Human Reproduction vol.13 no.10 pp.2797–2804, 1998
Human sperm function in co-culture with human,
macaque or bovine oviduct epithelial cell monolayers
J.E.Ellington1,5, A.E.Jones1, C.M.Davitt2,
C.S.Schneider3, R.S.Brisbois4, G.A.Hiss4 and
R.W.Wright Jr3
1Washington
State University, Health Research and Education
Center, Spokane, WA 99201-3899, 2Electron Microscopy Center
and 3Department of Animal Sciences, Washington State University,
Pullman, WA 99164 and 4Northwest OB/GYN, P.S., 105 W. Eighth
#6020, Spokane, WA 99204 USA
5To
whom correspondence should be addressed
Human sperm function was compared in co-culture with
monolayers of oviduct epithelial cells (OEC) from three
species, human, macaque and bovine. For all species,
freeze–thawed and passaged OEC from females in the
periovulatory phase were used. OEC cultured on an extracellular matrix (Matrigel) formed a monolayer which supported human sperm attachment to OEC from all three
species. Spermatozoa in co-culture with OEC from all three
species showed prolonged survival and improved motility
characteristics over those cultured in medium alone. This
paper describes an efficient, repeatable co-culture system
for human spermatozoa which supports sperm attachment
to OEC and subsequently improves sperm function over
that seen in control medium cultures. Because the improved
sperm function in co-culture did not differ significantly
between human and bovine OEC for those attributes
studied, it is proposed that bovine OEC could be used as
an alternative to human OEC in certain human sperm coculture studies. Follicular phase bovine OEC from reproductively normal donors are far more accessible than their
human counterparts, thus making this co-culture system
more widely available for the study of human spermatozoa–
female tract interactions.
Key words: Fallopian or uterine tube/sperm function/sperm–
oviduct co-culture
Introduction
The mammalian Fallopian tube or oviduct is thought to play
an important role in post-ejaculatory, pre-fertilization sperm
function (Hunter, 1995; Tulsiani et al., 1997). During the past
decade, significant gains in our understanding of sperm storage
and oviductal interactions for animal species have been
achieved. Similar studies on human tubal physiology have
been much more difficult to complete due to ethical and
logistical constraints. However, recently acquired data suggest
that specific interactions between human spermatozoa and
the oviductal environment are important in early human
© European Society for Human Reproduction and Embryology
reproduction (Pacey et al., 1995a; Baillie et al., 1997; Murray
and Smith, 1997). Specifically, studies with human spermatozoa
have shown a beneficial effect on sperm function of exposure
to oviduct epithelial cells (OEC) or their secretory products
in vitro, including prolonged survival time, increased motility
and velocity, delay of capacitation, stabilization of chromatin
structure and improved fertilization rates (Zhu et al., 1994;
Pacey et al., 1995a; Lai et al., 1996; Akhondi et al., 1997;
Kervancioglu et al., 1997; Murray and Smith, 1997; Ellington
et al., 1998). In addition to improving sperm function, in-vivo
studies in women also suggest that the oviductal environment
may be important in the formation of a pre-fertilization sperm
reservoir (Mansour et al., 1993; Williams et al., 1993; Kunz
et al., 1996). Specifically, during the late follicular phase,
sperm sequestration in the oviduct appears to be directed to
the tube ipsilateral to ovulation, apparently to optimize the
presence of viable spermatozoa for participation in fertilization.
Taken together, the above studies suggest that sperm storage
and survival in the oviduct before fertilization involves a
dynamic interaction between spermatozoa and the oviductal
environment to optimize reproductive outcomes. However,
such interactions remain poorly understood, particularly in
humans (Hunter, 1995).
Studies of spermatozoa–oviduct interaction in other species
have benefited from the use of in-vitro co-culture to evaluate
the physiology of sperm function during exposure to OEC
(Ellington et al., 1993a; Dobrinski et al., 1996; Suarez et al.,
1997). Notwithstanding recent progress (Pacey et al., 1995a;
Baillie et al., 1997; Murray and Smith, 1997), this system has
been difficult to adapt for human tissues in large part due to
the difficulty in obtaining physiologically normal Fallopian
tube tissue. Specifically, alterations of the tubal epithelium and
its secretory products occur during physiological pre- and postmenopausal ageing and after many types of reproductive
pathology (Schultka et al., 1993; Adachi et al., 1995). Although
gross disruption of the tube is common in women with histories
of infertility (Speroff et al., 1994), many other significant
changes may not be obvious on gross examination, such as
changes in the extracellular matrix of the OEC or loss of cell
surface receptors which occur as women age (Schultka et al.,
1993; Adachi et al., 1995). In spite of such potential tubal
changes, human tubal specimens made available by surgery
are often obtained from older women, or from those with
endocrine or reproductive tract pathology, and therefore may
not be physiologically representative for use in studying
‘normal’ spermatozoa–OEC interactions. In addition, the endocrine status of a woman’s cycle significantly affects aspects
of OEC physiology such as ciliary and secretory activity
(Donnez et al., 1985; Jansen, 1995). As in other species,
2797
J.E.Ellington et al.
changes in the secretory products and epithelial surface of the
woman’s oviduct occur during the oestrogenic influence of the
late follicular phase. Therefore, donors of OEC to be used in
co-culture studies should also be selected based on their
endocrine status to ensure that oestrogen-specific cellular
aspects of the OEC are present.
Unlike other species studied, the use of OEC monolayers
to study human spermatozoa–OEC interactions has been unrewarding, as human spermatozoa do not readily attach to human
OEC in routinely cultured monolayers (Bongso et al., 1993;
Pacey et al., 1995a,b; Baillie et al., 1997). This attachment,
which appears to be a conserved trait across all mammalian
species studied to date (Hunter, 1995), has been reported for
human spermatozoa incubated with fresh OEC explants (Pacey
et al., 1995a; Baillie et al., 1997). However, the use of fresh
tissue explants for co-culture studies limits the usefulness of
this system. An ideal human sperm co-culture system would
utilize pools of banked, cryopreserved OEC cultured in monolayers which could be used across all replicates of an experiment, given that a significant individual female effect in OEC
quality exists (Ellington et al., 1993b).
In contrast to the difficulty in accessing human OEC from
reproductively normal donors, oviducts from young, pathologyfree cows in the follicular phase of their cycle can easily and
readily be accessed in large numbers from slaughterhouse
material (Ellington et al., 1990). It is not currently well
understood if the beneficial effects of OEC seen on human
sperm function require homologous species OEC. The current
study was performed to identify a co-culture system for human
spermatozoa using monolayers of cryopreserved and passaged
OEC, and to determine if OEC from a more accessible species
other than human could be utilized in this system.
Materials and methods
Oviduct epithelial cell recovery
Human OEC were recovered as surgical specimens from eight women
undergoing surgical interventions including sterilization reversal,
hysterectomy due to endometriosis, dysmenorrhoea or focal carcinoma
(not involving the oviducts). Specimens were obtained from women
in the mid to late follicular phase as determined visually at the time
of surgery. Macaque (monkey) OEC were recovered from six females
within a 24 h periovulatory time period, as determined by overt signs
of oestrus. Bovine OEC were collected from ten slaughterhouse
animals in the late follicular phase, as determined visually. For all
three species, only samples with grossly normal epithelium, no signs
of active uterine or oviductal infection, and vigorously active OEC
cilia (shown microscopically) were included.
To isolate OEC for co-culture, the presumptive isthmic portion of
the oviduct from each species was isolated and rinsed in phosphatebuffered saline (PBS) containing 10% fetal calf serum (FCS) and 2%
antibiotic/antimycotic premix. A sterile 25-gauge needle was then
introduced into one end of the tube and held in place. Several
millilitres of PBS (as above) were then slowly rinsed through the
tube into a Petri dish, while gently massaging the tube to loosen
OEC. Large clumps of retrieved OEC were then dissected apart by
tearing between two needles. The recovered OEC were washed
twice by centrifugation in 15 ml of PBS medium. OEC were then
resuspended in a base culture medium of 50:50 Dulbecco’s minimal
essential medium (DMEM):Ham’s F12 containing 5 µg/ml insulin,
2798
5 µg/ml transferrin, 5 ng/ml selenium, 10 ng/ml epidermal growth
factor (EGF), 1% antibiotic/antimycotic premix and either 15% FCS
(for human and macaque OEC) or 10% FCS (for bovine OEC).
Suspensions of OEC from each species were then placed into
2 cm2 tissue culture wells in small volumes (175 µl medium per
well) in order to optimize OEC contact with the well (for attachment)
in spite of their ciliary activity. Plates were left undisturbed for 24 h
in an incubator with 5% CO2 and 95% humidity, after which an
additional 200 µl medium was added to each well. At this time, samples
with poor OEC ciliary activity were discarded (e.g. insufficient ciliary
activity to move unattached OEC clumps in the culture wells). After
48 h of culture, all unattached OEC and medium were gently removed
from each well and 400 µl fresh medium was applied to cover the
OEC which had attached to the plastic wells and had begun to divide.
Once OEC confluency in the well occurred (within 4–5 days),
samples were further screened to confirm that vigorous ciliary activity
was observed in a majority of the fields. The OEC from three to five
females of each species were then pooled. Immunocytokeratin staining
confirmed .90% epithelial cell content in the pools (Ellington et al.,
1990). Successful pools were then cryopreserved in DMEM containing
50% FCS and 10% DMSO by placing OEC-containing vials in a
styrofoam container inside a –70°C freezer for 24 h before plunging
into liquid nitrogen. This freezing protocol does not preserve ciliary
activity of the OEC in the species evaluated.
Establishment of OEC for co-culture
OEC from each species were thawed, cultured in 100 mm2 tissue
culture plates until confluent, and passaged onto precoated Biocoat
Thin Layer Matrigel 24-well plates (Becton Dickinson, Bedford, MA,
USA) for use in the co-culture studies. Before use, the dried Matrigel
membranes were reconstituted and further rinsed with DMEM:F12
medium. OEC from each species were then separately passaged into
the 2 cm2 wells (23105 cells/well). Unattached OEC were removed
from the culture wells 18–24 h after plating. Cells were confluent in
2–3 days and used in the following studies at that time.
Oviduct growth in culture
The quantitative MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; thiazolyl blue; Sigma Chemical Co., St Louis,
MO, USA] mitochondrial assay measures mitochondrial activity of
viable cells through a spectrophotometric colour reaction (Mosmann,
1983). Absorbancy levels correlate to numbers of viable cells present.
To compare growth rates in culture, OEC from each species were
added to 96-well plates (13105 cells/well). On day 1 (24 h after
plating) and day 6 of culture, MTT assays were carried out to
determine the number of cells present in each well to assess cell
division rates for the three OEC types. MTT stock was incubated
with OEC wells for 4 h, removed, and 200 µl propanol added to
dissolve formazan crystals which had formed. Plates were then read
by a microplate reader set at 570 nm. A standard curve for each
species of OEC was determined comparing manual cell counts
on trypan blue-stained cells versus absorbance readings for wells
containing 53103 to 43104 OEC/well.
Sperm preparation
Freshly ejaculated semen from four fertile men (normal semen
analysis and children aged ,2 years) was used throughout these
studies. Spermatozoa were washed in human tubal fluid (HTF)
containing 5 mg/ml human serum albumin (HTF1). After washing,
the spermatozoa were analysed initially for computer-assisted sperm
analysis (CASA) motility (HTM-C, Version 8), viability (eosin–
nigrosin stain), hypo-osmotic swelling test (HOS) and count. Counts
on each sample were performed in triplicate using a Makler chamber.
Sperm function with oviduct cell monolayers
CASA data evaluated included percentage progressively motile,
straight-line velocity (VSL), curvilinear velocity (VCL), amplitude
of lateral head displacement (ALH), linearity (LIN) and beat cross
frequency (BCF).
Co-culture studies
Experimental wells were established by placing suspensions of
spermatozoa (143106/ml) in 500 µl of HTF1 into the treatments of:
(i) control HTF1 alone (no OEC); (ii) human OEC monolayers; (iii)
macaque OEC monolayers; (iv) bovine OEC monolayers; or (v)
Matrigel membranes alone (no OEC). All cultures were maintained
at 37°C in a 5% CO2/air mixture.
After 2 h, spermatozoa which had not attached to the OEC in coculture treatments were removed by rinsing the wells. Rinsing
involved pipetting up and down the full 500 µl of medium in each
well five times, followed by recovery of medium and spermatozoa
in suspension. An additional 500 µl medium was then vigorously
rinsed across each co-culture well by pipetting up and down five
more times before removal. For comparison, an aliquot of spermatozoa
was also removed from vigorously mixed control media wells after
2 h of culture. These 2-h sperm samples from each treatment were
evaluated for CASA motility, viability, HOS and Makler chamber
counts. Sperm attachment to OEC in co-culture treatments was
estimated by subtracting the number of spermatozoa (millions/ml)
removed from the co-culture wells at 2 h from the number of
spermatozoa in the control medium wells. Spermatozoa which
remained in co-culture by having attached to OEC were evaluated
qualitatively for flagellar beat scores, which denoted the vigour of
flagellar motion (from 5 5 fast to 1 5 slow).
Subsequently, spermatozoa were evaluated in each treatment after
4 h of culture and at 24 h intervals thereafter. At these times, all
spermatozoa released from OEC into suspension were removed from
the wells and fresh HTF1 was added. Flagellar beat scores were
then assigned to spermatozoa which remained attached to OEC in
each co-culture treatment and the percentage progressive motility was
determined for spermatozoa in the control wells. The ‘end of survival
time’ for spermatozoa in this study was defined as being when ,5%
of the spermatozoa remaining in the well showed active flagellar beat
(co-culture treatments), or remained progressively motile (control
treatment). Evaluation of motility characteristics for spermatozoa
which had been released from OEC between 4 and 24 h of co-culture
was also made using the CASA.
A modified sperm penetration assay using zona-free hamster eggs
(Rogers et al., 1979) was performed for spermatozoa in co-culture
with OEC from each of the three species compared with those in
control medium alone. For this experiment, spermatozoa which had
not attached to OEC were removed from the wells at 2 h after coculture initiation. Sperm numbers remaining (those attached to OEC)
were then calculated and a volume of spermatozoa was removed
from the control medium for use in the penetration assay; thus,
approximately equal numbers of spermatozoa were present in both
the co-culture and control wells. Commercially frozen hamster eggs
were then placed into Biggers, Whitten and Whittingham (BWW)
medium in co-culture or control wells and incubated routinely for
18 h (at least six oocytes/well/replicate for a total of 36 or more
oocytes per treatment). After incubation, oocytes were removed and
stained to determine the percentage of oocytes penetrated and the
mean number of spermatozoa which had penetrated each oocyte in
each treatment.
Electron micrography
Scanning electron micrographs of spermatozoa and OEC co-cultures
were prepared for each of the three OEC types after the initial 2 h
Figure 1. The growth rate of oviduct epithelial cells (OEC) from
different species in culture for 6 days, expressed as a ratio of cell
numbers (determined by absorbence) of day 6/day 1 on an
expanded scale of 0–1000.
of incubation. Co-cultures were fixed with 3% glutaraldehyde in 0.1
M cacodylate buffer pH 7.4 for 2 h at room temperature, rinsed twice
with 0.1 M cacodylate buffer, postfixed with 2% osmium tetroxide
in cacodylate buffer for 2 h, dehydrated in a graded ethanol series
(30, 50, 70, 95 and 100%) and then placed into two changes of pure
acetone. Samples were then immersed in 50:50 acetone:hexamethyldisilazane (HMDS) twice for 15 min each and subsequently into
100% HMDS twice for 15 min each followed by air-drying (Bray
et al., 1993). Samples were then coated with 20–30 nm of gold using
a Hummer V sputter coater.
Statistical analysis
Analysis of variance was used to evaluate OEC growth in culture
(MTT studies) and to compare sperm function among the five culture
treatments. The percentage of motile spermatozoa were arcsine
transformed before analysis. Unless otherwise stated, data are presented as mean values 6 SEM.
Results
Oviduct cell growth
Analysis of standard curves which compared levels of MTT
absorbence and manual cell counts for OEC from each of the
three species showed high levels of correlation in all cases
(r ù0.94). Bovine OEC consistently had much higher rates of
growth in culture compared with human or macaque OEC, as
determined by the ratio of cell numbers present on day 6 to
those present on day 1 after plating (Figure 1). Based on these
data and microscopic evidence of monolayer confluency, the
protocol for establishing monolayers was altered to initiate
culture of human and macaque OEC 1.5 days before initiation
of bovine OEC monolayers in order to utilize equivalent OEC
numbers for co-culture experiments. The MTT assay also
identified pools of OEC which grew relatively poorly in culture
and thus were not utilized in the following studies.
Sperm cell attachment
Attachment of human spermatozoa to all OEC types began to
occur within 15 min of co-culture initiation. This attachment
appeared to involve the sperm head, with the flagellum
remaining active. The flagellar beat score in all cases was
initially quite vigorous, with a gradual decline noted over
the time in co-culture, which consistently differed between
2799
J.E.Ellington et al.
spermatozoa from different donors. The numbers of spermatozoa attaching to OEC from each of the three species during
the first 2 h of co-culture were similar (Table I). Some sperm
attachment was also seen for spermatozoa in the Matrigelonly control. In all co-culture systems, spermatozoa were
occasionally noted to release from the OEC and become freeswimming, or to reattach to OEC in another location. Scanning
electron micrographs of spermatozoa attached to OEC from
all species showed the sperm heads in close association with
microvilli of the OEC and a mucus-like substance present in
the co-culture wells (Figure 2A–C).
Motility characteristics
Spermatozoa in each of the OEC co-culture treatments which
did not attach to the OEC within 2 h showed significantly
poorer motility than that observed for spermatozoa in control
media wells (27 6 4% versus 57 6 6%). This apparent
selectivity, with poorer quality spermatozoa not attaching to
the OEC, did not differ between OEC types (P 5 0.25).
Spermatozoa which had been released from each of the three
OEC treatments between 4 and 24 h of co-culture were more
progressively motile than were spermatozoa in the control
medium (Table II). Additionally, the VSL of these motile
spermatozoa was significantly improved at this time in both
the human OEC and the bovine OEC co-culture treatments
compared with the medium control. Other CASA variables
did not differ between treatments, although VCL and ALH
tended to be higher with the bovine OEC compared with the
medium control (P 5 0.1).
Sperm survival in culture
Survival time for spermatozoa in co-culture with each of the
three OEC types was significantly increased over that in
medium-only controls (Figure 3). In general, OEC co-culture
prolonged the survival of freshly ejaculated spermatozoa by
at least 24 h.
Table I. Numbers of spermatozoa (mean 6 SEM) attaching in co-culture to
oviduct epithelial cells (OEC) from the three species
Treatment
Number of spermatozoa
(3106) in suspension after
2 h of culture
Presumptive number of
spermatozoa (3106)
attached/well
Medium alone
(control)
Human OEC
Macaque OEC
Bovine OEC
Matrigel alone
14 6 2
0
11 6 1
12 6 2
12 6 1
13 6 2
3
2
2
1
6
6
6
6
0.5
0.3
0.1
0.1
Figure 2. (A) Human spermatozoa attached to human oviduct
epithelial cell (OEC) after 2 h of co-culture. (B) Human sperm
head in close association with microvilli of bovine OEC in coculture. Note the copious ‘mucus-like‘ substance covering the
portion of the OEC in contact with the spermatozoon. (C) Multiple
human spermatozoa attached to bovine OEC after 2 h of co-culture.
Again, note the mucus-like substance on the spermatozoa and OEC.
Scale bars: (A) 5 20 µm; (B) 5 2.0 µm; (C) 5 10 µm.
2800
Sperm function with oviduct cell monolayers
Table II. Motility characteristics of spermatozoa released from oviduct epithelial cells (OEC) between 4 and
24 h of co-culture or in control medium alone. Values are mean 6 SEM
Human
OEC
% Progressively motile
VSL (µm/s)
VCL (µm/s)
ALH (µm)
71
58
98
5.1
6
6
6
6
11a
8a
9a
0.6a
Macaque
OEC
76
49
104
6.2
6
6
6
6
9a
7a,b
11a
1.0a
Bovine
OEC
Matrigel 1
medium
74 6 10a
57 6 3a
118 6 10a
7.4 6 0.9a
59
47
101
6.3
6
6
6
6
Medium
alone
15a,b 47
6a,b
42
8a
97
0.8a 6.8
6
6
6
6
9b
4b
9a
10a
a,bDifferent superscripts within rows show means which differ at P , 0.05.
VSL 5 straight-line velocity.
VCL 5 curvilinear velocity.
ALH 5 amplitude of lateral head displacement.
Sperm penetration assay
The number of spermatozoa able to penetrate zona-free hamster
oocytes was lower for those co-cultured with human or
macaque OEC, than for spermatozoa in bovine OEC co-culture
or medium controls (Figure 4). Penetration rates did not differ
for spermatozoa in bovine OEC co-culture or control medium,
although 54 6 9% of the spermatozoa in the co-culture wells
remained attached to OEC during the penetration assay, and
thus were probably not available for oocyte penetration. The
overall percentage of oocytes penetrated by spermatozoa did
not differ among treatments (human OEC 73%, macaque OEC
85%, bovine OEC 90% and medium control 89%).
Discussion
The human spermatozoa and OEC monolayer co-culture system
described in this report supported spermatozoa attachment to
OEC which was similar to that seen in other mammalian species.
Furthermore, the beneficial effect of OEC co-culture on sperm
motility and survival did not require species homology between
the spermatozoa and OEC. In this study, bovine and human
OEC improved sperm function in vitro over that observed for
spermatozoa in control medium alone. In fact, for the sperm
penetration assays, spermatozoa in co-culture with bovine OEC
showed a greater ability to penetrate hamster oocytes than either
human or monkey OEC. This may have been due to the source
of human and monkey OEC, where older females in a broader
range of cycle stages were used as donors compared with bovine
OEC donors. It is not clear if the lower zona penetration rates
found for spermatozoa exposed to human and monkey OEC
were due to inferior explant quality, or whether these OEC
provided factors that stabilized sperm against capacitation which
were absent from the bovine OEC co-culture.
In general, the ability to utilize bovine OEC from young,
healthy animals in the oestrus phase of their cycle further adds
to the efficiency of the OEC co-culture system described here.
The ability to collect adequate numbers of pathology-free OEC
from women in the late follicular phase has previously limited
the use of a homologous co-culture system. Women in this study
became OEC donors due to a variety of medical conditions.
Because OEC were pooled together from several women it is
not possible to identify any specific effects which these various
conditions may have had on spermatozoa–OEC interactions.
Even though aspects of sperm physiology in the female ovi-
duct differ between species, we have found improved sperm
function in cross-over designs with species as varied as the
equine, canine, bovine, feline and human (Ellington et al., 1993c;
and unpublished data). Others have also reported an effect of
heterologous OEC on human sperm function, including changes
in spermatozoa–zona binding, and a prolongation of motility
during culture (Guerin et al., 1991; Wetzels et al., 1995). In
contrast, others have found certain sperm function changes
induced by oviductal products to require homologous human
OEC (DeJonge et al., 1993; O’Day-Bowman et al., 1996).
Recent work has shown that the carbohydrate moieties involved
in sperm attachment to OEC may differ between species (Demott
et al., 1995; Lefebvre et al., 1997). However, sperm attachment
to heterologous species OEC readily occurs, even for species
with apparently different lectins involved in spermatozoa to
OEC attachment (e.g. stallion sperm to bovine OEC; Samper
et al., 1995). Therefore, it is possible that some aspects of
spermatozoa–OEC interaction function generally across species,
whereas other components function through more speciesspecific mechanisms.
In the oviductal environment, as well as in the OEC co-culture
system, a variety of factors may exert differing effects on sperm
function (Aitken, 1997). Prolongation of motility for spermatozoa in co-culture with OEC and other reproductive cell types
has been widely reported (Bongso and Trounson, 1996; Akhondi
et al., 1997; Guerin et al., 1997). This may be due to a stabilizing
effect for spermatozoa against capacitation changes in co-culture
(Murray and Smith, 1997). However, concomitant increases in
the populations of spermatozoa showing capacitation changes
have also been reported during co-culture (Guerin et al., 1991;
Ellington et al., 1993a; Guerin et al., 1997). It is possible that
separate OEC surface and/or secretory products work together
to stabilize a cohort of spermatozoa for prolonged storage, as
well as to capacitate released spermatozoa for participation in
fertilization. This may explain discrepancies found between
studies with regards to how specific motility parameters are
affected during co-culture.
Previously, attachment of human spermatozoa to OEC has
been variably reported in the literature (Williams et al., 1993;
Pacey et al., 1995a; Baillie et al., 1997). Co-culture systems
before this study have required fresh explants of OEC in order for
such attachment to occur consistently. We propose that culturing
OEC on an extracellular matrix, as described in this study,
allowed for greater differentiation of the OEC which supported
2801
J.E.Ellington et al.
Figure 3. Survival time of spermatozoa in culture. Survival time is
defined as the time until ,5% of spermatozoa remain progressively
motile (medium control) or have flagellar beat activity
(co-cultures).
Figure 4. Numbers of spermatozoa penetrating zona-free hamster
oocytes (n ù 36 oocytes/treatment).
the expression of surface lectins required for the attachment of
human spermatozoa to the freeze–thawed, passaged OEC. We
have found recently that human spermatozoa will also attach to
OEC monolayers which have been routinely cultured in plastic
wells (with no exogenous extracellular matrix) for a week or
more. This time may allow the OEC in such cultures to establish
a more elaborate endogenous extracellular matrix for the expression of OEC surface components required for sperm attachment
(J.E.Ellington et al., unpublished data).
Human sperm attachment to OEC in co-culture seems to
involve a lower percentage of the overall sperm sample than that
observed in other species studied to date; however, as in other
species, this attachment does appear to select the higher-quality
spermatozoa in a sample (Thomas et al., 1994). The current
studies of human spermatozoa–OEC interactions in vitro suggest
that sperm detachment and reattachment to OEC may occur
readily (Pacey et al., 1995a; Murray and Smith, 1997). Such
transient attachment was observed during the 4 h co-culture time
periods of these studies. Recent results in our laboratory suggest
that human sperm attachment continues to increase (both in
2802
terms of numbers and stability) throughout the first 24 h of coculture (data not shown).
In vivo, storage of human spermatozoa in the Fallopian tube
probably involves not only sperm attachment to OEC, but also
entrapment of spermatozoa in the viscous, copious mucus
present in the isthmus near the time of ovulation (Donnez et al.,
1985; Jansen, 1995). This combination of sperm attachment to
the OEC and sequestration in tubal mucus to form a prefertilization reservoir has also been reported in other species
(Suarez et al., 1997). Formation of the hypothetical isthmic
reservoir of spermatozoa in women may be directed in part
by the dominant follicle on the side of ovulation, potentially
through paracrine effects (Kunz et al., 1996). Oviductal lumen
glycoconjugates are also likely to be involved in the creation
of such a sperm reservoir through mediation of both sperm
attachment and survival (Wu et al., 1993). In the human tube,
OEC glycoconjugates differ between ciliated OEC with sialic
acid residues, and secretory OEC with fucose residues (Jansen,
1995). Further, these fucose residues have been found to
be evenly distributed throughout the human tube, whereas
galactosyl residues are found only in the isthmus of the human
oviduct, suggesting a specific role for these isthmic residues
at the presumptive site of sperm storage (Wu et al., 1993).
Studies are currently under way to identify the specific glycoconjugates involved in human sperm attachment to OEC.
The ability to utilize OEC monolayers for the study of
human sperm function in the tubal environment offers several
advantages. Freeze–thawed pools of passaged OEC can be
screened for cytokeratin content as well as for the ability to
divide readily in culture, thus confirming the use of vigorous
epithelial cells in sperm co-culture studies. Passaged lots of
OEC also allow a more standardized environment for replicating sperm function experiments by controlling for OEC variation across studies. We have successfully utilized OEC from
each of the species presented here through three passages.
Studies with bovine OEC show a significant decrease in the
numbers of spermatozoa attaching and how long these survive
in co-culture after five passages of the OEC (J.E.Ellington
et al., unpublished results).
Co-culture of human spermatozoa with the Matrigel matrix
alone also had beneficial effects on sperm function over that
seen in medium-only culture, including sperm attachment to
the matrix and prolonged survival and motility changes.
However, the results were not equivalent to the benefit gained
for spermatozoa with OEC present. These findings are consistent with interactions reported for stallion spermatozoa in coculture with Matrigel, in which the enhanced sperm function
declined over time, or was less marked than that seen with
OEC co-culture (Thomas et al., 1994; Dobrinski et al., 1996).
Quantitative determination of the numbers of spermatozoa
attaching to OEC in co-culture is difficult. In this study, sperm
cell counts should be viewed as relative, as the ability to
distinguish accurately between treatments in the working
ranges found here (millions of sperm/ml) is not possible.
Our laboratory has utilized various systems for counting
spermatozoa in these lower numbers (haemocytometers,
Makler and Coulter counters), and have found the smallest
coefficient of variation to occur with carefully prepared Makler
Sperm function with oviduct cell monolayers
chambers (data not shown). A recent study supports the
accuracy of the Makler chamber, particularly for low-density
sperm suspensions as being down to 53106/ml (Shiran
et al., 1995).
In conclusion, a system for establishing human spermatozoa
and oviduct epithelial cell co-cultures is described in this
study. This system utilizes freeze–thawed and passaged OEC
cultured in monolayers on an extracellular matrix. Such OEC
have a beneficial effect on sperm function and support sperm
attachment to the OEC, possibly mimicking post-ejaculatory,
pre-fertilization changes observed for spermatozoa in the
woman’s Fallopian tube. Spermatozoa–OEC interactions
observed in this study were similar in co-culture with both
bovine and human OEC, suggesting that some applications of
the co-culture system for human spermatozoa could make use
of the more readily accessible bovine OEC.
Acknowledgements
This work was funded by the NICHHD #HD 32851. The authors
thank Dr S.A.Oliver for critical comment and review and D.Kaiser
for excellent manuscript and figure preparation.
References
Adachi, K., Kurachi, H., Adachi, K. et al. (1995) Menstrual cycle specific
expression of epidermal growth factor receptors in human fallopian tube
epithelium. J. Endocrinol., 147, 553–563.
Aitken, R.J. (1997) Molecular mechanisms regulating human sperm function.
Mol. Hum. Reprod., 3, 169–173.
Akhondi, M.A., Chapple, C. and Moore, H.D.M. (1997) Prolonged survival
of human spermatozoa when co-incubated with epididymal cell cultures.
Hum. Reprod., 12, 514–522.
Baillie, H.S., Pacey, A.A., Warren, M.A. et al. (1997) Greater numbers of
human spermatozoa associate with endosalpingeal cells derived from
the isthmus compared with those from the ampulla. Hum. Reprod., 12,
1985–1992.
Bongso, A. and Trounson, A. (1996) Evaluation of motility, freezing ability
and embryonic development of murine epididymal sperm after co-culture
with epididymal epithelium. Hum. Reprod., 11, 1451–1456.
Bongso, A., Ho, J., Fong, C.Y. et al. (1993) Human sperm function after coculture with human fallopian tubal epithelial cell monolayers: in vitro model
for studying cell interactions in early human conception. Arch. Androl., 31,
183–190.
Bray, D.F., Bagu, J. and Koegler, P. (1993) Comparison of
hexamethyldisilazane (HMDS), Peldri II, and critical-point drying methods
for scanning electron microscopy of biological specimens. Microsc. Res.
Tech., 26, 489–495.
DeJonge, C.J., Barratt, C.L.R., Radwanska, E. et al. (1993) The acrosome
reaction-inducing effect of human follicular and oviductal fluid. J. Androl.,
14, 359–365.
Demott, R.P., Lefebvre, R. and Suarez, S.S. (1995) Carbohydrates mediate
the adherence of hamster sperm to oviductal epithelium. Biol. Reprod., 52,
1395–1403.
Dobrinski, I., Suarez, S.S. and Ball, B.A. (1996) Intracellular calcium
concentration in equine spermatozoa attached to oviduct epithelial cells
in vitro. Biol. Reprod., 54, 783–788.
Donnez, J., Casanas-Roux, F., Caprasse, J. et al. (1985) Cyclic changes in
ciliation, cell height, and mitotic activity in human tubal epithelium during
reproductive life. Fertil. Steril., 43, 554–559.
Ellington, J.E., Carney, E.W., Farrell, P.B. et al. (1990) Bovine 1-2-cell
embryo development using a simple medium in three oviduct epithelial cell
co-culture systems. Biol. Reprod., 43, 100–107.
Ellington, J.E., Ball, B.A., Blue, B.J. et al. (1993a) Capacitation-like membrane
changes and prolonged viability in vitro of equine spermatozoa cultured
with uterine tube epithelial cells. Am. J. Vet. Res., 54, 1505–1510.
Ellington, J.E., Varner, D.D., Burghardt, R.C. et al. (1993b) Cell-to-cell
communication of equine uterine tube (oviduct) cells as determined by
anchored cell analysis and sorting in culture. Anim. Reprod. Sci., 30,
313–324.
Ellington, J.E., Meyers-Wallen, V.N. and Ball, B.A. (1993c) Establishment of
a co-culture system for canine sperm and uterine tube epithelial cells. Vet.
Rec., 136, 542–543.
Ellington, J.E., Evenson, D.P., Fleming, J.E. et al. (1998) Co-culture of human
sperm with bovine oviduct epithelial cells decreases sperm chromatin
structural changes seen during culture in media alone. Fertil. Steril., 69,
643–649.
Guerin, J.F., Ouhibi, N., Regnier-Vigouroux, G. and Ménézo, Y. (1991)
Movement characteristics and hyperactivation of human sperm on different
epithelial cell monolayers. Int. J. Androl., 14, 412–422.
Guerin, J.F., Merviel, P. and Plachot, M. (1997) Influence of co-culture with
established human endometrial epithelial and stromal cell lines on sperm
movement characteristics. Hum. Reprod., 12, 1197–1202.
Hunter, R.H.F. (1995) Human sperm reservoirs and fallopian tube function: a
role for the intra-mural portion? Acta Obstet. Gynecol. Scand., 74, 677–681.
Jansen, R.P.S. (1995) Ultrastructure and histochemistry of acid mucus
glycoproteins in the estrous mammal oviduct. Microsc. Res. Tech., 32, 29–49.
Kervancioglu, M.E., Saridogan, E., Atasti, T. et al. (1997) Human fallopian
tube epithelial cell co-culture increases fertilization rates in male factor
infertility but not in tubal or unexplained infertility. Hum. Reprod., 12,
1253–1258.
Kunz, G., Beil, D., Deininger, H. et al. (1996) The dynamics of rapid sperm
transport through the female genital tract: evidence from vaginal sonography
of uterine peristalsis and hysterosalpingoscintigraphy. Hum. Reprod., 11,
627–632.
Lai, Y.M., Chang, F.H., Lee, C.L. et al. (1996) Co-culture of human
spermatozoa with reproductive tract cell monolayers can enhance sperm
functions better than co-culture with Vero cell monolayers. J. Assist. Reprod.
Genet., 13, 417–422.
Lefebvre, R., Lo, M.C. and Suarez, S.S. (1997) Bovine sperm binding
to oviductal epithelium involves fucose recognition. Biol. Reprod., 56,
1198–1204.
Mansour, R.T., Aboulghar, M.A., Serour, G.I. et al. (1993) In vivo survival
of spermatozoa in the human fallopian tube for 25 days: a case report. J.
Assist. Reprod. Genet., 10, 379–380.
Mosmann, T. (1983) Rapid colorimetric assays for cellular growth and
survival: application to proliferation and cytotoxicity assays. J. Immunol.
Methods, 65, 55–63.
Murray, S.C. and Smith, T.T. (1997) Sperm interaction with fallopian tube
apical membrane enhances sperm motility and delays capacitation. Fertil.
Steril., 68, 351–357.
O’Day-Bowman, M.B., Mavrogianis, P.A., Reuter, L.M. et al. (1996)
Association of oviduct-specific glycoproteins with human and baboon
(Papio anubis) ovarian oocytes and enhancement of human sperm binding
to human hemizonae following in vitro incubation. Biol. Reprod., 54, 60–69.
Pacey, A.A., Hill, C.J., Scudamore, I.W. et al. (1995a) The interaction in vitro
of human spermatozoa with epithelial cells from the human uterine
(Fallopian) tube. Hum. Reprod., 10, 360–366.
Pacey, A.A., Davies, N., Warren, M.A. et al. (1995b) Hyperactivation may
assist human spermatozoa to detach from intimate association with the
endosalpinx. Hum. Reprod., 10, 2603–2609.
Rogers, B.J., Van Campen, H., Ueno, M. et al. (1979) Analysis of human
spermatozoal fertilizing ability using zona-free ova. Fertil. Steril., 32,
664–670.
Samper, J., Ellington, J.E., Burnett, K. et al. (1995) Use of sperm and oviduct
cell co-culture as a test for stallion field fertility. Am. Assoc. Equine Proc.,
41, 3–5.
Schultka, R., Gopel, C., Schuppan, D. et al. (1993) Age-dependent changes
of the immunohistochemical distribution of various collagen types and
structural glycoproteins in the human uterine tube. Acta Histochem. (Jena),
95, 139–153.
Shiran, E., Stoller, J., Blumenfeld, Z. et al. (1995) Evaluating the accuracy
of different sperm counting chambers by performing strict counts of
photographed beads. J. Assist. Reprod. Genet., 12, 434–442.
Speroff, L., Glass, R.H. and Kase, N.G. (1994) Clinical Gynecologic
Endocrinology and Infertility. Williams & Wilkins, Baltimore.
Suarez, S.S., Brockman, K. and Lefebvre, R. (1997) Distribution of mucus
and sperm in bovine oviducts after artificial insemination: the physical
environment of the oviductal sperm reservoir. Biol. Reprod., 56, 447–453.
Thomas, P.G.A., Ball, B.A., Miller, P.G. et al. (1994) A subpopulation of
morphologically normal, motile spermatozoa attach to equine oviductal
epithelial cell monolayers. Biol. Reprod., 51, 303–309.
2803
J.E.Ellington et al.
Tulsiani, D.R., Yoshida-Komiya, H. and Araki, Y. (1997) Mammalian
fertilization: a carbohydrate-mediated event. Biol. Reprod., 57, 487–494.
Wetzels, A.M.M., Van der Auwera, I., Bastiaans, B.A. et al. (1995) Sperm
functional changes and fertilization in vitro in co-culture with human skin
fibroblasts. Hum. Reprod., 10, 137–141.
Williams, M., Barratt, C.L.R., Hill, C.J. et al. (1993) Recovery of artificially
inseminated spermatozoa from the fallopian tubes of a woman undergoing
total abdominal hysterectomy. Hum. Reprod., 7, 506–509.
Wu, T.C., Lee, S.M., Jih, M.H. et al. (1993) Differential distribution of
glycoconjugates in human reproductive tract. Fertil. Steril., 59, 60–64.
Zhu, J., Barratt, C.L.R., Lippes, J. et al. (1994) The sequential effects of
human cervical mucus, oviductal fluid, and follicular fluid on sperm
function. Fertil. Steril., 61, 1129–1135.
Received on January 22, 1998; accepted on July 3, 1998
2804