Journal of Andrology, Copyright

Vol.

@ Amencan

16, No.

of

Society

Expression Spermatozoa JITH-SHYAN

1995

of Mannose-Binding and Their Role CHEN,

ANIBAL

AND

1, January/February Andrology

GUSTAVO

Sites on Human in Sperm-Zona Pellucida

F. DONCEL,t

CRISTINA

Binding

ALVAREZ,t

A. ACOSTAt

From the *Depa?fnt of Obstetrics and Gynecology, Taichung Veterans General Hospital, Taichung, Taiwan, R.O.C.; tThe Jones Institute for Reproductive Medicine, Department of Obstetrics and Gynecology, Eastern Virginia Medical School, Norfolk, Virginia; and the Hospital Universitari “La Fe” H. Maternal, Servicio de Reproduccion, Avd. Campanar 21, 46009 Valencia, Spain.

albumin (DMA) neoglycoprotein was a probe capable of detecting mannose-binding sperm receptors involved in human sperm-egg interaction. DMA specifically blocked zona binding of swim-up human spermatozoa in a concentration-dependent manner. While no considerable effect was observed on sperm-zona initial contact, almost 50% of spermatozoa bound to the zona during a 2-hour period detached from it when DMA was introduced in the incubation medium. DMA inhibition was evident when 10% fetal bovine serum, but not 3.5% human serum albumin (HSA), was used as Ham’s FlO medium supplementation. This may be due to the amount of free

ing capacitation, appearing on an average of 20% of the sperm after overnight incubation. They also increased, especially the bar pattern, following calcium ionophore treatment. Nearly all of methanol-fixed spermatozoa displayed the fluorescent label at the head level. Concomitant assessment of sperm membrane integrity and DMA fluo-

calcium in the medium since addition of 40 mM CaCI2 to F10-HSA restored DMA inhibition. Furthermore, the higher the calcium concentration in the incubation buffer, the greater the DMA blockage of sperm-zona binding. Unfixed sperm presented fluorescent DMA label over the entire acrosomal area (cap pattern), or concentrated at the equatorial segment (bar pattern). These patterns increased dur-

cise location on human spermatozoa, however, warrants further investigation. Key words: Sperm receptors, sperm-egg interaction, carbohy-

ABSTRACT:

assessed

A D-mannosylated

to validate

rtilization

experimentally

is a complex

spermatozoon its

eventual

fusion

One crucial interaction

face

in this

step

process complementary

ally,

by binding and Shur, numerous

hydrates

which

proteins

have

mediate

reported

inhibit

drate-binding J Androl

added

to

that

sperm-zona

ase (Dun

carbowhen

coincubation

et al,

have

been

Reproductive

Eastern

Virginia

Medicine, Medical

23507. Received for publication August 26, 1994.

F. Doncel,

Department School,

601

December

of Colley

The Jones

Obstetrics Avenue,

20, 1994; accepted

and

Institute

for

plasma

Gynecology,

Norfolk,

1977),

and

suggested

Mannosidase to: Dr. Gustavo

medium

reaction.

(Ahuja,

1982;

(Macek 1981),

Shalgi sperm macro-

and Shur,

sialytransfer-

fucosyltransferase

(Ram

et al,

it was

Virginia

to

be

involved

in some

aspect

recognition.

sperm-egg Correspondence

acrosome

1989) were identified in mouse spermatozoa. A novel a-Dmannosidase, with different characteristics from other mannosidases previously described, e.g., acrosomal and hepatic, was isolated from rat sperm plasma membranes and characterized (Tulsiani et al, 1989). All these proteins

of the ZP Additionsome

capacitation,

such as galactosyltransferase trypsin-like protease (Saling,

1988),

rec-

binding

to the

molecules

and on

gamete

sites, sperm I 995;16:55-63

et al, 1986; Oehninger et al, 1991). Zona-binding proteins may act as lectins or enzymes. Several

In different mamthat sperm sur-

to complex glycoconjugates 1988; O’Rand et al, 1988).

studies

competitively

appear to require free calcium concentrations to operate, and their expression changes with capacitation and acrosome reaction. Pre-

the

is the recognition molecules present

the zona pellucida (ZP). it has been demonstrated

carbohydrate-binding

ognition (Macek

during

undergoes a cascade of events prior with the oocyte plasma membrane.

between

the sperm and malian species,

process

rescent patterns revealed that DMA fluorescence coincided mostly with permeabilized or altered sperm plasma membrane. In condusion, DMA is a suitable probe to identify human sperm mannosebinding sites crucially involved in sperm-zona interaction. These sites

activity

membranes reported

was also

(Tulsiani that

pretreatment

found

et al,

in human

sperm

1990).

Furthermore,

of human

spermatozoa

the monosaccharide r-mannose inhibited sperm penetration through the zona (Mon et al, 1989). From a clinical standpoint, expression of mannosewith

for publication

55

of

Journal of Andrology

56 binding sites on human spermatozoa has been recently correlated with sperm fertilizing potential (Tesarik et al, 1991; Benoff et al, 1 993a). Human spermatozoa that fail

to fertilize conditions

mature oocytes may lack zona

Aiming

to detect

have

studied

these the

in vitro fertilization

under

receptors

alterations

involvement

on subfertile of

ones.

sperm,

several

we

molecules

control.

effect

in

sperm-zona interactions (Doncel et al, 1993). The main goal of the present study was to validate experimentally a probe capable of detecting mannose-binding sperm receptors involved in human sperm-egg interaction.

In some

of DMA

MgCI2) + 2 or 20 mM Cad2. Short time (15-minute) incubations hemizonae were employed to assess DMA To

study

and zonae

DMA

Source of Spermatozoa

(Ca2

Semen

at 200 g/ml nonspecific

After

liquefaction,

each

semen

proven-fertile

specimen

was

donors.

diluted

with

buffer)

action

on

were allowed

Fl 0+ 3.5% HSA. Then, to droplets of sperm-free

by healthy

(control)

Ca2

samples

used

in any given experiment, e.g., zona binding or fluorescent localization, belong to different donors. The number of donors/ samples (n) employed per experiment is indicated in the corresponding figures or tables.

Binding Assay

aliquots Ham’s

or medium 300 ag/mi)

containing different concentrations of n-mannosylated albumin (DMA)

Co.)

Subsequently,

(test).

the hemizona

termine potential inhibition et al, 1988). Briefly, salt-stored an IVF

program

incubated l0 motile one period, bound index

zona

were

microbisected in control

sperm/mi.

One zona

for each hemizonae

experimental were

to the outer surface (HZ!) was obtained

=

no.

of sperm

no. of sperm

assay

of sperm-zona prophase

for 4 hours

washed

(100, (Sigma

and

200, and Chemical

was used

to de-

binding (Burkman I human oocytes from matching

hemizonae

or test droplets containing was used per sperm sample, replicate. extensively,

After and

of the ZP were counted. as follows: bound in test sample bound in control sample

the

5 x

i.e.,

incubation

sperm

tightly

The hemizona

ml, and sperm motion of a computer-assisted

DMA

also

used

as a

binding

100

given in percentages are w/v unless otherwise indinot apply for gases, which are all expressed in % v/v.

4 hours

in supple-

DMA

at 300

parameters were assessed with the semen analyzer (Cell Soft#{174},Labsoft York, New York). and test droplets

g/ help Di-

Motility from the

experiments.

Acrosomal Status Spermatozoa 3.5% HSA

were incubated for 5 or 18 hours in Ham’s Fl 0 + at 37#{176}C under 5% CO2 in water-saturated air, cen-

out of the medium, in Ca2

400 Mg/mI DMA ionophore

diluted

A23

with in Ca2

100 mM

the incubations,

buffer; sperm

and fixed with methanol. using fluoresceinated Pisum ma

Chemical

Co.)

of acrosomal

status

experiments

were

donors

A23

and then

incubated

terns

or percentages

and test samples

in dimethyl sulfoxide La Jolla, California). spotted on glass slides,

interfere

±

described

containing

out

this

were with

Ca2

1.6 and 42.8

by Cross

for a-o-mannosyl with PSA determination DMA,

possibility. stimulated

buffer

res-

additional

Five with

(control)

sperm 10

iM

or Ca2

Acrosomal status was could be found in pat-

of acrosome-reacted

(43.2

DMA,

or 10 MM calcium

method

PSA affinity

in experiments

different

incubated

acrosomal status was assessed agglutinin (FITC-PSA; Sig-

to the

to rule

from

187

sativum

run

samples

D-mannose,

Their

Since could

and

200 or 400 ag/mI

stock solution Calbiochem, were washed,

according

and co-workers (1986). idues of glycoproteins

resuspended

containing

187 (10 mM

1:1,000

After

and then

buffer

buffer containing 200 Mg/mI DMA (test). assessed using FITC-PSA. No differences

Experiments were also carried out in which DMA was washed off after the 30-minute preincubation. D-Glucosamide-albumrn (Sigma Chemical Co.) at 300 g/ml was used as nonspecific

§ Concentrations cated. This does

containing was

were incubated for 2 and FlO or Ca2 buffer containing

for 90 minutes

fertile donors were preincubated for 30 of Ham’s Fl 0 + 10% v/v FBS (control)

Swim-up sperm from mm in 1 00-id droplets

buffer

Sperm Motion Parameters

trifuged

Sperm-Zona

binding,

in Ham’s

an

vision of Cryo Resources, Ltd., New was visually checked on all control

The

secondary

for 2 hours

ci-Glucosamide-albumin in the latter experiment.

spectively to remove seminal plasma. The resulting pellet was overlaid with 0.3 ml Ham’s FlO containing 3.5% HSA or 10% v/v FBS for 1 hour at 37#{176}C under 5% C02, and the motile fraction was retrieved.

with initial

were washed and transferred buffer with 20 mM CaC12

Sperm mented

up in the supernatant

gamete

hemizonae MgCl/Hepes or the same

(test). control

of spermatozoa effect on gamete

to interact

equal volume of Ham’s FlO medium (GIBCO, Grand Island, New York) supplemented with 3.5% human serum albumin (HSA) (Sigma Chemical Co., St. Louis, Missouri) or 10% v/v fetal bovine serum (FBS) (GIBCO). Sperm were washed twice by centrifugation at 400 x g for 7 minutes and 5 minutes, re-

swimming

the with sper-

for 4 hours.

Materials and Methods provided

to determine

were preincubated coincubated with

establish

sperm

were

in order

hemizonae and finally

the role of Ca2 in DMA inhibition of spermzona binding, various media were used including Ham’s FlO + 3.5% HSA, Ham’s FlU + 3.5% HSA + 40 mM CaC12, and MgCl/Hepes buffer (containing 30 mM Hepes, pH 7.0, 150 mM NaC1, 10 mg/mi bovine serum albumin [BSA], and 0.5 mM To

contact.

samples

experiments,

on the zona, ig/ml, washed,

DMA at 300 matozoa

(IVF)

or bear defective

related

1995

Januaiy/Februaiy

.

sperm ±

between

control

1.1, respectively).

Ruorescent Localization of Sperm DMA-Binding Sites sites, a fluoresceinated DMA no. A7790) was used. Sperm were studied after swim up (Ti) or incubated in supplemented at 37#{176}C, 5% CO2 for 5 (T6) or 17 (T18) hours. At

To localize

the sperm

(FITC-DMA)

(Sigma

immediately Ham’s FlO

DMA-binding

chen et al

Mannose-Binding

.

Sites on Human Sperm

57 in a drop of regular buffer (not antiquenching) and observed immediately. DMA fluorescent patterns were differentially identified on motile or immotile spermatozoa. To confirm these results, sperm were simultaneously incubated with FITC-DMA as described above, and a supravital stain, Hoechst 33258 (Sigma Chemical Co.), at I Mg/mi (Cross et al, 1986). On every spermatozoon analyzed, the uptake of each marker was fresh-mounted

150

evaluated

‘-4

using

too

200

0-mannosyl

300

3000

300G

at ed

albumin

the appropriate

fluorescence

filters.

A sequential

ofFITC-DMA labeling and a hypoosmotic was also performed. Briefly,afterstaining

sweffing the sperm with FITC-DMA, they were washed and resuspended. A 100-pl aliquot was then transferred to 1 ml of hypoosmotic medium, containing 7.35 g sodium citrate and 13.51 g fructose in 1 L of distilled water (Jeyendran et al, 1984), and incubated at 37#{176}C

combination test (HOST)

300HZ

( uq/mL)

FIG. I. Inhibition of human sperm-zone binding by DMA. Ham’s FlO + 10% v/v FBS was used. Test sperm (n = 4/mean) were preancubated with DMA for 30 minutes and then further coincubated with the hemizona for 4 hours. 300 W: DMA (300 pg/mI) was washed off prior to spermzona coincubation. 300 G: 300 pg/mI of D-glycoslated albumin was used Instead of DMA. 300 HZ: Hem,zonae were preincubated with DMA (300 pg/mi), then washed, and finally coincubated with sperm. HZls for 100, 200, and 300 pg/mI are statistically different (P = 0.003). HZIs of 300W, 300 G, and 300 HZ are statistically different from 300 HZI (all P = 0.02).

and 5% CO2 for 1 hour. Finally, ferentially recorded on spermatozoa As described, “coiled” tails would

fluorescent patterns were difwith or without curled tail. be indicative of intact sperm

membranes.

Statistical Analysis In Figure 1, comparison among the inhibitory effect (HZIs) of various DMA concentrations (100,200, and 300) was done using the ANOVA test. For comparisons between 300 W or 300 G, or 300

HZ

and

300,

a Mann-Whitney

test

was

used.

This

test

was also used

to compare appropriate pairs in Figures 2-4 and Table 1. In Table 2, changes in percentages of fluorescent patterns were analyzed using the ANOVA test. Selected pairs were then compared using post-tests with the Bonferroni’s correction. The level of significance chosen for all tests was P 0.05.

the end of incubation, sperm were washed twice in MgCI/Hepes buffer + 20 mM CaCl2 (Ca2 buffer) and labeled with FITCDMA 200 Mg/mi for 15 minutes at 37#{176}C under 5% CO2. Subsequently, sperm were washed twice with MgCI/Hepes buffer without

calcium,

spotted

onto

glass

slides,

briefly

air-dried

and

with antiquenching buffer. Some experiments were done on methanol-fixed spermatozoa to determine probe recognition of internal binding sites. Determination of fluorescent patterns

Results

was performed

Inhibition

mounted

microscope

at 1,000 (Nikon).

uated. In competition

x magnification

At least

200

an epifluorescence

per specimen

with

different

spermatozoa

were eval-

and mM, 100 mM, 200 mM, or 400 mM) and FITC-DMA (200 pg/mI). To observe potential changes in the DMA-binding patterns after the acrosome reaction, spermatozoa were treated with 10 incubated

experiments,

using

sperm

concentrations

pM Ca2 ionophore A23 187 then labeled with FITC-DMA. uated

either

on a different

were pretreated

of r-mannose

(50

for 1 hour at 37#{176}C, washed, and Sperm acrosomal status was eval-

aliquot

or on the

same

sample

using

a double-labeling technique with rhodamine-conjugated PSA (TRITC-PSA, Sigma Chemical Co.). In this latter case, after the

FITC-DMA

step, sperm were washed,

fixed with methanol,

cubated with TRITC-PSA, again washed, and finally with an antiquenching buffer (100 mg O-phenylenediamine 10 ml phosphate-buffered carbonate/bicarbonate the

buffer).

appropriate

+ 90 ml glycerol,

saline

filters

Observation

for each

in-

mounted in

pH 8.6 with

was performed

using

fluorochrome.

sperm

membrane

concomitantly,

motility

served.

were

lower

Sperm speed

(290

integrity

and

of the sperm centrifuged

x g) for

A D-mannosylated blocked the tight

shorter

mannose-binding

samples

only time

once

was per

carefully

washing

(5 minutes).

step,

medium

They

entire

When

interaction.

sperm but its inhibitory

manner test sperm for coincubation

washed effect

DMA

the

sperm,

the hemizonae

were

was

to spermeven when

off prior was lost,

the maximum assayed concentration (300 pg/ml) The same lack of activity was observed when, preincubated

was tried. instead of with

DMA.

In regard assayed

imally

to specificity, a D-glycosylated albumin was under the same conditions that proved to be maxinhibitory for DMA. No blocking activity could

be detected in this case. The above experiments

inhibitory

if, as demonstrated

effect

er cell systems to facilitate

were

run

employing

Ham’s

FlO

supplemented with 10% v/v FBS. InterestFBS was replaced by 3.5% HSA, the DMA

prewere

the

during

preincubated with zona coincubation,

sites at

albumin (DMA) neoglycoprotein binding of swim-up human spermatozoa

to human zonae in a concentration-dependent (Fig. 1). The DMA was preincubated with 30 minutes and then kept in the sperm-zona

as medium ingly, when

Sperm Membrane Integrity To assess

of Human Sperm-Zona Pellucida Binding

disappeared

(Fig.

2). In order

for mannose-binding (Pontow

ligand-receptor

et ai, 1992),

receptors free Ca2

interaction,

to verify in oth-

was required

Ham’s

FlO

sup-

58

Journal of AndroJogy

Janua,y/Februarj

1995

=

rio

rio NSA

3.5%

3.5%

Duller NSA

1.0%

40mMCo

Butter BSA

2Co

1.0%

initial

2OmMCa

FIG. 2. Influence of Ca#{176}concentration on DMA-mediated spermzona binding Inhibition. Sperm were preincubated with DMA (200 pg/mI) for 30 minutes and further coincubated with hemizonae for 4 hours. Different medium supplements and Ca2* concentrations were used. HZI of FlO + 3.5% HSA (n = 10) Is statistically different (P = 0.007) from HZI of 200 paJmI in FlO + 10% FBS (Fig. 1); the other three variants, n 4; F10/HSA #{247} 40mM Ca#{176} versus F1O/HSA (P= 0.008); buffer/BSA, 2mM Ca#{176} versus 20mM Ca#{176} (P = 0.02). All buffer/medium containing extra Ca2’ were significantly different from Fl 0/HSA without extra Ca#{176}.

HSA and 40 mM CaC12 was used. sperm-zona binding inhibition was thus

with

3.5%

DMA-mediated

recovered. The influence of Ca2 concentration on such activity was confirmed by the utilization of a MgClIHepes buffer Ca2

with 2 or 20 mM CaCl2. As predicted, the higher concentration permitted a stronger DMA-induced

inhibition. As for the aforementioned DMA was able to impede

6 versus

secondary

siderable interaction

sperm-zona

we knew

interaction

ferred

to a sperm-free

Table

1. E ifect

medium

of n-mannos

containing

ylated albumin on

DMA.

50% of spermatozoa

almost

bound

to the

ployed.

Sperm Motion Parameters of sperm

in a

motility was visually lets from the binding

No con-

0.05).

=

zona during the 2-hour incubation detached from itwhen DMA was introduced in the incubation medium. This effectwas not seen when n-glucosamide-albumin was em-

zona binding

time was modified. In an attempt to assess only a 15-minute sperm-zona interaction whereas to study secondary binding, zonae with sperm for 2 hours and then trans-

(P

secondary

that

4-hour incubation. However, in order to ascertain whether this inhibition was primarily effected through the initial sperm-zona binding or through secondary interactions, the incubation initial binding, was allowed, were incubated

C

inhibition was observed when the sperm-zona was mostly limited to the initial contact (Fig.

Alterations experiments,

Secondary

FIG. 3. DMA-mediated inhibition of inItial (30-minute) and secondary (120-minute) sperm-zona Interactions. For testing initial binding, sperm were preincubated with DMA (200 pg/mI) and then further coincubated with hemizonae for 30 mInutes. For secondary binding, sperm were coincubated with hemizonae for 2 hours, then hemlzonae were transferred to a sperm-free medium containing DMA at 200 pg/mI. Secondary 6: D-glycosylated albumIn was used Instead of DMA. All n =4; inItial or

3). Conversely, plemented

Secondary

BSA

ences were detected. objective and detailed incubated

with

ious motion assisted ences out

DMA

can be the cause of spermrule out this possibility, checked on all control and test dropexperiments. No noticeable differ-

DMA

To

To support this manner, swim-up for 2 and

analyzer

found

comparing

at any

time

sperm

(Table

finding in a more spermatozoa were

4 hours

were assessed (CASA). No

parameters

semen were

motility

impairment.

at 37#{176}C and

var-

with a computersignificant

samples

differ-

with

or with-

1).

motion parameters*

Time after swim-up

(hours)

DMA

medium

0

-

2

-

+

4

-

+

in

Velocity (pm/second) 104.6 119.9

120.5 107.3 98.9

± ± ± ± ±

6.8

Uneanty

ALH mean (pm)

Motility

5.39 ± 0.68 4.67 ± 0.54

4.47 ± 0.24 5.20 ± 1.14

17.60

17.1 17.2

4.80

5.05

± 0.76

17.21

14.2 13.1

4.54 ± 0.68 4.35 ± 0.68

4.80 ± 0.70 4.81 ± 0.25

19.48 16.56

* Data represent mean ± SE, n = 3. Sperm were Incubated BICF: flagellar beatJcross frequency. No statistical differences

± 0.81

(%)

B/CF (Hz)

18.01

± ± ± ± ±

0.31 1.02 0.92

94.6 85.5 87.6

0.82

81.5

0.13

82.6

± ± ± ± ±

4.4 8.7 7.5

4.7 5.9

with DMA (200 pg/mI) for 2 and 4 hours. ALK: amplitude of lateral head displacement; between medium with or without DMA for any of the motion parameters.

Chen et al

Mannose-Binding Sites on Human Sperm

‘

59

a V

DMA

did

not

modify

ment

(10

iM

in

substantially ed sperm,

a V.,

both

#{149}0

this result.

Ca2

buffer)

Ca2 used

ionophore as

treat-

positive

control

increased the percentage of acrosome-reactat 5 and 18 hours of incubation.

a, V C

Ruorescent Localization Sites on Spermatozoa

E 0

To localize

of DMA-Binding

DMA-binding

sites

in unfixed

swim-up

sperm,

0

FITC-DMA seen at the acrosomal concentrated Post

-Swi

m

Up

FIG. 4. Acrosomal status of sperm coincubated with DMA. Sperm (n =8 except for 400 with or without o-mannose, n =4) were preancubated for 5 and 18 hours in Ham’s FlO + 3.5% HSA and further challenged with Ca2 buffer 4, DMA 200 pg/mI 4, DMA 400 pg/mI 4, DMA 400 pg/mI plus 100 mM mannose 4, or A23187 10pM (tl for 90 minutes. A231 87 is the only treatment statistically different from Ca2 buffer (P =

0.02).

Incubated with DMA The sperm

acrosome reaction event whose main

mediated teins,

especially

ZP3

To determine rosome or

(Wassarman,

1988;

the potential of DMA

reaction,

18 hours

is thought to be a receptornatural triggers are ZP pro-

swim-up

in Ham’s

sperm FlO

+

Saling,

3.5%

HSA

at

further challenged with Ca2 buffer containing 400 cg/ml of DMA. The neoglycoprotein did a statistically rosome-reacted

significant sperm

controls

4). Addition

(Fig.

1991).

for inducing an acwere capacitated for 5

increase in the when compared of 100 mM

37#{176}C, and

200 and not

induce

percentage of acto corresponding D-mannose

preincubation and incubation of sperm

during

with 400 ig/ml

area (cap pattern) at the equatorial

and the segment

orescent labeling and in combination

of neck with

some spermatozoa. or neck and tail

Staining only was

grouped The bar,

increased

with

1.5%

of the

icant

increase

sperm

were

label Flu-

was present alone head patterns in

of the entire spermatozoon considered nonspecific

and

2.

the incubation

of fertile donors in Ham’s FlO

were entire

other with the (bar pattern).

andlor tail the described

as “others” in Table incidence of FITC-DMA

samples cubation

Acrosomal Status of Spermatozoa

was used. Two main fluorescent patterns head level (Fig. 5). One covering the

head

patterns,

time

of swim-up

cap

and

sperm

(Table 2). After 5 hours of in3.5% HSA at 37#{176}C,13.1 ±

+

head-labeled,

compared

to

indicating

aliquots

a signif-

studied

right

after

swim-up (6.4 ± 0.81%). Spermatozoa bearing these patterns represented around 20% of the sperm incubated for 18 hours under capacitating conditions. Negative sperm decreased

while

nificant changes. In an attempt

pattern to verify

increasing concentrations were preincubated with incubation with FITC-DMA labeled sperm diminished, be obtained. sperm were To

study

At head the

400 mM positive. location

“others”

did

specificity of the

show

of the

D-mannose sperm and

no

not

head

any patterns,

(50-400 mM) kept during the

(Table 2). Although complete blockage

D-mannose, of DMA-binding

sig-

still

10% sites

head-

could of the after

the

FIG. 5. Ruorescent localization of DMA-binding sites on spermatozoa. Unfixed sperm with the label (HTC-DMA) covering the entire acrosomal area (cap pattern; Fig. 5b) or concentrated at the equatorial segment (bar pattern; Fig. 5d). Methanol-fixed membrane-permeabilized sperm show the label in a disperse pattern all over the head with a brighter band at the equatorial segment (Fig. 5f). Neck and tail were also fluorescent. Corresponding phase-contrast pictures (Fig. 5a,c,e) are included (sperm magnification: 600 x).

Journal of Andrology

60 Table

2. Fluorescent

localization

of DMA-binding

sites

Fluorescent

Mannose

1 6

18

patterns

(%)

(mM)

n

Bar

None

14 14

89.0 ± 1.6

4.2 ± 0.8

None

4.8 ± 0.8

8.3 ± 0.9

82.2

± 1.9

50 100 200 400

2 8 10 6

8.0

±

14

2.3 ± 3.0 ± 3.2 ± 5.5 ±

50 100 200 400

2 8 10 6

81.5 89.2 88.7 90.3 75.2 75.9 84.0 84.6 87.4

13.1 13.6

None

5.6 3.9 3.2 3.5 7.8 5.7 4.7 4.9 5.7

4.7 ± 0.6 4.9 ± 0.7

8.4 ± 1.5 8.3 ± 0.8 7.3 ± 1.5 19.3 ± 3.1 17.3 ± 4.7 12.8 ± 2.1 11.5 ± 1.7 10.5 ± 1.4

1.8

± ±

Cap

0.4

4.7

0.9 0.6

± 1.2 ± 1.1

11.5 11.6

0.7 ± 0.8

0.5

±

±

2.0

4.1 8.0 ± 1.7

± ±

±

0.9

±

6.6

± 1.1

4.8 ± 0.9

± 1.3

Bar + Cap

Otherst

Negative

1.9 4.5 ± 0.8 5.1 ± 0.5 3.8 ± 0.9

± 1.1 ±

1995

on spermatozoa’

Incubation

time (hours)

January/Februaty

2.7 ± 1.8 ± 0.9 ± 1.8 ± 4.1

0.4 0.3 0.8 1.2

3.3

6.9 ± 1.5

± 2.5 ± 2.2 ± 1.6

3.2 ± 0.6 3.8 ± 0.7 2.9 ± 0.6

±

6.4 ± 0.8 ± 1.5 ± 2.8

mean ± SE. Sperm were labeled with FITC-DMA (200 pg/mI) in Ca2 buffer (20 mM Cad2) with or without D-mannose for 15 5 hours or 17 hours after swim-up (total incubation time: 1, 6, and 18 hours, respectively). Spermatozoa were incubated in Ham’s FlO + 3.5% HSA at 37’C, 5% Co2. t Staining of neck and tall or the entire spermatozcon. Except for Others, the bar and the cap patterns significantly Increased their incidence along the incubation time (Bar: P = 0.0001; Cap: P = 0.003; Negative: P = 0.004; Bar + Cap: P = 0.0003). The most significant changes occurred between 1 and 18 hours (P < 0.01). The addition of mannose produced some inhibition of FITC-DMA binding; however, it did not significantly change the Incidence of the fluorescent patterns. *

Data represent

minutes

immediately,

acrosome treated

reaction

has

taken

place,

swim-up

sperm

were

with Ca2 ionophore (10 NM), incubated with FITC-DMA, fixed, and finally stained with TRITC-PSA. This double labeling allowed for a sequential determination of DMA patterns and acrosomal status on the same spermatozoon. As shown in Table 3, the bar pattern strongly

associated

good

with

correlation,

cap pattern and acrosome-intact spermatozoa pattern

acrosome-reacted

however,

could

not

sperm. be found

acrosome-intact sperm. Five and 10% of acrosome-reacted

treated

with

Ca2

in double-staining

ionophore

experiments.

bore In sperm

Such between

percent of swim-up the

cap

samples

labeled only with FITC-DMA, there was always an increase in both bar and cap patterns after the induction of acrosome reaction with Ca2 ionophore. To find out if this observation could be due to the recognition of an internal molecule by FITC-DMA, methanol-fixed (membrane-permeabilized) sperm were studied. Almost all of them were fluorescent at the head level displaying a cap pattern with a brighter band at the equatorial segment (Fig. 5). Neither this pattern nor its mci-

Table

3. Influence

of acrosorne

reaction

Ca2

DMA fluorescent pattern

Negative

Bar

0.37

±

Cap Negative Others

3.20

± 0.37 ± 0.41 ± 0.18

*

90.80 0.32

Data represent

incubated

(AR) on the fluorescent

mean percentages

with FITC-DMA

0.22

patterns

dence were changed when spermatozoa treated with Ca2 ionophore and then fixed were studied. Aware that plasma membrane integrity and, therefore, sperm viability could be confounding factors for DMAbinding site location and incidence, experiments using FITC-DMA and a positive identification of viable sperm were run. Capacitated sperm after 18 hours of incubation were labeled with FITC-DMA while motility was carefully preserved.

A fresh

drop

of the

suspension

was

placed

on a

prewarmed slide, cover-slipped, and observed under an epifluorescence microscope. Sperm head DMA-fluorescent patterns could not be detected on any motile sperm (Table 4). Conversely, 14% of the immotile population displayed

these

patterns.

In another set of experiments, a HOST, which assesses plasma membrane integrity through the coiling of the tail of sperm subjected to hypoosmotic medium, was run together with the DMA fluorescence. Again, no head patterns were observed on sperm bearing intact plasma membranes (coiled tails) (Table 4).

of DMA-binding

sites’

buffer AR

A 23187 AR Positive

2.61 ± 0.43 1.38 ± 0.20 1.27 ± 0.53 1.01

± 0.35

± SE, n = 6. Swim-up sperm were treated with Ca2 (200 pg/mI) for 15 minutes, fixed, and stained with TRITC-PSA.

Negative

1.47 4.95

±

Positive

0.35

36.60

39.60 ± 3.31 1.47 ± 0.34 ionophore

(10 pM) in Ca2

± 3.41

9.87 ± 1.80

± 0.73

3.43 3.41 buffer

± 1.50 ± 0.53

or with

buffer

alone,

Chen et al

Mannose-Binding Sites on Human Sperm

.

Table 4. Influence of sperm fluorescent patterns’ Sperm membrane

membrane

integrity

DMA-fluoresce

on DMA-

(%)

nt patterns

Others

Bar

Cap

Negative

0.0 ± 0.0 8.1 ± 0.5

0.0 ± 0.0 6.1 ± 0.5

52.6 ± 5.7 30.5 ± 5.7

0.0

0.6 ± 0.3 7.4 ± 1.1

0.6 ± 0.2 5.8 ± 0.6

57.1

0.4 ± 0.2 2.5 ± 0.3

assessment Motility Motile Immotile

± 0.0 2.7 ± 0.4

HOS test Positive

Negative

± 4.1

25.6 ± 3.7

33258

Hoechst

Negative

0.0 ± 0.0 6.1 ± 0.6

Positive

83.9 ± 1.4

0.0 ± 0.0 4.3 ± 0.4

0.0 ± 0.0 5.3 ± 0.7

0.4 ± 0.1

* Data represent mean ± SE; motility, n = 8; HOS test, n = 6; Hoechst 33258, n = 12. Sperm incubated under capacitating conditions for 18 hours were labeled with FITC-DMA (200 pg/mI) for 15 minutes and fluorescent patterns were assessed. Concomitantly, sperm membrane permeability was evaluated through motility, hypoosmotic swelling test (HOST), or staining with Hoechst 33258.

Finally, marker

the that

stain

supravital

cell membranes

concomitantly could be seen 4).

Hoechst

33258,

are impermeable

with FITC-DMA. No DMA in Hoechst 33258-negative

a DNA

to, was used

head patterns sperm (Table

authors

pation of ZP the mammalian Swenson

1989).

have demonstrated glycoproteins and zona

and

Yet,

fertilization Dunbar,

remains

the crucial particisperm receptors in

process 1982;

the nature

component

(Yanagimachi,

Wassarman,

of the sperm

1981;

1987;

Saling,

carbohydrate-binding

elusive.

In this work, we have shown evidence that a mannoseenriched neoglycoprotein (DMA) specificallyinhibits, in a concentration-dependent manner, human sperm-zona binding. as well as fucose, galactose, N-acetylglucossialic acid are prominent constituents of studied mammalian ZPs (Shalgi et al, 1986; Wassannan, 1988; Mon et al, 1989; Yurewicz et al, 1991; Tulsiani et al, 1992). D-Mannose proved to be an effective inhibitor of sperm-zona binding in rats (Shalgi et al, 1986). In turn, an a-i>mannosidase activity was demonstrated in rat and Mannose, amine, and

human

though only few oocytes were assessed, each of them with low numbers of bound sperm. Since ZP carbohydrates are presented to the sperm in a defined spatial arrangement, most likely in special clusters, we chose to use the mannose coupled to a backbone of albumin in an attempt to resemble these conditions. In order to be inhibitory, the neoglycoprotein DMA had to be present in the coincubation medium (Ham’s FlO + 10% v/v FBS). No activity was detected when DMA was preincubated with either sperm or zonae and then washed off. Mori and co-workers (1989) also concluded

that

pretreatment

of oocytes

with

D-mannose

did

not block penetration of untreated spermatozoa. Surprisingly, when fetal serum was replaced by HSA as medium supplement, the DMA inhibitory effect was lost. The

possibility

existed

that

the

sperm

sites would be similar to the mannose rophages and hepatic endothelial cells, osaccharides

but

displays

much

mannose-binding

receptor of macwhich binds mon-

higher

affinity

for

mul-

tivalent oligosaccharides in the presence of free calcium (Pontow et al, 1992; Taylor and Drickamer, 1993). Therefore, to overcome the chelating effect of 3.5% HSA, 40

mM CaCl2 was added to the medium. The percent binding was significantly reduced revealing a recovery of the DMA inhibitory

activity.

To confirm

buffer with fully utilized

lower

protein

impairment mind, we

in sperm-zona checked motility

this

content

finding, that

had

a MgClJHepes been

success-

with the same probe was used (Benoff et al, 1 993a). Again, the higher the calcium concentration, the greater the sperm-zona binding inhibition. Alteration of sperm motility can be the cause of binding

Discussion Several

61

sperm

plasma

membranes

(Tulsiani

et al,

1989,

1990). Using various and co-workers nose in human monosaccharide

lectins (1989)

with different reported the

specificities, Mori presence of D-man-

ZP. They also demonstrated a role for the in the sperm-zona binding process al-

sperm in all experiments tion parameters were tem

after

various

times

binding assays. With of coincubation-remaining

herein analyzed

reported. Moreover, with a computerized

of sperm-DMA

incubation.

this

in mosysNo

significant changes were observed with respect to controls. To localize the mannose-binding sites on sperm, a fluoresceinated DMA was used. The same probe had been previously employed on human sperm (Tesarik et al, 1991; Benoffet al, 1993a), but results concerning incidence and type of fluorescent patterns were not in agreement. In our experiments and in accordance with the binding assay findings, when Ham’s Fl 0 + 3.5% HSA was used as medium for FITC-DMAJsperm incubation, the resulting fluorescent patterns were expressed in a low percentage of the spermatozoa (data not shown). In addition, fluorescence was faint and inconsistent. This could have been the case in Tesarik’s experiments. Conversely, when a buffer containing 20 mM calcium was used, a defined bar and cap patterns, similar to those observed by Benoff and co-workers, were seen at the sperm head level. Consistently with above-mentioned results, 100 mM D-mannose completely blocked sperm fluorescence when incorporated to FITC-DMA-containing Ham’s FlO.

Journal

62 be achieved when even higher concentrations of the monosaccharide were used in an FITC-DMA calcium-supplemented buffer. Sperm-ZP interaction progresses through a cascade of events involving multiple binding steps. From our data changing the incubation time and conditions of the HZA, it seems that the DMA probe would preferentially block secondary binding, that taking place after 2 hours of spermzona coincubation. A similar probe where mannose is replaced by glucosamine did not interfere with such binding, endowing specificity to the DMA effect. This type of binding would involve sperm receptors relocated or exposed during the acrosome reaction, which are responsible for binding to secondary ligands on the ZP, e.g., ZP2 (Bleil et al, 1988) during sperm penetration. DMA could block sperm penetration as well. In this regard, Mori et al (1993), confirming earlier results, have recently reported that the monosaccharide D-mannose inhibits zona penetration while it does not affect the spontaneous acrosome reaction in human spermatozoa. Using sperm capacitated during 6 hours or overnight incubation in Ham’s Fl 0 + 3.5% HSA and a rhodamineconjugated PSA for acrosomal status assessment, we were unable to detect any acrosome reaction-inducing effect of DMA. Addition of 100 mM D-mannose to DMA-containing calcium-supplemented challenging buffer neither facilitated nor affected this fact. The same sperm population typically responded to the calcium ionophore A23 187. Mannose-binding sites were located at the sperm acrosomal area in a disperse manner (cap pattern) or concentrated at the equatorial segment (bar pattern). As with other carbohydrate-binding sites and other better-characterized sperm antigens (Myles and Primakoff, 1984; Saxena et al, 1986; Lopez and Shur, 1987; Cowan et al, 1987), the bar pattern incidence increased after calcium ionophore treatment. Interestingly, the cap pattern also increased after such treatment. The possibility that acrosome-reacting sperm would allow the probe to enter and recognize cytoplasmic/acroHowever,

somal

only

partial

mannose-binding

inhibition

sites

immediately

after

swim-up

tion. Almost level.

all of them

could

was tested

and

presented

after

using

overnight

fluorescence

fixed

sperm

incubaat the head

Verification of plasma membrane integrity in experiments using unfixed spermatozoa was then warranted even under the assumption that previous FITC-DMA localization experiments had been carried out with “good motility” samples. No fluorescent head patterns could be detected on motile, or HOST-positive, or Hoechst 33258negative sperm when techniques were combined to assess DMA fluorescence and membrane integrity concomitantly in overnight-incubated spermatozoa. The existence of sperm subplasmalemmal storage mod-

of Andrology

January/Febrary

1995

ules for mannose-binding receptors has been recently suggested (Benoff’ et al, l993b). They would externalize during capacitation in association with membrane cholesterol efflux and spontaneous acrosomal loss. These sperm mannose receptors would be similar to those already described in human macrophages, which belong to calcium-dependent animal lectins with C-type carbohydrate-recognition domains (Taylor et al, 1990). Although displaying different requirements and kinetics (Tulsiani et al, 1990) both human sperm acrosomal and surface a-i>mannosidases are inhibited by D-mannose and could well represent binding sites for DMA. The authors have clearly suggested that a-i>.mannosidase activity may have a role in the interactions of mammalian gametes. Another possibility, however, would be the existence of intra-acrosomal mannose-binding proteins with limited surface expression (Tesarik et al, 1990). Proacrosin, the zymogen form of the acrosomal protease acrosin, would fit within this category. It has already been described as a potential zona-ligand molecule on boar spermatozoa (Jones et al, 1988). Proacrosin has a definite requirement for a polysaccharide structure or for “clustering” of saccharides on a protein backbone. It recognizes carbohydrate moieties of ZP glycoproteins as well as neoglycoproteins like BSA-fucose and BSA-mannose. As in our experiments with human sperm, fluoresceinated neoglycoproteins labeled 100% of permeabilized boar spermatozoa. Fluorescence and ultrastructural studies using fucosylated probes revealed that, although some of these

proteins

were

located

on

the

plasma

membrane

overlying the anterior tip of the sperm head, the number of fucose-binding sites increased substantially after induction of the acrosome reaction (Friess et al, 1987). This could be an explanation for our results showing an unexpected increment in the incidence of cap pattern in the sperm treated with Ca2 ionophore. Since the surface density of receptors would be low, it is also understandable that the DMA could not stimulate the acrosome reaction. Using sperm extracts from various mammals including humans, and [‘25I]BSA-fucose and mannose, Jones (1989) demonstrated that proacrosin is the major protein species recognized on western blots. The author proposed a hypothesis for sperm-egg interaction in mammals in which proacrosin, released during early stages of the acrosome reaction, mediates secondary or consolidated binding of spermatozoa to the ZP by virtue of its carbohydrate-binding capacity. Again our observation that DMA preferentially blocks secondary binding would support this mechanism in the human system. From our experiments, it is clear that mannose-binding receptors on sperm are critically involved in human sperm! egg interaction. It would be important, however, to determine their exact location to better understand the physiological role they play.

Chen et al

Mannose-Binding Sites on Human Sperm

Acknowledgment The authors express their appreciation to Ms. Pauline M. Clynes of the Jones Institute for Reproductive Medicine for her excellent editorial contribution and assistance in manuscript preparation.

References Ahuja KK. Fertilization studies in the hamster: the role of cell surface carbohydrates. Exp Ce// Res 1982;140:353-362. Benoff S, Cooper OW, Hurley I, Napolitano B, Rosenfeld DL, Scholl GM, Hershlag A. Human sperm fertilizing potential in vitro is correlated with differential expression of a head-specific mannose-ligand receptor. Fertil Steril 1993a;59:854-862. BenoffS, Hurley I, Cooper GW, Mandel FS, Rosenfeld DL, Hershlag A. Human sperm head-specific mannose-ligand receptor expression is dependent on capacitation-associated membrane cholesterol loss. Hum Reprod 1993b;8:2141-2154. Bleil JD, Greeve JM, Wassarman PM. Identification of a secondary sperm receptor in the mouse egg zona pellucida: role in maintenance of binding of acrosome-reacted sperm to eggs. Dev Biol 1988; 128: 376-385. Burkman Li, Coddington CC, Franken DR, Kruger TF, Rosenwaks Z, Hodgen GD. The hemizona assay (HZA): development of a diagnostic test for the binding of human spermatozoa to the human hemizona pellucida to predict fertilization potential. Fertil Steril 1988;49:688-697. Cowan AE, Myles DO, Koppel DE. Lateral diffusion of the PH-20 protein on guinea pig sperm: evidence that barriers to diffusion maintain plasma membrane domains in mammalian sperm. J Cell Biol 1987; 104:917-923. Cross NL, Morales P, Overstreet JW, Hanson FW. Two simple methods for detecting acrosome-reacted human sperm. Gamete Res 1986;! 5: 213-226. Doncel GF, Valerio C, Alvarez C, Chen iS, Acosta AA. In search of biomarkers for zona receptors in human sperm. JAndrol 1993; 14(1): (Suppl)27. Durr R, Shur B, Roth S. Sperm-associated sialyltransferase activity. Nature (Lond) 197 7;265:547-548. Friess AE, Topfer-Petersen E, Nguyen H, Schill WB. Electron microscopic localization of a fucose-binding protein in acrosome reacted boar spermatozoa by the fucosylperoxidase-gold method. Histochemistry 1987;86:297-303. Jeyendran RS, Van der Ven HH, Perez-Palaez M, Crabo BO, Zaneveld LJD. Development of an assay to assess the functional integrity of the human sperm membrane and its relationship to other semen characteristics. J Reprod Fertil1984;70:219-228. Jones R. Identification of carbohydrate-binding proteins in mammalian spermatozoa (human, bull, boar, ram, stallion and hamster) using [‘251]fucoidin and [1231]neoglycoprotein probes. Hum Reprod l989;4: 550-557. Jones R, Brown CR, Lancaster RT. Carbohydrate-binding properties of boar sperm proacrosin and assessment of its role in sperm-egg recognition and adhesion during fertilization. Development 1988;l02: 781-792. Lopez LC, Shur BD. Redistribution of mouse sperm surface galactosyltransferase after acrosome reaction. J Cell Biol 1987; 105:1663-1670. Macek MB, Shur BD. Protein-carbohydrate complementarity in mammalian gamete recognition. Gamete Res 1988;20:93-109. Mon K, Daitoh T, Irahara M, Kamada M, Aono T. Significance of n-mannose as a sperm receptor site on the zona pellucida in human fertilization. Am J Obstet Gynecol 1989;161:207-21 1. Mori K, Daitoh T, Kamada M, Maeda N, Maaegawa M, Hirano K, Irahara M, Aono T. Blocking of human fertilization by carbohydrates. Hum Reprod 1993;8:1729-1732.

63 Myles DO, Primakoff P. Localized surface antigens of guinea pig sperm migrate to new regions prior to fertilization. J Cell Biol 1984;99: 1631-1641. Oehninger 5, Clark OF, Acosta AA, Hodgen GD. Nature of the inhibitory effect of complex saccharide moieties on the tight binding of human spermatozoa to the human zona pellucida. Fertil Steril 1991 ;55: 165169. O’Rand MG, Widgren EE, Fisher Si. Characterization of the rabbit sperm autoantigen RSA as a lectin-like zona binding protein. Dev Biol 1988;129:23 1-240. Pontow SE, Kery V, Stahl PD. Mannose receptor. mt Rev Cytol 1992;! 37B: 221-244. Ram PA, Cardullo RA, Millette CF. Expression and topographical 10calization of cell surface fucosyltransferase activity during epididymal sperm maturation in the mouse. Gamete Res 1989;22:32l-332. Saling PM. Involvement of trypsin-like activity in binding of mouse spermatozoa to zona pellucida. Proc Nail Acad Sci USA 1981 ;78: 6231-6235. Saling PM. Mammalian sperm interaction with extracellular matrices of the egg. Oxf Rev Reprod Biol 1989;! 1:339-388. Saling PM. How the egg regulates sperm function during gamete interaction: facts and fantasies. Bid Reprod 1991 ;44:246-25 1. Saxena NK, Russell LD, Saxena N, Peterson RN. Immunofluorescence antigen localization on boar sperm plasma membranes: monoclonal antibodies reveal apparent new domains and apparent redistribution of surface antigens during sperm maturation and ejaculation. Anal Rec 1986;214:238-252. Shalgi R, Matityahu A, Nebel L. The role of carbohydrates in spermegg interaction in rats. Biol Reprod 1986;34:446-452. Swenson CE, Dunbar B. Specificity of sperm-zona interaction. J Exp Zool 1982;219:97-104. Taylor ME, Conary JT, Lennartz MR, Stahl PD, Drickamer K.. Primary structure of the mannose receptor contains multiple motifs resembling carbohydrate-recognition domains. J Biol Chem 1990;265: 12146-12162. Taylor ME, Drickamer K. Structural requirement for high affinity binding of complex ligands by the macrophage mannose receptor. JBiolChem 1 993;268(1):399-404. Tesarik J, Drahorad J, Testart J, Mendoza C. Acrosin activation follows its surface exposure and precedes membrane fusion in human sperm acrosome reaction. Development 1990;! 10:391-400. Tesarik I, Mendoza C, Carreras A. Expression of D-mannose binding sites on human spermatozoa: comparison of fertile donors and infertile patients. Fertil Steril l991;56:l 13-118. Tulsiani DRP, Nagdas SK, Cornwall GA, Orgebin-Crist MC. Evidence for the presence of high-mannose/hybrid oligosaccharide chain(s) on the mouse ZP2 and ZP3. Biol Reprod 1992;46:93100. Tulsiani DRP, Skudlarek MD, Orgebin-Crist MC. Novel a-D-mannosidase of rat sperm plasma membranes: characterization and potential role in sperm-egg interactions. J Cell Biol l989;109:1257-1267. Tulsiani DRP, Skudlarek MD, Orgebin-Crist MC. Human sperm plasma membranes possess a-D-mannosidase activity but no galactosyltransferase activity. Biol Reprod 1990;42:843-858. Wassarman PM. Early events in mammalian fertilization. Annu Rev Cell Biol l987;3: 109-142. Wassarman PM. Zona pellucida glycoproteins. Annu Rev Biochem 1988;57:415-442. Yanagimachi R. Mechanisms of fertilization in mammals. In: Mastroianni L, Biggers JD, eds. Fertilization andEmbryonic Development In Vitro New York: Plenum Press; 1991:8 1-1 82. Yurewicz EC, Pack BA, Sacco AG. Isolation, composition, and biological activity of sugar chains of porcine oocyte zona pellucida 55K glycoproteins. Mo! Reprod Dev 199 l;30(2): 126-134.