| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From the Centre for Advanced Technologies in Animal Genetics and Reproduction (ReproGen), Faculty of Veterinary Science, The University of Sydney, Australia.
| Correspondence to: Chis Maxwell, Faculty of Veterinary Science, The University of Sydney, NSW 2006, Australia (e-mail: chism{at}vetsci.usyd.edu.au). |
| Received for publication June 25, 2006; accepted for publication August 21, 2006. |
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
The present study had 3 objectives. The first was to separate membrane vesicles from ram seminal plasma and compare their protein content with the remaining supernatant fraction and with whole seminal plasma. The second aim was to determine the effect of seminal plasma and its 2 fractions on the motility characteristics, membrane status, and in vitro fertility of frozen-thawed ram spermatozoa. The third aim was to assess the value of sperm motility characteristics for predicting in vitro fertilization (IVF) rate.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Experimental Animals, Housing, and Locations— Procedures described herein were approved by The University of Sydney's Animal Ethics Committee. For experiments 1 and 2, semen was collected from 8 rams: 1 Suffolk x Merino cross, 4 Merino, and 3 Awassi x Merino cross. Four of the 8 rams (2 Merino and 2 Awassi x Merino cross) were vasectomized after sedation with diazepam (2 mL Pamlin IM; 4 mg/kg diazepam; Parnell Laboratories Pty Ltd, Alexandria, Australia) and ketamine (1 mL Ketamil IM; Parnell) and under local anaesthesia (10 mL lignocaine SC; Troy Laboratories Pty Ltd, Smithfield, Australia). The rams were restrained in a laparotomy cradle and the vas deferens palpated through the inguinal canal. The vasa deferentia were exteriorized through a small incision and a 2–4 cm length removed using surgical scissors. The remaining 4 rams were not vasectomized and acted as controls (entire rams).
For experiment 3, semen was collected from 3 entire Merino rams. The rams were kept either on natural pasture with supplementary feeding of hay and grain at the Sheep Unit, Faculty of Veterinary Science, Camden, Australia, or on a fully prepared chaff-based ration (oaten chaff–lucerne chaff, 1:1) with supplementary lupin grain in an animal house at the Faculty of Veterinary Science, Camperdown, Australia.
Ovine ovaries collected at 2 abattoirs (located at Picton and Goulburn, Australia) from Merino sheep or crossbred lambs were used for recovery of oocytes for in vitro maturation (IVM) and fertilization.
Semen Collection and Freezing— For experiments 1 and 2, semen was collected every second day or twice a week from the 8 rams by artificial vagina (Evans and Maxwell, 1987) before and after vasectomy. Ejaculates collected before vasectomy were washed 2 times with Salamon buffer (Tris [hydroxymethyl] aminomethane [300 mM], glucose [30 mM], and citric acid [monohydrate] [94.7 mM]) after centrifugation (200 x g, 38°C) for 10 minutes on a Spintron GT-15FR centrifuge (Spintron Pty/Ltd, Sydney, Australia) to precipitate spermatozoa and eliminate seminal plasma. Washed spermatozoa were diluted (1:4, semen-diluent) with Salamon buffer containing egg yolk (15% vol/vol), glycerol (5% vol/vol), penicillin (100 000 IU), and streptomycin (73 500 IU) and frozen in pellet form before storage in liquid nitrogen as described by Evans and Maxwell (1987).
For experiment 3, semen was collected from the 3 entire rams by artificial vagina. The semen was washed twice with Salamon buffer and centrifuged each time (600 x g, 30°C, 10 minutes) to concentrate the spermatozoa. The concentrated spermatozoa were diluted (1:1, spermatozoa-diluent) with Salamon buffer supplemented with egg yolk, glycerol, and antibiotics and frozen-stored as for experiments 1 and 2.
Fractionation of Seminal Plasma— For experiments 1 and 2, 12 ejaculates collected from each ram before vasectomy and 8 ejaculates after vasectomy were pooled within ram and used for seminal plasma separation. For experiment 3, seminal plasma was prepared by the same methods using the entire rams. Thus, autologous plasma and its fractions (pooled from several ejaculates within male added to spermatozoa from the same male) were used throughout this study.
Whole seminal plasma. Semen was centrifuged (2000 x g, 4°C) for 20 minutes on a Spintron GT-15FR centrifuge. Pellets of spermatozoa were discarded, and supernatants were centrifuged again (2500 x g, 4°C, 30 minutes) to eliminate any remaining spermatozoa and cell debris. Supernatants were measured using a 1-mL measuring disposable pipette (Interpath Services Pty Ltd, Sydney, Australia) and stored in Eppendorf tubes (Interpath) in a –20°C freezer until use.
Supernatant and vesicle fractions of seminal plasma. For the 3 experiments, aliquots of 650 µL thawed whole seminal plasma were centrifuged (100 000 x g, 4°C, 80 minutes) on a Beckman ultracentrifuge using special adaptors for such small volumes in a 40.1 Ti swing head rotor (Beckman, Sydney, Australia). Pellets of vesicles were washed 2 times with Salamon buffer and then resuspended to 650 µL with the same buffer for storage in Eppendorf tubes (–20°C).
Experimental Procedures and Design
Experiment 1: Ejaculate Characteristics and Protein Concentration and SDS-PAGE Patterns of Seminal Plasma and Its Fractions—
The motility and concentration of spermatozoa were assessed for all
ejaculates collected from the 8 rams before and after vasectomy. Subjective
assessment of forward progressive motility of spermatozoa was determined
visually by placing 10 µL of semen on a clean, warm (37°C) slide held
on a warm stage under 200x magnification, and concentration of
spermatozoa was determined by duplicate hemocytometer counts as described by
Evans and Maxwell (1987).
Aliquots from whole seminal plasma, supernatant, and pellet of vesicles (1.5 mL from each of the 8 rams before and after vasectomy; n = 18 samples) were thawed and tested for their protein concentration using the BIO-RAD reagent. The samples were also analyzed by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) for one-dimensional (1D) protein patterns. For the latter analysis, 10 mL of 15% resolving gel (milliQ water, 1.5 M Tris pH 8.8, 30% T/2.67% C acrylamide, 10% SDS, 10% APS, and TEMED) was prepared, and protein separation according to molecular weight was performed on a Mini-protean II apparatus (8-cm width and 7-cm length; 1–2 per run; BIO-RAD). Twelve-microliter aliquots of whole seminal plasma, supernatant, and pellet of vesicles were prepared in 1x loading buffer, adjusted to a standard amount of protein (20 µg), heated at 95°C for 5 minutes, centrifuged briefly, and loaded onto the gel. Each gel was used for samples from 2 rams. Protein marker (6 µL; PageRuler Protein Ladder; Progen Industries Ltd, Darra, Australia) was loaded into the first column (A). Whole seminal plasma, supernatant, and pellet of vesicles were loaded, respectively, into columns B, C, and D (first ram) or E, F, and G (second ram). Electrophoresis was carried out at 200 V for 45 minutes. The gels were then treated for 1 hour with 50% methanol, stained overnight in 0.125% Coomassie (brilliant) blue R-250, 25% methanol, and 10% acetic acid and destained in 25% methanol and 10% acetic acid until the background became clear. The gel was scanned on an Image Scanner (Amersham Biosciences, Castle Hill, Australia) using LabScan v5.0 (Amersham).
Experiment 2: Motility Characteristics and Membrane Status of Spermatozoa— Addition of seminal plasma fractions to frozen-thawed semen. Frozen-thawed (dry test tubes, 37°C water bath) spermatozoa from each ram were resuspended in 12 different tubes by the addition of aliquots of whole seminal plasma, supernatant, and pellet of vesicles from the same ram. The volume of each aliquot was adjusted to provide the same concentration of protein to contribute the equivalent of 20% whole seminal plasma in the final volume of resuspended frozen-thawed spermatozoa. The volumes in each tube were then adjusted to 500 µL with Salamon buffer to provide a concentration of 20 x 106 spermatozoa per milliliter. The latter sperm concentration is the minimum for accurate assessment of sperm motility characteristics by computer-assisted sperm analysis (CASA). In addition to spermatozoa, 10 of the tubes contained whole seminal plasma, supernatant, pellet, 2x pellet volume, or 3x pellet volume from the samples collected before vasectomy or after vasectomy. In 2 additional tubes, spermatozoa were resuspended with Salamon buffer (control) and Salamon buffer containing BSA (fraction V) at the same protein concentration as the whole seminal plasma (protein control). The resuspended spermatozoa were then incubated in a water bath at 37°C for 6 hours.
Assessment of sperm motility characteristics— The motility of frozen-thawed spermatozoa was assessed subjectively (SUBJ) and by CASA at 0, 3, and 6 hours after thawing and resuspension. Subjective assessment of forward progressive motility of spermatozoa was determined visually as described earlier.
Spermatozoa were assessed by CASA (HTM-IVOS version 12; Hamilton Thorne, Beverly, Mass) for total motility (TOT), progressive motility (PROG), average path velocity (VAP), average straight line velocity (VSL), average track speed (VCL), amplitude of lateral head displacement (AHL), beat cross frequency (BCF), straightness (STR), linearity (LIN), and elongation (ELONG). Resuspended sperm samples (5.5 µL) were placed on slides (Cell Vu, Millennium Sciences Inc, New York, NY; prewarmed to 37°C) and enclosed using a 22 x 22 mm coverslip before immediate transfer to the CASA. Motility characteristics were determined by assessment of at least 3 randomly selected microscopic fields (more than 300 spermatozoa per sample) using factory CASA settings (ram) at an image sampling frequency of 60 Hz.
Assessment of membrane status of spermatozoa— The membrane status of frozen-thawed spermatozoa was also assessed at 0, 3, and 6 hours after thawing and resuspension using the CTC staining method as described by Gillan et al (1997). Briefly, 45 µL of resuspended spermatozoa was added to 45 µL of CTC solution and mixed well. Ten microliters of glutaraldehyde solution was added with mixing and centrifuged for 30 seconds to precipitate spermatozoa. A 5 µL drop of DABCO was placed on a microscope slide and mixed with 5 µL of the fixed spermatozoa to retard fading of fluorescence. A coverslip was placed on the suspension, excess fluid removed by compression, and the edges of the coverslip sealed with colorless nail varnish. A total of 200 spermatozoa were classified according to CTC patterns using the nomenclature described by Gillan et al (1997). The 3 patterns were F (characteristic of uncapacitated, acrosome-intact cells), B (characteristic of capacitated, acrosome-intact cells), and AR (characteristic of capacitated, acrosome-reacted cells).
Experiment 3: In Vitro Fertility of Spermatozoa— Oocyte preparation. This IVF experiment was replicated 3 times. For replicates 1, 2, and 3, respectively, 500, 680, and 276 oocytes were aspirated from follicles measuring 2–4 mm in diameter from 200, 270, and 110 ovaries using a 2.5-mL syringe and a 21-g needle. Oocytes with homogenous cytoplasm and surrounded by 3–4 layers of cumulus cells, 344, 380, and 221, respectively, for the 3 replicates, were selected and washed twice in H199+ (Medium 199 [25 mM HEPES; Sigma-Aldrich] supplemented with 10 µg/mL heparin [Sigma-Aldrich], 125 IU/mL penicillin [Sigma-Aldrich], 50 µg/mL streptomycin [Sigma-Aldrich], and 2% [vol/vol] heat-inactivated sheep serum [H199+; Trace Biosciences, Victoria, Australia]) and twice in IVM wash (Cook IVF, Brisbane, Australia) before equilibration. The oocytes were transferred into maturation medium consisting of Medium 199 (with Earles salts, l-glutamine, 2.2 g/L sodium bicarbonate; Gibco-BRL, Grand Island, NY) supplemented with 20% (vol/vol) heat-inactivated sheep serum (Trace), 10 µg/mL porcine follicle stimulating hormone (p-FSH) (Folltropin-V; Bioniche Australasia, Armidale, Australia), and 10 µg/mL porcine luteinizing hormone (p-LH, Lutropin-V; Bioniche) under oil (Sydney IVF culture oil; Cook IVF) and cultured at 38.6°C for 22–24 hours in 5% O2, 5% CO2, and 90% N2 in a MINC incubator (K-MINC; Cook IVF). After maturation, the oocytes were washed 3 times with H199+ until the excess of cumulus cells had been removed and then twice with IVF complete medium. Fifteen minutes before insemination, the oocytes were placed into 4-well dishes (Nalge Nunc, Naperville, Ill) containing IVF complete medium (Cook IVF) supplemented with 2% heat-inactivated sheep serum and equilibrated under mineral oil. Each well contained the total number of aspirated oocytes divided by 9 (number of treatments in each tube x number of rams).
Preparation of spermatozoa— Frozen-thawed spermatozoa from the 3 entire rams were resuspended in seminal plasma fractions using the same methods described for experiment 2, except that only the supernatant and pellet of vesicle fractions were used. Briefly, the frozen spermatozoa were thawed quickly in dry test tubes (37°C water bath) and resuspended to a concentration of 20 x 106/mL in a final volume of 500 µL. The spermatozoa were resuspended in Salamon buffer (control), supernatant (20% of final volume adjusted to the mean protein concentration for the 3 rams), and pellet of vesicles (3 times the volume of supernatant added in the second tube). The motility of frozen-thawed spermatozoa was assessed subjectively and by CASA at 0 and 3 hours after thawing and resuspension as described for experiment 2. After 3 hours of incubation, all the tubes were centrifuged (750 x g, 38°C, 10 minutes) to precipitate the spermatozoa.
Insemination. After centrifugation, 390 µL of supernatant were removed with a micropipette and discarded. The sperm pellet was resuspended in the remaining 100 µL medium. Different volumes containing 1 x 106 motile spermatozoa per milliliter were taken from each tube for in vitro insemination of equilibrated oocytes. Oocytes and spermatozoa were cocultured in 4-well dishes for 2, 6, and 18 hours.
Oocyte/zygote fixation— At 2, 6, and 18 hours after coincubation, 10–15 oocytes were removed from each treatment, transferred to clean medium, and washed to remove excess spermatozoa. Oocytes were then fixed with ethanol (60%), glacial acetic acid (30%), and chloroform (10%) for 22–24 hours and then stained with 0.5% orcein. The fixed and stained oocytes (315, 341, and 198 for the 3 replicates) were observed under a phase contrast microscope at 600x magnification for evidence of penetration and fertilization. Oocytes were classified as immature unfertilized (germinal vesicles and Metaphase I), mature unfertilized (Metaphase II + 1 polar body), mature fertilized (Metaphase II + penetrated spermatozoon, Anaphase II / Telophase II + penetrated spermatozoon, Telophase II + decondensed spermatozoon, 2 pronuclei forming + tail, 2 pronuclei + tail, and Syngamy + tail) or in an advanced stage (Cleavage or 4 blastomeres to a morula stage).
Statistical Analysis
Experiment 1—
Ejaculate characteristics and protein concentration. A combined
analysis of all the data using analysis of variance (ANOVA) was used to assess
differences among the 3 samples used (Sample: whole seminal plasma,
supernatant and pellets of vesicles) as well as compare vasectomized and
control rams (Treatment). Ram was specified as a blocking term. The model
fitted was as follows: Yijk = µ+
Treatmenti + Samplej +
(Treatment.Sample)ij + Ramk +
ijk.
Experiment 2—
Motility characteristics of spermatozoa. Motility characteristics
of spermatozoa were analyzed to compare the 12 different treatments and their
time profiles (0, 3, and 6 hours) using a restricted maximum likelihood (REML)
procedure in GenStat (Release 8; Ceanet, Brisbane, Australia). The model
fitted was as follows: Yijk = µ +
Treatmenti + Timej +
(Treatment.Time)ij + Ramk +
Runl + (Ram.Run)kl +
(Ram.Run.Treatment)ikl + ijkl,
where Yijk is the measure being modeled and
Treatment is the particular treatment (tube). Treatment and Time were
designated as fixed effects, while Run, Ram, Ram.Run, and Ram.Run.Treatment
were random effects, that is,
|
|
2
tests. Subsequently, the treatments were grouped into 2 groups (TRTGRP1 and TRTGRP2) of treatments that showed similar effect, and the REML models for each sperm motility measure were refitted. TRTGRP1 included Control (tube control buffer), SP (seminal plasma and supernatant from the samples collected before and after vasectomy), and Vesicles (pellet of vesicles from the samples collected before and after vasectomy). TRTGRP2 included Control (control buffer), 2x-Pre (double-vesicle concentration from the samples collected before vasectomy), 2x-Post (double-vesicle concentration from the samples collected after vasectomy), and 3x (triple vesicle concentration from the samples collected before and after vasectomy).
Membrane status of spermatozoa assessed by CTC staining—
The response to the 12 treatments was the frequency of each of 3 categories
(F-, B-, and AR-pattern spermatozoa). However, because there were extremely
few Type F spermatozoa (0.35%), these were omitted from the analysis. Logistic
regression was used to model the probability of a sperm being of Type B
(relative to Type AR), the specific form of the model being as follows:
|
|
ijk is the probability of the acrosome being
Type B; and Treatmenti, Timej, and
Ramil are as defined in the previous model. Note that
Ramk and (Ram.Treatment)ik are modeled
as random effects, so the class of model fitted is a generalized linear mixed
model (GLMM), using GenStat 8. Consequently, the data were reanalyzed using a
grouping of the tube treatments: TRTGRP1 (2 treatments with the control),
TRTGRP2 (4 treatments with seminal plasma and supernatant from the samples
before and after vasectomy), and TRTGRP3 (2 treatments with vesicles at triple
concentration from the samples before and after vasectomy), and a GLMM was
refitted for the new grouped data.
Experiment 3—
In vitro fertilization. The analysis was divided into 3 parts. The
first part analyzed sperm motility characteristics after incubation of
spermatozoa for 3 hours with the 3 treatments using the same model and REML
procedure in GenStat 8 as described in experiment 2. The second part was an
analysis of IVF success in which the binary data were in the form of
y successful fertilizations in a batch of n ova, and
consequently the following logistic regression model was fitted to the
following data:
|
|
ijkl is probability of a successful
fertilization and Run refers to replicate. Tube and Time were regarded as
fixed effects, while Ram and Run (and hence Ram.Run) were regarded as random
effects. Thus, the model fitted was a GLMM using GenStat 8. The third part of
the analysis examined IVF success for a 3-level variable: 1) not fertilized,
2) fertilized, or 3) advanced stage. An ordinal logistic regression was used
to model this 3-level outcome variable. This was an extension to the binary
logistic regression used in the second part of the analysis, except that the
response was taken as ordinal in the sense that this was an ordered sequence:
Not fertilized < Fertilized < Advanced. The form of the model fitted to
these data was as follows:
|
|
Prediction of in vitro fertility. A separate series of logistic
regression models was fitted to the fertilization data, this time relating the
fertilization success to the sperm motility characteristics. For initial
exploration, individual models were fitted for each motility measure
(subjective and CASA as listed previously), but Time was also included, given
its overriding importance. The models fitted had the following form:
|
|
j and Timej are as defined
in the previous model, xmt is motility measure
m (SUBJt, TOTt,
PROGt,..., or ELONGt) recorded at
either t = 0 hours or t = 3 hours after semen collection,
and ßmt is the corresponding regression coefficient
for this measure. As an aid to interpreting the importance of each
characteristic, "standardized" odds ratios (ORs) were obtained for
the effect of each, namely,
, where
SDmt is the standard deviation of measure m at
t hours after collection. The OR may be interpreted as the relative
change in the odds of a successful fertilization when the motility measure is
increased by 1 SD. Basing the measure on the SD allows comparisons to be made
across motility measures.
For predictive purposes, a combined model with Time and all 22 motility
measures was fitted:
|
|
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
The mean protein concentrations for the 3 seminal plasma fractions (whole, supernatant, and pellet of vesicles) for 2 vasectomized and 2 control rams are summarized in Table 2. The whole seminal plasma and supernatant had significantly higher protein levels than the pellets (P = .016 vs .018, respectively), but no significant difference was detected between the former 2 fractions (P = .86). There were no differences in protein concentration in the seminal plasma fractions between entire and vasectomized rams (P = .63).
|
SDS-PAGE. Figure 1 shows 1D gels of proteins from the seminal plasma, supernatant, and vesicles (representative gels from 2 intact and 2 vasectomized rams are presented because the 8 rams displayed the same protein profiles). Similar patterns were identified for whole seminal plasma and supernatant from intact and vasectomized rams (lanes B, C, E, and F). The most intensely stained spots (ranging from 10–60 kd) were at approximately 15 and 20, equivalent to the molecular weight regions of the major proteins in ram seminal plasma previously identified by Bergeron et al (2005; ram seminal plasma protein [RSP]-15 kd, RSP-16 kd, 15.5 kd spermadhesin, RSP-22 kd, and RSP-24 kd) and Fernández-Juan et al (2006; ram seminal vesicle protein [RSVP]-14 and RSVP-20).
|
Experiment 2: Motility Characteristics and Membrane Status of Spermatozoa
Sperm Motility and Characteristics (Grouped Treatments)—
The mean sperm motility characteristics, assessed subjectively and by CASA,
grouped for ram, replicate, and treatment (TRTGRP1 and TRTGRP2), are presented
in Figure 2 (data shown for
TOT, PROG, VAP, VSL, and VCL only). All the treatments were compared with the
control buffer. The treatment containing Salamon buffer with added BSA
fraction V (BSA) was not different from the control buffer without BSA for any
of the parameters assessed and was excluded from the grouped analysis. Because
in the TRTGRP1 analysis whole seminal plasma and supernatant had the same
treatment effect for all parameters assessed, these treatments were combined
in 1 line in Figure 2. In the
TRTGRP2 analysis, because double-vesicle concentration from the samples
collected before and after vasectomy (2x-Pre and 2x-Post) had no
significant effect on most of the sperm motility characteristics assessed (all
P > .10 except SUBJ at 3 hours: P < .001), these have
been omitted from Figure 2.
|
The fastest decline in most motility measures with incubation time after thaw was observed in control and vesicle (at normal concentration) treatments, and these were mostly not significantly different from each other (P > .05), bearing in mind that vesicles were absent after vasectomy and this treatment was similar to the control buffer. In general, spermatozoa resuspended in whole seminal plasma or supernatant had the slowest rate of decline in motility measures, followed by those exposed to triple vesicle concentration (Figure 2). No significant differences were detected between treatments for the sperm parameters immediately after thawing (0 hours; all P > .05), and for most measures there was a significant treatment x time interaction (P < .05), requiring an evaluation of the treatment effect at each post-thaw incubation time (0, 3, and 6 hours).
At 3 and 6 hours of incubation, a significant difference from the control was obtained when semen was treated with whole seminal plasma or supernatant (before and after vasectomy) and with vesicles at triple protein concentration for SUBJ, TOT, and PROG (Figure 2A through C; P < .05). For all other treatments and parameters, no significant difference from the control was obtained for any of the motility measures (P > .05; Figure 2D through F; data not shown for ALH, BCF, STR, LIN, and ELONG).
No significant differences between the control and any treatments were obtained at any incubation time for VAP, VCL, or ALH (P > .05; Figure 2D and F; data not shown for ELONG).
Membrane Status After CTC Staining— Overall, 0.4%, 36.5%, and 63.1% of spermatozoa displayed the F, B, and AR CTC patterns, respectively. There were no detectable treatment effects for F-pattern spermatozoa at any time, largely because these occurred at a low frequency, and the response for AR pattern spermatozoa was the complement of that for the B-pattern spermatozoa, so only the results for B pattern spermatozoa are presented in Figure 3.
|
Progressively over time, the lowest frequency of B pattern and the highest frequency of AR pattern was obtained for spermatozoa resuspended in buffer (controls). There was no detectable treatment group x time interaction (P = .462; Figure 3), indicating that the time course profiles for all treatment groups were similar. However, there were significant differences in the percentage of B-pattern spermatozoa across the 3 treatment groups (P < .001). There was also a reduction in the percentage of B-pattern (P < .001) and an increase in AR-pattern spermatozoa over time. Pairwise comparisons showed that whole seminal plasma or supernatant resulted in a higher proportion of B-pattern than control and vesicles, but no significant difference between control and vesicles was detected.
Experiment 3: In Vitro Fertility of Spermatozoa
In experiment 2, no significant differences were obtained when semen was
incubated with either whole seminal plasma or supernatant. Therefore, only the
supernatant and pellet of vesicle fractions were incubated with frozen-thawed
spermatozoa in experiment 3.
|
There were no differences between treatments (control buffer, supernatant, and vesicles at 3x normal protein concentration) for the sperm motility characteristics measured at 0 hours (all P > .05). After 3 hours of incubation, spermatozoa resuspended in supernatant and vesicles at 3x protein concentration had significantly higher subjective motility (SUBJ) and total progressing motility (TOT) than the controls (all P < .001), whereas only spermatozoa treated with the supernatant were superior to controls for CASA progressive motility (PROG) (P < .001). There were no differences between treatments and the control after 3 hours of incubation for the other sperm parameters (all P > .05).
IVF— Only the oocytes classified as 1) mature unfertilized, 2) mature fertilized, and 3) at an advanced stage of fertilization, from the 3 replicates, were integrated in the analysis of successful fertilization. The probability of fertilization expressed in percent and fertilization rates after incubation of frozen-thawed spermatozoa with the 3 treatments are presented in Figure 5 and Table 3, respectively.
|
|
Table 3 shows that the
percentages of mature fertilized and advanced-stage oocytes were higher for
spermatozoa incubated with supernatant, followed by the pellet of vesicles at
3x protein concentration and the control buffer (38.14% and 49.55% vs
19.40% and 48.07% vs 27.12% and 31.25%, respectively). There were significant
differences in fertilization outcomes across the 3 treatments
(2 = 50.61, df = 2, P < .001), with all treatments
leading to fertilization outcomes that were statistically significant from
each other (all P < .05). There was an increase in the percentages
of both mature fertilized and advanced-stage oocytes as coincubation time
elapsed (2 = 48.67, df = 2, P < .001).
The overall (model-based) probabilities for successful fertilization were 0.173, 0.363, and 0.255 for control buffer, supernatant, and vesicles, respectively (P < .001). Supernatant was significantly better than the control (P < .001), but vesicles at 3x protein concentration were only marginally better than the control (P = .032; Figure 5).
The probabilities for successful fertilization were 0.183, 0.328, and 0.272 at 2, 6, and 18 hours, respectively (P = .004). Fertilization success rate was significantly higher at 6 and 18 hours than at 2 hours (P < .001 and P = .029, respectively), but the decline from 6 to 18 hours was not significant (P = .257). There was no significant interaction between treatment and time (P = .934), and there were no differences across the 3 rams or replicates.
Prediction of in vitro fertility— There were 23 variables considered for inclusion in the predictive model, namely Time and the 11 motility variables measured at 0 and 3 hours after thawing and resuspension of spermatozoa. Time was included in each individual motility model because of its overriding importance. Motility measures associated with a significant increase (P < .05) in fertility were SUBJ, TOT, PROG, VAP, and ELONG, and these were all at t = 3 hours. Significant reductions in fertility were associated with VAP, VSL, VCL, and ALH, and these were all at t = 0 hours. The most significant predictor, VAP at 3 hours (P < .001), had a standardized OR of 1.48, indicating that for an increase in VAP by 1 SD (14.76 units) the odds of a successful fertilization increases by 1.48 times.
The combined (full) model with Time and all motility variables was significantly predictive of the fertilization success rate (F = 3.21; df = 24, 56; P < .001; pseudo R2 = 58%). Comparisons with models using Time and all 0-hour measures or using Time and all 3-hour measures are shown in Table 4. They indicate that both sets of variables may be used to develop a predictive model, although the 3-hour measures provided better predictions than the 0-hour measures (pseudo R2 = 29% for 0 hours, 35% for 3 hours). However, the combined model (0 hours + 3 hours) provided significantly better predictions than either individual one (0 hours: P = .006; 3 hours: P < .001).
|
However, there were many redundant variables in the full model, primarily
due to the strong correlations among the CASA measures. Consequently, a
stepwise procedure resulted in the following model with terms for Time and 7
motility measures:
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
In experiment 1, the total protein concentration of whole seminal plasma was compared with its vesicle-containing and vesicle-free (supernatant) fractions in entire and vasectomized rams. In addition, the protein profiles as detected by SDS-PAGE were determined for these fractions. There was no difference in the protein concentration between whole seminal plasma and its supernatant fraction (range, 24–37 mg/mL), but this was much higher than the protein concentration measured in the pellet of vesicles in both intact and vasectomized rams (2–5 mg/mL, respectively; Table 2). These data suggest that most of the protein component of ram seminal plasma is contained in its nonvesicle fraction and that this may provide the beneficial effects reported previously (Maxwell et al, 1999). This outcome in the ram supports findings in other species, for example, the buffalo, where the mean protein concentration in the seminal plasma was 29 ± 2.7 mg/mL and cauda epididymal sperm function was improved by the addition of 0.4 µg/mL of protein (Arangasamy et al, 2005). In the present study, the protein concentration in the pellet of vesicles for intact rams was higher than for vasectomized rams, but in both cases the values were low compared with whole seminal plasma and supernatant. The protein concentration for the vesicle fraction was especially low after vasectomy because the vesicles were absent and only faint protein bands were detectable by SDS-PAGE (Figure 1B, lanes D and G).
The SDS-PAGE protein profiles obtained in the present study were similar to those published for ovine seminal plasma by Jobim et al (2005), Bergeron et al (2005), and Fernández-Juan et al (2006). The latter authors termed the major seminal plasma proteins RSVP-14 and RSVP-20. Bergeron et al (2005) isolated and characterized a 15.5-kd spermadhesin together with 4 RSP proteins (15, 16, 22, and 24 kd) with fibronectin-2 domains. The similarity in protein concentrations and 1D gel protein profiles for whole seminal plasma and supernatant obtained in our study confirm that most seminal plasma proteins are associated with the vesicle-free supernatant fraction. This supports the finding that most of the protein constituents of seminal plasma are derived from accessory sex glands, namely the vesicular gland in rams as recently confirmed by Fernández-Juan et al (2006) using immunohistochemical localisation, and not from the epididymis or testis. Furthermore, the vesicle-free protein component was responsible for the beneficial effects of the plasma on sperm function examined in experiments 2 and 3, because it was necessary to increase the concentration of the vesicle fraction by threefold in the post-thaw medium to detect any effect on the motility characteristics and fertilizing capacity of frozen-thawed ram spermatozoa. The beneficial components in the supernatant fraction were specific to ram seminal plasma and not just due to the total protein concentration in this fraction, because no benefit was obtained by including a similar protein concentration of BSA in the post-thaw medium as a protein control.
High-speed centrifugation was used to precipitate high-molecular-weight protein and vesicles, but the high-molecular-weight proteins (100–200 kd) could not be detected clearly in the 1D gel profiles. It may be that these proteins were integrated into the membranes of vesicles in whole seminal plasma but were separated by high-speed centrifugation and appeared separately in the SDS profile of the pellet of vesicles (Figure 1A, lanes D and G).
We have previously demonstrated that vesicles are absent from the ram ejaculate after vasectomy, suggesting that they are derived from the testis or epididymis (El-Hajj Ghaoui et al, 2006). The present study confirms our findings that the pellet of vesicles fraction of seminal plasma obtained from vasectomized rams showed insignificant protein migration on SDS-PAGE. Thus, the faintly visible bands shown in columns D and G of Figure 1B could represent contamination by the supernatant that remained after washing the pellet. Vesicles were absent after vasectomy, and the high-molecular-weight proteins (100–200 kd) present in samples analyzed before vasectomy, which may have been integrated in the membranes of the vesicles, disappeared from the profiles of the vesicle fraction after vasectomy.
In experiment 2, the effect of autologous whole seminal plasma and its fractions on the motility characteristics and membrane status of washed frozen-thawed spermatozoa was examined. Whole seminal plasma and the supernatant significantly improved the motility of spermatozoa compared with the control buffer. The high-molecular-weight proteins and the vesicles precipitated by high-speed centrifugation (vesicle fraction) had no effect on sperm motility or membrane status at their normal concentration in seminal plasma. Nevertheless, they marginally improved sperm motility and membrane status when the concentration was increased to 3 times normal in the post-thaw incubation medium. This effect was observed for subjective and CASA assessment of total sperm motility but not forward progressive motility as assessed by CASA. This difference could be due to the active proteins adhering to the sperm membrane and improving the percentage of motile spermatozoa while having no direct effect on their forward motility. These protein components did not benefit sperm function at low concentrations. However, when the vesicles were added to the sperm at a higher concentration, they had a similar effect on sperm motility to seminal plasma or supernatant, suggesting that small amounts of the beneficial protein were either segregated from the vesicles or remained as contaminants from the other seminal plasma fraction.
The findings of the present study with respect to seminal plasma vesicles disagree with some reports in other species. Davis (1974) demonstrated sperm decapacitation activity associated with membrane vesicles isolated from rabbit seminal plasma. Stegmayr and Ronquist (1982) suggested that prostasomes have specific effects on human sperm progressive motility, and Arienti et al (1997) reported that prostasomes had a protective effect and increased the percentage of motile spermatozoa. In addition, it has been hypothesized that membrane vesicles could interact with the sperm membrane (horse: Minelli et al, 1998, 1999) and transfer some anchored protein to the sperm surface, which might play a role in membrane stabilzation (hamster: Légaré et al, 1999; bull: Frenette and Sullivan, 2001; and Frenette et al, 2002). Our findings suggest that the membrane vesicles found in ram seminal plasma have no role in sperm function.
In sheep and cattle, the specific roles of seminal plasma proteins have not been defined, but those with fibronectin-2 domains (RSP proteins) are believed to interact with choline phospholipids on the sperm membrane, with high- and low-density lipoproteins, and with heparin, playing major roles in membrane stabilization (decapacitation) and destabilization (capacitation) (Manjunath and Thérien, 2002). Constituents of whole seminal plasma have been shown to delay capacitation-like changes and the acrosome reaction as well as improve the motility of ram spermatozoa after freezing and thawing (Maxwell et al, 1999). We have been unable to find a role for membrane vesicles in ram seminal plasma, but we have demonstrated that the vesicle-free fraction of seminal plasma delays membrane changes in frozen-thawed ram spermatozoa during post-thaw incubation.
Whole seminal plasma and its vesicle-free fraction improved the IVF rate of oocytes in addition to improving sperm motility and membrane status in comparison with spermatozoa treated with the control buffer. Most of the oocytes had been fertilized by 6 hours and all by 18 hours after insemination. The fertilization rate was better after 6 and 18 hours than 2 hours of coincubation with the oocytes. This is to be expected because 2 hours is probably too short a coincubation time for the spermatozoa to penetrate and fertilize a normal proportion of oocytes, even when fully capacitated, and is in agreement with Gillan et al (1997), who demonstrated that at 0.5, 1, and 2 hours of coincubation of frozen-thawed spermatozoa with oocytes, the fertilization rate and the proportion of advanced-stage zygotes were lower than after 4, 6, 8, and 21 hours. Given that most the of the frozen-thawed spermatozoa assessed in experiment 2 exhibited the B-pattern of fluorescence after CTC staining (indicative of capacitated, acrosome-intact cells) and that the progression of spermatozoa from the B pattern to the AR pattern (indicative of acrosome reacted cells) was slowed by the supernatant fraction of seminal plasma (Figure 3), it is not surprising that this fraction also extended the fertilizing lifespan of the frozen-thawed spermatozoa (Figure 5).
In the present study, we observed an increase in the proportion of zygotes at an advanced stage of fertilization (2, 4, or 16 blastomeres) with increasing coincubation time up to 18 hours, mainly for oocytes inseminated with the spermatozoa incubated with the supernatant or vesicles at 3 times their normal concentration. Thus, spermatozoa exposed to the beneficial proteins present in the seminal plasma, even at the low concentrations present in the vesicle fraction, were able to penetrate oocytes earlier than those not exposed to seminal plasma proteins.
To understand the value of the different sperm motility characteristics for predicting the success of IVF, descriptive multivariable analyses were undertaken using the subjective motility and CASA data. From this analysis, a model was developed for prediction of IVF success rate. There have been a number of other studies attempting to relate both subjective and objective measures of sperm motility to fertility both in vitro and in vivo (reviewed by Larsson and Rodríguez-Mártinez, 2000; Gillan et al, 2006). Few single in vitro sperm parameters show a reliable and repeatable correlation with field fertility (Rodríguez-Mártinez, 2000), but clear associations have been reported between multivariable mixed models incorporating sperm motility characteristics assessed by CASA and in vivo (eg, cattle: Gillan et al, 2006) and in vitro fertility (eg, human: Liu et al, 1991). An interesting observation in the present study was the additive value of measurements made after 3 hours after thaw incubation, as well as immediately after thawing, in the predictive value of the model. Nevertheless, the most significant individual predictors of fertility (subjective motility, total and progressive motility, and the main velocity measurements assessed by CASA: VAP, VSL, and VCL) are in agreement with other studies relating CASA assessments with in vivo fertility of bull spermatozoa (Gillan et al, 2006). The most important components of the full predictive model obtained after dropping redundant variables were those associated with total motility and straightness, largely due to the strong correlations among CASA-based measurements. To our knowledge, this is the first report of a relationship between sperm motility characteristics and in vitro fertility in sheep. The relationship between in vitro an in vivo fertility in sheep has been established, provided low concentrations of spermatozoa are used for IVF (O'Meara et al, 2005), so further investigations are justified using simpler alternatives based on the function of the spermatozoa themselves.
In conclusion, autologous ram seminal plasma was split into 2 fractions, supernatant and pellet of vesicles, added back to washed frozen-thawed spermatozoa, and their effect determined after incubation on sperm motility characteristics, membrane status, and in vitro fertilizing capacity. The seminal plasma and the supernatant fraction from entire and vasectomized rams, containing similar proteins, improved sperm motility and membrane stability as well as facilitating their IVF of oocytes. The other fraction of seminal plasma, the membrane vesicles, was not important for sperm function or fertilizing capacity. A predictive model of IVF success could be developed based on subjective and objective measurements of frozen-thawed sperm motility characteristics. Further work is needed to determine whether particular proteins present in ram seminal plasma could be characterized, synthesized, and used as additives to sperm preservation media to improve fertility of ewes after AI.
|
Acknowledgments |
|---|
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Arangasamy A, Singh LP, Ahmed N, Ansari MR, Ram GC. Isolation and characterization of heparin and gelatin binding buffalo seminal plasma proteins and their effect on cauda epididymal spermatozoa. Anim Reprod Sci. 2005;90: 243 –254.[CrossRef][Medline]
Arienti G, Carlini E, Palmerini CA. Fusion of human sperm to prostasomes at acidic pH. J Membr Biol. 1997; 155: 89 –94.[CrossRef][Medline]
Bergeron A, Villemure M, Lazure C, Manjunath P. Isolation and characterization of the major proteins of ram seminal plasma. Mol Reprod Dev. 2005;71: 461 –470.[CrossRef][Medline]
Breitbart H, Rubinstein S. Characterization of Mg2+-ATPase and Ca2+-ATPase activity in membrane-vesicles from ejaculated ram seminal plasma. Arch Androl. 1982;9: 147 –157.[Medline]
Breitbart H, Stern B, Rubinstein S. Calcium-transport and Ca2+-ATPase activity in ram spermatozoa plasma-membrane vesicles. Biochim Biophys Acta. 1983; 728: 349 –355.[Medline]
Davis BK. Decapacitation and recapacitation of rabbit spermatozoa treated with membrane vesicles from seminal plasma. J Reprod Fertil. 1974;41: 241 –244.[Medline]
El-Hajj Ghaoui R, Thomson PC, Evans G, Maxwell WMC. Characterization and localization of membrane vesicles in ejaculate fractions from the ram, boar and stallion. Reprod Domest Anim. 2004; 39: 173 –180.[CrossRef][Medline]
El-Hajj Ghaoui R, Thomson PC, Evans G, Maxwell WMC. The origin of membrane vesicles in ram seminal plasma. Reprod Domest Anim. 2006;41: 98 –105.[CrossRef][Medline]
Evans G, Maxwell WMC. Salamon's Artificial Insemination of Sheep and Goats. Sydney, Australia: Butterworths; 1987 .
Fernández-Juan M, Gallego M, Barrios B, Osada J, Cebrián-Pérez JA, Muiño-Blanco T. Immunohistochemical localization of sperm-preserving proteins in the ram reproductive tract. J Androl. In press.
Frenette G, Lessard C, Sullivan R. Selected proteins of
"prostasome-like particles" from epididymal cauda fluid are
transferred to epididymal caput spermatozoa in bull. Biol
Reprod. 2002;67: 308
–313.
Frenette G, Sullivan R. Prostasome-like particles are involved in the transfer of P25b from the bovine epididymal fluid to the sperm surface. Mol Reprod Dev. 2001; 59: 115 –121.[CrossRef][Medline]
Gillan L, Evans G, Maxwell WMC. Capacitation status and fertility of fresh and frozen-thawed ram spermatozoa. Reprod Fertil Develop. 1997;9: 481 –487.[CrossRef][Medline]
Gillan L, Kroetsch T, Maxwell WMC, Evans G. Assessment of in vitro sperm characteristics in relation to fertility in dairy bulls. Anim Reprod Sci. In press.
Jobim MIM, Oberst ER, Salbego CG, Wald VB, Horn AP, Mattos RC. BSP A1/A2-like proteins in ram seminal plasma. Theriogenology. 2005; 63: 2053 –2062.[CrossRef][Medline]
Larsson B, Rodríguez-Mártinez H. Can we use in vitro fertilization tests to predict semen fertility? Anim Reprod Sci. 2000;60-61: 327 –336.
Légaré C, Bérubé B, Boué F, Lefièvre L, Morales CR, El-Alfy M, Sullivan R. Hamster sperm antigen P26h is a phosphatidylinositol-anchored protein. Mol Reprod Dev. 1999;52: 225 –233.[CrossRef][Medline]
Liu DY, Clarke GN, Baker HWG. Relationship between sperm motility
assessed with the Hamilton-Thorn motility analyzer and fertilization rates in
vitro. J Androl. 1991; 12: 231
–239.
Manjunath P, Thérien I. Role of seminal plasma phospholipid-binding proteins in sperm membrane lipid modification that occurs during capacitation. J Reprod Immunol. 2002; 53: 109 –119.[CrossRef][Medline]
Maxwell WMC, de Graaf SP, El-Hajj Ghaoui R, Evans G. Seminal plasma effects on sperm handling and female fertility. Reproduction. In press.
Maxwell WMC, Evans G, Mortimer ST, Gillan L, Gellatly ES, McPhie CA. Normal fertility in ewes after cervical insemination with frozen-thawed spermatozoa supplemented with seminal plasma. Reprod Fertil Dev. 1999;11: 123 –126.[CrossRef][Medline]
Minelli A, Allegrucci C, Mezzasoma I, Ronquist G, Lluis C, Franco
R. CD26 and adenosine deaminase interaction: its role in the fusion between
horse membrane vesicles and spermatozoa. Biol Reprod. 1999; 61: 802
–808.
Minelli A, Moroni M, Martinez E, Mezzasoma I, Ronquist G. Occurrence of prostasome-like membrane vesicles in equine seminal plasma. J Reprod Fertil. 1998; 114: 237 –243.[Abstract]
Ollero M, Cebrián-Pérez JA, Muiño-Blanco T.
Improvement of cryopreserved ram sperm heterogeneity and viability by addition
of seminal plasma. J Androl. 1997; 18: 732
–739.
O'Meara CM, Hanrahan JP, Donovan A, Fair S, Rizos D, Wade M, Boland MP, Evans ACO, Lonergan P. Relationship between in vitro fertilisation of ewe oocytes and the fertility of ewes following cervical artificial insemination with frozen-thawed ram semen. Theriogenology. 2005; 64: 1797 –1808.[CrossRef][Medline]
Rodríguez-Mártinez H. Evaluation of frozen semen: traditional and new approaches. In: Chenoweth PJ, ed. Topics in Bull Fertility, Recent Advances in Veterinary Medicine [monograph on the Internet]. International Veterinary Information Services (IVIS); 2000 [modified 2001 Nov 7; cited 2006 Oct 11]. (Document No. A0502.0600, 2000a). Available from: http://www.ivis.org/advances/Repro_Chenoweth/Rodriguez_Martinez/chapter_frm.asp?LA=1.
Stegmayr B, Ronquist G. Distribution of cyclic AMP in human seminal plasma and its relation to sperm progressive motility. Scand J Urol Nephrol. 1982;16: 91 –95.[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE |
), vesicles at their normal concentration (
), and vesicles at 3
times their concentration (
). Data are means ± SEM for each
post-thaw incubation assessment time (0, 3, and 6 hours) and were pooled for
the samples collected before and after vasectomy from entire and vasectomized
rams.
