67 encoded a 533-amino acid
Mr 59,768 protein (Fig. 6A).
Following removal of a predicted 28-amino acid hydrophobic secretory
signal sequence from the N-terminal region of the protein, the
polypeptide was predicted to have a Mr of
56,851. Glycosylation at one or more of the seven consensus sites for
asparagine-linked glycosylation probably accounts for the difference
between the predicted (56,851) and experimentally determined (67,000)
molecular weights.
Comparison of the sequence of AM67 with GenBankTM data bases using the
BLAST algorithm demonstrated that AM67 is most closely related to mouse
sperm protein sp56 and other members of the C4BP The observation that the amino acid sequence of AM67 was most similar
to the sequence of mouse sperm protein sp56 was of great interest since
this protein has been identified as a candidate cell-surface receptor
for the ZP3 glycoprotein of the egg zona pellucida (15). There are
large stretches of identical residues and conservative substitutions
throughout the sequence (Fig. 7A). For example, following
removal of the signal peptides (AM67, residues 1-28; and sp56,
residues 1-32), 12 out of the first 13 amino acids of mature mouse
sp56 were identical to mature AM67, with the one differing residue
being a conservative substitution. A 50-amino acid stretch conserved in
guinea pig AM67 (residues 346-395) and mouse sp56 (residues 351-400)
appeared to have diverged extensively from the C4BP Given the similarity of amino acid sequences, one would predict that
AM67 and sp56 would be localized to the same cellular compartment.
Using standard immunofluorescence and immunoelectron microscopic
procedures, AM67 was localized within the acrosome (Figs. 4 and 5). To
the contrary, sp56 was previously reported to be localized to the sperm
surface over the acrosome using the 7C5 monoclonal antibody and an
unconventional immunogold surface replica approach (15, 36). Thus, we
re-examined the subcellular localization of sp56 in mouse sperm using
monoclonal antibody 7C5 and a standard post-embedding immunoelectron
microscopic approach that has been used extensively in previous studies
(5, 12, 37). Under these conditions, sp56 was not detected on the
plasma membrane of mouse sperm as was previously reported. Instead, it was detected abundantly within the acrosome and did not appear to
associate preferentially with the acrosomal membranes (Fig. 8, A-C). The acrosomal localization of sp56
by this method was consistently reproducible and was identical both in
mature epididymal sperm and in sperm incubated under capacitating
conditions (data not shown). Control sperm incubated with an equal
concentration of nonspecific whole IgG showed no cross-reactivity (data
not shown). To enhance the possibility of detecting sperm-surface sp56,
we also used a pre-embedding immunoelectron microscopic approach. Under
these conditions, sp56 was not detected on the surface of
acrosome-intact sperm, but was associated with the acrosomal contents
in sperm with damaged acrosomes (data not shown). Furthermore, using
indirect immunofluorescence for localizing sp56 in mature mouse sperm
with the 7C5 monoclonal antibody, bright acrosomal staining was
observed in paraformaldehyde-fixed sperm (Fig. 9,
A and B), but no acrosomal staining
was observed in unfixed sperm (Fig. 9, C and
D).
This study demonstrates that the guinea pig sperm acrosome
contains a testis-specific protein related to mouse sperm sp56 and
other members of the complement regulatory protein superfamily. This
result is reminiscent of our previous reports concerning the acrosomal
matrix protein AM50, which was found to be a member of the pentraxin
superfamily (11, 13). Like AM50, AM67 was found to be a testis-specific
member of a protein superfamily, was localized to a specific domain
within the acrosome, and also formed high molecular weight
homopolymers. Similar to C4BP Of all the sequences found to be homologous to AM67, mouse sperm
protein sp56 and the serum complement component 4-binding proteins
within the complement regulatory protein superfamily ranked the highest
in a BLAST search of GenBankTM sequences (29). The significance of this
is not clear at the present time; but it is noteworthy that serum C4BP
has been shown to interact with serum pentraxins (C-reactive protein
and serum amyloid P component) (38-40), and thus, AM67 might be
predicted to interact with AM50. Although immunoelectron microscopy
demonstrated that AM67 and AM50 are localized in distinct regions of
the acrosomal matrix of acrosome-intact sperm, it is possible that
these components could interact following the acrosome reaction.
C4BP AM67 is not the first member of the complement regulatory protein
superfamily to be found in sperm. Additional members recently identified include another guinea pig sperm C4BP distinct from AM67
(13) as well as decay-accelerating factor (CD55), membrane cofactor
protein (CD46), and protectin (CD59), which have been shown to be
membrane-bound components of the sperm acrosomal region (41).
Clusterin, another regulator of complement, is secreted by the Sertoli
cells and the epididymal epithelium and has also been shown to be
associated with sperm (42, 43). The roles of these complement
regulatory proteins in fertilization and reproductive immunology are
debatable at this point, but several of them appear to be present in
sperm-specific forms. The testis-specific expression of AM67 is also
intriguing considering C4BP The very remarkable similarity of the amino acid sequence of guinea pig
AM67 to that of mouse sp56 suggests that these proteins are
orthologues. Although the apparent molecular weights of the native AM67
and sp56 multimeric proteins were ~380,000 and 110,000, respectively
(14, 15), these sizes are similar to the differences in the properties
of the C4BPs from other mammalian species compared with mouse (44).
Furthermore, AM67 and sp56 are soluble at low pH, a commonly used
condition for the extraction of acrosomal proteins (45). However, our
finding that AM67 is intracellular contrasts with previous reports
characterizing mouse sp56 as a cell-surface protein (15, 46).
To examine the discrepancy in the localizations of AM67 and sp56, we
used the post-embedding immunoelectron microscopic method, but found
that both AM67 and sp56 were localized to the acrosomal contents in
mature guinea pig and mouse sperm, respectively, while neither was
observed on the plasma membrane. It seems unlikely that there is a pool
of sp56 on the mouse sperm surface that was not detected by this method
since both surface and intracellular antigens should be detected
equally well. This conclusion was further substantiated by results from
pre-embedding immunoelectron microscopy, a standard method used to
localize cell-surface antigens, showing that sp56 was associated with
acrosomal material in damaged sperm, but not on the plasma membrane
(data not shown). It is possible that the colloidal gold labeling of
sp56 interpreted as surface labeling by the immunogold surface replica
method (15, 36) could, in fact, be due to the exposure of acrosomal
sp56 on the surface by permeabilization of the outer acrosomal and plasma membranes. The outer acrosomal membrane and the plasma membrane
over the acrosome are known to be susceptible to mechanical and
chemical disruption, and these membranes are also destabilized during
capacitation (47). Furthermore, the mouse sperm utilized for the
localization of sp56 by the immunogold surface replica technique were
capacitated and lightly fixed with 1% paraformaldehyde, but not
embedded prior to immunolabeling (36). Since formaldehyde fixation is
reversible (48), it appears that the "surface labeling" of sp56 may
actually be attributable to the exposure of intra-acrosomal sp56 caused
by the permeabilization, rupture, or partial fusion of the destabilized
outer acrosomal and plasma membranes during the course of the sample
manipulations.
These results do not negate the results showing the ability of mouse
sp56 to bind to the zona pellucida, but should cause us to reassess our
concepts regarding sperm capacitation, sperm-egg recognition, and the
acrosome reaction. Evidence from a variety of systems indicates that
small dynamic fusion pores form in a common bilayer created by the
apposing leaflets of the vesicular and plasmalemmal membranes (reviewed
by Monck and Fernandez (49)). These small pores are thought to
occasionally close after the release of small non-quantal amounts of
secretory products. In an application of this "dynamic fusion pore"
hypothesis to sperm, capacitation may represent a state where the small
fusion pores are dynamically opening and closing between the outer
acrosomal and plasma membranes, thereby exposing acrosomal components.
Thus, it might be possible that a capacitated sperm encountering an egg
may actually bind to the zona via the transiently exposed acrosomal
proteins, not a cell-surface molecule. Binding to the zona in this
manner might "lock" the fusion pore into the static open state,
driving exocytosis to completion.
In consideration of the resistance of the acrosomal matrix to Triton
X-100 solubilization, we propose that acrosomal matrix components such
as AM67 and AM50 help form the dense core of the acrosome and establish
the different compartments (M1, M2, and M3 domains) within this
organelle. As proposed for the dynamic fusion pore hypothesis, the
relative distribution of these proteins may position them for binding
to other molecules such as the zona pellucida glycoproteins, and the
affinity of AM67 and AM50 for other acrosomal components may regulate
the differential release of certain secretory products following the
acrosome reaction. Other acrosomal proteins such as dipeptidyl
peptidase, autoantigen 1, and soluble hyaluronidase may not associate
tightly with the matrix, and thus, they would be freely released
following the acrosome reaction (5, 6, 8, 34). Proteins such as
proacrosin would be more firmly associated with certain components of
the matrix (e.g. AM67, AM50, and proacrosin-binding protein)
(50-52). The gradual dissolution of the matrix may result from the
dissociation of matrix components resulting from partial proteolysis
following the progressive conversion of the zymogen proacrosin to
enzymatically active acrosin. With the activation of proteolysis, AM50
would be processed to AM50AR by proteolysis, leading to the
destabilization of the AM50 oligomers and continued dissolution of the
matrix (12). This working hypothesis opens up new avenues for examining the release of materials following acrosomal exocytosis. In studies in
progress, the binding properties of AM67 and AM50 are being examined to
identify ligands for these large oligomers and to determine their roles
in acrosomal biogenesis and fertilization.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) U75654[GenBank]. We thank Drs. Stuart B. Moss, Gregory S. Kopf, and Bayard T. Storey for comments and critical evaluation of the
manuscript. We also greatly appreciate the preparation of capacitated
mouse sperm by Dr. Cindi Ward and Sonya Martin.
Center for Research on Reproduction and
Women's Health,
Department of Cell and Developmental Biology, University of
Pennsylvania Medical Center, Philadelphia, Pennsylvania 19104-6080, the
§ Department of Cellular Biochemistry and Biophysics,
Memorial Sloan-Kettering Cancer Center, New York, New York 10021, and the ¶ Department of Cell Biology, Vanderbilt University,
Nashville, Tennessee 37232
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
gt11 guinea pig testis cDNA
library was screened to yield two cDNA clones that encode the AM67
peptides. Northern analysis revealed that AM67 is transcribed as a
1.9-kilobase testis-specific mRNA. The complete AM67 sequence encodes a prepropolypeptide of 533 amino acids with a calculated Mr of 59,768. Following cleavage of a probable
signal sequence, the polypeptide was predicted to have a
Mr of 56,851 and seven consensus sites for
asparagine-linked glycosylation. The deduced amino acid sequence of
AM67 is most similar to those of the mouse sperm protein sp56 and the
-subunits of complement component 4-binding proteins from various
mammalian species. Although mouse sp56 has been reported to be a
cell-surface receptor for the murine zona pellucida glycoprotein ZP3,
standard immunoelectron microscopy using the anti-sp56 monoclonal
antibody 7C5 detected sp56 within the mouse sperm acrosome, but failed
to detect sp56 on the surface of acrosome-intact mouse sperm.
Furthermore, acrosomal labeling was detected in mouse sperm prepared
for immunofluorescence using paraformaldehyde fixation, but was not
observed with live unfixed sperm. Thus, the finding that sp56 is
present within the acrosome provides further support that sp56 and AM67
are orthologues and suggests that sp56 may function in acrosomal
matrix-zona pellucida interactions during and immediately following the
acrosome reaction in the mouse.
Materials
gt11 cDNA library with degenerate primers based upon conserved
regions in cDNAs encoding C4BP
from various mammals and the
continuous peptide sequence (MALEVYKLSLEIEQLEKEKY) previously reported
for AM67 (11). The primer pair used was 5
-TTTGGATCACA(A/G)ATAGAATTCAGC-3
and
5
-(C/T)AGTTGTTCAAT(C/T)TCCAGAGACAG-3
. Cycling conditions were
95 °C for 3 min followed by 30 cycles of 95 °C for 1 min,
42 °C for 2 min, and 72 °C for 2 min and a final extension at
72 °C for 6 min. Preliminary DNA sequencing of the ~1200-base pair
PCR product confirmed that the product corresponded to a fragment of
the AM67 mRNA.
Protein Composition of Acrosomal Matrices
Fig. 1.
Composition of acrosomal matrices from guinea
pig cauda epididymal sperm. Acrosomal matrices (right
lane) were analyzed by SDS-PAGE (10 µg of protein with 40 mM DTT) on a gel containing 7% polyacrylamide. The major
proteins detected by silver staining are proacrosin, AM50, and AM67.
Molecular weight standards are shown in the left lane.
[View Larger Version of this Image (53K GIF file)]
200,000 disulfide-bonded complex. Although initially reported to have a
molecular weight approximating 520,000 (14), a reanalysis of these data
provided an estimate of Mr ~380,000 for the
AM67 multimer (data not shown). In the absence of reducing agent, AM67 barely migrated into a 7.0% running gel. With increasing amounts of
DTT, this complex was reduced to smaller complexes and eventually to
monomers (Fig. 2).
Fig. 2.
Immunoblot analysis of AM67 from isolated
guinea pig acrosomal matrices treated under nonreducing and reducing
conditions. Acrosomal matrices were extracted with 1% acetic acid
containing 50 mM benzamidine. The extracts were then
diluted in sample buffers containing increasing amounts of DTT, and 5 µg of protein were analyzed per lane by SDS-PAGE (7% polyacrylamide
in the separating gel). The numbers listed above each lane represent
the amount of DTT (mM) added to each sample. After transfer
to nitrocellulose, the membrane was probed with anti-AM67 and developed
by enhanced chemiluminescence. In the absence of reducing agent, AM67
barely migrated into the running gel, demonstrating that AM67 is part of a high molecular weight complex. With increasing amounts of DTT,
this complex began to be reduced to smaller complexes and, eventually,
to monomers.
[View Larger Version of this Image (47K GIF file)]
Fig. 3.
Immunoblot analysis of AM67 in various
extracts of sperm and germ cells. The soluble fraction
(SF) and acrosomal matrices (AM) were prepared as
described under "Experimental Procedures." Whole cauda epididymal
sperm (wS) were extracted in Laemmli sample buffer. Mixed
germ cells (GC) were prepared as described (25) and
extracted with 1% Triton X-100 and 1% sodium deoxycholate in PBS
containing a protease inhibitor mixture (10 µg/ml leupeptin, 10 µg/ml aprotinin, and 1 mM aminobenzamidine). Cauda
epididymal sperm (eS) were also extracted with 1% acetic
acid containing 5 mM acetic acid. Each lane contained 5 µg of protein separated by SDS-PAGE following reduction with 40 mM DTT. After electrophoresis, the proteins were
transferred to Immobilon-P, and the membrane was probed with anti-AM67
and developed by enhanced chemiluminescence methods.
[View Larger Version of this Image (81K GIF file)]
Fig. 4.
Indirect immunofluorescence demonstrating
localization of AM67 to the apical segment of the acrosome. Sperm
were attached to polylysine-coated coverslips. Following fixation with
4% paraformaldehyde, the sperm were permeabilized with methanol
(20 °C). The coverslips were treated with anti-AM67 (1:100
dilution) and fluorescein isothiocyanate-conjugated goat anti-rabbit
IgG and viewed by fluorescence microscopy to localize AM67 protein.
AM67 protein was exclusively restricted to the apical segment of the
acrosome. No staining was observed in the principal or equatorial
segment of the acrosome. In addition, staining was not observed in
sperm treated with preimmune serum instead of anti-AM67 or in sperm
that had not been permeabilized prior to antibody treatment.
A and B, paired phase-contrast and fluorescence
micrographs of anti-AM67-treated sperm; C and D, paired phase-contrast and fluorescence micrographs of sperm treated with preimmune serum. Arrows in A and
B indicate the apical segment region of the sperm
acrosome.
[View Larger Version of this Image (94K GIF file)]
Fig. 5.
Immunoelectron microscopic localization of
AM67 in the M1 (dorsal bulge) domain of guinea pig sperm. Guinea
pig sperm were fixed and embedded in LR White resin. Thin sections were processed with anti-AM67 as described under "Experimental
Procedures." The immunogold particles (arrows) were found
solely within the M1 (dorsal bulge) domain of the lumen of the
acrosome. In general, the gold particles were excluded from spherical
bodies within this region. No staining was observed in the M2
(intermediate zone) or M3 (ventral zone) region. Magnification × 51,000.
[View Larger Version of this Image (94K GIF file)]
, diverging only at
the last four amino acids (KEKY in AM67 and LQRD in human C4BP
)
(11). Four additional tryptic peptides were subsequently
sequenced and also found to be homologous to C4BP
(Figs.
6 and 7).
Fig. 6.
Cloning of cDNAs encoding AM67 and
testis-specific expression of AM67 mRNA. A, the deduced
amino acid is shown below the nucleotide sequence numbered in the 5 to
3
direction. The amino-terminal amino acids of AM67 tryptic peptides
are underlined. Potential sites of asparagine-linked
glycosylation are indicated by asterisks at amino acids 68, 77, 185, 191, 200, 433, and 455. B, AM67 is a
testis-specific gene product. A message of ~1.9 kilobases was found
in testis (T), but not in heart (H), brain
(Br), liver (L), or kidney (K).
[View Larger Version of this Image (46K GIF file)]
Fig. 7.
Analysis of AM67 amino acid sequence.
A, comparison of AM67, mouse sp56 (mSP56;
GenBankTM accession number A56740[GenBank]), human C4BP (hC4BP;
GenBankTM accession number P04003[GenBank]), and mouse C4BP (mC4BP;
GenBankTM accession number P08607[GenBank]) aligned by the Clustal V program and
additional visual adjustments. Identical residues among AM67 and the
other proteins are shaded. Homologous regions (conservative
substitutions) are boxed using the Kyte-Doolittle algorithm
of the SeqVu Version 1.0.1 program. The bar above residues
346-395 indicates a region of high homology between AM67 and mouse
sp56. The corresponding region in human C4BP is much less homologous,
and there is no corresponding region in mouse C4BP. B,
identification of SCRs in AM67, a characteristic feature of members of
the complement-binding superfamily. AM67 has seven complete SCRs and
one incomplete SCR, with the cysteine residues being completely
conserved among the seven complete SCRs. The first cysteine of the
incomplete SCR is also present. The numbers to the left and right of
each SCR represent the amino acid residues.
[View Larger Version of this Image (118K GIF file)]
gt11 guinea pig testis
cDNA library as the template. A product near the expected size
(1197 base pairs) was obtained and found by DNA sequencing to encode a
fragment homologous to C4BP
. The PCR product was then labeled by
random priming using digoxigenin-labeled dUTP. The digoxigenin-labeled
probe was then used to screen the cDNA library using an
anti-digoxigenin antibody and the ECL detection system. From a screen
of 120,000 plaques, six positives were obtained after three rounds of
screening and were subcloned into the NotI site of
pGEM-11Zf(+). Preliminary DNA sequencing of these plasmids indicated
that all six of the clones were true positives, encoding a C4BP
-like
protein. Two of the plasmids containing the largest inserts, p67-9 and
p69-10, were chosen for complete sequencing of the insert and found to
encode the complete open reading frame of the protein (Fig.
6A). Plasmid p67-9 represented bases 1-1735, and p67-10
represented bases 83-1864 in the composite sequence. Overall, the
composite sequence encoded 83 nucleotides of 5
-untranslated region and
82 nucleotides of 3
-untranslated region. Although no poly(A) tail was
observed in these clones, a possible poly(A) signal (TATAAA) was found
at positions 1851-1856. An in-frame stop codon was found 9 bases
upstream from the predicted translation start codon, demonstrating that
the open reading frame of the cDNA was full-length. Northern
analysis of RNAs extracted from various tissues indicated that the
1.9-kilobase AM67 mRNA was specific to the testis (Fig.
6B).
family of proteins
(Fig. 7A). Although guinea pig C4BP
has
yet to be cloned and sequenced from the liver (the tissue that
synthesizes bona fide C4BP
), the conclusion that AM67 is different
from C4BP
was confirmed by partial DNA sequence analysis of a
reverse transcription-PCR product from guinea pig liver using
degenerate primers for C4BP
(data not shown). The highest homology
of AM67 to C4BP
proteins from different species was to the human
protein, followed by bovine, rat, and mouse. Based upon the molecular
weight of the unreduced form of AM67 and its homology to C4BP
, we
hypothesize that six to eight monomers oligomerize to form the
unreduced, high molecular weight aggregate (Mr
~380,000) of AM67. There was also significant homology to porcine
apolipoprotein R and gene fragments from a putative C4BP
pseudogene,
human C4BPAL1, as well as other members of the complement
regulatory protein superfamily (data not shown). Like other members of
the complement regulatory protein family, AM67 contains a repetitive
structure called a short consensus repeat (SCR). Although human and
bovine serum C4BP
proteins contain eight SCR units, AM67 was found
to contain seven complete SCR units and an eighth incomplete SCR.
Alignments of the SCR units indicate that the sixth SCR of AM67
approximately corresponds to the first half of the sixth SCR and the
last half of the seventh SCR of human C4BP
(Fig. 7B).
Each SCR of AM67 (except the eighth incomplete SCR) contained four
cysteine residues, which have been shown to be important in
establishing and maintaining the three-dimensional structure of each
SCR in human C4BP
(35).
sequences (Fig.
7A, bar). However, another region in mouse sp56
(residues 414-453) had no corresponding counterpart in guinea pig AM67
and was divergent from the corresponding regions in human and mouse
C4BP
proteins. More important, mouse sp56 was more homologous to
guinea pig AM67 than to mouse C4BP
.
Fig. 8.
Ultrastructural localization of sp56 in
mature mouse sperm by post-embedding immunoelectron microscopy.
Shown is the localization of mouse sp56 within the dorsal region of the
acrosome (asterisks) overlying the nucleus (n) in
longitudinal (A; magnification × 30,000), cross
(B; magnification × 31,000), and oblique
(C; magnification × 26,000) thin sections through
mature mouse sperm treated with sp56-specific monoclonal antibody 7C5.
Control incubations with an equal concentration of nonspecific whole
IgG demonstrated the lack of nonspecific antibody interactions with
sperm (not shown).
[View Larger Version of this Image (90K GIF file)]
Fig. 9.
Indirect immunofluorescence of
paraformaldehyde-fixed and live mouse sperm. Shown are
corresponding phase-contrast (A and C) and
immunofluorescence (B and D) micrographs of
mature mouse sperm immunostained with anti-sp56 monoclonal antibody
7C5. Paraformaldehyde-fixed sperm (A and B)
showed bright acrosomal fluorescence localization of sp56
(B), whereas live unfixed sperm (C and
D) showed no immunofluorescence (D).
Magnification × 1400.
[View Larger Version of this Image (65K GIF file)]
, we estimate that six to eight
monomers of AM67 oligomerize to constitute the unreduced form of the
protein (Mr ~380,000). Based upon the C4BP
sequence homology, AM67 might be predicted to form
calcium-dependent complexes with other ligands within the
acrosome or, after exocytosis, with extracellular ligands. Comparable
calcium-dependent binding of AM50 to the apical segment
complex (containing the acrosomal matrix) has already been
demonstrated (11).
and the pentraxins are known to bind to plasma membranes, and
based upon the homologies to these proteins, AM67 and AM50 might also
be expected to interact with components of the acrosomal membranes. If
AM67 and AM50 bind to different ligands of the acrosomal membrane, aggregation of these proteins with their ligands during acrosome biogenesis may help to establish the M1 and M3 domains. Additionally, AM50 and AM67 may interact with membranes or react with each other upon
their release following the acrosome reaction and the dissolution of
the acrosomal matrix.
proteins are generally considered to be
liver-specific proteins.
*
This work was supported in part by National Institutes of
Health Grants HD06274 and HD22899 (to G. L. G.), Grants HD-07305 and
HD-07792 (to J. A. F.), and Grant HD-20419 (to G. E. O.) and by a
grant from the Korea Research Foundation (to K.-S. K.).The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement" in
accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
**
To whom correspondence should be addressed: Center for Research on
Reproduction and Women's Health, University of Pennsylvania Medical
Center, 306 John Morgan Bldg., Philadelphia, PA 19104-6080. Tel.:
215-662-6062; Fax: 215-349-5118; E-mail:
ggerton{at}obgyn.upenn.edu.
1
The abbreviations used are: PAGE, polyacrylamide
gel electrophoresis; HPLC, high-performance liquid chromatography; PBS,
phosphate-buffered saline; PCR, polymerase chain reaction; C4BP,
complement component 4-binding protein; DTT, dithiothreitol; SCR, short
consensus repeat.
2
C. Blobel, unpublished observation.
3
Available by anonymous ftp at
gimr.garvan.unsw.edu.au or 129.94.136.101, directory/pub.
©1997 by The American Society for Biochemistry and Molecular Biology, Inc.