Human Reproduction vol.15 no.6 pp.1322–1328, 2000
Objective evaluation of hyperactivated motility in rat
spermatozoa using computer-assisted sperm analysis*
Aida M.Cancel1, Danelle Lobdell2, Pauline Mendola3
and Sally D.Perreault4,5
1Toxicology Program, University of North Carolina, Chapel Hill,
NC 27599, 2Department of Social and Preventive Medicine,
The State University of New York at Buffalo, Buffalo, NY 14214,
3US EPA, Human Studies Division, Chapel Hill, NC 27599 and
4US EPA, Reproductive Toxicology Division, Chapel Hill
NC USA 27711, USA
5To
whom correspondence should be addressed at: US EPA,
NHEERL MD 72, Research Triangle Park, NC 27711, USA.
E-mail: [email protected]
The aim of this study was to use computer-assisted
sperm analysis (CASA) to examine changes in motion
parameters of rat spermatozoa incubated under culture
conditions that support IVF. Rat cauda epididymal
spermatozoa were evaluated in six replicate experiments,
at 0 and 4 h of incubation. CASA was conducted at 60
Hz on digital 1 s tracks (~100 spermatozoa/rat). Mean
values of CASA parameters that describe the vigour of
spermatozoa [curvilinear velocity (VCL), amplitude of
lateral head displacement (ALH) and beat cross frequency
(BCF)] increased, while those indicating progressiveness
[straight line velocity (VSL), linearity (LIN) and straightness (STR)] decreased between 0 and 4 h. Visual
inspection of sperm tracks after 4 h of incubation
revealed classical hyperactivation patterns. Bivariate
models were evaluated to objectively define the subpopulation of hyperactivated (HA) spermatozoa. Of all models
considered, ALH and LIN, VCL and LIN, BCF and
LIN, VCL and BCF, and VCL and ALH showed
significant changes in the percentage of HA spermatozoa
after the 4 h incubation period. The efficacy of detecting
HA spermatozoa was evaluated using sperm tracks that
were visually classified as HA or progressive. VCL and
LIN provided the most accurate prediction of HA
spermatozoa. It was concluded that analysis of CASA
data using bivariate models could be used to detect and
monitor hyperactivation in rat spermatozoa.
Key words: computer-assisted sperm analysis/hyperactivation/
rat/spermatozoa
*The research described in this article has been reviewed by the
Health Effects Research Laboratory, US Environmental Protection
Agency, and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the agency
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
1322
Introduction
Hyperactivated (HA) motion is a type of vigorous non-linear
motion that mammalian spermatozoa exhibit as they progress
through the female oviduct (Yanagimachi, 1994; ESHRE,
1996; Suarez, 1996). During hyperactivation, the pattern and
vigour of the sperm track undergo dramatic changes, which
are best characterized by movements in a random path and
result in a non-progressive circular movement. These changes
in motion are described as ‘whiplash’ or ‘figure-8’. Several
studies have suggested that HA allows spermatozoa to detach
from the oviductal epithelium and provides increased thrust
for penetration of the cumulus (Suarez et al., 1991; Suarez
and Dai, 1992; Pacey et al., 1995). Recently, it has been shown
that HA also facilitates penetration of spermatozoa through
the oocyte zona pellucida (Stauss et al., 1995). Consequently,
objective measures of HA can serve as biological end-points
to evaluate the functional capabilities of spermatozoa.
Computer-assisted sperm analysis (CASA) provides the
means for an objective classification of a given population of
spermatozoa. Using digital images of each sperm track, CASA
machines are able to analyse, by processing algorithms, the
motion properties of spermatozoa. The commonly reported
CASA parameters include curvilinear velocity (VCL), amplitude of lateral head displacement (ALH), beat cross frequency
(BCF), average path velocity (VAP), straight line velocity
(VSL), straightness (STR) and linearity (LIN). These CASA
parameters have been modelled and refined mathematically to
describe best the motion parameters of each spermatozoon as
it travels through a microscopic field (Boyers et al., 1989).
Until recently, CASA of HA spermatozoa was difficult
because of poor tracking efficiency. The relatively new use of
60 Hz combined with increased resolution of the optical system
and better computers to allow longer tracking has made it
possible to evaluate spermatozoa with high velocities and
erratic tracks. Therefore vigour and pattern changes associated
with HA spermatozoa can be detected using CASA. The tracks
of HA spermatozoa would be expected to have increased
amplitude of the ALH, increased VCL and decreased LIN. In
addition, distribution based statistical methods are being
applied successfully to characterize HA motility of hamster
spermatozoa (S.D.Perreault, unpublished results). This
distribution-based analysis has proven to be more powerful in
determining changes in sperm motion that may not be detected
using conventional mean-based statistical outcomes.
CASA has been used to characterize the motion of rat
epididymal spermatozoa and changes in CASA parameters have been interpreted as indicators of testicular and
epididymal toxicity (Toth et al., 1989; Slott et al., 1993;
© European Society of Human Reproduction and Embryology
Hyperactivation of rat spermatozoa
Perreault, 1998). As such, these CASA measures are made
shortly (10–30 min) after collection and dilution of epididymal
or ejaculated spermatozoa in medium that is not designed to
support sperm capacitation and/or IVF. Freshly diluted rat
epididymal spermatozoa swim in a progressive manner (fast
and straight) and these are the end-points of interest in routine
toxicology protocols (Perreault, 1998). However, examination
of freshly diluted spermatozoa does not necessarily predict whether those spermatozoa can undergo HA and other
changes required for successful fertilization. Therefore, it is
desirable to develop assays for rat spermatozoa that can be
reliable indicators of sperm function.
Hyperactivation of rat spermatozoa in the oviduct has
been described (Shalgi and Phillips, 1988). However, limited
information is available regarding the motion of rat
spermatozoa during IVF and to date there are no published
reports demonstrating HA motility of rat spermatozoa during
in-vitro incubation. In this report, it was postulated that rat
spermatozoa would exhibit HA motility after incubation under
conditions that support IVF and that HA rat spermatozoa could
be classified in an objective manner by using CASA parameters
to identify those spermatozoa showing both increased vigour
and decreased progression.
Materials and methods
Animals
Sexually mature (⬎90 days old) Sprague–Dawley rats (Charles River
Breeders, Raleigh, NC, USA) were allowed to acclimatize in an
AAALAC-approved animal care facility (12 h light:12 h dark, 72°F)
for at least 2 weeks prior to use. Animals were housed two per cage
and provided access to Purina Laboratory Chow and tap water ad
libitum. All investigations were conducted in accordance with the
Guiding Principles for the Care and Use of Research Animals.
Preparation of rat spermatozoa and in-vitro culture conditions
Animals were killed by CO2 asphyxiation followed by cervical
dislocation. The epididymis was removed, trimmed of fat and clamped
at the corpus–cauda junction. Rat spermatozoa were obtained by
puncturing the distal cauda epididymidis and allowing the
spermatozoa to disperse in 3 ml of Armstrong’s Rat Fertilization Medium (ARFM) supplemented with 4 mg/ml bovine serum
albumin (BSA) (Fraction V: A-4503, Sigma Chemical Co., St Louis,
MO, USA) (Evans and Armstrong, 1984). This Krebs–Ringer, a
bicarbonate-buffered medium, has been shown to support rat IVF
(Miyamoto and Chang, 1973; Toyoda and Chang, 1974; Perreault
and Jeffay, 1993), and has been used successfully in hamster zonafree oocyte IVF using rat spermatozoa (A.M.Cancel, unpublished
results). After culturing (37°C, 5% CO2 in air) for 30 min, to allow
dispersion of the spermatozoa, the concentration of spermatozoa was
determined by haemocytometer and aliquots of the sperm suspension
(⬍100 µl) were transferred to 3 ml of fresh medium to give a final
concentration of 3.3⫻104/ml. The incubation (37°C, 5% CO2 in air)
was continued for 4 h, which is the minimum time for attachment of
rat spermatozoa to and penetration of the zona pellucida during IVF
(Shalgi et al. 1988). Aliquots of rat spermatozoa were withdrawn at
the start (0 h) and end (4 h) of incubation, and analysed using CASA.
A complication that arises when studying HA spermatozoa during
CASA analysis is that spermatozoa may stick to the glass cannulae
producing tracking errors and preventing the accuracy of the motility
analysis. Recommendations for using higher concentrations of BSA
in the medium, to minimize sticking of spermatozoa to the glass
cannula, have been published (ESHRE, 1998). To test whether a
higher concentration of BSA in the medium may affect the detection
and accurate measurement of HA motility, the rat fertilization medium
was supplemented with 15 mg/ml BSA, which has been used for the
evaluation of hyperactivation of hamster spermatozoa (Perreault and
Jeffay, 1993).
CASA analysis
Six replicate experiments were conducted on 6 separate days.
Suspensions of spermatozoa were loaded into flat 100 µm deep
cannulae (HTR1099, VitroCom Inc., Mountain Lakes, NJ, USA) for
CASA analysis using an HTM-IVOS motility analyser (Hamilton
Thorne Research, Beverly, MA, USA), with software version 10.6.
One-second tracks were captured at 60 Hz under ⫻4 dark-field
illumination, and saved as digital images to an optical disc. Instrument
settings were: temperature ⫽ 37°C, video frequency ⫽ 60, magnification 0.82. Tracks were recalled later for analysis at 60 Hz, with
minimum size ⫽ 7 pixels, minimum contrast ⫽ 65, low VSL cutoff ⫽ 20 µm/s, and low VAP cut-off ⫽ 30 µm/s. During analysis,
the playback feature was used to identify and delete obviously
mistracked spermatozoa as may occur due to collisions. Tracks of
30–60 points were accepted, while track fragments of ⬍30 points
were eliminated by setting the minimum track length at 30 points.
This setting was justified because previous work has shown a high
correlation between CASA measures based on 30 or 60 points for
rat spermatozoa tracked under identical conditions (Dunson et al.,
1999). Observations based on tracks with fewer than 30 points showed
systematic biases. For each of six rats and each time-point, ~100
sperm tracks (mean, 96; range, 53–170) were imported into a Statistics
Package for Social Sciences (SPSS) database for statistical analysis.
The percent motile spermatozoa and percent HA spermatozoa were
also determined subjectively using the playback option. To be
classified HA by visual analysis, a spermatozoon had to clearly
exhibit whiplash motion.
CASA parameters evaluated in this study were those provided by
the HTM-IVOS motility analyser and defined as follows (HTM-IVOS
Operation Manual): VAP: a smoothing of the path of the centre of
brightness of the spermatozoon, which reduces the effect of the lateral
head displacement; VSL: the distance between the first and last
tracked point of the spermatozoon trajectory divided by the time
elapsed; VCL: the sum of the distances between each centre of
brightness, during each frame, divided by the time elapsed; ALH:
the maximum value of the distance of any point on the track from
the corresponding average path, multiplied by two; BCF: frequency
with which the cell track crosses the cell path in either direction;
STR: measures the departure of the cell path from a straight line.
STR is derived from the ratio of VSL/VAP multiplied by 100; LIN:
measures the straightness of the path. LIN is derived from the ratio
of VSL/VCL multiplied by 100.
Analysis plan
Means for each CASA end-point were calculated for each rat and
treatment at each time point. This preserved the rat as the unit of
measure in the statistical analysis. These means were compared across
time (0 versus 4 h) to detect motility changes in CASA end-points
during in-vitro culture conditions in medium that has been shown to
support rat IVF. Appropriate pairs of CASA end-points were selected
for bivariate analysis, by including one measure of vigour (VCL,
BCF or ALH) and one of progression (VSL, VAP, STR and LIN).
Cut-off gates were selected based on the distribution of CASA data
in six control samples (ARFM-4 mg/ml BSA) at 0 h. Since HA
1323
A.M.Cancel et al.
Table I. Mean CASA values of rat spermatozoa incubated in ARFM with 4 mg/ml or 15 mg/ml BSA
4 mg/ml BSA
15 mg/ml BSA
Incubation
period (h)
Motility (%)
VAP (µm/s)
VSL (µm/s)
0
4
0
4
63.4
64.5
69.6
72.3
⫾
⫾
⫾
⫾
137.1 ⫾ 2.2
128.5 ⫾ 3.6
133.7 ⫾ 3.2
119.9 ⫾ 3.2
98.9
82.7
94.7
73.1
2.5
6.9
4.9
3.4
⫾
⫾
⫾
⫾
3.8
5.7
6.4
6.0
VCL (µm/s)
322.8
367.2
309.0
345.2
⫾
⫾
⫾
⫾
7.5
11.6
13.3
13.1
ALH (µm)
22.8
24.2
21.9
23.4
⫾
⫾
⫾
⫾
0.7
1.4
0.5
0.9
BCF (Hz)
23.6
28.1
22.4
30.2
⫾
⫾
⫾
⫾
0.8
1.2a
0.6
1.1a
STR (%)
72.7
65.4
70.9
62.1
⫾
⫾
⫾
⫾
2.2
3.1
3.5
4.1
LIN (%)
30.8
23.5
30.7
22.0
⫾
⫾
⫾
⫾
1.1
1.8a
1.2
1.9a
Data are expressed as means ⫾ SEM (n ⫽ 6).
aIndicates significant differences (P ⬍ 0.01) between 0 and 4 h in the same group.
CASA ⫽ computer-assisted sperm analysis; ARFM ⫽ Armstrong’s rat fertilization medium; BSA ⫽ bovine serum albumin; VAP ⫽ average path velocity;
VSL ⫽ straight line velocity; VCL ⫽ curvilinear velocity; ALH ⫽ amplitude of lateral head displacement; BCF ⫽ beat cross frequency; STR ⫽ straightness;
LIN ⫽ linearity.
Table II. Changes (∆) in CASA parameters after 4 h incubation in ARFM
4 mg/ml BSA
15 mg/ml BSA
VAP (µm/s)
VSL (µm/s)
VCL (µm/s)
ALH (µm)
BCF (Hz)
STR (%)
LIN (%)
⫺8.5 ⫾ 4.0
⫺13.5 ⫾ 5.0
⫺16.3 ⫾ 6.9
⫺21.5 ⫾ 6.7
⫹44.3 ⫾ 13.4
⫹36.3 ⫾ 23.0
⫹1.4 ⫾ 0.9
⫹1.5 ⫾ 0.9
⫹4.5 ⫾ 1.7
⫹7.8 ⫾ 1.1
⫺7.3 ⫾ 3.4
⫺8.9 ⫾ 3.5
⫺7.3 ⫾ 2.2
⫺8.8 ⫾ 1.5
Values are calculated as: ∆ ⫽ t (4 h) – t (0 h). Data are expressed as means ⫾ SEM (n ⫽ 6). The direction of the change is expressed by a (⫹) plus or (⫺)
minus sign. There were no significant differences (P ⬍ 0.01) between 4 mg/ml BSA and 15 mg/ml BSA.
For definitions, see Table I.
spermatozoa have high vigour and low progression, the cut-offs for
vigour parameters were set at the 90th percentile and for progression
at the 10th percentile. Spermatozoa were considered HA if they
exceeded the cut-off for vigour and fell below the cut-off for
progression. The percent HA spermatozoa for each bivariate model
was defined as the number of spermatozoa that passed both criteria
for the bivariate analysis divided by the number of tracked spermatozoa
⫻100. Finally, the accuracy of track classification was examined
using subjectively selected HA and progressive tracks (n ⫽ 35 each).
Accuracy was defined as the percentage of HA (or progressive) tracks
classified correctly by the bivariate model.
Statistical analysis
CASA measurements for each spermatozoon were imported into a
statistical software package (SPSS 7.5 for Windows). Means for each
treatment by time were calculated, as was the average change in
each CASA parameter over time. A generalized linear model with
repeated measures was used to analyse the main effects of time and
treatment for each of the CASA parameters and for the bivariate
indicators of hyperactivation. Paired sample t-tests were conducted
to test the average change in each CASA parameter over time
by treatment (4 mg/ml and 15 mg/ml BSA). Pearson correlation
coefficients were calculated from the individual sperm tracks. Significance was set at P ⬍ 0.01 in consideration of the multiple
comparisons in these analyses, unless otherwise specified.
Results
Changes in individual CASA motility parameters
Motility parameters were measured in samples of spermatozoa at the beginning of the experiment (0 h) and after
incubation (4 h) in ARFM with 4 mg/ml or 15 mg/ml BSA.
The percentage of motile spermatozoa was maintained during
the 4 h incubation. Mean values of CASA parameters changed
as expected for spermatozoa with high vigour and low progres1324
sion, characteristics of HA spermatozoa. The vigour parameters
VCL, ALH and BCF increased, while progression parameters
(VSL, VAP, STR and LIN) decreased (Tables I and II).
However, only mean BCF and LIN showed statistically significant changes over time. There was no significant difference
in any parameter at either time when BSA was increased from
4 to 15 mg/ml.
Subjective analysis of HA motion
Subjective evaluation of sperm tracks at 4 h yielded
a small number of spermatozoa (~3%) exhibiting whiplash
motion typical of hyperactivation (Figure 1C, Table III), but
most of the spermatozoa either had progressive tracks typical
of those visualized at 0 h (Figure 1A), or showed a transitional pattern (Figure 1B). In addition, it was common to see
a spermatozoon switch between transitional and HA motion.
This made it difficult to classify intermediate forms and
suggests that classification based solely on whiplash motion
may be overly strict. These subjectively determined HA sperm
tracks were tallied and the percent of HA spermatozoa was
determined for each animal. There was no significant difference
in the percent HA spermatozoa between treatments with 4 mg/
ml or 15 mg/ml BSA (Table III).
Bivariate analysis of hyperactivated tracks
Bivariate models were applied in an attempt to use CASAbased end-points and provide an objective definition of
hyperactivation for rat spermatozoa. Since CASA parameters
LIN, VSL and STR had a high correlation coefficient among
each other (r 艌 0.80, P ⫽ 0.0001), and changes in the mean
CASA values for LIN were statistically significant over time,
only LIN was included in the bivariate analysis. Cut-off points
for bivariate analyses using the remaining end-points were
Hyperactivation of rat spermatozoa
determined using distribution analysis of the spermatozoon
tracks at 0 h (ARFM-4 mg/ml BSA). Gates were set at the
10th percentile for progression terms (LIN 艋22% and VAP
艋107.0 µm/s) and the 90th percentile for vigour terms (VCL
艌388.7 µm/s, ALH 艌29.6 µm and BCF 艌33.3 Hz). Spermatozoa that exceeded the cut-off for vigour and fell below the
cut-off for progressiveness were classified as HA, as illustrated
in Figure 2.
Three bivariate models combining vigour and progression
terms (LIN paired with VCL, ALH or BCF) showed statistically
significant differences in the percentage of HA spermatozoa
between 0 h and 4 h (Table III). Interestingly, bivariate models
combining two vigour terms (‘VCL and ALH’ and ‘ALH and
BCF’) also showed significant increases in hyperactivation
over time (Table III). The concentration of BSA did not
Figure 1. Representative tracks and individual computer-assisted
sperm analysis (CASA) parameters of rat spermatozoa during invitro capacitation. Sperm trajectories were reconstructed from the
x,y co-ordinates obtained during CASA analysis.
VAP ⫽ average path velocity; VCL ⫽ curvilinear velocity;
VSL ⫽ straight line velocity; STR ⫽ straightness; LIN ⫽ linearity;
ALH ⫽ amplitude of lateral head movement; BCF ⫽ beat cross
frequency.
significantly affect the percentage of HA spermatozoa at either
0 or 4 h.
The accuracy of these five bivariate models was evaluated
against sperm tracks that were previously classified as HA or
progressive (n ⫽ 35 each). Table IV shows the percent HA
spermatozoa detected by these models. Models ‘LIN and BCF’
and ‘ALH and BCF’, identified only 42.9% HA spermatozoa;
whereas ‘VCL and LIN’, ‘ALH and LIN’ and ‘ALH and VCL’
correctly identified 94.3, 85.7 and 91.4% HA spermatozoa
respectively.
Discussion
Hyperactivation has been described in spermatozoa from many
species after incubation under in-vitro culture conditions, but
not in the rat. Several investigators have studied the movement
of rat spermatozoa during in-vitro culture conditions (Moore
and Akhondi, 1996; Oberländer et al., 1996) but did not report
on HA motility. This study provides the first evidence of
hyperactivation of rat spermatozoa observed during in-vitro
culture conditions typically used for rat IVF. A potentially
important component of the analysis strategy was the use of
a higher image sampling frequency (60 Hz) than was used in
most previous studies (25 or 30 Hz). The increased image
sampling frequency in the current studies provided improved
track definition. Indeed, for uncapacitated rat spermatozoa, all
CASA measures except VSL have been documented to be
extremely sensitive to tracking rate, comparing 60 and 30 Hz
(Dunson et al., 1999). Furthermore, it has been shown that the
use of 60 Hz allows better discrimination than 30 Hz between
forward progressive motility and HA motility in human
spermatozoa (Mortimer and Swan, 1995).
In an attempt to define HA rat spermatozoa objectively,
current CASA guidelines were followed which recommend
the use of multiparametric kinematic definitions to classify
individual spermatozoa into specific subpopulations (ESHRE,
1996, 1998). In doing so, it was postulated that one or more
bivariate models that included both a vigour term and a
progression term would provide a better definition of hyperactivation than reliance on a single parameter. Using a distribution approach, cut-off gates were defined for each CASA
parameter in an objective manner; these were used in bivariate
analysis and it was found that several bivariate models could
be used to define and quantify a subpopulation of HA rat
spermatozoa. The use of such a distribution-based analysis to
Table III. Mean ⫾ SEM percentages of spermatozoa that meet the hyperactivated criteria using various bivariate models
4 mg/ml BSA
15 mg/ml BSA
Incubation
period (h)
Hyperactivated
(subjective)
0
4
0
4
0.0
3.6
0.1
3.7
⫾
⫾
⫾
⫾
0.0
1.0a
0.1
0.8a
VCL and LIN
2.8
24.6
2.3
18.9
⫾
⫾
⫾
⫾
0.9
5.0a
1.0
3.1a
ALH and LIN
4.1
21.6
3.3
18.3
⫾
⫾
⫾
⫾
1.0
4.5a
1.0
2.5a
BCF and LIN
5.6
16.5
3.6
22.0
⫾
⫾
⫾
⫾
1.6
3.9a
1.0
5.4a
VCL and ALH
4.3
19.3
3.1
15.1
⫾
⫾
⫾
⫾
0.6
4.3b
1.5
2.5b
ALH and BCF
3.9 ⫾ 0.7
11.2 ⫾ 2.3a
2.3 ⫾ 0.7
9.9 ⫾ 1.4a
significant differences (P ⬍ 0.01) between 0 and 4 h in the same group.
significant differences at P ⫽ 0.011, between 0 and 4 h in the same group.
For definitions, see Table I.
aIndicates
bIndicates
1325
A.M.Cancel et al.
characterize subtle changes in the motion characteristics of
spermatozoa has been recommended previously (Toth et al.,
1989; Gladen et al., 1991).
The model containing VCL and LIN had the highest
accuracy for defining HA sperm tracks that agreed with visually
classified whiplash sperm tracks. Models containing VAP,
which represents the general trajectory of the spermatozoa
corrected for time, did not identify changes in HA between 0
and 4 h incubation. Indeed, while CASA means for VAP
decreased over time, VAP values from tracks of selected HA
spermatozoa increased during the 4 h incubation. It is important
to note that VAP may impact on other CASA measurements
that are defined with reference to the average path of spermatozoa. The two CASA measurements most likely to be influenced
by VAP are ALH and BCF. In addition, ALH is sometimes
defined as the mean amplitude rather than the maximum
amplitude of the track (Mortimer, 1997). Since VAP and ALH
are defined by an instrument-specific algorithm, and VAP
influences BCF and ALH, bivariate models containing these
parameters may need to be verified when using other CASA
instruments. Based on this study, use of the model containing
VCL and LIN is recommended, since it is very accurate in
defining HA tracks and is based on end-points that are less
likely to vary across different CASA instruments.
The bivariate models described herein categorize as HA
about 20% of the sperm tracks analysed after a 4 h incubation.
These CASA-derived estimates of HA spermatozoa were
higher than the subjective classification (~3%). The most likely
explanation for this difference in percentage HA is that the
subjective evaluation was based on the single, strict criterion
that spermatozoa exhibit whiplash motion during the 1 s
playback option. Therefore, rat spermatozoa exhibiting transitional motion were not scored as HA during the subjective
classification. The transition from progressive to whiplash
motion occurs as sperm tracks become circular and circumscribe a smaller and smaller circumference; at the same time,
the amplitude of the track increases. Furthermore, spermatozoa
may switch back and forth between transitional and whiplash
Figure 2. Bivariate model curvilinear velocity (VCL) and linearity (LIN). Representative sample of a bivariate analysis used to characterize
hyperactivated rat spermatozoa. Open circles (m) represent CASA sperm values at the beginning of the experiment (t ⫽ 0 h) and closed
circles (l) represent CASA sperm values after 4 h incubation. Lines represent 90th or the 10th percentile cut-off points. Hyperactivated
spermatozoa located in the lower right quadrant exceeded the cut-off for VCL and fell below the cut-off for LIN.
Table IV. Accuracy of bivariate models to correctly identify hyperactivated spermatozoa*
VCL and LIN
Hyperactivated (%) 94.3
Progressive (%)
0
ALH and LIN
BCF and LIN
VCL and ALH
ALH and BCF
85.7
0
42.9
0
91.4
0
42.9
0
*Values represent the percentage of subjectively classified hyperactivated or progressive sperm tracks defined
as hyperactivated by each bivariate model.
For definitions, see Table I.
1326
Hyperactivation of rat spermatozoa
motion (Suarez, 1996). Transitional forms of motion are well
described in hamster and human spermatozoa (Suarez, 1988;
Mortimer and Swan, 1995) and were clearly evident in rat
spermatozoa during the conduct of this study. The bivariate
analyses, with the cut-off points set at a statistically defined
place, allow inclusion of transitional spermatozoa based on
their kinematic characteristics. Indeed, a similar approach
using two end-points has been used to evaluate hyperactivation
of mouse (Neill and Olds-Clarke, 1987) and bovine (McNutt
et al., 1994) spermatozoon tracks, while three or more endpoints have been used to characterize HA primate (Mahony
and Gwathmey, 1999), human (reviewed in Mortimer, 1997),
boar (Abaigar et al., 1999) and ram (Vulcano et al., 1998)
sperm tracks. The percentage of rat HA sperm tracks obtained
was low relative to mouse and hamster, where ⬎50% HA
spermatozoa have been reported (Suarez et al., 1991; Suarez
and Dai, 1992; Yanagimachi, 1994; Suarez, 1996). A low
percentage of HA rat sperm was also observed at earlier and
later time points up to 6 h (data not shown). However, the
percentage of HA rat sperm was similar to that observed for
human spermatozoa, where low percentages of HA spermatozoa can be found at any given time during in-vitro culture
conditions.
Increasing the concentration of BSA to 15 mg/ml during
in-vitro culture does not appear to alter the frequency of
detection of HA spermatozoa during CASA analysis. Although
BSA appears to be an important component for rat spermatozoa
capacitation, it does not affect the percentage HA spermatozoa,
indicating that a range of BSA concentrations can be used
when studying hyperactivation.
In summary, it was shown that rat spermatozoa undergo
changes in motion, during in-vitro culture conditions in media
that has been shown to support rat IVF, that are characteristic
of hyperactivation as described in other species. Using CASA
analysis of rat spermatozoa, an increase in mean vigour and a
decrease in mean progression over a 4 h period of time were
found under conditions that also support rat IVF. Finally, it
was demonstrated that bivariate models using two CASA
parameters, one for progression (LIN) and one for vigour
(VCL) can be used to define HA tracks in an objective
and accurate manner. This subpopulation probably represents
spermatozoa with fertilizing ability.
Significant relationships have been successfully determined
between HA of human spermatozoa and IVF or pregnancy
rates in donor insemination (Wang et al., 1993; Johnston et al.,
1994). Indeed, evaluation of HA human spermatozoa has been
suggested as a prognostic end-point to test sperm fertility
(Mortimer, 1997). Standardization of methods, such as the one
described here, allow accurate and objective identification of
HA spermatozoa. These methods can now be further developed
and applied to studies of sperm function and may be useful in
toxicological studies after reproductive toxicant exposure.
Acknowledgements
The authors thank Susan Jeffay for her technical assistance. We also
thank Dr David F.Katz (Department of Biomedical Engineering, Duke
University, Durham, NC) for his helpful discussions. This work was
funded by the EPA/UNC Toxicology Research Program, Training
Agreement CT902908, with the Curriculum in Toxicology, University
of North Carolina at Chapel Hill.
References
Abaigar, T., Holt, W.V., Harrison, R.A.P. et al. (1999) Sperm subpopulations
in boar (Sus scrofa) and gazelle (Gazella dama mhorr) semen as revealed
by pattern analysis of computer-assisted motility assessments. Biol. Reprod.,
60, 32–41.
Boyers, S.P., Davis, R.O. and Katz, D.F. (1989) Automated semen analysis.
Curr. Probl. Obstet. Gynecol. Fertil., 12, 173–199.
Dunson, D.B., Winberg, C.R., Perreault, S.D. et al. (1999) Summarizing the
motion of self-propelled cells: applications to sperm motility. Biometrics,
55, 537–543.
ESHRE Andrology Special Interest Group (1996) Consensus workshop on
advanced diagnostic andrology techniques. Hum. Reprod., 11, 1463–1479.
ESHRE Andrology Special Interest Group (1998) Guidelines on the application
of CASA technology in the analysis of spermatozoa. Hum. Reprod., 13,
142–145.
Evans, G. and Armstrong, D.T. (1984) Reduction in fertilization rat in vitro
of oocytes from immature rats induced to superovulate. J. Reprod. Fert.,
70, 131–135.
Gladen, B.C., Williams, J. and Chapin, R.E. (1991) Issues in the statistical
analysis of sperm motion data derived from computer-assisted systems. J.
Androl., 12, 89–97.
Johnston, R.C., Mbizvo, M.T., Summerbell, D. et al. (1994). Relationships
between stimulated hyperactivated motility of human spermatozoa and
pregnancy rate in donor insemination: a preliminary report. Hum. Reprod.,
9, 1684–1687.
Mahony, M.C. and Gwathmey, T.Y. (1999) Protein tyrosine phosphorylation
during hyperactivated motility of cynomolgus monkey (Macaca fascicularis)
spermatozoa. Biol. Reprod., 60, 1239–1243.
McNutt, T.L., Olds-Clarke, P., Way, A.L. et al. (1994) Effect of follicular or
oviductal fluids on movement characteristics of bovine sperm during
capacitation in vitro. J. Androl., 15, 328–336.
Miyamoto, H. and Chang, M.C. (1973) Fertilization of rat eggs in vitro. Biol.
Reprod., 9, 384–393.
Moore, H.D.M. and Akhondi, M.A. (1996) Fertilizing capacity of rat
spermatozoa is correlated with decline in straight line velocity measured
by continuous computer-aided sperm analysis: epididymal rat spermatozoa
from the proximal cauda have a greater fertilizing capacity in vitro than
those from the distal cauda or vas deferens. J. Androl., 17, 50–60.
Mortimer, S.T. (1997) A critical review of the physiological importance and
analysis of sperm movement in mammals. Hum. Reprod. Update, 3, 403–439.
Mortimer, S.T. and Swan, M.A. (1995a) Kinematics of capacitating human
spermatozoa analysed at 60 Hz. Hum. Reprod., 10, 873–879.
Mortimer, S.T. and Swan, M.A. (1995b) Variable kinematics of capacitating
human spermatozoa. Hum. Reprod., 10, 3178–3182.
Neill, J.M. and Olds-Clarke, P. (1987) A computer-assisted assay for mouse
sperm hyperactivation demonstrates that bicarbonate but not bovine serum
albumin is required. Gamete Res., 18, 1221–1240.
Oberländer, G., Yeung, C.H. and Cooper, T.G. (1996) Influence of oral
administration of ornidazole on capacitation and the activity of some
glycolytic enzymes of rat spermatozoa. J. Reprod. Fert., 106, 231–239.
Pacey, A.A., Davies, N., Warren, M.A. et al. (1995) Hyperactivation may
assist human spermatozoa to detach from intimate association with the
endosalpinx. Hum. Reprod., 10, 2603–2609.
Perreault, S.D. (1998) Gamete toxicology: the impact of new technologies. In
Korach, K.S. (ed.), Reproductive and Developmental Toxicology. Marcel
Dekker, New York, pp. 635–654.
Perreault, S.D. and Jeffay, S. (1993) Strategies and methods for evaluating
the functional capacity of oocytes and zygotes in vitro. In Heindel, J.J. and
Chapin, R.E. (eds), Methods in Toxicology: Female Reproductive Toxicology.
vol. 3B, Academic Press, New York, pp. 92–109.
Shalgi, R. and Phillips, D.M. (1988) Motility of rat spermatozoa at the site
of fertilization. Biol. Reprod., 39, 1207–1213.
Slott, V.L., Suarez, J.D., Poss, P.M. et al. (1993) Optimization of the Hamilton–
Thorn computerized sperm motility analysis system for use with rat
spermatozoa in toxicological studies. Fundam. Appl. Toxicol., 21, 298–307.
Stauss, C.R., Votta, T.J. and Suarez, S.S. (1995) Sperm motility hyperactivation
facilitates penetration of the hamster zona pellucida. Biol. Reprod., 56,
1280–1285.
Suarez, S.S. (1988) Hamster sperm motility transformations during
1327
A.M.Cancel et al.
development of hyperactivation in vitro and epididymal maturation. Gamete
Res., 19, 51–65.
Suarez, S.S. (1996) Hyperactivated motility in sperm. J. Androl., 17, 331–335.
Suarez, S.S. and Dai, X. (1992) Hyperactivation enhances mouse sperm
capacity for penetrating viscoelastic media. Biol. Reprod., 46, 686–691.
Suarez, S.S., Katz, D.F., Owen, D.H. et al. (1991) Evidence for the function
of hyperactivated motility in sperm. Biol. Reprod., 44, 375–381.
Toth, G.P., Stober, J.A., Read, E.J. et al. (1989) The automated analysis of
rat sperm motility following subchronic epicholorohydrin administration:
methodologic and statistical considerations. J. Androl., 10, 401–415.
Toyoda, Y. and Chang, M.C. (1974) Fertilization of rat eggs in vitro by
epididymal spermatozoa and the development of eggs following transfer.
J. Reprod. Fert., 36, 9–22.
Vulcano, G.J., Moses, D.F., Valcárcel, A. et al. (1998) A lineal equation for
the classification of progressive and hyperactive spermatozoa. Math. Biosci.,
149, 77–93.
Wang, C., Lee, G.S., Leung, A. et al. (1993) Human sperm hyperactivation
and acrosome reaction and their relationships to human in vitro fertilization.
Fertil. Steril., 59, 1221–1227.
Yanagimachi, R. (1994) Mammalian fertilization. In Knobil, E., Neill, J.D.
(eds), The Physiology of Reproduction. Raven Press, New York, pp. 189–317.
Received on October 20, 1999; accepted on February 29, 2000
1328