Editor-in-Chief
Prof. Yi-Fei WANG,
Resazurin
reduction and other tests of semen quality and fertility of bulls
Robert
H. Foote
Department
of Animal Science, Cornell University, Ithaca, NY 14853-4801, USA
Asian J Androl 1999 Sep; 1: 109-114
Keywords:
fertility;
semen; spermatozoa; cattle; resazurin reduction
Abstract
Aim: This study was undertaken to compare the reduction in color of two dyes methylene blue (MBRT) and resazurin dye (RRT) with other tests of bull semen quality and to examine their relationship to fertility. Methods: One hundred sixty-four ejaculates from 59 bulls were examined, processed, and used for 30 016 inseminations. Results: Bulls used in artificial insemination have been selected for high semen quality and fertility, and semen from these bulls averaged 80.6% unstained sperm, only 11 % had abnormalities, and fertility ranged from 64% to 76%. The MBRT and RRT were run with standardized sperm numbers to prevent sperm concentration from influencing the dye reduction time. A short RRT was correlated with higher fertility (r=-0.26) but MBRT was not correlated (r=-0.05, P>0.05). The correlations were low, but are typical and reflect the statistical effect of the large binomial variance associated with pregnancy or nonpregnancy being coded as 1 or 0. Conclusion: The fact that the RRT was significantly correlated with fertility when sperm numbers were standardized, and published reports that resazurin is useful for monitoring other semen characteristics, indicate that this dye may provide valuable information not given by other simple laboratory tests.1 Introduction
The
routine testing of semen quality in humans[1] and animals[2,3]
is usually limited
to tests that can be sun quickly and simply, such as monitoring sperm
motility, concentration, and morphology. Morphology appears to be particularly
important with human sperm because the proportion of abnormals is high[4].
With the development
of computer-assisted sperm analysis (CASA), sophisticated measurements
of sperm motion also can be performed rapidly[5,6]. Metabolic
tests such as fructolysis and oxygen consumption are important measures
of sperm function[2,7], but these tests are not conducted routinely
because of their complexity[1,3]. However these assays are
highly correlated with a simple test that reflects sperm metabolism, such
as the reduction of methylene blue (MBRT) to the reduced colorless form
by accepting two hydrogen atoms during sperm metabolism[7].
Furthermore, the MBRT of bull sperm is correlated with fertility[2,8].
Erb
et al[9,10] evaluated sperm, utilizing the redox dye,
resazurin, which changes
from blue to pink as it is reduced to resorufin, and to colorless hydroresorufin
upon further reduction. This resazurin reduction test (RRT), like the
MBRT, requires little equipment and both are simple to carry out. These
workers reported a significant correlation between RRT and fertility (r=-0.17),
of high fertility
bulls, but not MBRT with fertility. When bull fertility ranged from 22%
to 63% the correlation with RRT was -0.74 (P<0.05). Despite
the promising results published with bull sperm these tests have not been
widely used.
2 Materials and methods
Three
ejaculates of semen were collected with an artificial vagina at 3-4-d
intervals from each of 59 bulls in regular service at the New York Artificial
Breeders' Cooperative Inc. (now Genex Coop. Inc., Ithaca, NY). Following
various tests of semen quality, 164 ejaculates were selected for use which
contained a sufficient number of motile sperm to be processed to meet
the needs for insemination. Semen volume was measured and the ejaculate
was held at 37℃[17] while subsamples were taken to estimate
the sperm concentration, the percentage of motile sperm, and to prepare
slides for staining with fast green FCF-eosin Y. A subsample of sperm
was diluted and the concentration measured with a calibrated spectrophotometer.
The correlation between sperm concentration measured with the spectrophotometer
and the Coulter Counter is r=0.96. The percentage of motile sperm
was estimated subjectively on semen diluted so that individual sperm could
be visualized when placed on a slide, coverslipped, and examined at 37℃
at a magnification
of 430×[17]. The rate of progressive movement was estimated
in 0.5 increments
on an arbitrary scale of 0 to 4, with 4 being the most rapid movement.
The repeatability of these subjective estimates is 0.75. Two slides were
prepared with a mixture of sperm and fast green FCF-eosin Y dye spread
thinly over both slides. They were dried quickly on a warming plate at
40℃. Abstracts of these procedures are published[18].
The
semen extender was composed of eighty parts of 2.9% (wt/vol) sodium citrate
dihydrate, 0.3% (wt/vol) of sulfanilamide, 500 μg of streptomycin and
500 IU of penicillin per mL and 20 parts of egg yolk. Semen was mixed
1:4 (vol/vol) at 37℃ and cooled to 5℃ over a period of 2 h. After cooling
to 5℃ it was further extended with 5℃ extender so as to usually provide
10 million motile sperm per insemination for the majority of bulls that
were Holsteins. For non-Holstein breeds
semen was only extended to meet daily needs, resulting in more than 10
million sperm per insemination.
All
of the inseminations were done with unfrozen semen. This has a potential
advantage over frozen semen in that the relationship of initial semen
quality to fertility is not complicated by any interaction with freeze-thaw
damage.
The
extender used for the resazurin test was the same egg yolk-citrate as
used for insemination, except that it contained resazurin (Allied Dye
Corp., NY, USA). During preliminary tests it was determined that 22 mg
of resazurin per 100 mL
of extender gave more rapid and sharper endpoints than the 11 mg of resazurin/200
mL of phosphate buffer used by Erb et al[10]. Based
upon the initial sperm
concentration in the ejaculate, the sperm were diluted to 200×106
total sperm per mL of egg yolk-citrate extender containing 22 mg of resazurin.
Tubes containing 2 mL were filled, sealed and incubated in a waterbath
at 45℃ to measure the resazurin reduction time (RRT). This temperature
promoted a short practical RRT time from the pink oxidized resazurin to
the yellow color of the egg yolk, with a minimal effect on motility of
the sperm. Likewise, the MBRT was
conducted with 50 mg of methylene blue per 100 mL of extender and 200×106 sperm/mL.
Control tubes with only the yellow egg yolk extender were included and
the test for each
sample of semen was terminated when the color matched the control. The
time was recorded to the nearest second. Repeatability of duplicate measurements
of both RRT and MBRT was 0.94. The other tests of semen quality were performed
as described here and by Bratton et al[2].
After
finishing the daily semen processing the FCF-eosin Y stained slides were
examined as soon as possible after preparation. One hundred sperm on each
of two slides were
classified as having completely stained heads, partially stained heads
and no staining of the head. Also 200 sperm were classified as having
normal or misshapen heads, tailless heads, or coiled or bent tails. All
examinations were done at 440×.
The
cooled semen (5℃) was insulated, packaged, and distributed rapidly to
a cross section
of professional inseminators. All cows requiring one or two inseminations
and not requiring reinsemination within 4-5 weeks (28-35 d), or within
2-3 months (60-90 d) were recorded. A total of 30 016 inseminations were
performed.
Insemination
of 30 016 cows is a large experiment. However, when distributed over a
total of 164 ejaculates of semen the repeatability of fertility measurements
on each ejaculate is not high. Much of this random variation in estimating
fertility is due to the binomial variance as nonpregnancy is coded as
0 and a pregnancy as 1. When fertility is 50%, this binomial variance
is equivalent to that of
flipping coins. While the true coin ratio of heads to tails is 50%, with
small sample size the deviations from 50:50 are great.
Sample
variation from the true value, expressed as a standard deviation (s)
is calculated by√pq
/n, where p is the pregnancy rate, q=1-p,
and n=sample size. If the pregnancy rate is 50%, the s with
sample sizes of 10, 25, 100, and 200 are 16.8, 10.0, 5.0 and 3.5. As the
mean±2s includes
about 95% of a normal distribution it is obvious that an accurate estimate
of fertility requires hundreds of breedings.
A
dummy uniformity trial was undertaken with 9 000 inseminations distributed
equally across 25 bulls, so that there were 360 inseminations per bull.
These inseminations
were split randomly into groups containing 180, 60, 30, and 15 per group,
and correlations among groups of different sample size were calculated.
The correlations between samples from the same pool decreased as sample
size decreased, being 0.70, 0.42, 0.14, and 0.05, respectively. Therefore,
tests of semen quality with a high biological relationship to fertility
will have a lower statistical relationship because the repeatability of
the fertility measurement is much less
than 1.
3 Results
Several
semen quality tests are characterized
with the standard errors
and ranges given based on individual ejaculates (Table 1). Because the bulls
were highly selected for use in artificial insemination, the semen usually
was of high quality with good sperm motility, only 1% primary head abnormalities,
and a high percentage of unstained cells. The tailless heads (3.7%) were
primarily an artefact
of slide preparation, as very few tailless heads were seen in wet smears
viewed at 440×. The fertility rates were very good, with the decline
from the 28-35-d percentages
to the 60-90-d values representing typical early losses of pregnancies.
These losses are not associated with semen from individual bulls selected
to be free from genetic defects and disease, but are due to cow, management
and other factors in the field.
Two
sets of correlations are given in Table 2. On the left of the table are
the correlations between
measurements of different traits based upon individual ejaculate data.
The highest correlations are between MBRT and RRT, and the percentage
of motile sperm with MBRT and RRT. The sperm concentration of the ejaculate was
not correlated with MBRT and RRT because sperm concentration was standardized
for these tests. The low correlation with fertility partly reflects sampling variation
and the large binomial variance previously described associated with a
pregnancy being 0 or 1.
Table
1. Various tests of semen quality and measures of fertility based upon
164 ejaculates.
|
Tests
of semen quality and fertility |
mean±SEM |
Rangeb |
| Variable
measured |
||
| X1
Volume of ejaculate (mL) |
8.80±0.21 |
3.8-18.0 |
| X2
Motile sperm (%) |
64.90±0.55 |
40.0-80.0 |
| X3
Motile sperm rate |
3.40±0.02 |
3.0-4.0 |
| X4
Sper m concentration (106/mL) |
1593±33 |
752.0-2759.0 |
| X5
Extension rate |
73.1±2.6 |
13.0-156.0 |
| X6
Final motile sperm (106/mL) |
14.1±1.0 |
8.8-36.0 |
| X7
MBRT (min) |
6.40±0.08 |
4.5-11.0 |
| X8
RRT (min) |
4.60±0.06 |
3.3-7.3 |
| X9
Tail abnormalities (%) |
6.50±0.27 |
2.0-24.0 |
| X10
Tailless heads (%) |
3.70±0.23 |
1.0-24.0 |
| X11
Head abnormalities (%) |
1.00±0.07 |
0.0-5.0 |
| X12
Total abnormalities (%) |
11.30±0.38 |
4.0-31.0 |
| X13
Unstained sperm (%) |
80.60±0.41 |
66.0-91.0 |
| X14
Wholly stained sperm (%) |
11.60±0.35 |
3.0-28.0 |
| X15
Partially stained sperm (%) |
7.70±0.23 |
2.0-18.0 |
| Fertilitya |
||
| 28-35-day
(%) |
81.30±0.47 |
47.0-99.0 |
| 60-90-day
(%) |
70.60±0.51 |
35.0-91.0 |
aBased
upon cows not requiring reinsemination within 5 weeks (28-35 d) or 2-3
months (60-90 d).
bRanges are for individual ejaculates of semen. Averaged over
bulls, ranges are much less. For example bull fertility ranged from 64%
to 76%.
Table
2. Correlation coefficients between different tests of semen quality and
fertility.
|
Variable |
Variable
correlateda |
||||||
|
Motile
sperm(%) |
Initial
sperm per mL |
MBRT |
RRT |
Total
abnormals |
Unstained |
Bull
fertility(%)b |
|
|
X2
Motile sperm (%) |
1.00 |
0.08 |
0.19 |
0.25 |
-0.38 |
0.26 |
0.08 |
|
X4
Sperm per mL |
-0.05 |
1.00 |
0.12 |
0.03 |
0.14 |
-0.10 |
0.36 |
|
X7
MBRT (min) |
0.51 |
0.09 |
1.00 |
0.88 |
0.02 |
-0.40 |
-0.05 |
|
X8
RRT (min) |
0.41 |
0.11 |
0.86 |
1.00 |
-0.04 |
-0.34 |
-0.26 |
|
X12
Total abn (%) |
-0.02 |
0.15 |
0.14 |
-0.08 |
1.00 |
-0.17 |
-0.09 |
|
X13
Unstained (%) |
0.01 |
0.01 |
-0.13 |
-0.17 |
-0.24 |
1.00 |
-0.08 |
|
Fertilityb
(%) |
0.06 |
0.04 |
0.10 |
-0.20 |
-0.08 |
0.10 |
1.00 |
aLeft
side of the table are correlations on totals for all 164 ejaculates (n-2=162;
r≥0.15,P<0.05), and on the right side are correlations
based on the means for 59 bulls (n-2=57; r≥0.26, P<0.05
for among-bull values).
bFertility is based on cows not requiring reinsemination within
2-3 months and assumed to be pregnant.
The
fertility was averaged across the ejaculates of each bull to estimate
bull fertility and
this was correlated with the average semen characteristics for each bull
(right half of Table 2). The initial sperm concentration in the ejaculate and
the RRT were significantly correlated with fertility (P<0.05).
As the number of sperm inseminated was standardized, the fact that the
among-bull correlation between initial sperm concentration and fertility(r=0.36)
was significant indicates
that bulls producing a higher concentration of sperm in their semen tend
to produce more fertile sperm. The MBRT was not correlated with fertility.
While these correlations are low, for reasons discussed, they are consistent
with other reports[2,8] and a vast published literature[18],
particularly where the
bulls selected for use in artificial insemination have a narrow range
in fertility. In this study the range in fertility was 64% to 74%.
4 Discussion
Among
the many tests of semen quality studied, the most commonly used ones for
many years have been sperm concentration and total sperm count, the percentage
of motile sperm and
sperm morphology[1,3,18,19]. When sperm are stained for morphology
a supravital or“live-dead” stain can be used to measure plasma membrane
integrity as well. Although the percentage of motile sperm was estimated
by an experienced worker, the repeatability of these subjective estimates
is only about 0.75. This contrasts with the repeatability of the quantitative
estimate of sperm concentration which exceeds 0.95 repeatedly, and most
other characteristics estimated quantitatively, where r≥0.90.
The
semen quality of these bulls retained for use in artificial insemination
was generally high(Table
1), as bulls with poor semen quality are not retained for use.
However, with 164 ejaculates of semen collected there was a substantial
range of values. To eliminate the variation in RRT and MBRT due to sperm
numbers, the RRT and MBRT were conducted with semen diluted to a standard
number of motile sperm.
The
estimated percentage of progressively motile sperm (Table 1) was conservative
compared with the percentage of unstained sperm cells. This is consistent
with the published literature. However, since the present study was completed
we have been able, for the first time, to account for most of the difference
between unstained
sperm and the lower percentage of progressively motile sperm cells[20].
Unstained sperm conventionally have been based on the appearance of the
sperm head. When the proportion of sperm with stained midpieces is added
to the stained heads the percentage of unstained sperm is decreased, more
comparable to the percentage of motile cells.
The
percentage of abnormal sperm was very low, totaling only 7.6 % when the
tailless heads are excluded as probably artefacts (Table 1). There was
no relationship between abnormal sperm and fertility, as bulls with abnormal
sperm and low fertility are culled. This contrasts with human sperm[4],
where the reports often involve a series of subfertile patients with many
abnormal sperms.
The
MBRT and RRT were the criteria of special interest in the present study.
Earlier work[7] had shown that the MBRT, much as was run here,
was highly correlated with bull sperm concentration (-0.81), the percentage
of motile sperm (-0.63), glucose loss (-0.78), and lactic acid gain (-0.75).
However, the significant
correlations with fertility of -0.22 and -0.28 were low[2].
The
low correlations between any measurements of semen quality and fertility
are typical of the
studies where the males are selected to produce good quality semen and
the range in fertility is small[2]. This could be interpreted
as meaning that there is only a small biological relationship. However,
the statistical scatter of fertility estimates with moderate sample size,
decreases the correlations, as previously discussed. With 100 inseminations
per subsample the 95% confidence limits on fertility due to binomial variance
alone are ±10%, so the true fertility of a sample estimated to be 70%
probably is somewhere between 60% and 80%. This range is wider than the
range of fertility estimates of the bulls in this study. Erb and Ehlers[9]
obtained low correlations between RRT until Erb et al[10]
located bulls ranging in fertility from 24% to 63%. Then Erb et al[10]
obtained a correlation between RRT and fertility of -0.736.
The
RRT is highly correlated with sperm concentration and the percentage of
motile sperm from several species[12,14,15]. Although fertility
information usually has been lacking in these reports, insofar as low
sperm numbers and low sperm motility are responsible for reduced fertility,
the test is a valuable monitor of potential fertility of an ejaculate.
Fuse et al[13] reported that the RRT was more useful
than most other criteria in evaluating fertility potential of human semen.
The
ATP regenerating capacity of sperm is another indication of semen quality[11].
Mahmoud et al[15], found that the RRT was as accurate
and predictive of men producing high quality semen as was measuring ATP.
Zalata et al[16] found that the RRT was correlated with
γ-glutamyltransferase and
that the test could distinguish between semen samples in which sperm produced
varying amounts of reactive oxygen species. The reactive oxygen species
can cause lipid peroxidation of the sperm membrane[21], and
high reactive oxygen species production has been reported to be associated
with poor sperm function, including DNA fragmentation[22-24].
The DNA fragmentation is of concern where intracytoplasmic sperm injection
is used with sperm from subfertile males.
Neutrophils
appear to be a major source of the reactive oxygen species produced in
human semen[25]. Resazurin is a useful indicator of neutrophil
activity[26] as it serves as an electron receptor from nicotinamide
adenine dinucleotide phosphate (NADPH) produced by neutrophil activity.
The blue resazurin can be assayed
with great sensitivity fluorometrically as it is reduced to pink fluorescent resorufin.
Where fluorometers are available the assay can be run with high precision.
In
summary, the RRT has been reported to be highly correlated with sperm
concentration and the proportion of motile sperm in semen. It is a test
that reflects metabolic activity of the sperm cell, such as ATP, and the
RRT test is simpler to run. It has been reported to be a useful measure
of undesireable components or contaminants of semen, such as reactive
oxygen species and neutrophils. Also, when
bulls ranged greatly in fertility the RRT was highly correlated with their fertility.
In the present report it was one of the few statistically significant
semen characteristics correlated with fertility, where only a narrow range
in fertility existed.
Although further work is desirable, valuable information may be obtained
by clinical and research andrology laboratories adding this simple test
to augment the current battery of tests used to evaluate semen.
5
Acknowledgements
References
[1]
World Health Organization. WHO laboratory manual for the
examination of human
semen and sperm-cervical mucus interaction. 3rd ed. Cambridge: Cambridge
Univ Press; 1992. p 107.
[2] Bratton RW, Foote RH, Henderson CR, Musgrave SD, Dunbar RS Jr,
Dunn HO, et al.
The relative usefulness of combinations of laboratory tests for predicting
the fertility of bovine semen. J Dairy Sci 1956; 39: 1542-9.
[3] Seed J, Chapin RE, Clegg ED, Dostal LA, Foote RH,
Hurtt ME, et al. Methods for assessing sperm motility, morphology,
and counts in the rat, rabbit, and dog: a consensus report. Reprod Toxicol
1996; 10: 237-44.
[4] Kruger TF, Acosta AA, Simmons KF, Swanson RJ, Matta JF, Oehinger S.
Predictive value of abnormal sperm morphology in in vitro fertilization.
Fertil Steril 1988; 49: 112-17.
[5] Farrell PB, Foote RH, Zinaman MJ. Motility and other characteristics
of human sperm can be measured by computer-assisted sperm analysis of
samples stained with Hoechst 33342. Fertil Steril 1996; 66: 446-53.
[6] Tardif AL, Farrell PB, Trouern-Trend V, Simkin ME, Foote RH. Use of
Hoechst 33342 stain to evaluate live fresh and frozen bull sperm by computer
assisted analysis. J Androl 1998; 19: 201-6.
[7] VanDemark NL, Mercier E, Salisbury GW. The methylene blue reduction
test and its relation to other measures of quality in bull semen. J Dairy
Sci 1945; 28: 121-8.
[8] Branton C, James CB, Patrick TE, Newsom MH. The relationship between certain
semen quality tests and fertility and the interrelationship of these tests.
J Dairy Sci 1951; 34: 310-6.
[9] Erb RE, Ehlers MH. Resazurin reducing time as an indicator of bovine
semen fertilizing capacity. J Dairy Sci 1950; 33: 853-64.
[10] Erb RE, Ehler MH, Flerchinger FH. Modified resazurin reduction test
for estimating fertilizing capacity of bull semen. J Dairy Sci 1952; 35:
881-8.
[11] Comhaire FH, Vermeulen L, Schoonjans F. Reassessment of accuracy of
traditional sperm characteristics and adenosine triphosphate (ATP) in
estimating the fertilizing
potential of human semen in vivo. Int J Androl 1987; 41: 653-62.
[12] Dart MG, Mesta J, Crenshaw C, Ericsson SA. Modified resazurin reduction
test for determining the fertility potential of bovine spermatozoa. Arch
Androl 1994; 33: 71-5.
[13] Fuse H, Ohta S, Katayama T. Varicocelectomy: changes in resazurin
reduction test. Arch Androl 1993; 31: 49-54.
[14] Glass RH, Ericsson SA, Ericsson RJ, Drouin MT, Marcoux LJ, Sullivan
H. The resazurin reduction test provides an assessment of sperm activity.
Fertil Steril 1991; 56: 743-6.
[15] Mahmoud AM, Comhaire FH, Vermeulen L, Andreon E. Comparison of the
resazurin test, adenosine triphosphate in semen, and various sperm parameters.
Hum Reprod 1994; 9: 1688-93.
[16] Zalata A, Hafez T, Mahmound A, Comhaire F. Relationship between reduction
test, reactive oxygen species generation, and γ-glutamyltransferase.
Hum Reprod 1995; 1136-40.
[17] Liu Z, Foote RH. Bull sperm motility and membrane integrity in media
varying in osmolality. J Dairy Sci 1998; 81: 1868-73.
[18] Foote RH. Artificial Insemination to Cloning. Tracing 50 Years of
Research. Ithaca: Cornell University Resource Center; 1998. p 231.
[19] MacLeod J, Wang Y. Male fertility potential in terms of semen quality:
review of the past, a study of the present. Fertil Steril 1979; 31: 103-16.
[20] Nagy S, Házas G, Bali Papp Á, Iváncsics J, Szász F, Szász
F Jr, et al. Evaluation of sperm tail membrane integrity by light
microscopy. Theriogenology 1999; 52: in press.
[21] Alvarez JG, Touchstone JC, Blasco L, Story BT. Spontaneous lipid peroxidation
and production of hydrogen peroxide and superoxide in human spermatozoa.
J Androl 1987; 8: 338-48.
[22] Aitken RJ, Buckingham D. Enhanced detection of reactive oxygen species
produced by human spermatozoa with 7-dimethylamino-naphthalin-1,2-dicarbonic
acid hydrazide. Int J Androl 1992; 15: 211-9.
[23] Aitken RJ, Clarkson JS. Cellular basis of defective sperm function
and its association with the genesis of reactive oxygen species by human
spermatozoa. J Reprod Fertil 1987; 81: 459-69.
[24] Aitken RJ, Gordon E, Harkiss D, Twigg JP, Milne P, Jennings Z, et
al. Relative impact of oxidative stress on the functional competence
and genomic integrity of human spermatozoa. Biol Reprod 1998; 59: 1037-46.
[25] Kessopoulou E, Tomlinson MJ, Barratt CLR, Bolton AE, Cooke ID. Origin
of reactive oxygen species in semen: spermatozoa or leukocytes? J Reprod
Fertil 1992; 94: 463-70.
[26] Fang W, Myllys V, Sandholm M. Resazurin reduction as a function of
respiratory burst in bovine neutrophils. Am J Vet Res 1997; 58: 601-7.
Correspondence
to Dr. Robert H. Foote.
Tel: +1-607-255 2866 Fax: +1-607-255 9829
E-mail: dgb1@cornell.edu
Received
1999-03-06 Accepted 1999-08-24
