This article appeared in a journal published by Elsevier. The attached
copy is furnished to the author for internal non-commercial research
and education use, including for instruction at the authors institution
and sharing with colleagues.
Other uses, including reproduction and distribution, or selling or
licensing copies, or posting to personal, institutional or third party
websites are prohibited.
In most cases authors are permitted to post their version of the
article (e.g. in Word or Tex form) to their personal website or
institutional repository. Authors requiring further information
regarding Elsevier’s archiving and manuscript policies are
encouraged to visit:
http://www.elsevier.com/copyright
Author's personal copy
Available online at www.sciencedirect.com
Theriogenology 69 (2008) 1033–1038
www.theriojournal.com
Technical note
Comparison of three staining techniques for the morphometric
study of rainbow trout (Oncorhynchus mykiss) spermatozoa
V.M. Tuset a,b,*, G.J. Dietrich b, M. Wojtczak b, M. Słowińska b,
J. de Monserrat c, A. Ciereszko b
a
Departamento de Biologı́a Pesquera, Instituto Canario de Ciencias Marinas, P.O. Box 56, E-35200 Telde (Las Palmas),
Canary Islands, Spain
b
Department of Semen Biology, Institute of Animal Reproduction and Food Research of Polish Academy of Science,
ul. Tuwina 10, 10-747 Olsztyn, Poland
c
Unidad de Análisis por Imagen, Laboratorio de análisis Dr. Echevarne, Barcelona, Spain
Received 27 April 2007; received in revised form 20 December 2007; accepted 12 January 2008
Abstract
This study was designed to compare the performance of the kits Diff-Quick, Hemacolor and Spermac for staining the
spermatozoa of rainbow trout. Automated sperm morphology analysis (ASMA) was performed using two image analysis programs
to determine the sperm measurements: head size (length, width, area and perimeter), shape (ellipticity, rugosity, elongation and
regularity) and tail length. Diff-Quick was found to be the best procedure for staining the trout spermatozoa. The use of this method
rendered the highest number of cells correctly analyzed, and provided good colour intensity and contrast of the sperm head. No
differences among the methods were detected in terms of tail length measurements. Mean values established using Diff-Quick for
the main morphometric variables were: head length 2.93 0.13 mm; head width 2.33 0.15 mm and tail length 34.16 1.66 mm.
Based on these findings, we recommend the Diff-Quick staining kit for its accurate and reproducible morphometric results.
Notwithstanding, when analyzing the sperm tail of the rainbow trout, the Spermac method offers improved contrast.
# 2008 Published by Elsevier Inc.
Keywords: Staining Techniques; Morphometry; Spermatozoa; Rainbow trout; ISAS1
1. Introduction
The rainbow trout Oncorhynchus mykiss (Walbaum,
1792) is the most commonly farmed fish species.
Numerous studies have focussed on several of its
reproduction features, particularly sperm biochemistry
and physiology, as well as the short- and long-term
storage of rainbow trout semen [1–6]. The results of
* Corresponding author at: Departamento de Biologı́a Pesquera,
Instituto Canario de Ciencias Marinas, P.O. Box 56, E-35200 Telde
(Las Palmas), Canary Islands, Spain. Tel.: +34 928 132 900;
fax: +34 928 132 908.
E-mail address: victorta@iccm.rcanaria.es (V.M. Tuset).
0093-691X/$ – see front matter # 2008 Published by Elsevier Inc.
doi:10.1016/j.theriogenology.2008.01.012
these investigations have improved production efficiency through broodstock selection or milt cryopreservation based on identifying the highest quality sperm
in terms of their motility, speed and fertilizing capacity.
Despite sperm quality in fishes being also determined
by their morphology, data on fish sperm morphology are
scarce due to methodological limitations.
Sperm morphological variables are usually established by staining sperm samples, and examining the
slides under a microscope with the 100 non-phase
contrast lens and correctly adjusting field brightness
followed by analysis of captured images [7]. Contrast
techniques, especially when automatic detection
systems are employed, have to be optimized before
Author's personal copy
1034
V.M. Tuset et al. / Theriogenology 69 (2008) 1033–1038
designing an efficient protocol (depending, for example,
on the species, whether the semen is fresh or
cryopreserved and whether an extender is used [8]).
When examining mammalian sperm, this type of
preliminary analysis has proved essential [8–10].
However, few studies have tried to assess the
efficiency of the staining techniques available for fish
sperm. The purpose of this study was to establish the
optimal staining technique for morphometric analysis
of rainbow trout spermatozoa using an automated
computerized integrated semen analysis system
(ISAS1).
2. Material and methods
2.1. Sperm collection
Milt was obtained from five 3-year-old rainbow
trout (spring spawning) reared at the Department of
Salmonid Research, Rutki, Poland. Milt was collected
by striping the fish anesthetized with propiscin
(1 ppm, IRS, Żabieniec, Poland). Special care was
taken to avoid contamination of milt with urine.
Samples were stored in an insulated biocontainer
(TMVLN EPS, Termovial, Kern Frio, S.A., Barcelona, Spain) until analysis. The study protocol was
approved by the Animal Experiments Committee in
Olsztyn, Poland.
2.2. Staining techniques
Semen was diluted 1:100 in 3% citrate sodium. The
diluted sperm was deposited in an Eppendorf tube and
centrifuged for 15 s at 300 g. For each of the five
individual milt samples, nine smears were prepared by
placing a 5 ml aliquot onto a slide and pulling out into
a smear using a second slide followed by air drying for
20–30 s. The staining kits used were: Diff-Quick1
(DQ) (Medion Diagnostics GmbH, Düdingen, Germany), Hemacolor1 (HC) (Merck KGaA, Darmstadt,
Germany) and Spermac1 (Stain Enterprises Inc.,
Wellington, RSA). Each kit was used to stain three
smears of the nine prepared from each milt sample.
The following modifications were made to the
procedures recommended by the manufacturers: for
Hemacolor, the fixing time was 10 min and staining
time was 5 min; for Diff-Quick fixing and staining
times were 5 min. In all cases, smears were washed in
distilled water to eliminate excess stain, air dried,
covered with a coverslip and permanently sealed with
Eukitt mounting medium (Kindler & Co., Freiburg,
Germany).
2.3. Head morphology
In each smear, 100 spermatozoa were randomly
captured and subjected to automated sperm morphology
analysis (ASMA) using the sperm morphometry
module of the ISAS1 (Proiser R+D SL, Buñol, Spain).
Slides were viewed under an Olympus BX50 microscope equipped with a 100 bright field objective and
images were captured by a digital video camera (Basler
A310, Vision Technologies, Basler AG, Germany). The
sperm head measurements calculated automatically by
ISAS1 included the size variables: length (L, in mm),
width (W, in mm), area (A, in mm2), and perimeter (P, in
mm); and shape variables: ellipticity (L/W), rugosity
(4pA/P2), elongation ((L W)/(L + W)) and regularity
(pLW/4A). The best staining technique was determined
in terms of the percentage of cells correctly analyzed,
variability of parameters and correlations among stains
for each variable [8].
2.4. Tail morphology
Using the 100 lens, flagellum length was measured
as the distance from its insertion point to the end of the
main section (in salmonids the flagellum is comprised
of two sections: a main, longer and visible segment, and
an end piece, which is narrower, shorter and less visible
by light microscopy [11]). Thirty tails were measured
per milt sample (10 per smear) [12] and staining
technique. Images were analyzed using the Image-Pro
Plus version 4.1.0 package (Media Cybernetics L.P.,
Carlsbad, USA).
2.5. Statistical analysis
To determine if the measurements made were
influenced by the handling procedures, coefficients of
variability (%) were calculated for the sperm head size
variables (length, width, area and perimeter) among
smears corresponding to each individual milt sample
and staining technique.
For multiple comparisons among staining techniques, normality distributions and variance homogeneity
were checked by the Kolmogorov–Smirnov and Levene
tests, respectively. For data showing a normal distribution, one-way ANOVA was performed, followed by a
Tukey post hoc test. For non-normally distributed
variables, Kruskal–Wallis analysis was performed and
the Mann–Whitney test applied for pairwise comparisons with Bonferroni correction (P < 0.017). Finally, a
subsample consisting of measurements made on 450
randomly selected sperm heads for each staining
Author's personal copy
V.M. Tuset et al. / Theriogenology 69 (2008) 1033–1038
1035
technique was used to calculate Pearson’s correlation
coefficients.
3. Results
Sperm heads stained with Diff-Quick were more
susceptible to morphometric analysis in that measurements could be made in 98.8% of the sperm heads
compared to 90.7% for Hemacolor and 76.1% for
Spermac. Spermac was unable to stain all cells with a
similar intensity, which hindered morphometric analysis. In contrast, sperm heads stained with Diff-Quick
and Hemacolor showed a good intensity of staining and
good contrast (Fig. 1).
Coefficients of variability revealed that sperm size
did not vary among the five smears for each staining
technique, indicating the handling procedure did not
affect the final results. Similar coefficients were
obtained for Hemacolor and Diff-Quick, while Spermac
showed greater variability (Table 1). Using a nonparametric test, all the morphometric variables were
observed to vary significantly according to the staining
method with the exception of the factors sperm head
length and regularity provided by Diff-Quick and
Spermac. The Hemacolor technique gave rise to the
highest head size variable values, while using this stain
the shape of the head appeared slightly less elliptical
and elongated (Table 2). Our Pearson’s correlation
analysis revealed that head parameters determined after
staining with Hemacolor were less correlated to values
provided by the other techniques, while the data
obtained using Diff-Quick and Spermac were strongly
correlated (Table 3).
In many cases, it was impossible to distinguish
between the mid-piece and flagellum. Using Hemacolor
and Spermac, contrast for examining the tail was
improved over the contrast obtained using Diff-Quick
(Fig. 1). However, tail length measurements failed to
differ among the three staining techniques (ANOVA,
F 2, 447 = 2.957, P > 0.05) (Table 2). Thus, tail lengths
ranged from 30.04 to 39.26 mm, with a coefficient of
variability of 24.5%.
Fig. 1. Sperm morphology of rainbow trout using the three staining
techniques, Hemacolor, Diff-Quick and Spermac.
4. Discussion
Our study demonstrates that fish sperm can be
stained and morphometrically analyzed using an
automated sperm morphology system. To select an
appropriate staining procedure, preliminary tests need
to consider the effects of different extenders, and
drying, fixation and staining times. Moreover the results
of staining, such as appropriate grey-level contrast for
accurate morphometric analysis [13–15], need to be
assessed. Current knowledge on staining techniques for
fish sperm is scarce. Thus, Howell and Butts [16]
stained fish spermatozoa using a silver stain with
limited success. Wirtz and Steinmann [17] described the
use of Diff-Quick on perch sperm, although they only
measured sperm tail length under phase contrast.
Author's personal copy
1036
V.M. Tuset et al. / Theriogenology 69 (2008) 1033–1038
Table 1
Coefficient of variation (CV, %) for morphometric measurements of
spermatozoa head by staining technique
Size parameters
Individual
Hemacolor
Diff-Quick
Length (mm)
1
2
3
4
5
3.050
3.025
3.199
3.562
3.399
3.462
3.525
3.627
3.997
3.455
4.989
5.736
5.775
4.656
3.882
3.245
3.615
5.023
3.936
5.234
5.141
5.936
4.816
4.287
6.002
5.456
5.000
5.328
5.799
8.511
8.718
7.910
8.134
5.013
5.221
7.795
4.574
5.501
5.669
6.820
5.986
4.646
6.967
6.025
6.205
5.975
7.174
11.266
11.269
9.773
9.191
5.701
5.987
9.697
2.282
2.705
2.876
3.254
2.914
2.386
3.342
2.978
3.048
2.942
3.753
5.429
5.487
4.810
4.554
2.805
2.944
4.801
Total
Width (mm)
1
2
3
4
5
Total
2
Area (mm )
1
2
3
4
5
Total
Perimeter (mm)
1
2
3
4
5
Total
Spermac
We speculate that the limited success of the
procedures used to stain fish spermatozoa is attributable
to numerous artefacts arising during the preparation of
smears. In this study, we observed that by reducing the
air drying and fixation times recommended by the
manufacturers of the staining kits, the morphology of
rainbow trout spermatozoa could be assessed. We
believe that these steps are critical for the successful
staining of fish spermatozoa. Optimizing the air drying
and fixation steps for each fish species should be a
primary focus of future studies.
Another way of performing a morphometric analysis
of fish sperm is to fix the spermatozoa in glutaraldehyde
and examine the sperm using the contrast phase lens
under low magnification (20 or 40) [11,12,18–21].
Thus, phase contrast acts as a ‘stain’, enabling
morphometric measurements. The main technical drawbacks of this method are floating cells and the appearance
of a phase contrast ring around the sperm head [7].
Marco-Jiménez et al. [15] recently obtained good results
using negative phase contrast at 100 to examine
European eel spermatozoa. Accordingly, phase contrast
and staining techniques can be used for morphometric
studies on teleost fish spermatozoa. However, for lower
groups of fish such as sharks, lampreys or sturgeons, the
staining method is recommended since these spermatozoa have an acrosome [22–23].
Using Hemacolor and Diff-Quick, the sperm head
was evenly colored and its shape was also uniform. The
head was accurately delineated such that the measurements made using the ISAS1 showed good precision
and reproducibility. Notwithstanding, the Spermac
staining kit provided the best contrast for examining
the tail and it was even possible to measure the length of
the end piece. Although similar head shape variables
were provided by all the stains, low correlation was
observed among stains. This was due to the effects of
Table 2
Shape and size variables of spermatozoa head and tail lengths by staining technique
Variables
Head
n
Length (mm)
Width (mm)
Area (mm2)
Perimeter (mm2)
Ellipticity
Rugosity
Elongation
Regularity
Tail
n
Length (mm)
Staining technique
Hemacolor
Diff-Quick
Spermac
1360
3.22 0.13a
2.67 0.17a
7.25 0.59a
9.99 0.39a
1.21 0.07a
0.91 0.01a
0.09 0.03a
0.93 0.03a
1482
2.93 0.13b
2.33 0.15b
5.82 0.49b
8.95 0.37b
1.26 0.07b
0.91 0.01b
0.11 0.03b
0.92 0.03b
1141
2.95 0.20b
2.50 0.24c
6.30 0.82c
9.37 0.64c
1.16 0.09c
0.89 0.03c
0.08 0.04c
0.92 0.04b
150
34.57 1.84a
150
34.16 1.66a
150
34.16 1.54a
Superscript letters (a, b and c) indicate significant differences (P < 0.05).
Author's personal copy
V.M. Tuset et al. / Theriogenology 69 (2008) 1033–1038
1037
Table 3
Correlation values between shape and size morphometric data of different stain techniques from 450 cells randomly selected from each pool of data
Shape parameters
Staining technique
Hemacolor/Diff-Quick
Length (mm)
Width (mm)
Area (mm2)
Perimeter (mm2)
Ellipticity
Rugosity
Elongation
Regularity
Significant differences *P < 0.05;
0.045
0.008
0.079
0.075
0.054
0.153**
0.049
0.024
**
Hemacolor/Spermac
0.034
0.002
0.007
0.035
0.055
0.177**
0.045
0.056
Diff-Quick/Spermac
0.478**
0.095*
0.224**
0.210**
0.029
0.078
0.026
0.181**
P < 0.01.
the fixative and dyes during the staining protocol.
Hence, Hemacolor produced clear morphologic
changes including rounding of the sperm head, which
led to low correlation with the measurements obtained
using the other two stains. Spermac provided similar
means to Diff-Quick but variability was high. In effect,
only the head size variables and not the shape
parameters were significantly correlated with those
recorded using Diff-Quick.
In conclusion, our findings indicate that Diff-Quick
staining is suitable for the morphometric analysis of
rainbow trout sperm. The sperm shapes obtained using
this staining kit are similar to those previously described
for scanning and transmission electron micrographs
[24]. The method exhibits good reproducibility and
requires a short drying time (20 s). Notwithstanding,
because of the improved tail contrast obtained using
Spermac, we recommend this technique for morphometric analysis of the rainbow trout sperm tail.
Acknowledgments
We thank Halina Karol and Wieslaw Demianowicz
for their excellent technical assistance. Also Mrs. Ana
Burton for proof reading and the reviewers for their
suggestions. This project was funded by grant number
PR2006-0354 from the Education and Science Ministry
of the Spanish Government.
References
[1] Lahnsteiner F, Patzner RA, Weismann T. Physiological and
biochemical determination of Rainbow Trout, Oncorhynchus
mykiss, semen quality for cryopreservation. J Appl Aquat
1996;6:47–73.
[2] Lahnsteiner F, Berger B, Weismann T, Patzner RA. Determination of semen quality of rainbow trout, Oncorhynchus mykiss, by
sperm motility, seminal plasma parameters, and spermatozoa
metabolism. Aquaculture 1998;163:163–81.
[3] Glogowski J, Kwasnik M, Piros B, Dabrowski K, Goryczko K,
Dobosz S, Kuzminski H, Ciereszko A. Characterization of
rainbow trout milt collected with a catheter: semen parameters
and cryopreservation success. Aqua Res 2000;31(3):289.
[4] Cabrita E, Robles V, Álvarez R, Herráez MP. Cryopreservation
of rainbow trout sperm in large volume straws: application
to large scale fertilization. Aquaculture 2001;201(3–4):
301–14.
[5] Gage MJG, Macfarlane CP, Yeates S, Ward RG, Searle JB,
Parker GA. Spermatozoal traits and sperm competition in Atlantic salmon: relative sperm velocity is the primary determinant of
fertilization success. Curr Biol 2004;14:44–7.
[6] Dietrich GJ, Kowalski R, Wojtczak M, Dobosz S, Goryczko K,
Ciereszko A. Motility parameters of rainbow trout (Oncorhynchus mykiss) spermatozoa in relation to sequential collection
of milt, time of post-mortem storage and anesthesia. Fish Phys
Biochem 2005;31(1):1–9.
[7] NAFA and ESHRE-SIGA. Manual on basic semen analysis;
2002. p. 1–34.
[8] Soler C, Gadea B, Soler AJ, Fernandez-Santos MR, Esteso MC,
Nuñez J, Moreira PN, Nuñez M, Gutierrez R, Sancho M, Garde
JJ. Comparison of three different staining methods for the
assessment of epididymal red deer sperm morphometry
by computerized analysis with ISAS1. Theriogenology
2005;64(5):1236–43.
[9] Sancho M, Perez-Sanchez P, Tablado L, de Monserrat JJ, Soler
C. Computer-assisted morphometric analysis of ram sperm
heads: evaluation of different fixative techniques. Theriogenology 1998;50:27–37.
[10] Iguer-Ouada M, Verstegen JP. Validation of the sperm quality
analyzer (SQA) for dog sperm analysis. Theriogenology
2001;55:1143–58.
[11] Vladić TV, Afzelius BA, Bronnikov GE. Sperm quality as
reflected through morphology in salmon alternative life histories.
Biol Reprod 2002;66:98–105.
[12] Burness G, Schulte-Hostedde AI, Casselman SJ, Moyes CD,
Montgomerie R. Sperm swimming speed and energetics vary
with sperm competition risk in bluegill (Lepomis macrochirus).
Behav Ecol Sociobiol 2004;56:65–70.
[13] Gravance CG, Liu IK, Davis RO, Hughes JP, Casey PJ. Quantification of normal head morphometry of stallion spermatozoa. J
Reprod Fertil 1996;108:41–6.
[14] Verstegen J, Iguer-Ouada M, Onclin K. Computer-assisted
semen analysers in andrology research and veterinary practice.
Theriogenology 2002;57:149–79.
Author's personal copy
1038
V.M. Tuset et al. / Theriogenology 69 (2008) 1033–1038
[15] Marco-Jiménez F, Pérez L, Viudes de Castro MP, Garzón DL,
Peñaranda DS, Vicente JS, Jover M, Asturiano JF. Morphometry
characterisation of European eel spermatozoa with computerassisted spermatozoa analysis and scanning electron microscopy.
Theriogenology 2006;65:1302–10.
[16] Howell WM, Butts D. Silver staining of spermatozoa from living
and preserved museum fishes: a new taxonomic approach.
Copeia 1983;4:974–8.
[17] Wirtz S, Steinmann P. Sperm characteristics in perch Perca
fluviatilis L.. J Fish Biol 2006;68:1896–902.
[18] Gage MJG, Stockley P, Parker GA. Sperm morphometry in the
Atlantic salmon. J Fish Biol 1998;53:835–40.
[19] Gage MJG, MacFarlane C, Yeates S, Shackelton R, Parker GA.
Relationship between sperm morphometry and sperm motility in
the Atlantic salmon. J Fish Biol 2002;61:1528–39.
[20] Balshine S, Leach BJ, Neat F, Werner NY, Montgomerie R.
Sperm size of African cichlids in relation to sperm competition.
Behav Ecol 2001;12(6):726–31.
[21] Van Look KJW, Kime DE. Automated sperm morphology
analysis in fishes: the effect of mercury on goldfish sperm. J
Fish Biol 2003;63:1020–33.
[22] Jaana H, Yamamoto TS. The ultrastructure of spermatozoa with
a note on the formation of the acrosomal filament in the maprey,
Lampetra japonica. Jpn J Ichthyol 1981;28:135–47.
[23] DiLauro MN, Kaboord W, Walsh RA. Sperm-cell ultrastructure
of North American sturgeons. III. The lake sturgeon (Acipenser
fulvescens Rafinesque, 1817). Can J Zool 1998;78(3):
438–47.
[24] Jamieson BGM. Fish evolution and systematics: evidence from
spermatozoa. Cambridge: University Press; 1991.