BIOLOGY O F REPRODUCTION 36, 1170-1176 (1987)
Effect of Aryl 4-Guanidinobenzoates on the
Acrosin Activity of Human Spermatozoa'
J . M. KAMINSKI,2,3 D. SMITH,3 D. S. REID,3
W. KENNEDY,3 R. S. JEYENDRAN4 and L.J.D. ZANEVELD3
Department o f Obstetrics and Gynecology3
Rush University
Rush-Presbyterian-St. Luke's Medical Center
Chicago, Illinois 60612
and
Institute f o r Reproductive Medicine4
Chicago, Illinois 60602
ABSTRACT
Certain aryl 4panidinobenzoates ( A G s ; inhibitors of proteinuses, including the sperm enzyme acrosin) have
been shown to be more potent vaginal contraceptives in rabbits and less toxic than nonoxynol-9, the active
ingredient of most marketed vaginal contraceptive formulations. To determine if these AGs can contact sperm
and inhibit acrosin when mixed with the entire human ejaculate for a short period of time (roughly imitating
clinical conditions), the inhibitors were added t o semen at various concentrations for 2 min, after which the
seminal plasma and unbound inhibitor were removed fr om the sperm by Ficoll centrifugation. Subsequently,
the total arginine amidolytic activity of the spermatozoa was determined spectrophotometrically after a combined treatment that resulted in extraction, proacrosin activation, and reaction with substrate. Dose-response
cuyves were prepared. All A G s studied were effective inhibitors of the amidolytic activity under these conditions, with EDs0 values (the dose levels at which half of the acrosin associated with lo6 sperm is inhibited)
ranging f r o m lo-' t o
M. To determine the effect on the proteolytic activity of individual spermatozoa, the
experiment was repeated with 4'-acetamidophenyl 4-guanidinobenzoate ( A G B ) , and the protease released f r o m
the sperm was measured b y the gelatinplate assay. The inhibition results were similar t o those obtained b y
extraction of the spermatozoa and measurement of amidolytic activity. Thus, when mixed with the human
ejaculate, AGs interact rapidly with spermatozoa t o inhibit both their arginine amidolytic and proteolytic
activity (probably due primarily or only t o inhibition of acrosin) and remain bound even after removal of the
seminal plasma. These data encourage further study of the compoitnds f o r contraceptive purposes.
INTRODUCTION
Acrosin, a sperm-specific acrosomal proteinase
with trypsin-like specificity and inhibitor sensitivity,
has an essential role in the fertilization process,
although its exact site of action remains to be firmly
established (Rogers and Bentwood, 1982). A variety
of experiments have been performed by a number of
laboratories to show that naturally occurring and
synthetic proteinase (including acrosin) inhibitors can
prevent both in vitro and in vivo fertilization in
different animal species (Zaneveld, 1976). Additional-
Accepted December 1. 1986.
Received September 4, 1986.
'The research was supported by PARFR (AID) 3 3 8 and NIH H D
19555.
'Reprint requests: J. M. Kaminski, Ph.D., Ob/Gyn Research,
Rush-Presbyterian-St. Luke's Medical Center, 175 3 West Congress
Parkway, Chicago, IL 60612.
ly, it was proposed that synthetic acrosin inhibitors
may be potent vaginal contraceptives (Zaneveld,
1982), but the inhibitors employed until several years
ago were generally too toxic for clinical use. One
of the most effective inhibitors of human acrosin is
the aryl 4-guanidinobenzoate (AG) containing 4nitrophenol (NPGB), which was shown to be equally
as contraceptive as nonoxynol-9 in the primate, but
at much lower concentrations (Zaneveld e t al., 1979).
Nonoxynol-9 is the active ingredient of most marketed
vaginal contraceptives.
Using NPGB as the lead compound, a number of
AGs were synthesized possessing phenols already
marketed for human use, i.e., methyl salicylate,
acetaminophen, etc. (Kaminski et al., 1986). These
compounds were shown t o be highly reactive, probably
pseudo-irreversible inhibitors of extracted, purified
human acrosin, exhibiting acylation rates in the order
1170
ACROSIN INHIBITORS
of 10-2--10-3 s-’ and Km values of approximately
10-7-10-9 M (Kaminski and Diao, 1984). The AGs
also prevent mouse in vitro fertilization (Beyler and
Zaneveld, 1982) and the fertilizing capacity of human
spermatozoa, as assessed by their inability to penetrate
zona-free hamster eggs (Van der Ven et al., 1985).
Vaginal contraceptive studies in rabbits have shown
that a number of the AGs are orders of magnitude
more potent than nonoxynol-9 (Kaminski et al.,
1985a). In addition, the acute toxicity of the
synthesized AGs containing marketed phenols was
generally much less than that of nonoxynol-9
(Kaminski et al., 1986). Based on their contraceptive
activity and low toxicity, four of the AGs possessing
acetaminophen (AGB), ethyl paraben (EGB), methyl
salicylate
(MSGB) and 4-methylumbelliferone
(MUGB) as the phenol groups, were selected for
further study as vaginal contraceptives.
At midcycle, spermatozoa begin to pass from the
human vagina into the cervix within a few min. Thus,
only a short time period is available for vaginal agents
to contact spermatozoa. Additionally, the fluid
portion of the ejaculate does not pass into cervical
mucus, i.e., the spermatozoa are “washed” upon
entry into the cervix (Hafex, 1976). For these reasons,
the ability of the AGs t o bind to human spermatozoa
and inhibit acrosin under in vitro conditions simulating
the clinical situation was determined by adding the
compounds t o whole ejaculates for 2 min, washing
the spermatozoa over Ficoll, and determining their
acrosin activity. Two different tests were employed: a
spectrophotometric assay, which measures the total
arginine amidolytic activity in extracts of a relatively
large number of spermatozoa simultaneously, and the
gelatin-plate assay, which approximates the activity
of protease released from individual spermatozoa.
Preliminary data have been published in abstract form
(Kaminski et al., 1985b).
MATERIALS A N D METHODS
Materials
We synthesized 4’-acetamidophenyl 4guanidinobenzoate hydrochloride (AGB), 4’-carbethoxyphenyl
4-guanidinobenzoate hydrochloride (EGB), and 2’carbomethoxyphenyl 4panidinobenzoate hydrochloride (MSGB) as previously reported, using a
modified dicyclohexylcarbodiimide (DCC)-coupling
procedure employing 4-toluenesulfonic acid as an
acid catalyst (Kaminski et al., 1986). Their purity was
1171
determined by thin-layer chromatography (TLC),
recrystallization t o constant melting point, nuclear
magnetic resonance spectra, and elemental analysis
for carbon, hydrogen, nitrogen and halogen. We
purchased 4’-methylumbelliferyl4-guanidinobenzoate
hydrochloride (MUGB) and 4’-nitrophenyl 4-guanidinobenzoate hydrochloride (NPGB) from Sigma
Chemical Company (St. Louis, MO). Dimethylsulfoxide, Ficoll (Type 400), a-N-benzoyl-DLarginine p-nitroanilide (BAPNA), methylene blue,
HEPES (N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid), Trizma base (Tris [hydroxymethyl]
aminomethane) and Triton X-100 were also obtained
from Sigma. Kodak Ar-10 plates were obtained from
Kodak (Rochester, NY).
Test Procedure
Semen was obtained from apparently normal,
healthy volunteers by self-masturbation. For each
series of experiments, at least 5 ejaculates were
pooled after complete liquefaction had taken place.
Only those pooled ejaculates were used that contained
50-70 X lo6 sperm/ml with greater than 60%
motility and 85% normal morphology. The aryl
4-guanidinobenzoates were initially dissolved in 100%
dimethyl sulfoxide (DMSO) and subsequently diluted
with H,O t o the desired concentrations. If the dilution resulted in less than 1% DMSO, DMSO was
added so that its final concentrations in solution was
always 1%. The inhibitors were mixed with pooled
semen, so that the final concentrations of inhibitor
t o 1.0 X
M. This
ranged between 1.0 X
mixture was incubated at room temperature for 2
min, after which time an amount of the ejaculate
containing 32 X lo6 sperm (varying from 250 to 368
~ 1 depending
,
on the sperm concentration of the
pooled sample) was carefully layered over 500 pl 11%
Ficoll (in 0.12 M NaC1, 0.025 M HEPES, pH 7.5) in a
centrifuge tube. A DMSO control (1% DMSO in
water) and a medium control (water only) were
always performed simultaneously. The tube was
centrifuged at 1000 X g for 30 min. This procedure
caused the sedimentation of the spermatozoa, whereas
the fluid portion of the ejaculate, containing seminal
plasma and unbound inhibitor, remained above the
Ficoll. The Ficoll and supernatant fluid were subsequently removed from the sperm pellet with a
plastic pipette. A background control was also run,
which consisted of layering untreated seminal plasma
over the Ficoll and centrifugation. The seminal
1172
KAMINSKI ET AL.
plasma was obtained from a sample of the same
pooled semen by centrifugation at 1000 X g for 15
min, and collection of the supernatant solution. After
Ficoll centrifugation, any precipitate from the
seminal plasma was treated identically to the spermatozoa, and the resultant solution was used as background against which the absorbance of the solutions
was determined.
Spectrophotometric Assay
The test system used was basically the same as the
arginine amidolytic assay developed for clinical
purposes (Kennedy et al., 1985). The washed sperm
pellet was suspended in 1.0 ml of a TRIS/Triton
buffer, pH 8.0, containing 0.1 M Trizma Base, 0.01%
Triton X-100, 0.05 M NaCI, 10% DMSO, and 23 mM
BAPNA. The mixture was incubated for 3 h at 25"C,
which caused dispersal of the sperm membranes, the
release of most of the acrosomal contents including
acrosin and proacrosin (the inactive form of acrosin),
the activation of proacrosin, and substrate hydrolysis.
After incubation, the sperm mixture was centrifuged
at 1000 X g for 5 min and the supernatant collected.
The absorption of the supernatant was measured with
either a Hitachi Model 100-40 or a Gilford 250
spectrophotometer at 410 nm. The percentage of
inhibition of amidolytic activity was taken as the
difference in optical density at 410 nm between the
control and inhibitor-treated sperm, divided by the
optical density of the control and expressed as
percent.
Gela tin-Pla te Assay
A modification of the method of Schill (Schill et
al., 1981) was used. Kodak AR-10 plates were fixed
in a darkroom for a minimum of 20 s, after which
they were washed in running cold water for at least
20 min. The plates were dried overnight at room
temperature and cut into 0.5- X 0.5-cm squares. The
squares were floated in cool, distilled water until
they were approximately double in size. Three
squares were glued to a microscope slide and dried
overnight. They were subsequently stained with 0.1%
methylene blue for 4 min, washed repeatedly (first in
tap water and then in distilled water), and dried
overnight, at which point they were ready for use.
After Ficoll centrifugation and removal of the supernatant, the pelleted sperm were resuspended in 1 ml
Biggers, Whitten and Whittingham (BWW) medium,
pH 7.4, the sperm concentration was determined with
a Neubauer hemacytorneter, and the sperm concentration was adjusted with BWW medium to 3 X lo6
sperm/ml. A small amount (2.5 pl) of the sperm
suspension was spread evenly over one of the small
square plates with a glass rod. After incubation for 30
min at 37"C, the squares were air-dried and studied
with the light microscope for the occurrence of
gelatin lysis ("halo" formation) around the sperm
head. One hundred randomly selected spermatozoa
were studied and observed for the presence or absence
of lysis around the sperm head. For statistical purposes,
no differentiation between the amount of lysis
formation was made, i.e., small vs. large.
RESULTS
Spectrophotometric Assay
AGs are probably pseudo-irreversible inhibitors of
human acrosin (Kaminski and Diao, 1984), and can
be completely or partially removed from the enzyme
by extensive dialysis (unpublished results). Therefore,
the standard method (Goodpasture et al., 1981) of
sperm extraction at acidic pH, extensive dialysis of
the extracts, and measurement of the activity before
and after activation could not be used. The assay
system employed herein contained detergent (Triton
X-100) to extract the acrosomal contents (including
acrosin and proacrosin) from the spermatozoa, a
buffer at pH 8.0 to allow activation of extracted
proacrosin, and substrate (BAPNA) to measure the
arginine amidolytic activity so that no dialysis was
required and no loss in bound inhibitor could occur.
The results with this assay system are comparable to
those obtained with the standard test method
described by Goodpasture et al. (1981), and no
substrate hydrolysis occurs in the presence of
benzamidine (Kennedy et al., 1985).
The inhibitors used for these studies were the lead
compound (NPGB) and its four analogs (AGB,
MSGB, EGB and MUGB), which were selected
because of their potential clinical application. All the
inhibitors were tested at a minimum of six concentrations in 1%DMSO and water. This DMSO concentration had no effect upon the total arginine amidolytic
activity of the spermatozoa, as compared to treatment with water only. Each concentration of inhibitor
was tested on a minimum of four different pooled
ejaculates. Dose-response curves were established for
each of the AGs according to the method of Litch-
1173
ACROSIN INHIBITORS
field and Wilcoxon (1949) by plotting percentage of
inhibition of arginine amidolytic activity versus the
log dose (pg/ml) (Fig. 1). The E D l 6 , E D s 0 , and EDg4
of each inhibitor (the dose at which 16%, 50%, and
84% of the amidolytic activity is inhib.ited) were
calculated and are listed in Table 1.
All five AGs effectively decreased the arginine
amidase levels in the spermatozoa. The EDs0 values
M. EGB
of the inhibitors ranged from
to
was particularly effective, causing a 50% decrease in
activity of the sperm at a concentration of 6.15 X
M in semen. NPGB and MUGB were somewhat
less effective, inactivating 50% of the total activity
M. AGB and MSGB
at concentrations of about
were the least active of the inhibitors, with ED50
values of approximately lo-' M. The ED,,,s of the
inhibitors varied to a lesser extent, ranging from
about
t o l O P M, but were still about 10 times
higher for EGB, MUGB, and NPGB than for AGB and
MSGB.
I /
-
-0s
0.0
0-0
Gelatin-Plate Assay
To determine the effect of the AGs on individual
spermatozoa, the tests were repeated with one of the
AGs (AGB), and the ability of the Ficoll-centrifuged
spermatozoa to digest gelatin was assessed. Three
to lo-' M)
concentrations of AGB were tested
and the results are summarized in Table 2. Studies were
initially performed to determine the optimal length of
gelatin incubation using untreated, Ficoll-centrifuged
spermatozoa. Gelatin digestion increased up to 30
min of incubation, at which point 14-25% of the
spermatozoa showed large areas of digestion and 3368%of the sperm showed some lysis. Longerincubation
periods (up t o 2 h) did not increase the percentage of
sperm that lysed the gelatin or enhanced the magnitude of digestion, so that the 30-min period
was selected as the standard incubation time.
The experiments were repeated four times. The
three AGB concentrations, the 1% DMSO control,
and the medium control were tested simultaneously.
Because of the inconsistency of the assay, the results
varied somewhat between the experiments. However,
even at lo-' M AGB, large digestion areas were
formed by the spermatozoa in only one of the
four experiments. Only a single sperm in one of the
experiments showed a large area of lysis at lo-, M
AGB, and none at
M. Some lysis was observed in
all experiments by the spermatozoa incubated with
lo-' and
M AGB, but only in two of the experi-
0.5
1.0
LOG DOSE ( p ~ / m O
NPGB
J+-*
MUGB
1.s
2.0
/
-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0
LOG DOSE ( pg/ml)
FIG. 1. Inhibition of acrosin activity by the aryl 4-guanidinobenzoates using the spectrophotometric assay (see t e x t for details).
Dose-response curves were constructed for each inhibitor according to
Litchfield and Wilcoxon (1949). Top graph: EGB = 4'-carbethoxyphenyl 4-guanidinobenzoate hydrochloride; AGB = 4'-acetamidophenyl
4-guanidinobenzoate hydrochloride; MSGB = 2'-carbomethoxyphenyl
4-guanidinobenzoate hydrochloride. B o t t o m graph: NPGB = 4'-nitrophenyl 4-guanidinobenzoate hydrochloride; MUGB = 4'methylumbelliferyl4-guanidinobenzoate hydrochloride.
1174
KAMINSKI ET AL.
TABLE 1 . Effect of aryl 4-guanidinobenzoates o n the acrosin activity of human spermatoz0a.a
~
~
~~
Inhibitor
ED16 (M)
ECB
NPGB
MUGB
AGB
MSGB
1.70 X
2.37 x
1.60 X
6.80 X
1.93 x
lo-*
lo-’
EDSO (M)
ED84 (M)
6.15 X l W 7
1.86 X
6.72 X
3.61 x 1 0 - 5
5.38 x
3.30 x 10-5
1.67 x
2.87 x 10-5
2.13 x
2.02 x
aSee text for experimental detail. Confidence limits of 95% were calculated for the ED,, values according to the method of Litchfield and
M), AGB ( 3 . 5 3 M), MUGB ( 4 . 4 9 - 13.7 X
NPGB ( 1 . 3 2 - 2.28 X
Wilcoxon (1949) and were as follows: EGB ( 4 . 4 6 - 7.92 X
3.66 X l o - ’ M), MSGB ( 4 . 3 2 6 . 8 6 X
M)
~
ments when incubated with
M AGB. The DMSO
control tended to decrease the number of sperm that
lysed gelatin, although this varied between the
experiments. At
M AGB, mean lysis of the
spermatozoa in all four experiments was about 46%
of the DMSO control; at
and
M, it was
about 16%.
DISCUSSION
During the last decade, several laboratories have
attempted to develop acrosin inhibitors as vaginal
contraceptives, using sterol sulfates (Burck and
Zimmerman, 1980; Burck et al., 1982; Zimmerman et
al., 1983), N-carbobenzoxy amino acid esters (Hall et
al., 1979; Drew et al., 198l), and the aryl 4-guanidinobenzoates (Kaminski et al., 1986). At least in
nonhuman species, the aryl 4-guanidinobenzoates are
much more potent and less toxic than nonoxynol-9,
encouraging their further development for clinical
use. Under optimal estrogen stimulation, cervical
mucus properties of the human are such that sperm
penetration starts within a few min. As the spermatozoa enter the cervical mucus, they leave the
fluid portion of the ejaculate behind. Thus, all soluble
material tends to remain in the vagina. For these
reasons, it was important to establish that 1.)the AGs
can rapidly bind to human spermatozoa when mixed
with the entire ejaculate, 2.) such binding can occur
at concentrations low enough to be of contraceptive
use, and 3 .) the AGs remain bound to the spermatozoa
when the sperm are separated from the fluid portion
of the ejaculate containing the inhibitors. In vitro
conditions were developed to simulate approximately
the natural condition by incubating the ejaculate with
inhibitor for 2 min maximally and immediately
separating the sperm from the fluid portion of the
ejaculate by Ficoll centrifugation. Molecular material
including soluble inhibitor will not precipitate at the
concentration of Ficoll used in the experiments. As
far as we know, such studies have never been performed previously with any other of the acrosin inhibitors.
The five AGs tested included the AGB, EGB,
MSGB, MUGB, and NPGB derivatives. All these
phenol moieties, with the exception of 4-nitrophenol,
are already marketed for clinical use. The first three
compounds were selected from the other synthesized
aryl 4-guanidinobenzoates because of their high
vaginal contraceptive activity in the rabbit and their
low acute toxicity. At 100 pg/ml (approximately 2.9
X
M) (2 mlhabbit), these AGs caused a 90-99%
mean reduction in fertilization, as compared to
vehicle-treated controls. Even 100-fold higher concentrations of nonoxynol-9 did not decrease fertilization
to the same extent (a mean decrease of 83.7% was
obtained) under exactly the same conditions (Kaminski et al., 1985a). These results also show that although the AGs are probably pseudo-irreversible
inhibitors of human acrosin, and can be completely
or partially removed by extensive dialysis, their
removal from the sperm in the female genital tract is
apparently so slow that the contraceptive effect is
TABLE 2 . Gelatin-plate test with 4‘-acetamidophenyl 4panidinobenzoate (AGB).*
_____~
~
Percent of sperm that caused
gelatin digestion (halo formation)
Experiment no.
Incubator
1
2
3
4
Mean
Controla
1% D M S O ~
M AGBC
82
21
23
3
70
ND
15
11
18
57
80
50
9
74
50
5
8
16
70.8
50.3
23.3
7.8
8.5
M AGBC
M AGBC
0
0
*See text for experimental detail. All data were normalized to 100
sperm. ND = not determined.
a’b’cThe treatments marked with a different superscript are statistically significantly different from each other (Student-Newman-KeuIs
Multiple Range Test, p < 0 . 0 5 ) .
ACROSIN INHIBITORS
retained. The LDsO (dose level that will produce
death in 50% of the animals) of these AGs on i.p.
injection in mice was 300 mg (EGB), 750 mg (MSGB),
and 850 mg (AGB). By contrast, the LDsO of
nonoxynol-9 was found to be 190 mg/kg (Kaminski
et al., 1986). MUGB was less effective than the other
AGs on vaginal application to rabbits, but it has very
low acute toxicity (LDsO > 2000 mg/kg) and may
also be potentially useful. NPGB was employed in the
present study because it was the lead compound.
However, NPGB is of no clinical interest, because
4-nitrophenol is released upon reaction with
proteinase, which is potentially toxic.
The present results indicate that the AGs satisfy
the requirements for human vaginal contraceptives as
stated above, at least under the in vitro conditions
used. At relatively low concentrations, the AGs bind
rapidly to human spermatozoa when added to the
ejaculate and remain bound even after separation of
the sperm from the supernatant solution. Although
some of the inhibitor is probably removed from the
sperm during Ficoll centrifugation, it is likely that
most remains associated with the sperm, because
linear dose-response curves were obtained. The site at
which the AGs bind to the sperm is not known. One
can speculate that the compounds interact directly
with acrosin/proacrosin, but the experiments do not
eliminate the possibility that the agents become
membrane-bound and interact with the enzyme only
after membrane rupture and release of the acrosomal
contents. However, in either case, the proteolytic
activity of the spermatozoa is inhibited, as the
gelatin-plate assay confirms. Of the four inhibitors
tested that have potential clinical application, EGB
was the most potent vaginal contraceptive in rabbits,
and was also the most active under the present in
vitro conditions. AGB and MSGB were about equally
as contraceptive in rabbits, and gave similar enzyme
inhibition on addition to the human ejaculate.
Interestingly, MUGB, which is a potent inhibitor of
acrosin (Beyler, 1980), was more effective under the
present conditions than AGB or MSGB, but was less
effective as a contraceptive in the rabbit (Kaminski
et al., 1985a).
In our laboratory, we have not been able t o find
any evidence for another serine proteinase in human
spermatozoa (besides acrosin) that has arginine
amidolytic or esterolytic activity and is inhibited by
typical trypsin inhibitors (e.g., Anderson et al.,
1981). Additionally, only low arginine esterase and
1175
amidase activity is present in human sperm extracts
until proacrosin activation takes place (Goodpasture
et al., 1981). In other species, the presence of another
arginine esterase besides acrosin (ninhibin) that is
inhibited by typical trypsin inhibitors has been
suggested (Anand-Srivastava et al., 1985; Johnson et
al., 1985), but the evidence is not convincing and the
enzyme may represent only a different molecular
form of acrosin, which is known t o exist (e.g., Parrish
and Polakoski, 1979). Although it cannot be excluded that such an enzyme occurs, it must constitute
only a very minor portion of the arginine esterolytic
or amidolytic activity, at least in the case of human
spermatozoa. Thus, minimally, the large majority of
the enzyme measured by the spectrophotometric
assay is acrosin. Finally, from a contraceptive standpoint, it does not matter how many arginine amidases
are inhibited, as long as the functionally active one
(acrosin) loses activity. Since linear dose-response
curves were obtained, it appears that all arginine
amidolytic activity of the sperm is inhibited, whether
this represents acrosin alone or acrosin and another
enzyme.
The spectrophotometric assay technique used to
measure total acrosin on the sperm eliminates possible
loss of inhibitor during extraction, centrifugation and
dialysis, which are part of the standard acrosin/
proacrosin extraction and assay procedure (Goodpasture et al., 1981). However, a drawback of this
method is that the amount of proacrosin cannot
be estimated, only the total acrosin after proacrosin
activation. Thus, it is not known whether the decrease in total acrosin is due t o the inhibition of
proacrosin conversion t o acrosin, t o the inhibition of
the acrosin that was formed after proacrosin activation, or to both. In either case, it appears that the
acrosin that can ultimately be formed by the sperm
(and potentially used for its functional activity) is
inhibited by the AGs because linear dose-response
curves were obtained and because the standard
extraction and assay technique, which has been
shown to extract all the proacrosin and acrosin from
human sperm (Goodpasture et al., 198l), gives the
same total acrosin activity (after activation) as the
present method (Kennedy et al., 1985).
The spectrophotometric assay does not differentiate
whether, e.g., at the EDs0, 50% of the acrosin from
all the sperm is inhibited or whether 100% of the
acrosin from half of the sperm is inhibited. No good
assay system is currently available to measure the
1176
KAMINSKI ET AL.
acrosin activity of individual spermatozoa. The
gelatin-plate test is the best, but measures only released protein protease, which probably constitutes
no more than 20-30% of the total acrosin on the
sperm (Bhattacharyya and Zaneveld, 1982). The
assay is not precise, and as a result variations were
found among individual experiments. Even so, the
results with the gelatin-plate test confirmed those
obtained with the spectrophotometric assay. At lo-’
M AGB, gelatin lysis of approximately half of the
spermatozoa was inhibited, as compared to the
DMSO control, and at
M. lysis of approximately
85% of the sperm was prevented. N o further decrease
in lysis occurred with an increase of AGB to
M.
In contrast to the medium and DMSO controls, large
areas of lysis around the sperm were rarely found
after incubation with AGB. This tends to suggest that
the inhibitors can decrease part of the acrosin activity
of individual spermatozoa as well as inhibit all of the
acrosin, the latter being more prevalent. DMSO had
no effect on the amidolytic activity of the spermatozoa, but did tend to decrease their proteolytic
activity. I t is possible that DMSO interferes with the
binding of acrosin to gelatin or that it stabilizes the
membranes, allowing less spontaneous release of
acrosin.
I t is estimated that 1 to 10 mg/ml (approximately
3-30 X
M) of the AGs will ultimately be
applied to the human for contraceptive purposes.
Much lower dose levels were contraceptive in the
rabbit. The present experiments indicate that these
concentrations should lead to complete inhibition of
acrosin on all or almost all the spermatozoa when
mixed with the human ejaculate. Together with the
rapid binding and continued association with the
sperm after washing, these results encourage further
development of the AGs for clinical use.
ACKNOWLEDGMENTS
The authors appreciate the manuscript preparation of Ms. Barbara
Hartrampf.
REFERENCES
Anand-Srivastava MB, Johsnon RA, Picard S, Cantin M, 1985. Ninhibin: a
sperm factor attenuates the atrial natriuretic factor mediated
inhibition of adenylate cyclase: possible involvement of inhibitory
guanine nucleotide regulatory protein. Biochem Biophys Res
Commun 129: 171-78
Anderson RA, Beyler SA, Mack SR, Zaneveld LJD, 1981. Characterization of a high-molecular-weight form of human acrosin. Biochem
J 199:307-16
Beyler SA, 1980. Evaluation of acrosin inhibitors as antifertility agents.
Ph.D. Dissertation. University of Illinois a t Chicago, Health
Sciences Center
Beyler SA, Zaneveld LJD, 1982. Inhibition of in vitro fertilization of
mouse gametes by proteinase (acrosin) inhibitors J Reprod Fertil
66:425-31
Bhattacharyya AK, Zaneveld LJD, 1982. The sperm head. In: Zaneveld
LJD, Chatterton R T (eds.), Biochemistry of Mammalian Reproduction. New York: J. Wiley, pp. 119-51
Burck PJ, Thakkar AL, Zimmerman RE, 1982. Antifertility action of a
sterol sulfate in t h e rabbit. J Reprod Fertil 66:109-12
Burck PJ, Zimmerman RE, 1980. Th e inhibition of acrosin by sterol
sulfates. J Reprod Fertil 58:121-25
Drew JH, Loeffler LJ, Hall IH, 1981. Antifertility activity of Nprotected glycine activated esters. J Pharm Sci 70:60-63
Goodpasture JC, Polakoski KL, Zaneveld LID, 1981. Acrosin, proacrosin, and acrosin inhibitor of human spermatozoa: extraction,
quantitation, and stability. J Androl 1:16-27
Hafez ESE, 1976. Transport and survival of spermatozoa in t he female
reproductive tract. In: Hafez ESE (ed.), St. Louis: Human Semen
and Fertility Regulation of Men, pp. 107-21
Hall IM, Drew JH, Sajadi 2, Loeffler LJ, 1979. Antifertility and antiproteolytic activity of activated N-carbobenzoxy amino acid
esters. J Pharm Sci 68:696-98
Johnson RA, Jakobs KH, Schultz G, 1985. Extraction of the adenylate
cyclase-activating factor of bovine sperm and its identification as a
trypsin-like protease. J Biol Chem 260:114-21
Kaminski JM, Bauer L, Mack SR, Anderson RA, Waller DP, Zaneveld
LJD, 1986. Synthesis and inhibition of human acrosin and trypsin
and acute toxicity of aryl 4guanidinobenzoates. J Med Chem
2 9 : s 14-19
Kaminski JM, Diao X-H, 1984. Kinetics of interaction of
aryl 4-quanidinobenzoates with human acrosin. Biol Reprod
3O(Suppl. 1):137
Kaminski JM, Nuzzo NA, Bauer L, Waller DP, Zaneveld LID, 1985a.
Vaginal contraceptive activity of aryl 4-guanidinobenzoates
(acrosin inhibitors) in rabbits. Contraception 32: 183-88
Kaminski JM, Reid DS, Kennedy W, Jeyendran RS, Zaneveld LJD,
1985b. Effect of aryl 4-guanidinobenzoates on t h e acrosin activity
of whole human spermatozoa. Biol Reprod 32(Suppl. 1):102
Kennedy WP, Reid DS, Kaminski JM, Jeyendran RS, Zaneveld LJD,
1985. Development and evaluation of a clinical assay for t he
determination of the acrosin activity of human spermatozoa. 41st
Meeting of the American Feriility Society, Chicago, IL
Litchfield FT, Wilcoxon FJ, 1949. A simplified method of evaluating
dose-effect experiments. J Pharmacol Exp Ther 96:99-113
Parrish RF, Polakoski KL, 1979. Minireview. Mammalian sperm proacrosin-acrosin system. Int J Biochem 10:391-95
Rogers BJ, Bentwood B, 1982. Capacitation, acrosome reaction and
fertilization. In: Zaneveld LJD, Chatterton RT (eds.), Biochemistry
of Mammalian Reproduction. New York: J. Wiley, pp. 203-30
Schill WB, Feifel M, Fritz H , Hammerstein J, 1981. inhibitors of
acrosomal proteinase as antifertility agents. A problem of
acrosomal membrane permeability. Int J Androl4:25-38
Van der Ven HH, Kaminski JM, Bauer L, Zaneveld LJD, 1985. Inhibition of human sperm penetration into zona-free hamster oocytes
by proteinase inhibitors. Fertil Steril 4 3 :609-16
Woolf CM, 1968. Introduction to analysis of variance: completely
randomized design-model I. In: Woolf CM (ed.), Principle of
Biometry. New York: D. Van Nostrand Co. Inc., pp. 83-115
Zaneveld LJD, 1976. Sperm enzyme inhibitors as antifertility agents.
In: Hafez ESE (ed.), Human Semen and Fertility Regulation in
Men. St. Louis: C. V. Mosby, pp. 570-82
Zaneveld LJD, 1982. Sperm enzyme inhibitors for vaginal and other
contraception. In: Zatuchni GI (ed.), Research Frontiers in
Fertility Regulation, Chicago: PARFR, pp. 1-14
Zaneveld LJD, Beyler SA, Kim DS, Bhattacharyya AK, 1979. Acrosin
inhibitors as vaginal contraceptives in t h e primate and their acute
toxicity. Biol Reprod 20: 1045-54
Zimmerman RE, Nevin RS, Allen DJ, Jones CD, Goettel ME, Burck PJ,
1983. Antifertility effects of tetradecyl sodium sulfate in rabbits.
J Reprod Fertil68: 257-61