Deleterious Actions of Gossypol on Bovine Spermatozoa, Oocytes, and Embryos

Prévia do material em texto

BIOLOGY OF REPRODUCTION 57, 901-907 (1997)
Deleterious Actions of Gossypol on Bovine Spermatozoa, Oocytes, and Embryos
C. Brocas,3 ,6 R.M. Rivera,3 F.F. Paula-Lopes, 3 L.R. McDowell, 4 M.C. Calhoun,7 C.R. Staples, 3 N.S. Wilkinson, 4
A.J. Boning,4 P.J. Chenoweth,5 and P.J. Hansen 2,3
Department of Dairy and Poultry Sciences,3 Animal Science Department,4 and Department of Large Animal Clinical
Sciences,5 University of Florida, Gainesville, Florida 32611-0920
Ecole Nationale d'lng6nieurs de Travaux Agricoles,6 Clermont-Ferrand, France
Agricultural Research Center,7 Texas A&M University, San Angelo, Texas 76901
ABSTRACT
Gossypol (50 and 100 ig/ml) decreased the percentage of
sperm that completed the swim-up procedure. This effect was
not blocked by glutathione monoethyl ester. Cleavage rates were
not different between oocytes inseminated with gossypol-treated
spermatozoa (10 or 50 ig/ml) and oocytes inseminated with
control spermatozoa. Development to the blastocyst stage at
Day 7 after insemination was reduced when spermatozoa treat-
ed with 50 Ig/ml gossypol were used for fertilization. Gossypol
toxicity was evident in cows fed cottonseed meal because eryth-
rocyte fragility was greater than for control cows. However,
there were no differences between cottonseed meal and control
groups in number of oocytes collected per cow, cleavage rate
after in vitro maturation and fertilization, or the proportion of
oocytes or embryos that developed to blastocysts. Similarly, ex-
posure of oocytes to 2.5-10 iug/ml gossypol during in vitro mat-
uration did not affect cleavage rates or subsequent development.
In contrast, addition of 10 jig/ml gossypol to embryos reduced
cleavage rate. Moreover, development of cleaved embryos was
reduced by culture with 5 or 10 Lg/ml gossypol and tended to
be reduced by 2.5 g/ml gossypol. In conclusion, bovine ga-
metes are resistant to gossypol at concentrations similar to those
in blood of cows fed cottonseed meal. In contrast, the devel-
oping embryo is sensitive to gossypol.
INTRODUCTION
Gossypol is a polyphenolic yellow pigment present in
cotton plants (Gossypium spp.) that has been implicated as
a reproductive toxicant [1]. Feeding cottonseed high in gos-
sypol content to bulls has been associated with reduced
sperm production, increased sperm abnormalities, de-
creased sperm motility, and decreased thickness of the ger-
minal epithelium of the testis [2, 3]. Gossypol has been
reported to compromise function of cultured spermatozoa
from cattle [4, 5] and other species [6, 7]. Exposure to gos-
sypol while in the female reproductive tract could conceiv-
ably result in damage to sperm. Additionally, gossypol
might also reduce female fertility by disrupting the oocyte
or embryo. Addition of gossypol to culture media inhibited
nuclear maturation of bovine oocytes [8] and subsequent
development of bovine embryos [9].
The overall objective of the present experiments was to
evaluate the effect of gossypol on bovine gametes and em-
bryos. In one series of experiments, the ability of gossypol to
inhibit motility and fertilizing capacity of cultured bovine
spermatozoa was evaluated. The aim of another experiment
was to determine whether the effect of gossypol on sperma-
Accepted June 10, 1997.
Received April 30, 1997.
'This is Journal Series No. R-05724 of the Florida Agricultural Exper-
imental Station. Research was partially supported by grants from the Flor-
ida Milk Checkoff Program and USDA-CBAG (Grant #95-34135-1860).
2Correspondence. FAX: (352) 392-5595; e-mail: hansen@animal.ufl.edu
901
tozoa could be reversed by a membrane-permeable form of
the antioxidant glutathione (GSH). This hypothesis was tested
because some toxicological actions of gossypol on sperma-
tozoa involve free radical generation [10-12]. Moreover, GSH
has been shown to reduce lipid peroxidation in boar sperm
[13] and increase motility of bull spermatozoa [14,15]. An
additional objective was to test whether feeding cattle a diet
supplemented with gossypol inhibited fertilization and sub-
sequent embryonic development when oocytes were subjected
to an in vitro maturation, fertilization, and culture system.
Finally, experiments were performed to determine whether
addition of gossypol to culture medium during oocyte matu-
ration or embryo culture affected cleavage rate or subsequent
embryonic development.
MATERIALS AND METHODS
Materials
Gossypol was purchased from Sigma Chemical Com-
pany (St. Louis, MO; lot 93H4014) and was determined to
be composed of 50.2% (+) isomer and 49.8% (-) isomer
using analytical procedures described below. Gossypol was
dissolved to a concentration of 20 mg/ml in ethanol before
dilution in culture medium. Reduced glutathione was ob-
tained from Sigma, and glutathione monoethyl ester (GSH)
was synthesized as previously described [16]. Frozen se-
men from various bulls was obtained from American
Breeders Service (De Forest, WI) or Southeast Semen
(Lake City, FL). Materials and media for in vitro matura-
tion, fertilization, and culture (IVM/IVF/IVC) were as de-
scribed previously [17] and included recipes for Hepes-
TALP, Sp-TALP, and IVF-TALP [18], and CRlaa medium
[19]. Standards for gossypol, -carotene, retinol, retinyl
palmitate, and DL-ot-tocopherol were obtained from Sigma.
Experiments with Cultured Spermatozoa
Effects of gossypol on the ability of sperm to undergo swi-
m-up purification. The swim-up procedure involved purifi-
cation of motile sperm by collection of those that could swim
from the bottom of a 1.25-ml column of medium to the top
0.8 ml. For each replicate, straws of semen from several bulls
were thawed, pooled, and purified by Percoll gradient cen-
trifugation [20] and then diluted to a concentration of 1.5-3.0
x 106 spermatozoa/ml in Hepes-TALP Aliquants of sper-
matozoa (0.5 ml) were mixed with an equal volume of gos-
sypol in Hepes-TALP containing 1.0% ethanol so that the
final concentrations of gossypol were 0, 10, 50, and 100 Rg/
ml, and all tubes contained 0.5% ethanol. All treatments were
applied in duplicate. Spermatozoa were incubated for 1 h at
39°C in an atmosphere of 5% CO2, centrifuged at 1000 x g,
and resuspended in 0.25 ml Hepes-TALP The resuspended
sperm were then layered under 1.0 ml of Hepes-TALP in a
BROCAS ET AL.
TABLE 1. Composition of experimental diets expressed as a percentage
of the total diet on a dry matter basis.
Control Cottonseed meal,
Ingredient Period 1 Period 2 Period 1 Period 2
Corn silage 45.1 38.9 35.9 31.0
Soybean meal 36.7 31.8
Cottonseed mealb 45.1 39.0
Alfalfa hay 17.3 28.5 17.3 28.5
Trace mineral salt 0.5 0.5 0.5 0.5
Limestone 0.4 0.3 1.2 1.0
a Alfalfa hay was fed at 2.2 kg dry matter/head per day before Days 21-
35 of feeding (period 1) and at 4.2 kg/head per day afterward (period 2);
for other dietary ingredients, the daily amount fed (kg/head) was the same
for both periods.
b Cottonseed meal contained 0.119% free gossypol and 1.09% total gos-
sypol: 62.6% of gossypol was the (+) isomer.
12 x 75-mm tube and incubated for 1 h at 390C and 5% CO2
in air. Afterward, the top 0.8 ml of Hepes-TALP was col-
lected, and the number of spermatozoa in this fraction was
determined by hemacytometer. Results were expressed as the
number of sperm undergoing swim-up (100 x [number of
sperm in 0.8 ml/number of sperm added to tube]). The ex-
periment was replicated four times with a different group of
donors for each replicate.
Interactions between gossypol and glutathione ester on
swim-up. This experiment was performed as described
above except that spermatozoa received gossypol (0 or 100
ixg/ml) and GSH ester (0, 0.1, or 1 mM) in a 2 X 3 factorial
arrangement of treatments. Each treatment was performed
in duplicate, and the experiment was replicated on five sep-
arate occasions.
Effect of gossypol on fertilizing ability of spermatozoa.
Semen fromseveral bulls was mixed and used for each
replicate. After purification by Percoll gradient centrifuga-
tion, spermatozoa were diluted to 3-3.5 x 106 spermato-
zoa/ml in Sp-TALP. Aliquants of spermatozoa were then
mixed with an equal volume of gossypol in Sp-TALP con-
taining 1.0% ethanol to achieve final gossypol concentra-
tions of 0, 10, and 50 pxg/ml. Spermatozoa were incubated
for 1 h at 39°C in an atmosphere of 5% CO2 , centrifuged
at 1000 x g, washed in Sp-TALP, and resuspended in IVF-
TALP to a concentration of 3-3.5 x 106 spermatozoa/ml.
Each preparation of spermatozoa was used to fertilize 1-
2 groups of -30 matured oocytes placed in 600 pl IVF-
TALP in 4-well plates. Sperm concentration was deter-
mined by hemacytometer and adjusted so that concentration
in the fertilization well was 250 000/ml. Fertilization drops
included 20 RIM penicillamine, 10 FM hypotaurine, and 2.5
p.M epinephrine. Fertilization was allowed to proceed for
9 h at 39°C in an atmosphere of 5% CO 2. Then oocytes/
zygotes were vortexed to remove cumulus, placed in groups
of -10 in 50-pl microdrops of CRlaa medium containing
3 mg/ml BSA, and cultured for 9 days at 39°C and 5%
CO2. Fetal calf serum (5 p1l) was added to each drop at Day
5 after insemination. Further details of oocyte collection
and maturation, IVF, and IVC are described by Edwards
and Hansen [17]. Cleavage rate was determined on Day 3
after insemination and development to blastocyst stage at
Days 7-9. The experiment was replicated on four separate
occasions.
Effect of Feeding Cottonseed Meal on Oocyte Quality
Cows and sampling. Twelve nonlactating Holstein cows
were assigned randomly to one of two dietary treatments
TABLE 2. Chemical analysis of experimental diets expressed on a dry
matter basis.
Control Cottonseed meal,
Item Period 1 Period 2 Period 1 Period 2
Net energy (Mcal/kg) 0.32 0.32 0.31 0.31
Crude protein (%) 26.5 27.8 26.6 26.8
Acid-digestible fiber (%) 24.3 24.1 26.8 25.1
Neutral-digestible fiber (%) 36.4 32.7 36.9 28.0
Calcium (%) 0.71 0.98 0.91 1.11
Phosphorous (%) 0.48 0.47 0.72 0.68
Magnesium (%) 0.28 0.30 0.43 0.43
Potassium (%) 2.09 1.98 1.85 1.63
Sodium (%) 0.24 0.24 0.33 0.30
Sulfur (%) 0.23 0.25 0.28 0.30
Iron (ppm) 240 212 167 198
Zinc (ppm) 55 67 57 67
Copper (ppm) 11 16 11 14
Manganese (ppm) 44 44 37 41
Molybdenum (ppm) 2.3 2.3 1.5 1.4
Total gossypol (g/cow/day) b ,, 0.00 0.00 62.68 62.68
Free gossypol (g/cow/day) c 0.00 0.00 6.84 6.84
, Alfalfa hay was fed at 2.2 kg dry matter per head per day before Days
21-35 of feeding (period 1) and at 4.2 kg/head per day afterward (period
2); for other dietary ingredients, the daily amount fed (kg/head) was the
same for both periods.
b Of the gossypol in cottonseed meal, 62.6% was the (+) isomer.
C Based on a daily intake of total mixed ration of 12.74 (period 1) or 14.74
kg dry matter (period 2).
(cottonseed meal or control) and one of three replicates.
The composition of diets is shown in Table 1, and their
chemical analysis is presented in Table 2. Diets were for-
mulated to be isocaloric and isonitrogenous. The cottonseed
meal diet provided 6.84 g free gossypol per cow per day.
Cows were group-fed the diets twice daily as a total mixed
ration. The amount of alfalfa hay in the diet was increased
once (at Days 21-35 of experimental feeding, depending
upon replicate) from 2.2 to 4.2 kg dry matter per cow per
day to allow ad libitum feeding (i.e., cows were consuming
all the feed provided when hay was fed at 2.2 kg dry matter
per cow per day).
Prior to the experiment, cows were maintained on pas-
ture and then fed a transition diet based on corn silage for
5-19 days. Feeding of experimental diets was initiated at
7-day intervals for each replicate of 4 cows (2 cottonseed
meal and 2 control) and continued for 9 weeks. Thus, rep-
licate 1 began experimental diets on May 15, replicate 2
on May 22, and replicate 3 on May 29. While being fed
experimental diets, cows were housed in a free stall facility
in which a high-pressure fogger system and fans had been
installed to provide relief from heat stress.
Each week, blood samples were collected from the coc-
cygeal vessels into heparinized tubes and subsequently used
for analysis of erythrocyte fragility and of plasma concen-
trations of gossypol, (x-tocopherol, 3-carotene, retinyl pal-
mitate, and retinol. Samples of the total mixed ration were
collected weekly and stored at -20 0 C. Each cow was
slaughtered after it received the experimental diet for 9 wk.
Ovaries were collected and oocytes subjected to IVM/IVF/
IVC.
Erythrocyte fragility tests. Aliquots of whole heparinized
blood (25 prl) were added to tubes containing 5 ml of either
distilled water or 0.45, 0.55, 0.65, 0.75, or 0.85% (w:v)
buffered NaCl [21], mixed by inversion, and incubated at
room temperature for 30 min. Samples were centrifuged at
1000 X g for 10 min, and the absorbance at 540 nm was
902
ACTIONS OF GOSSYPOL ON BOVINE GAMETES AND EMBRYOS
determined by spectrophotometer using distilled water as a
blank. Percent hemolysis was calculated as follows:
Percent hemolysis = 100 x (A540 of sample in NaCl/
A540 of sample in water)
Chemical assays and feed analysis. Two composite sam-
ples of each total mixed ration were subjected to chemical
analysis at the Dairy Herd Improvement Association For-
age Testing Laboratory (Ithaca, NY).
Samples of plasma, tissues, and cottonseed meal were
shipped on dry ice to the Texas A&M Agricultural Center
at San Angelo, TX, for gossypol analyses. HPLC proce-
dures were used to determine concentrations of (+) and (-)
gossypol in plasma and tissues [22], cottonseed meal [23],
and the gossypol from Sigma [23]. Free and total gossypol
concentrations in cottonseed meal were also determined by
the Official Methods of the Oil Chemists Society [24, 25].
Plasma concentrations of vitamins were analyzed for
HPLC. The procedure for ao-tocopherol has been published
[26]. Concentrations of -carotene, retinol, and retinyl pal-
mitate in plasma were analyzed as follows. Samples of
plasma (1 ml) were pipetted in a 16 x 125-mm glass tube,
deproteinized with 1 ml of ethanol and 0.5 ml of 25% (w:
v) ascorbic acid solution, vortexed, and then double-extract-
ed with 3 ml petroleum ether. The extract was dried by
evaporation under a stream of N2 in a 40°C water bath,
reconstituted in 1000 R.1 of a solution of 0.5% (v:v) acetic
acid, 29.5% (v:v) tetrahydrofuran, 69.8% (v:v) iso-octane,
and 0.2% (w:v) -hydroxytoluene. The sample was sepa-
rated in equal amounts into two sealed vials (one for 3-car-
otene and one for retinol and retinyl palmitate). The HPLC
system consisted of an ABI Spectroflow 400 (ABI Analyt-
ical, Ramsey, NJ), an ISS-100 auto sampler (Perkin-Elmer
Analytical Instruments, Norwalk, CT) with a 20-Rl loop,
and an EQC 10-pim I 60A SI, 4.6 mm x 250-mm, column
(Whatman International Ltd., Maidstone, UK). For 3-car-
otene, the mobile phase consisted of 95% (v:v) iso-octane,
4.5% (v:v) tetrahydrofuran, and 0.5% (v:v) acetic acid. The
ultraviolet detector was an ABI Analytical Spectroflow 757
(ABI Analytical) set at a wavelength of 450 nm and sen-
sitivity of 0.005 absorbance units full scale. Data were col-
lected by an LCI-100 Laboratory Computing Integrator
(Perkin-Elmer). Flow rate was 1 ml/min. The retention time
for -carotene was 3.57 min. Standard consisted of 50 ng
1-carotene. For retinol and retinyl palmitate, conditions
were similar except that the mobile phase consisted of 70%
(v:v) iso-octane, 29.5% (v:v) tetrahydrofuran, and 0.5% (v:
v) acetic acid; the ultraviolet detector was set at a wave-
length of 325 nm; standards consisted of 10 ng retinol and
10 ng of retinyl palmitate; and the retention time was 6.04
min for retinol and 3.32 min for retinyl palmitate. Identi-
fication and quantification of each molecule were done by
comparison of retention times and peakareas with stan-
dards. The procedure was validated with standard reference
material 968b, fat-soluble vitamins, and cholesterol in hu-
man serum (National Institute of Standards & Technology,
Gaithersburg, MD).
IVM/IVF/IVC. Within 5 min after slaughter, ovaries were
obtained and processing was started for oocyte collection.
Oocyte recovery, maturation, fertilization, and culture were
performed as described previously [17] except that, in this
experiment only, all oocytes were subjected to IVM/IVF/
IVC regardless of morphological appearance.
Effect of Gossypol on Oocyte Maturation
Ovaries were collected from an abattoir and used to col-
lect oocytes. Details of IVM/IVF/IVC were as reported
elsewhere [17] except that embryos were cultured in groups
of 25-30 per microdrop. Oocytes were matured in groups
of 10 in 50-pl microdrops of either oocyte maturation me-
dium (Tissue Culture Medium 199 supplemented with 10%
[v:v] bovine steer serum, 50 pig/ml gentamicin, 0.2 mM
sodium pyruvate, 2 pig/ml estradiol, and 20 pig/ml FSH),
or maturation medium containing 0.1% (v:v) ethanol (ve-
hicle) or vehicle plus 2.5, 5, or 10 Ipg/ml gossypol. After
maturation, oocytes were washed extensively to remove
gossypol and then subjected to IVF and IVC. The experi-
ment was replicated on four occasions with a total of 117-
151 oocytes per treatment.
Effect of Gossypol on Embryonic Development
Oocytes were collected from slaughterhouse ovaries,
matured, and fertilized according to standard procedures
[17] and then transferred in groups of 25-30 to 50-iil mi-
crodrops of CRlaa medium, CRlaa medium containing
0.1% (v:v) ethanol, or CRlaa medium containing 0.1% eth-
anol and 0, 2.5, 5, or 10 pg/ml gossypol. Heat-treated fetal
calf serum (5 1l) was added to each microdrop on Day 5
after insemination, and embryos were cultured until Day 9
after insemination. Cleavage rate was determined 7 days
after insemination and development to blastocyst stage at
Days 7-9. The experiment was replicated four times with
a total of 118-174 oocytes per treatment.
Statistical Analysis
Least-squares analysis of variance was performed using
the General Linear Models procedure of the Statistical
Analysis System [27]. Percentage data were subjected to
arcsin transformation before analysis. Cow was considered
a random effect, and other main effects were considered
fixed. Treatment effects were tested using the appropriate
error term after calculation of expected means squares. Or-
thogonal contrasts were performed when appropriate.
For IVM/IVF/IVC experiments, percentage oocytes that
cleaved and percentage cleaved embryos that developed to
the blastocyst stage were calculated within each replicate, and
these values were subjected to analysis of variance as de-
scribed above using main effects of treatment and replicate.
RESULTS
Effects of Gossypol on Sperm Swim-Up
At concentrations above 10 pxg/ml, gossypol caused a
concentration-dependent decrease in the percentage of
sperm successfully completing the swim-up procedure (Fig.
1). As compared to sperm without gossypol, incubating
sperm with 10 pIg/ml gossypol did not affect motility, but
both 50 and 100 pg/ml gossypol reduced the proportion of
sperm completing swim-up (0 and 10 jig/ml vs. 50 and 100
Rpg/ml gossypol, p < 0.02; 0 vs. 10 pxg/ml, p > 0.10; 50
vs. 100, p > 0.10). The deleterious effect of 100 Rpg/ml
gossypol on ability of sperm to complete the swim-up pro-
cedure was not blocked by the antioxidant GSH ester (Fig.
2). While gossypol reduced sperm completing swim-up (p
< 0.001) and the highest concentration of GSH ester tested
(1 mM) increased the percentage of sperm completing swim-
up (0 and 0.1 mM GSH ester vs. 1 mM GSH ester, p <
0.02; 0 vs. 0.1 mM GSH ester, p > 0.10), there was no
gossypol x GSH-ester interaction. Rather, deleterious ef-
fects of gossypol on swim-up were similar regardless of
concentration of GSH ester.
903
BROCAS ET AL.
14
0.
2 12
E
3' 10
c,
0. 6
, 4
E
CE 2
O,
0 10 50 100
Gossypol (g/ml)
FIG. 1. Effect of incubation of sperm with gossypol on percentage of
sperm completing swim-up purification. Data are least-squares means +
SEM of results from four separate replicates.
Effect of Gossypol on Fertilizing Ability of Spermatozoa
Cleavage rates of oocytes inseminated with gossypol-
treated spermatozoa were similar to those for oocytes in-
seminated with control spermatozoa (Table 3). At Day 7
after insemination, the percentage of embryos that were
blastocysts tended to be lowest for the 50 jig/ml group (10
vs. 50 jig/ml, p = 0.08). At Day 9 after insemination, how-
ever, differences among groups were nonsignificant.
Effects of Feeding Cottonseed Meal
Cows fed cottonseed meal had mean plasma concentrations
of gossypol of 3.9 + 0.6 and 4.2 + 0.4 jig/ml at Weeks 4
and 8 of feeding, respectively. At Week 4, 55% of the plasma
gossypol was the (+) isomer. At Week 8, 51% was the (+)
isomer. Gossypol concentrations were nondetectable in all
plasma samples from control cows at Weeks 0, 4, and 8 and
from cows fed cottonseed meal at Week 0.
Cows fed cottonseed meal had higher gossypol concen-
trations in the liver than control cows (145.1 5.6 vs. 2.7
+ 0.5 g/g dry matter; Table 4). Gossypol content was
measured in three samples of heart tissue: concentrations
were higher in two cows fed cottonseed meal (24.4 and
29.0 [jg/g dry matter) than in one cow fed the control diet
(1.6 jIg/g dry matter). In contrast with the liver, in which
the positive isomer predominated, the minus isomer was
the predominant form (74.8%) in the heart.
Cottonseed feeding caused an increase in erythrocyte
fragility. Analysis of hemolysis at Week 8 was performed
to determine differences between feeding groups at all
TABLE 3. Cleavage and development of embryos to the blastocyst stage
of development at Days 7 and 9 after IVF with gossypol-treated sperma-
tozoa. a
Gossypol (g/ml)
Item 0 10 50 SEM
Number of replicates 4 4 4
Number of oocytes 185 178 164
Oocytes cleaved (%) 76.5 80.0 76.8 3.0
Day 7 after insemination
Embryos to blastocyst (%)b 19.0 23.1 14.4 2.9
Oocytes to blastocyst (%)c 14.3 17.8 10.8 2.4
Day 9 after insemination
Embryos to blastocyst (%) 27.7 25.8 21.7 2.1
Oocytes to blastocyst (%) 20.8 20.2 16.0 1.7
a Data are least-squares means ± SEM of percentage values calculated
for each replicate.
b10 VS. 50 Ig/ml, p = 0.08.
10 vs. 50 txg/ml, p = 0.09.
30
' Control
m 25 -r Gossypol
E 0_
80
E5
0.0 0.1 1.0
Glutathione ester (mM)
FIG. 2. Effects of gossypol (100 pxg/ml) and glutathione monoethyl ester
(GSH ester) on percentage of sperm completing swim-up purification.
Data are least-squares means SEM of results from five separate repli-
cates.
NaCl concentrations tested (Fig. 3, top panel). Hemolysis
increased with decreasing NaCl concentration and was
greater for erythrocytes from cottonseed-fed cows at NaCl
concentrations of 0.45%, 0.55%, and 0.65% (treatment, p
< 0.01; NaCl concentration, p < 0.001; treatment x con-
centration, p < 0.01). When hemolysis at 0.55% NaCl was
examined across all weeks, erythrocytes from cows fed cot-
tonseed meal experienced greater hemolysis than erythro-
cytes from control cows beginning at 4-5 wk of feeding
(Fig. 3, bottom panel; treatment, p < 0.02; week of feeding,
p < 0.002; and treatment x week, p < 0.001).
Feeding cottonseed meal had no effect on plasma con-
centrations of a-tocopherol, retinyl palmitate, or retinol;
however, concentrations of -carotene were higher (p <
0.05) for cows fed cottonseed meal (Table 5). Concentra-
tions of 13-carotene and retinyl palmitate declined over the
course of the experiment (p < 0.001 for -carotene and p
< 0.01 for retinyl palmitate).
Results of IVM/IVF/IVC were excluded from one cot-
tonseed meal-fed cow and 1 control cow. Ovaries from
these cows were processed on the same day,and total fer-
tilization failure occurred for the oocytes from these cows.
Data from the remaining cows are summarized in Table 6.
There were no differences (p > 0.10) between cottonseed
meal-fed and control cows in numbers of oocytes collected
per cow, cleavage rate, or proportion of oocytes or embryos
that developed to the blastocyst stage.
Effects of Gossypol on Oocyte Maturation In Vitro
Addition of gossypol to maturation medium had no effect
(p > 0.10) on subsequent cleavage rate of oocytes (Table 7).
Similarly, subsequent development of cleaved embryos to the
blastocyst stage was not inhibited by gossypol at Day 7 or 9
after insemination (Table 7), and accordingly, the overall pro-
duction of blastocysts (percentage oocytes developing to blas-
tocyst) was not reduced by gossypol. In fact, the proportion
TABLE 4. Concentrations of gossypol in liver of cows fed cottonseed
meal.a
Item Control diet Cottonseed diet
Number of observations 4 4
Total gossypol (Lg/g dry matter) 2.7 + 0.5 145.1 + 5.6
(+) isomer (g/g dry matter) 2.7 ± 0.5 110.1 + 5.1
(-) isomer (g/g dry matter) NDb 35.0 ± 1.6
Percent (+) isomer 100 0 75.8 ± 0.01
a Data are means ± SEM; differences in concentrations of total gossypol,
(+) isomer, and (-) isomer were significant (p < 0.001).
b ND, nondetectable (considered as zero for statistical analysis).
904
ACTIONS OF GOSSYPOL ON BOVINE GAMETES AND EMBRYOS
_
E
IT:
NaCI (%, w/v)
F
5Zz
0
(U
U).021
E
0)
I
DU1 -
40 
30 
20 
2 3 4 5 6 7
Time (weeks)
TABLE 6. Effect of feeding cottonseed meal to oocyte-donor cows on in
vitro oocyte maturation, fertilization, and culture.a
Diet
Item Control Cottonseed SEM
Number of cows 5 5
Oocytes/cow 26.4 37.6 3.8
Oocytes cleaved (%) 73.1 71.3 1.1
Day 7 after insemination
Embryos to blastocyst (%) 25.5 23.0 7.4
Oocytes to blastocyst (%) 18.9 15.9 5.6
Day 9 after insemination
Embryos to blastocyst (%) 33.6 23.8 6.6
Oocytes to blastocyst (%) 24.4 16.3 4.9
aData are least-squares means + SEM of percentage values calculated on
a within-cow basis; all treatment effects were nonsignificant.
8 9 10
FIG. 3. Effect of feeding cottonseed meal on erythrocyte fragility. For
both graphs, closed circles represent control cows (n = 6) and open cir-
cles represent cows fed cottonseed meal (n = 6). The top panel shows
least-squares means SEM for hemolysis of erythrocytes collected at 8
wk after initiation of cottonseed meal feeding and exposed to 0.45 0.85%
(w:v) NaCI. The bottom panel shows least-squares means + SEM of he-
molysis of erythrocytes collected weekly from 3-9 wk after initiation of
feeding and exposed to 0.55% (w:v) NaCI.
of embryos giving rise to blastocysts at Day 9 after insemi-
nation tended to be greatest for the 2.5 pxg/ml group (0 vs.
2.5 pxg/ml, p = 0.09). The presence of ethanol in culture
medium caused a reduction in development of oocytes (p =
0.05) and cleaved embryos (p < 0.05) to the blastocyst stage
at Day 7 after insemination but not at Day 9.
Effects of Gossypol during Embryonic Development
Addition of 10 $pg/ml gossypol after fertilization caused
a reduction in cleavage rate (0 and 2.5 vs. 5 and 10 ,ug/ml,
p < 0.01; 5 vs. 10 ug/ml, p < 0.01; Table 8). Subsequent
development of embryos that did cleave also was reduced
by gossypol. At Day 7 after insemination, the proportion
of embryos that developed to the blastocyst stage of de-
TABLE 5. Effect of feeding cottonseed meal on concentrations of vita-
mins in plasma.a
Week of feeding
Vitamin (g/ml) Diet 0 4 8 SEM
a-tocopherol Control 4.1 4.3 4.2 0.40
Cottonseed 4.7 5.6 5.9 0.43
P-caroteneb Control 2.7 1.3 1.4 0.34
Cottonseed 4.0 1.9 2.1 0.34
Retinyl palmitatec Control 0.80 0.47 0.50 0.089
Cottonseed 1.02 0.58 0.61 0.089
Retinol Control 0.45 0.40 0.40 0.028
Cottonseed 0.39 0.36 0.43 0.028
a Data are least-squares means SEM; times are relative to initiation of
cottonseed meal feeding.
b Effects of treatment (p < 0.05) and week (p < 0.001).
c Effect of week (p < 0.01).
velopment was reduced by 5 and 10 lug/ml gossypol (0 and
2.5 pug/ml vs. 5 and 10 pg/ml, p < 0.001). At 10 pxg/ml
gossypol, no blastocysts were present at Day 7. Results
were similar at Day 9 after insemination except that even
the lowest concentration of gossypol tested (2.5 jpg/ml)
tended to reduce development to the blastocyst stage (0 and
2.5 pg/ml vs. 5 and 10 pg/ml, p < 0.001; 0 vs. 2.5 pLg/ml,
p = 0.08). Similar to what was observed at Day 7, no
blastocysts were seen at Day 9 after insemination when
embryos were cultured with 10 pag/ml gossypol. It was not-
ed that most embryos cultured with 5 and 10 ,ug/ml gos-
sypol did not develop past the 8- to 16-cell stage. The over-
all production of blastocysts (percentage oocytes develop-
ing to blastocyst) was also reduced by gossypol in a manner
similar to that for percentage development of cleaved em-
bryos to blastocyst.
DISCUSSION
The actions of gossypol on spermatozoa, oocytes, and
embryos were tested with several concentrations that fall
within the range of values found in blood plasma and tis-
sues of cows fed cottonseed meal. Concentrations in plasma
averaged 3.9-4.2 g/ml. It was not possible to measure
concentrations of gossypol in uterine fluid or oviductal fluid
because of the unknown degree of dilution associated with
collecting these fluids by lavage. The gossypol content in
the heart was slightly higher than in plasma (24.4-29.0
Vxg/g dry matter or 4.9-5.8 pxg/ml if one assumes a water
TABLE 7. Cleavage and development of embryos to the blastocyst stage
of development at Days 7 and 9 after IVF for oocytes treated with gossypol
during maturation.a
Gossypol (g/ml)
Item Control Ob 2.5 5 10 SEM
Number of replicates 4 4 4 4 4
Number of oocytes 151 119 117 126 151
Oocytes cleaved (%) 70.5 67.1 60.4 61.2 63.2 5.0
Day 7 after insemination
Embryos to blastocyst (%)c 31.0 10.4 21.5 15.8 11.0 6.1
Oocytes to blastocyst (%)d 22.7 7.0 13.8 9.7 7.3 4.5
Day 9 after insemination
Embryos to blastocyst (%)e 40.4 28.7 51.9 31.9 24.6 8.0
Oocytes to blastocyst (%) 29.8 20.3 29.4 19.1 16.6 5.7
aData are least-squares means + SEM of percentage values calculated
for each replicate.
b Control is without ethanol; 0-10 ig/ml contained 0.1 % ethanol.
cControl vs. 0 tg/ml, p < 0.05.
d Control vs. 0 )ig/ml, p = 0.05.
e0 vs. 2.5 pxg/ml, p = 0.09.
1U 
n I I I I I I I
905
BROCAS ET AL.
TABLE 8. Cleavage and development of embryos cultured with gossy-
po 01.
Gossypol (g/ml)
Item Control Ob 2.5 5 10 SEM
Number of replicates 4 4 4 4 4
Number of oocytes 151 118 166 174 121
Oocytes cleaved (%)' 70.5 66.0 61.9 59.7 38.0 4.3
Day 7 after insemination
Embryos to blastocyst(%)d 31.0 28.5 21.5 3.7 0.0 4.2
Oocytes to blastocyst (%)d 22.7 18.7 14.0 2.6 0.0 3.1
Day 9 after insemination
Embryos to blastocyst (/o 40.4 44.9 29.8 6.1 0.0 6.1
Oocytes to blastocyst (%)' 29.8 28.9 19.4 4.3 0.0 4.3
Data are least-squares means + SEM of percentage values calculated
for each replicate.
h Control is without ethanol; 0-10 jig/ml contained 0.1% ethanol.
0 + 2.5 vs. 5 + 10 iLg/ml, p < 0.01; 5 vs. 10 ILg/ml, p < 0.01.
d 0 + 2.5 vs. 5 + 10, p < 0.001.
e0 + 2.5 vs. 5 + 10, p < 0.001; 0 vs. 2.5, p = 0.08.
0 + 2.5 vs. 5 + 10, p < 0.001; 0 vs. 2.5, p = 0.10.
content of 80%), and gossypol was greatly elevated in the
liver (average of 145.1 jig/g dry matter or -29.0 pig/ml).
Spermatozoa and oocytes were resistant to deleterious ef-
fects of gossypol at concentrations within the range of val-
ues found in plasma and heart tissue. However, in vitro
embryonic development was disrupted by exposure to con-
centrations of 5.0 jig/ml gossypol, and there was some ev-
idence that gossypol concentrations as low as 2.5 txg/ml
were disruptive. Thus, feeding of diets containing similar
amounts of gossypol may disrupt fertility of cattle by in-
hibiting earlyembryonic development.
High concentrations of gossypol inhibit embryonic devel-
opment and block establishment of pregnancy in rodents [1].
Zirkle et al. [9] reported that development of bovine morulae
to blastocysts could be inhibited by gossypol, but the lowest
concentration tested (12.5 g/ml) was higher than concentra-
tions experienced by cows fed cottonseed meal. In the present
study, the actions of gossypol on embryonic development
were occurring within the range of physiologically relevant
values. Moreover, effects of gossypol became evident early in
development because, at high concentrations, embryos failed
to develop beyond the 8- to 16-cell stage. This stage of de-
velopment corresponds to a major period of embryonic ge-
nome activation [28]. Embryos are prone to experiencing a
block to development at this stage [29, 30] in part because of
free radicals [31]. Since one action of gossypol is to increase
free radical production [10-12, 32], the mechanism by which
gossypol inhibits embryonic development may involve in-
creased production of free radicals.
The severe effects of gossypol on embryonic develop-
ment observed in the present study are in contrast to results
from several reports indicating that feeding diets containing
gossypol had no effect on fertility of beef cows [33, 34].
These studies were not definitive, since few animals were
used (n = 6-17 per group). Given the magnitude of adverse
effects of gossypol on embryonic development observed in
the present study, further studies on gossypol feeding and
bovine fertility are warranted. It may be that effects of gos-
sypol on embryonic development are less in vivo than for
cultured embryos if gossypol does not enter the oviductal
or uterine lumen in high enough concentration to be inhib-
itory to embryonic development. Also, molecules in the
oviductal and uterine environment may reduce gossypol
toxicity. Bovine oviductal epithelial cells, for example, se-
crete the antioxidant amino acids hypotaurine and taurine
[35]. Taurine concentrations in human uterine fluid are up
to 400-fold greater than in plasma [36]. The observation
that gossypol feeding did not reduce plasma concentrations
of a-tocopherol, 3-carotene, retinyl palmitate, or retinol in
this study suggests that certain animal antioxidant systems
remain functional in the face of gossypol toxicity.
The decline in plasma concentrations of B-carotene, re-
tinyl palmitate, and retinol over time likely reflects the de-
crease in dietary 3-carotene intake after cows were moved
from pasture at the start of the experiment. The higher plas-
ma concentrations of 3-carotene seen in cows fed cotton-
seed meal, and the lack of effect of the feeding regimen on
ot-tocopherol concentrations, are in contrast to an earlier
study [34] in which beef cows consuming cottonseed meal
had lower serum concentrations of 3-carotene and ot-toco-
pherol than cows fed a diet in which soybean meal replaced
cottonseed meal. In a more recent study [37], feeding cot-
tonseed products to lactating dairy cows did not affect vi-
tamin A concentrations in plasma and increased concentra-
tions of vitamin E.
The lowest tested concentration of gossypol that inhib-
ited sperm motility as determined by the swim-up proce-
dure was 50 jIg/ml. Gossypol did not affect fertilizing abil-
ity of sperm even at 50 jig/ml. Possibly, effects of gossypol
on fertilizing ability were obscured by the high sperm:oo-
cyte ratio used in IVE In other studies, 9.3-10 jig/ml gos-
sypol caused a slight reduction in motility of ram [6] and
human sperm [7]. Differences between these studies and
the present ones may reflect differences between species or
methods for assessing motility. In any case, a concentration
of 10 g/ml is considerably higher than gossypol concen-
trations in the blood of cows fed cottonseed meal.
The inhibitory action of 100 [tg/ml gossypol on sperm
swim-up was not blocked by the antioxidant GSH ester. Glu-
tathione is a cytosolic antioxidant. The ester of GSH was used
because it more readily crosses cell membranes than does
GSH [38]. Administration of gossypol to rats caused lipid
peroxidation without affecting tissue GSH concentrations
[32], suggesting that gossypol is acting on membranes rather
than in the cytosol. Another possibility is that effects of gos-
sypol on sperm motility involve some actions independent of
damage involving free radicals. For example, gossypol can
inactivate certain spermatozoal enzymes [39, 40] and cause
structural damage to the sperm acrosome [3].
Glutathione ester increased sperm motility in a manner
independent of gossypol. Similar results have been reported
elsewhere [14, 15]. A recent report indicates that addition
of GSH during sperm purification enhanced the cleavage
rate when sperm were used to fertilize IVM oocytes [41].
Feeding cows a diet high in gossypol did not cause any
permanent defect in oocytes recovered for IVM/IVF/IVC.
The feeding regimen used delivered about 7 g/head per day
of free gossypol to each cow. This amount was clearly
enough to cause toxicological problems as witnessed by an
increase in erythrocyte fragility. However, feeding gossypol
had no significant effect on any measure of oocyte perfor-
mance in IVE There was a tendency for the proportion of
embryos developing to blastocysts at Day 9 after insemi-
nation to be lower for cows fed cottonseed meal, but further
studies are warranted to show whether this effect is real.
However, gossypol was also without deleterious effect on
oocytes that were matured in vitro in the presence of gos-
sypol. In fact, gossypol tended to be beneficial when added
at 2.5 jig/ml. Taken together, the present evidence indicates
that the oocyte is very resistant to gossypol. In contrast, it
was found in an earlier study [8] that 10 jig/ml gossypol
906
ACTIONS OF GOSSYPOL ON BOVINE GAMETES AND EMBRYOS 907 
inhibited the proportion of oocytes completing cumulus ex- 
pansion and nuclear maturation. 
The oocytes used for IVMAVFAVC were from tertiary 
follicles at various stages of growth and atresia. The pos- 
sibility remains that gossypol can affect oocytes from sec- 
ondary or primary follicles or exert specific effects on the 
oocyte in the follicle destined for ovulation. Additionally, 
cows were fed diets containing protein in excess of require- 
ments for nonlactating cows, -and it is possible that results 
would have been different if protein content of the diet had 
been lower. It is also possible that removing oocytes from 
their local microenvironment in the follicle caused a rever- 
sal of adverse effects of gossypol. 
In conclusion, male and female bovine gametes were 
resistant to deleterious effects of gossypol at concentrations 
similar to those in animals fed diets containing cottonseed 
meal. The developing embryo, in contrast, was very sen- 
sitive to gossypol. Thus, further experiments with large 
numbers of cattle are warranted to determine whether feed- 
ing diets high in gossypol leads to infertility. 
A C K N O W L E D G M E N T S 
Authors thank Mary Russell and the Dairy Research Unit farm crew 
for help in feeding animals; Carlos Arkchiga, who synthesized GSH ester; 
and Tommy Bryan, owner of Central Packing Co., Center Hill, FL, and 
his employees for their kindness and assistance with collection of ovaries. 
REFERENCES 
1. Randel RD, Chase CC Jr, Wyse SJ. Effects of gossypol and cottonseed 
products on reproduction of mammals. J h i m Sci 1992; 70: 1628-1638. 
2. Chase CC Jr, Bastidas P, Ruttle JL, Long CR, Randel RD. Growth 
and reproductive development in bulls fed diets containing gossypol. 
J Anim Sci 1994; 72:445-452. 
3. Chenoweth PJ, Risco CA, Larsen RE, Velez J, Tran T, Chase CC Jr. 
Effects of dietary gossypol on aspects of semen quality, sperm mor- 
phology and sperm production in young Brahman bulls. Theriogen- 
ology1994; 42: 1-13. 
4. Breitbart H, Rubinstein S, Nass-Arden L. Effect of gossypol-acetic 
acid on calcium transport and ATPase activity in plasma membranes 
from ram and bull spermatozoa. Int J Androl 1984; 7:439-447. 
5. Rovan E, Kalla NR, Frick J, Adam H. Effect of gossypol on bull 
spermatozoa in vitro. Urol Res 1984; 12:187-192. 
6. Breitbart H, Mayevsky A, Nass-Arden L. Molecular mechanisms of gos- 
sypol action on sperm motility. Int J Biochem 1989; 21:1097-1102. 
7. Cowart CL, London SN, Vernon MW, Pedigo NG. The effects of 
cyclic adenosine monophosphate, forskolin, and theophylline on mo- 
tility parameters in gossypol-treated human sperm. Fertil Steril 1994; 
61 :929-934. 
8. Lin YC, Coskun S, Sanbuissho A. Effects of gossypol on in vitro 
bovine oocyte maturation and steroidogenesis in bovine granulosa 
cells. Theriogenology 1994; 41:1601-1611. 
9. Zirkle SM, Lin YC, Gwazdauskas FC, Canseco RS. Effect of gossypol 
on bovine embryo development during the preimplantation period. 
Theriogenology 1988; 30:575-582. 
10. de Peyster A, Quintanilha A, Packer L, Smith MT. Oxygen radical 
formation induced by gossypol in rat liver microsomes and human 
sperm. Biochem Biophys Res Commun 1984; 118573-579. 
11. Kanwar, U, Kaur R, Chadha S, Sanyal S. Gossypol-induced inhibition 
of glucose uptake in human ejaculated spermatozoa may be mediated 
by lipid peroxidation. Contraception 1990; 42:573-587. 
12. Fornes MW, Barbieri AM, Burgos MH. Sperm motility loss induced 
by gossypol: relation with OH scavengers, motile stimulators and ma- 
londialdehyde production. Biochem Biophys Res Commun 1993; 195: 
1289-1293. 
13. Brezezinska-Slebodzinska E, Slebodinski AB, Pietras B, Wieczorek 
G. Antioxidant effect of vitamin E and glutathione on lipid peroxi- 
dation in boar semen. Biol Trace Elem Res 1995; 47:69-74. 
14. Lindemann CB, O'Brien JA, Giblin FJ. An investigation of the effec- 
tiveness of certain antioxidants in preserving the motility of reacti- 
vated bull sperm models. Biol Reprod 1988; 38:114-120. 
15. Monterroso VH, Drury KC, Ealy AD. Howell JD, Hansen PJ. Effect 
of heat shock on function of frozenithawed bull spermatozoa. Theri- 
ogenology 1995; 44:947-961. 
16. Arkchiga CF, Ealy AD, Hansen PJ. Evidence that glutathione is in- 
volved in thermotolerance of preimplantation mouse embryos. Biol 
Reprod 1995; 52: 1296-1301. 
17. Edwards JL, Hansen PJ. Elevated temperature increases heat shock 
protein 70 in bovine two-cell embryos and compromises function of 
maturing oocytes. Biol Reprod 1996; 55:340-346. 
18. Pamsh JJ, Susko-Parrish JL, Critser ES, Eyestone WH, First NL. Bo- 
vine in vitro fertilization with frozen-thawed semen. Theriogenology 
1986; 25:59 1-600. 
19. Rosenkrans CF Jr, First NL. Effect of free amino acids and vitamins 
on cleavage and developmental rate of bovine zygotes in vitro. J Anim 
Sci 1994; 72:434-437. 
20. Hernandez-Ledezma JJ, Villanueva C, Sikes JD, Roberts RM. Effects 
of CZB versus medium 199 and of conditioning culture media with 
either bovine oviductal epithelial cells or buffalo rat liver cells on the 
development of bovine zygotes derived by in vitro maturation-in vitro 
fertilization procedures. Theriogenology 1993; 39: 1267-1277. 
21. Fairley NM, Price GS, Meuten DJ. Evaluation of red blood cell fra- 
gility in Pygmy goats. Am J Vet Res 1988; 49:1598-1600. 
22. Kim HL, Calhoun MC. Determination of gossypol in plasma and tis- 
sues of animals. Inform 1995; 6:486 (abstr.). 
23. Hron RJ, Kim HL, Calhoun MC. Determination of gossypol in cot- 
tonseed products. Inform 1995; 6:485 (abstr.). 
24. American Oil Chemists Society. Determination of free gossypol. Of- 
ficial Method Ba 7-58. In: Official and Tentative Methods of Analysis, 
3rd ed. Chicago: AOCS; 1985. 
25. American Oil Chemists Society. Determination of total gossypol. Of- 
ficial Method Ba 8-58. In: Official and Tentative Methods of Analysis, 
3rd ed. Chicago: AOCS; 1985. 
26. Njeru CA, McDowell LR, Wilkinson, NS, Linda SB, Williams SN, 
Lentz EL. Serum a-tocopherol concentration in sheep after intramus- 
cular injection of DL-a-tocopherol. J Anim Sci 1992; 70:2562-2567. 
27. SAS. SAS User's Guide. Version 6, 4th edition. Cary, NC: Statistical 
Analysis System Institute, Inc.: 1989. 
28. Barnes FL, First NL. Embryonic transcription in in vitro cultured bo- 
vine embryos. Mol Reprod Dev 1991; 29: 1 17-123. 
29. Eyestone WH, First NL. Co-culture of early cattle embryos to the 
blastocyst stage with oviducal tissue or in conditioned medium. J Re- 
prod Fertil 1989; 82:7 15-720. 
30. Larson RR, Ignotz GG, Cunie WB. Transforming growth factor P and 
basic fibroblast growth factor synergistically promote early bovine 
embryo development during the fourth cell cycle. Mol Reprod Dev 
1992; 33:432-435. 
31. Liu Z, Foote RH. Development of bovine embryos in KSOM with 
added superoxide dismutase and taurine and with five and twenty 
percent 02 . Biol Reprod 1995; 53:786-790. 
32. Sheriff DS. Effect of gossypol on tissue levels of glutathione and 
thiobarbituric acid reactive products in rats. Ind J Exp Biol 1992; 30: 
734-735. 
33. Gray ML, Greene LW, Williams GL. Effects of dietary gossypol con- 
sumption on metabolic homeostasis and reproductive endocrine func- 
tion in beef heifers and cows. J Anim Sci 1993; 71:3052-3059. 
34. Willard ST, Neuendorff DA, Lewis AW, Randel RD. Effects of free gos- 
sypol in the diet of pregnant and postpartum Brahman cows on calf 
development and cow performance. J Anim Sci 1995; 73:49&507. 
35. Guerin P, Menezo Y. Hypotaurine and taurine in gamete and embryo 
environments: de novo synthesis via the cysteine sulfinic acid pathway 
in oviduct cells. Zygote 1995; 3:333-343. 
36. CasslCn BG. Free amino acids in human uterine fluid: possible role 
of high taurine concentration. J Reprod Med 1987; 32:181-184. 
37. Mena H, Huber JT, Santos J, Aquino J, Tarazon M, Nussio L. The 
effect of whole cottonseed and cottonseed meal supplementation on 
lactating dairy cows. J Dairy Sci 1996; 79(suppl ]):I81 (abstc). 
38. Anderson ME, Meister A. Glutathione monoesters. Anal Biochem 
1989; 183: 16-20. 
39. Nakamura M, Ikeda M, Komukai M, Suyemitsu T, Okinaga S, Arai K. 
Presence of a 16 kd protein in human seminal plasma counteracts the 
effects of the anti-fertility agent, gossypol. Hum Reprod 1991; 6:714- 
721. 
40. Yuan YY, Shi QX, Srivastava PN. Inhibition of rabbit sperm acroso- 
ma1 enzymes by gossypol. Mol Reprod Dev 1995; 40~228-232. 
41. Earl CR, Kelly J, Rowe J, Armstrong DT Glutathione treatment of 
bovine sperm enhances in vitro blastocyst production rates. Theri- 
ogenology 1997; 47:255 (abstr.).