sperm characteristics and zona pellucida binding in relation to field fertility of frozen–thawed...

Sperm characteristics and zona pellucida binding

in relation to ®eld fertility of frozen±thawed semen

from dairy AI bulls

B. R. ZHANG, B. LARSSON, N. LUNDEHEIM* and

H. RODRIGUEZ-MARTINEZ

Department of Obstetrics and Gynaecology, Faculty of Veterinary Medicine, and *Department of

Animal Breeding and Genetics, Faculty of Agriculture, Landscape Planning and Horticulture,

Center for Reproductive Biology, Swedish University of Agricultural Sciences,

Box 7039, SE-750 07, Uppsala, Sweden

SummaryThe present study examined the relationship between bull sperm characteristics

immediately post-thaw and some characteristics registered after swim-up, including the

ability of spermatozoa to bind to homologous zona pellucidae (ZP) in vitro, and fertility

after arti®cial insemination (AI) of 9426 females. Frozen±thawed semen from 22 AI bulls

of the Swedish Red and White Breed, represented by 43 different frozen batches

(1±4 batches/bull, 2 consecutive ejaculates/batch), was examined with the aim of

determining concentration, motility patterns, morphology and membrane integrity. In

addition, the frozen±thawed spermatozoa were subjected to a swim-up procedure and

those separated in this way were tested with two assays of sperm-binding to the ZP of

homologous oocytes in vitro (ZBA), using either a relative ZBA index against a control

bull of proven high fertility or absolute binding (Absolute ZBA). The correlations of the

various sperm traits and 56-day non-return rates (NRR) after ®eld AI were retrospec-

tively examined as single traits and as combinations of traits (combined measures),

including regression analysis of signi®cant traits. Among the sperm characteristics,

positively signi®cant ( p < 0.01) correlations with NRR were found for linear motility

post-thawing (r � 0.45±0.59) and the concentration of motile spermatozoa after swim-

up (r � 0.43±0.63). Results obtained with the absolute ZBA approach were signi®-

cantly ( p < 0.05) correlated with NRR (r � 0.50), whereas the correlation between

NRR and the ZBA index was not signi®cant. The use of combined measures of sperm

traits, including the ability to bind to ZP, showed a stronger predictive correlation with

NRR (r � 0.68±0.75), compared with single traits. The results suggest that the

combined analysis of sperm linear-motility patterns, swim-up separated sperm motility

and absolute ZBA can provide a valuable assessment of the fertilizing capacity of AI bull

semen.

Keywords: fertility, non-return rates, sperm characteristics, zona binding

Correspondence: B. R. Zhang, Department of Obstetrics and Gynaecology, Faculty of Veterinary Medicine,

Center for Reproductive Biology, Swedish University of Agricultural Sciences, Box 7039, SE-750 07,

Uppsala, Sweden.

International Journal of Andrology, 21:207±216 (1998)

Ó 1998 Blackwell Science Ltd.

IntroductionSire fertility is a basic component of arti®cial insemination

programmes. Most often, the fertility of the young bulls

entering the AI programme is evaluated after insemination of

a certain number of females with in vitro approved frozen±

thawed semen. Although this is a very reliable method, it is

expensive. Therefore, it is of both scienti®c and economic

interest to develop laboratory methods for accurately

predicting the potential fertility of AI semen.

The ability of a single spermatozoon to achieve fertilization

depends on several cellular characteristics that can be assessed

with laboratory tests. Only viable spermatozoa are able to

interact with the oocyte and pursue fertilization. Visual

estimation of the percentage of motile spermatozoa is the most

commonly used viability test. Signi®cant correlations between

motility and ®eld fertility have been found by some authors

(Lindford et al., 1976; Kjñstad et al., 1993; Correa et al., 1997)

but not by others (Graham et al., 1980; SoÈderquist, 1991;

Januskauskas et al., 1996a). The percentage of spermatozoa

with normal morphology in a semen sample is generally

related to its viability. Further, a negative relationship between

the content of morphologically abnormal spermatozoa and

fertility has been documented (Barth, 1989). Membrane

integrity has also been assessed on the presumption that only

viable spermatozoa maintain intact membranes. In this regard,

signi®cant correlations between membrane integrity and

fertility have been found by some authors using osmotic tests

(Correa et al., 1997), but not by others using ¯uorophores

(Eriksson et al., 1994; Januskauskas et al., 1995, 1996b;

Thomas et al., 1996; Garner et al., 1997). Bull rankings have,

however, indicated that semen samples from bulls with

decreased fertility tend to have a lower percentage of SYBR-

14 stained cells (Thomas et al., 1996).

The zona pellucida (ZP) of mammalian oocytes is a

critical site for sperm±oocyte interaction. Effective binding

of spermatozoa to ZP is an early step in the fertilization

process (Gould et al., 1983). The ability of sperm to bind to

the ZP may re¯ect multiple functions of spermatozoa such as

viability, motility, morphology, acrosomal status, and the

ability to penetrate the oocyte investments (Burkman et al.,

1988; Kaskar et al., 1994; Liu & Baker, 1994), thus having

diagnostic signi®cance (Oehninger, 1992). ZP binding assays

(ZBA) have been developed as diagnostic tests for several

species, including the bull (Burkman et al., 1988; Liu et al.,

1989; Fazeli et al., 1993). Both hemi-zona (bisected oocytes)

binding (HZA) and intact ZBA including ZBA index and

absolute ZBA (Fazeli et al., 1995, 1997; Zhang et al.,

1995b,c, 1996) have been tested. Fazeli et al. (1997) found

a positive relationship between the binding ability of the

spermatozoa and their fertility, measured as non-return rates

of the inseminated animals after AI, when using HZA but

not when using the ZBA index. However, experiments

using the ZBA index and absolute ZBA to evaluate

relationships within and among ZBA methods have not

been reported, indicating that it seems appropriate to

re-examine its value as a functional method to evaluate

frozen bovine semen.

Based on his own and previous results, Amann (1989)

suggested that the analysis of a combination of multiple

parameters of spermatozoa could be used to demonstrate

more precisely the fertilizing capacity of the semen.

Although more complex indices based on combinations of

several parameters have proven only slightly better (Hirao,

1975), the best prediction of fertility was obtained by

combining certain parameters (Wood et al., 1986). How-

ever, these aspects should be further investigated, including

tests of several aspects of sperm function.

The present study was therefore performed, using a

retrospective approach, to further investigate the relationship

between a series of sperm characteristics, including the ability

of sperm in frozen±thawed batches of semen from dairy AI

bulls to bind to homologous ZP (ZBA index and absolute

ZBA) and their 56-day non-return rate after AI. The data have

also been examined statistically to determine the correlations

between both individual and combinations of sperm param-

eters and the ®eld fertility of the semen batches assayed.

Materials and methods

Semen sourceFrozen semen from 22 Swedish Red and White AI bulls,

represented by 43 different batches (1±4 batches/bull, 2

consecutive ejaculates per batch) with a wide range in ®eld

fertility after AI, was used. All bulls were born within a

5-year period, and their ages ranged between 12 and

17 months during the period when the semen tested in this

study was collected. The intervals between semen collec-

tions, i.e. the different batches tested from the same bull,

varied between 7 and 85 days. Semen (with >70% initial

motility and a concentration of at least 300 million

spermatozoa/mL) was diluted with a commercial extender,

processed and frozen in 0.25 mL plastic straws, each

containing 15 ´ 106 spermatozoa. Only frozen semen doses

with >50% post-thaw motility were accepted for arti®cial

insemination. The semen was collected, processed and stored

at one bull station (Svensk Avel, Skara, Sweden). Non-

return rates 56 days after AI were recorded, including all

inseminations, and corrected statistically for differences

between the areas where and seasons during which the AIs

were performed (differences between geographic areas and

season). Further, a single frozen semen batch from a control

bull with high fertility (76.6% 56-day non-return rates, based

on 1572 arti®cial inseminations) was used as a control for the

relative sperm zona pellucida binding assay (ZBA index).

Post-thaw sperm characteristics

Sperm motility. For motility analysis, a 5 lL aliquot of

post-thawed semen was put in a Makler chamber (10 lm

208 B. R. Zhang et al.

Ó 1998 Blackwell Science Ltd, International Journal of Andrology, 21, 207±216

depth) and analysed using the `Stromberg-Mika' Cell Motion

analyser (SM-CMA, Stromberg-Mika, Bad Feilnbach, Ger-

many) at 38 °C (Rodriguez-Martinez & Berrosteguieta,

1994). The computer settings used were: number of frames

per analysis, 32; time between two video half pictures for

detection of immotile objects, 20 ms; cell size range, 35±300

pixels; threshold value for velocity to be classi®ed as

immobile objects, 10 lm/sec; threshold value for velocity

to be accepted as locally motile spermatozoa, 25 lm/sec;

maximum value for linearity, 90%; minimum number of

frames, 15; velocity class width, 5 lm/sec; maximum radius

for circles, 25 lm. Five ®elds in each sample were recorded.

Sperm morphology. To evaluate sperm morphology, the

frequencies of proximal and distal droplets, loose heads,

defects and abnormalities of the acrosome, nuclear pouches,

abnormal midpieces and coiled tails were registered after

counting 200 spermatozoa in wet smears (®xed in buffered

formol-saline) with a phase-contrast microscope (1000´).

The frequency of spermatozoa with normal morphology was

expressed as the percentage of the total number of

spermatozoa (Bane, 1952).

Membrane integrity. Sperm membrane integrity was

assessed using the method originally described by Harrison

& Vickers (1990) and modi®ed for frozen bull semen by

Eriksson et al. (1994). In brief, 0.1 mL post-thawed semen

was added to 0.3 mL staining medium containing 6 lL

formaldehyde, 6 lL propidium iodide (PI) and 6 lL

6-carboxy¯uorescein diacetate (C-FDA) (Calbiochem, Swe-

den) and incubated in darkness for 15 min at 30 °C.

Subsamples (5 lL) of the stained mixture were placed on

two clean microscope slides and overlaid carefully with

coverslips. Random ®elds were observed under epi¯uore-

scence illumination on a Leitz-Dialux 20 microscope (Leitz

Wetzlar, Germany). The counted spermatozoa (100 per

slide) were classi®ed as having an intact plasmalemma if they

were stained green with C-FDA but were not stained with

PI, and expressed as percentages.

Sperm concentration. The primary sperm concentration of

spermatozoa in the straws was con®rmed by manual

counting of thawed semen samples in a BuÈrker chamber

(Bane, 1952).

Intact zona pellucida binding assay (ZBA)

Oocyte source. Ovaries collected from a local slaughter-

house were transported to the laboratory within 3 h in a

thermos containing warm NaCl solution (0.9%). The

contents of antral follicles 2±6 mm in size were gently

aspirated with a needle into a vacutainer tube using a

vacuum-pump system. The cumulus±oocyte complexes

recovered were denuded by vortexing for 4 min. The

denuded oocytes were washed three times in TALP-HEPES

(Zhang et al., 1995a) and stored overnight in PBS containing

3% oestrus cow serum at +4°C.

Sperm preparation. The frozen semen straws were thawed

in a 37 °C water bath for 12 sec. Immediately after thawing,

semen from each straw was transferred to a silicon-coated

soda-glass test tube for swim-up separation. The semen was

covered with 0.45 mL of the fertilizing medium composed

of Fert-TALP (Parrish et al., 1988), supplemented with BSA

(fatty free acid, 6 mg/mL), sodium pyruvate (0.25 lM),

heparin (5 lg/mL), D-penicillamine (20 lM), hypotaurine

(10 lM), and epinephrine (1 lM) and incubated for 1 h at

39 °C in 5% CO2 in humidi®ed air. Thereafter, the upper

3/4 of the medium was collected and transferred to an

Eppendorf tube. The motility and concentration of the

swim-up, separated spermatozoa were assessed subjectively

in a phase-contrast microscope (250´) and determined by

haemocytometry, respectively. The concentration was ad-

justed to 0.625 ´ 106 spermatozoa/mL by dilution with

fertilization medium.

Sperm±oocyte incubation. The stored oocytes were washed

three times in TALP-HEPES and once in fertilization

medium. The oocytes were transferred, in groups of ®ve, to

drops (10 lL) of fertilization medium under paraf®n oil.

Spermatozoa in 40 lL drops were added to reach a ®nal

concentration of 0.5 ´ 106 spermatozoa/mL (5000 sperma-

tozoa/oocyte). The sperm + oocytes were co-incubated

(50 lL ®nal droplets) for 4 h at 39 °C in 5% CO2 in

humidi®ed air.

Evaluation of sperm binding to the zona pellucida. After

4 h of sperm±oocyte coincubation, the oocytes were

transferred to 0.5 mL drops of PBS supplemented with

3% ECS and pipetted 10 times with a pipette of 0.45-mm

inner diameter to remove loosely attached spermatozoa.

The oocytes were stained with 20 lL propidium iodide in

1 mL of a saline medium (140 mM NaCl, 10 mM glucose,

2.5 mM K+, 0.5 mg polyvinyl alcohol/mL and 20 mM

Hepes, pH 7.5, 300 mOsm/kg) for at least 15 min at

30 °C. Thereafter, the oocytes were placed on slides and

covered, being slightly compressed manually with a

coverslip that had a dot of a mixture of paraf®n wax and

vaseline (1:3) in each corner. The number of spermatozoa

bound to the zona pellucida was counted under epi¯uo-

rescence illumination.

Experimental design and presentation of dataSemen from the 22 test bulls, represented by 43 batches,

was examined in order to assess various sperm parameters

(concentration, motility, morphology, and membrane in-

tegrity) and their ability to bind to homologous zona

pellucida (ZP) in vitro in two experiments. Data are either

presented as the batch level or as the bull level (pooling the

different batches within bulls) data.

Sperm characteristics and ZP-binding relative to non-return rate 209

Ó 1998 Blackwell Science Ltd. International Journal of Andrology, 21, 207±216

In the sperm±zona pellucida binding index assay (exper-

iment 1), 17 harvests of oocytes were used. Each harvest of

oocytes was divided into two groups: one group was used to

test two or three different batches of semen from the 22 test

bulls (one straw per batch), and the other group was used to

test one straw with semen from one control bull. Twenty

oocytes/straw were used. The results were expressed by

calculating a ZBA index:

ZBA index � NT=NC ;

where NT is the total number of spermatozoa bound to the

oocytes using semen from the test bulls, and NC is the total

number of spermatozoa bound to the oocytes using semen

from the control bull.

In the absolute sperm±zona pellucida binding assay

(absolute ZBA, experiment 2), each of the 43 semen batches

was tested in three or four replicates using one straw per

replicate and batch. Each harvest of oocytes was used to test

2±4 batches of frozen semen, and 20±30 oocytes were used

for each replicate. Data are presented as mean values of the

replicates within batch (mean number of spermatozoa bound

to the ZP of each oocyte).

Statistical analysisThe handling of the data and statistical analyses were

undertaken using the SAS software (SAS Institute Inc., Cary,

NC, USA 1987). The VARCOMP procedure was used to

analyse the inter- and intra-harvest variation in numbers of

spermatozoa bound to the ZP for the control semen (17

harvests, experiment 1). In these analyses, each harvest was

regarded as one observation. In total, information was

obtained for 340 oocytes from the 17 harvests of oocytes.

The statistical model included the effect of oocyte harvest.

Correlations between sperm parameters tested and ®eld

fertility were calculated using Spearman rank correlation.

Stepwise multiple regression analysis (REG procedure)

was used for analysis of the variation in NRR, in order to

estimate regression equations to predict the non-return rates

on the basis of sperm parameters measured.

ResultsThe 56-day non-return rates (NRR) and the numbers of

arti®cial inseminations (AI) performed per batch ranged from

46.2 to 77.4% and from 89 to 441 (total 9426 AIs),

respectively. Considering the large amount of original data

for the 43 different frozen semen batches, the data are

presented as bull data (on a per bull basis), for example

pooled data from the different batches within bulls (Table 1).

Relationship between single tests and non-return rate (NRR)

Post-thaw spermatozoa, sperm characteristics and NRR. The

pooled data for bulls, consisting of concentration, total

motility and the three motility patterns, normal morphology

and membrane integrity of post-thaw spermatozoa are

shown in Table 1. The correlations between these seven

parameters and NRR are presented in Table 2. Signi®cant

positive correlations were found with the pattern of linear

sperm motility, at both the batch and bull levels, and with

normal sperm morphology at the batch level but not at the

bull level. Signi®cant negative correlations were found with

the circular motility pattern at both the batch and bull levels

and with non-linear motility pattern at the bull level but not

at the batch level.

Swim-up separated spermatozoa, sperm characteristicsand NRR. The pooled data for bulls comprising total

concentration, motility and concentration of motile, swim-

up separated, spermatozoa are presented in Table 1. The

correlations between the three parameters and NRR are

shown in Table 2. Signi®cantly positive correlations were

found with total concentration and concentration of motile,

swim-up separated spermatozoa at both the batch and bull

levels, but not with the motility of the swim-up separated

spermatozoa at either level.

Sperm±zona pellucida binding index and NRR. The mean

number of spermatozoa bound to each oocyte group with

control semen (20 oocytes each group) varied from 12.9 to

37.8. In the 17 harvests of oocytes, 23% of the variation was

explained by harvest ( p < 0.01). The mean number of

spermatozoa bound to each oocyte group in the test semen

(20 oocytes/batch) varied from 2.1 to 64.5. The correlation

between ZBA index and NRR is shown in Table 2. A

signi®cant positive correlation between ZBA index and

NRR was found at the batch level (Fig. 1), but not at the

bull level.

Absolute sperm±zona pellucida binding (absolute ZBA) andNRR. The mean number of spermatozoa bound to the zona

pellucida ranged between 4.9 and 38.3 for each batch

(pooling the replicates within batch), while for bulls the

variation ranged from 7.6 to 34.0 (Table 1). The correlations

between absolute ZBA and NRR are shown in Table 2.

A signi®cantly positive correlation between the absolute

ZBA and NRR existed at both the batch (Fig. 2) and bull

(Fig. 3) levels.

Relationship between sperm characteristics and ZBASigni®cant positive correlations were found between the

frequency of spermatozoa exhibiting a linear motility pattern

and absolute ZBA, at both the batch (r � 0.39, p � 0.010)

and bull (r � 0.54, p � 0.009) levels. The total concen-

tration and the concentration of motile, swim-up separated

spermatozoa also proved to be signi®cantly correlated with

absolute ZBA at both the batch (r � 0.40; p � 0.008 and

r � 0.42; p � 0.005, respectively) and bull (r � 0.55;

p � 0.008 and r � 0.52; p � 0.012, respectively) levels,



Table 1. Sperm characteristics post-thawing and after swim-up and results of relative and absolute sperm±zona pellucida binding (ZBA index and absolute ZBA) as well as 56-day non-return rates (NRR) in 22 dairy bulls (43 frozen batches)

Bulls(no of

Post-Thaw Swim-up ZBAIndex*

AbsoluteZBA

Field fertility

batches) Conc(106/mL)

Motility(%)

Motile Patterns(%)

Morph(%)

MI(%)

Conc(106/mL)

Motility(%)

M-conc(106/mL)

NRR(%)

No. of Al

Circ Non-L Linear

1 (3) 61.7 49.0 7.0 8.7 84.3 76.1 50.8 1.06 85.4 0.91 1.11 20.6 59.4 5352 (1) 66.0 53.0 3.0 3.0 94.0 81.7 42.0 1.05 85.0 0.89 1.62 26.0 72.8 1713 (1) 65.0 65.0 9.0 9.0 82.0 79.5 40.0 1.13 86.3 0.98 1.40 27.6 72.8 4414 (2) 63.5 53.5 7.5 11.0 81.5 80.8 46.0 1.47 86.8 1.28 1.00 23.1 71.9 5125 (2) 57.5 44.0 10.5 12.5 77.0 80.4 47.8 0.98 85.8 0.84 2.22 34.0 68.0 3266 (2) 68.0 57.0 8.5 13.5 78.0 80.5 44.3 1.10 83.0 0.91 3.49 21.2 72.5 3807 (3) 56.3 55.0 13.0 8.7 79.3 80.4 45.0 0.90 87.1 0.78 1.30 9.7 68.3 4978 (3) 58.0 60.0 6.7 4.7 88.7 81.6 49.7 0.85 82.5 0.70 2.34 19.0 72.0 8409 (2) 55.0 67.5 8.0 7.5 84.5 81.9 46.5 0.82 86.8 0.71 1.87 30.8 69.6 573

10 (2) 64.5 52.5 5.0 9.0 86.0 76.4 47.3 0.87 86.3 0.75 0.77 9.8 66.7 48711 (2) 58.6 62.0 28.5 9.5 62.0 80.0 61.7 0.85 88.0 0.74 2.01 14.9 58.9 19812 (1) 56.0 64.0 8.0 11.0 81.0 78.6 54.5 1.10 80.0 0.88 1.23 23.8 55.3 18913 (2) 65.5 42.5 21.0 16.5 62.5 78.4 45.8 0.68 84.2 0.57 1.11 7.6 53.0 53614 (2) 61.5 56.0 11.0 9.5 77.0 80.5 48.0 0.94 84.7 0.79 1.55 11.4 58.9 51815 (2) 60.5 62.5 7.5 8.5 85.0 80.4 50.0 0.89 85.0 0.75 1.81 19.9 57.3 35616 (2) 63.5 52.0 14.0 14.0 72.0 85.2 50.5 0.74 83.6 0.62 1.15 12.4 57.6 50817 (1) 56.0 48.0 17.0 13.0 70.0 77.4 55.5 0.86 83.3 0.72 1.62 15.6 54.0 21818 (1) 66.4 48.0 16.0 16.0 68.0 69.7 39.5 0.63 85.4 0.54 1.22 13.7 55.4 26619 (2) 73.3 47.5 24.0 16.0 60.0 72.5 50.0 0.73 85.1 0.62 0.75 8.8 57.6 41420 (2) 66.7 43.0 17.0 12.5 70.5 83.0 35.5 0.70 84.0 0.58 0.87 17.0 58.6 56221 (1) 62.9 66.0 28.0 8.0 64.0 80.0 65.0 1.02 87.5 0.89 0.95 10.2 60.5 24722 (4) 59.2 62.0 18.0 8.0 74.0 80.7 42.4 0.70 84.5 0.59 0.77 11.0 55.3 652

Note: Conc � concentration; Moti � motility; Circ � circular motility; Non-L � non-linear motility; Linear � linear motility; Morph � normal morphology; MI � membrane intact sperm; M-conc � motilespermatozoa concentration. *Indices of test/control. Mean number of sperm/oocyte.

Sperm

chara

cteristics

and

ZP-b

indin

gre

lativ

eto

non-re

turn

rate

21

1

Ó1998

Black

well

Scien

ceLtd

.In

ternation

alJou

rnal

ofA

ndrology,

21,

207±216

but not with the ZBA index. The correlation between ZBA

index and absolute ZBA was signi®cant at the both batch

(r � 0.38; p � 0.013) and bull (r � 0.56; p � 0.007)

levels.

Relationship between predictive non-return rate (PNRR)and observed non-return rateThe predictive equation included four parameters at the

batch level [absolute ZBA, linear motility pattern (Linear-

M), ZBA index, and swim-up separated sperm motility

(S-moti)] and ®ve parameters at the bull level [linear motility

pattern, ZBA index, post-thaw concentration (Conc),

concentration of motile after swim-up (SM-conc) and

concentration of swim-up separated motile sperm (S-conc)].

All variables selected by the Stepwise REG procedure and

left in the model are signi®cant at the 0.15 level. Since the

amount of material was rather small while the model was

comparatively large, an adjusted R2 was presented. The

predictive equations were as follows.

Table 2. Correlations between single and combined semen parameters and 56-day non-return rates (NRR)

Parameters 43 Batches 22 Bulls

r p-value r p-value

Single parameter

(A) Post-thaw spermatozoa

(1) Concentration )0.05 0.734 0.09 0.981(2) Motility 0.16 0.307 0.32 0.146(3) Circular motility )0.46 0.002 )0.55 0.009(4) Non-linear motility )0.17 0.271 )0.47 0.026(5) Linear motility 0.45 0.003 0.59 0.004(6) Normal morphology 0.35 0.022 0.37 0.090(7) Membrane integrity )0.19 0.213 )0.27 0.219(B) Swim-up separated spermatozoa

(1) Concentration 0.37 0.014 0.59 0.004(2) Motility 0.22 0.155 0.36 0.099(3) Concentration of motile spermatozoa 0.43 0.004 0.63 0.002(C) Intact ZBA

(1) ZBA index 0.34 0.027 0.36 0.102(2) Absolute ZBA 0.49 <0.001 0.50 0.018

Combined measures 0.68 <0.001 0.75 <0.001

Figure 1. Correlation between the zona pellu-cida (ZP) binding index (the ratio of the number ofspermatozoa from the test semen batches bound tothe ZP of oocytes and the corresponding numberfrom the control semen) and 56-day non-returnrates (NRR) (r � 0.34, p � 0.027, n � 43). Theline shows a trend in the data.



At the batch level:

PNRR � ÿ43:36� 0:27 (absolute ZBA)

� 0:21 (Linear-M)� 2:86 (ZBA index)

� 0:95 (S-moti)

with adjusted R2 � 0.36.

At the bull level:

PNRR � ÿ4:63� 0:44 (Linear-M)� 4:50 (ZBA index)

� 0:34 (Conc)� 157:56 (SM-conc)

ÿ 127:68 �Sÿ conc�

with adjusted R2 � 0.71.

The two equations were used to estimate the non-return

rates to be expected at both the batch and bull levels. By

comparing these observed non-return rates in the ®eld with

the predicted non-return rates (PNRR) obtained from the

above equations, the predictive Spearman correlation coef-

®cients between PNRR and NRR could be calculated and

they were 0.68 ( p < 0.001) at the batch level and 0.75

( p < 0.001) at the bull level.

DiscussionThe present study has investigated the relationship

between both sperm characteristics and sperm±oocyte

interactions in two zona pellucida binding tests in vitro

and ®eld fertility, expressed as 56-day non-return rates

(NRR) after arti®cial insemination using frozen±thawed

bovine semen. For sperm characteristics, positive correla-

tions were found between ®eld fertility and the pattern of

linear sperm motility post-thaw, and between ®eld

fertility and both the total concentration and the

concentration of motile spermatozoa after swim-up

separation. There were also positive correlations between

®eld fertility and the zona binding tests, although they

differed in value.

Field fertility of bulls used in this study ranged from 53 to

72.8%. There was no bull with very low fertility (e.g. <50%

56-day non-return rate). In Sweden, there are very few cases

in which ®eld fertility of AI bulls is <50%. A bull was

considered as subfertile when the 56-day non-return rate

was <60%, and immediately removed from the AI schemes.

The bulls used in the study represent therefore the widest

possible range of NRRs.

Figure 2. Correlation between mean numbers of sperma-tozoa bound to the zona pellucida (ZP) in each tested semenbatch and 56-day non-return rates (NRR) (r � 0.49,p � 0.001, n � 43). The line shows a trend in the data.

Figure 3. Correlation between mean numbers of sperma-tozoa bound to the zona pellucida (ZP) in tested bulls and 56-day non-return rates (NRR) (r � 0.50, p � 0.018,n � 22). The line shows a trend in the data.



The relationship between laboratory semen characteristics

and fertility has been discussed and reviewed (Woelders,

1990; den Daas, 1992; Graham, 1994; Rodriguez-Martinez

et al., 1996, 1997). In earlier investigations at our department

(SoÈderquist, 1991; Eriksson et al., 1994; Januskauskas &

Rodriguez-Martinez, 1995), single parameter tests of post-

thaw spermatozoa, using the same standard for approval of

the frozen semen for arti®cial insemination (e.g. ³50% post-

thaw motility), have been used. In these studies no

signi®cant correlations were found between these tests and

®eld fertility. In a recently published paper (Correa et al.,

1997), signi®cant differences in post-thaw motility, deter-

mined in an automated system (Sperm Quality Analyzer),

morphology, acrosome integrity and ®eld fertility between

bulls of high and low fertility were found, indicating that

differences in fertility level could be attributed to variations

in these qualitative sperm characteristics. Interestingly

enough, the values of motility provided by the computer-

based analyser were very high for both bull categories. In the

present study, a signi®cant positive correlation with ®eld

fertility was found for the linear-motility parameter. The

swim-up method used to separate spermatozoa is widely

used for in vitro fertilization. Lathrop & Foote (1986), who

®rst used the `swim-up' test to predict the fertility of bull

spermatozoa, found that the concentration of `swim-up'

separated spermatozoa had a higher correlation with non-

return rate than did the percentage of motile spermatozoa.

The present results also indicate that both sperm concentra-

tion and the number of motile spermatozoa after swim-up

separation are valuable parameters for evaluating ®eld

fertility, and the concentration of motile spermatozoa post

swim-up was found to be more powerful than the total

sperm concentration in this respect.

Sperm±zona pellucida binding assays are used to test

semen quality in men (WHO, 1992). Previous studies at our

department (Zhang et al., 1995b,c), based on an assay of

sperm binding to intact homologous zona pellucida (absolute

ZBA), found that there was a signi®cant difference in the

mean number of spermatozoa bound to the zona pellucida

between semen from bulls with high ®eld fertility and that of

bulls with low fertility, and that a signi®cant positive

correlation existed between the absolute ZBA and ®eld

fertility. These preliminary tests were, however, performed

on only a few bulls. To avoid the effects of differences

between individual oocytes on the numbers of spermatozoa

bound, Fazeli et al. (1993) used the index of the number of

bound spermatozoa from a control bull to the number of

bound spermatozoa from a test bull. Also, in that study the

authors suggested that there might be a relationship between

the sperm±zona pellucida binding capacity and the fertility of

bulls.

Other approaches to solving the problem of intra and

interassay variation in sperm±zona pellucida binding assays in

humans have also been used, such as labelling sperm (Liu

et al., 1988) and matching hemi-zona (Burkman et al., 1988).

In another study, Fazeli et al. (1997) used both a ZBA index

and matching hemi-zonas to test the binding capacity of

frozen bovine semen and found a signi®cant correlation

between sperm±zona pellucida binding and the 56-day non-

return rate in the assay using hemi-zona binding, but not in

the intact sperm±zona pellucida binding index. The hemi-

zona assay, although requiring fewer oocytes, is more

laborious than the intact zona assay. It has been shown that

the presumed oocyte variation that may exist in the latter

method is not signi®cant, provided that a large number of

oocytes per replicate is tested (Zhang et al., 1995a). The

present results indicate that the absolute ZBA can be used to

evaluate frozen bovine semen quality, while the ZBA index

value is of less certain accuracy. This could be explained by

the fact that more replicates (3±4) were tested using the

absolute ZBA than using the ZBA index (only 1 replicate).

In the present study (17 harvests of oocytes) the variation

obtained between single oocytes within harvest was more

than four times as high (77% vs. 23%) as that obtained

between harvests. This means that to achieve more precise

results from a sperm±zona pellucida binding assay more

oocytes or replicates would be needed (Zhang et al., 1995a).

Under in vivo conditions, only spermatozoa with normal

morphology and motility reach the site of fertilization

(Larsson, 1988; Roldan & Gomendio, 1992). Sperm±oocyte

interaction re¯ects the functionality of spermatozoa, includ-

ing several sperm parameters, such as viability, motility,

morphology and acrosomal status, in one functional test.

Sperm±oocyte interaction has been investigated in men

(Franken et al., 1989; Liu & Baker, 1992, 1994) in whom a

positive correlation existed between sperm morphology and

sperm±zona binding as well as between sperm±zona pellu-

cida penetration and in vitro fertilization. The present study

indicates that the absolute ZBA is signi®cantly correlated to

linear sperm motility post thaw as well as to the sperm

concentration and the concentration of motile spermatozoa

after swim-up separation.

Wood et al. (1986) suggested that when attempting to

predict bull fertility based on a semen assessment, 18±22

ejaculates are to be used per bull in order to ensure a

representative and accurate measure of a bull's individual

fertility (NRR). In Sweden, only about seven frozen semen

batches (ejaculates) (total of 7000 AI doses) are collected for

fertility testing by AI from each young AI dairy bull

(M. HaÊaÊrd, Svensk Avel AB, personal communication).

Further, it is impractical to undertake in vitro tests of all

ejaculates from each bull. When considering the use of

fertility tests, the relationship between age and fertility

should also be considered. In most laboratory tests, as well as

in most published studies, a single test of a semen sample was

used to represent the long-term or working-lifetime fertility

of a given bull (Hillery et al., 1990; Marquant-Le Guienne

et al., 1992; Shamsuddin & Larsson, 1993). Our former work

(Zhang et al., 1997) has shown that within-sire variation in

fertility exists between ejaculates. This implies that to obtain



representative results from a given bull, different frozen

batches of its semen should be analysed. The present results

indicate, furthermore, that the correlation with ®eld fertility

is higher at the bull level than at the batch level.

The use of functional tests could be a more precise way of

evaluating the fertility potential of bull spermatozoa com-

pared with routine spermiograms (Amann, 1989; Amann &

Hammerstedt, 1993). The correlation between both single

and combined tests of spermatozoa and in vivo fertility have

been investigated by some authors (Hirao, 1975; Linford

et al., 1976; Wood et al., 1986). No consistent conclusion

was obtained. The above authors assessed sperm character-

istics, such as motility, adenosine triphosphate content,

morphology and chemical constituents of fresh and/or

frozen semen. Methods for testing several parameters

(motility, viability, morphology, membrane integrity, and

acrosome status) of spermatozoa have also been developed

(e.g. sperm±zona pellucida binding assay and in vitro

fertilization systems). However, no investigations have been

carried out into the relationship between ®eld fertility, on

the one hand, and semen characteristics and the results of a

zona binding assay on a series of frozen batches of semen, on

the other hand, in an analysis of individual and combined

parameters. In the present study, a higher correlation

between predicted and observed non-return rates was found

at both the batch and bull levels, when compared with that

found between any single parameter and the observed non-

return rate. The valuable parameters in the regression

equations accounted for 75% (3/4) of the total parameters

at batch level and for 80% (4/5) at bull level, respectively.

This means that the use of combined tests and regression

analyses together is more powerful than the use of individual

parameters. It should be noted, however, that, even when a

signi®cant correlation between laboratory tests and in vivo

fertility was obtained, no standards of correlative strength

have yet been developed for practical use in a bull station

because they are retrospective. A prospective trial must

therefore be undertaken, i.e. the processed semen of bulls

should be tested in vitro, as performed here. The results

would then have to be analysed with multivariate statistics

before its predictive value could be assured, and eventually,

the trial procedure could be implemented by the AI station.

Such a study is being performed in our laboratory.

AcknowledgementsWe thank Dr M. HaÊaÊrd (Svensk Avel) for supplying the

frozen bovine semen and the ®eld fertility data. Thanks are

also due to Mrs A. Rikberg, K. Selin-Wretling, AÊ . Jansson,

A. Januskauskas, Mr B. Fred and Mr T. Jangby for their

assistance. This study received ®nancial support from the

Swedish Farmer Foundation for Agricultural Research

(Stiftelsen Lantbruksforsking) and the Swedish Council for

Forestry and Agricultural Research (SJFR), Sweden.

References

Amann, R. P. (1989) Can the fertility potential of a seminal sample

be predicted accurately? Journal of Andrology 10, 89±98.

Amann, R. P. & Hammersted, R. H. (1993) In vitro evaluation of

sperm quality: an opinion. Journal of Andrology 14, 397±405.

Bane, A. (1952) A study on the technique of hemocytometric

determination of sperm motility and sperm concentration in bull

semen. Cornell Veterinarian, XLII, 518±531.

Barth, A. D. (1989) In¯uence of abnormal sperm morphology on

conception rates in cattle. In: Proceedings of the Annual Meeting,

Society for Theriogenology, pp. 13±40.

Burkman, L. J., Kruger, T. F., Coddington, C. C., Rosenwaks,

Z., Franken, D. R. & Hodgen, G. D. (1988) The hemizona

assay (HZA): development of a diagnostic test for the binding

of human spermatozoa to the human hemizona pellucida to

predict fertilization potential. Fertility and Sterility 49, 688±

697.

Correa, J. R., Pace, M. M. & Zavos, P. M. (1997) Relationships

among frozen-thawed sperm characteristics assessed via the

routine semen analysis, sperm functional tests and fertility of

bulls in an arti®cial insemination program. Theriogenology 48,

721±731.

den Daas, N. (1992) Laboratory assessment of semen characteristics.

Animal Reproduction Science 28, 87±94.

Eriksson, B., Shamsuddin, M., HaÊaÊrd, M., Larsson, B., SoÈderquist,

L. & Rodriguez-Martinez, H. (1994) Motility and membrane

integrity of frozen/thawed bull spermatozoa in relation to IVF

and ®eld fertility. Proceedings of the XVII Nordic Veterinary

Congress, Reykjavik 2, 200±201. Nordic Veterinary Association.

Fazeli, A. R., Holt, C., Steenweg, W., Bevers, M. M., Hoil, W. V.

& Colenbrander, B. (1995) Development of a sperm hemizona

binding assay for boar semen. Theriogenology 44, 17±27.

Fazeli, A. R., Steenweg, W., Bevers, M. M., de Looves, F. A. M.,

van den Broek, J. & Colenbrander, B. (1993) Development of a

sperm zona pellucida binding assay for bull semen. Veterinary

Record 132, 14±16.

Fazeli, A. R., Zhang, B. R., Steenweg, W., Larsson, B., Bevers, M.

M., van den Broek, J., Rodriguez-Martinez, H. & Colen-

brander, B. (1997) Relation between sperm zona pellucida

binding assays with the 56-day nonreturn rates of frozen thawed

bull semen. Theriogenology 48, 853±863.

Franken, D. R., Oehninger, S., Burkman, L. J., Coddington, C.

C., Kruger, T. F., Rosenwaks, Z., Acosta, A. A. & Hodgen, G.

D. (1989) The hemizona assay: a predictor of human sperm

fertilizing potential in in vitro fertilization (IVF) treatment.

Journal of In Vitro Fertilization and Embryo Transfer 6, 44±50.

Garner, D. L., Thomas, C. A., Allen, C. H., Senger, P. L. & Sasser,

R. G. (1997) Effects of cryopreservation on bovine sperm

viability as determined by dual DNA staining. Reproduction in

Domestic Animals 32, 279±283.

Gould, J. E., Overstreet, J. W., Yanagimachi, H., Yanagimachi, R.,

Katz, D. F. & Hansonm, F. W. (1983) What functions of the

sperm cell are measured by in vitro fertilization of zona-free

hamster eggs? Fertility and Sterility 40, 344±350.

Graham, J. K. (1994) In vitro assays of bull fertility. In: Proceedings of

the 15th Technical Conference on Arti®cial Insemination and

Reproduction, 22±23 April, Wisconsin, pp. 74±81.



Graham, E. F., Schmehl, M. K. L., Evenson, B. K. & Nelson, D. S.

(1980) Problems with laboratory assays. Proceedings of the 8th

Technical Conference on Arti®cial Insemination & Reproduction,

National Association of Animal Breeders, Wisconsin, pp. 1±8.

Harrison, R. A. P. & Vickers, S. E. (1990) Use of ¯uorescent

probes to assess membrane integrity in mammalian spermatozoa.

Journal of Reproduction and Fertility 88, 343±352.

Hillery, F. L., Parrish, J. J. & First, N. L. (1990) Bull speci®c effect

on fertilization and embryo development in vitro. Theriogenology

33, 249 (abstract).

Hirao, K. A. (1975) A multiple regression analysis on six

measurements of bovine semen characteristics for fertility.

International Journal of Fertility 20, 204±208.

Januskauskas, A. & Rodriguez-Martinez, H. (1995) Assessment of

sperm viability by measurement of ATP, membrane integrity

and motility in frozen/thawed bull semen. Acta Veterinaria

Scandinavica 36, 571±574.

Januskauskas, A., SoÈderquist, L., HaÊaÊrd, M. G., HaÊaÊrd, M. C. h.,

Lundeheim, N. & Rodriguez-Martinez, H. (1996a) In¯uence of

sperm number per straw on the post-thaw sperm viability and

fertility of Swedish Red and White AI bulls. Acta Veterinaria

Scandinavica 37, 461±470.

Januskauskas, A., SoÈderquist, L., HaÊaÊrd, M. G., HaÊaÊrd, M. C. h.,

Lundeheim, N. & Rodriguez-Martinez, H. (1996b) Estimation

of sperm viability in frozen-thawed semen from Swedish A.I.

Bulls. Journal of Veterinary Medicine, Series A 43, 281±287.

Kaskar, K., Franken, R., van der Horst, G., Oehninger, S., Kryger,

T. F. A. & Hodgen, G. D. (1994) The relationship between

morphology, motility and zona pellucida binding potential of

human spermatozoa. Andrologia 26, 1±4.

Kjñstad, H., Ropstad, E. & Andersen Berg, K. (1993) Evaluation of

spermatological parameters used to predict the fertility of frozen

bull semen. Acta Veterinaria Scandinarica 34, 299±303.

Larsson, B. (1988) Distribution of spermatozoa in the genital tract

of heifers inseminated with large numbers of abnormal

spermatozoa. Journal of Veterinary Medicine, Series A 35, 721±728.

Lathrop, W. F. & Foote, R. F. (1986) The ``swim-up'' test to

predict fertility of bull spermatozoa. Journal of Dairy Science 69

(Suppl. 1), 239.

Linford, E., Glover, F. A., Bishop, C. & Stewart, D. L. (1976) The

relationship between semen evaluation methods and fertility in

the bull. Journal of Reproduction and Fertility 47, 283±291.

Liu, D. Y. & Baker, H. W. G. (1992) Sperm nuclear chromatin

normality: relationship with sperm morphology, sperm-zona

pellucida binding, and fertilization rates in vitro. Fertility and

Sterility 58, 1178±1184.

Liu, D. Y. & Baker, H. W. G. (1994) A new test for the assessment

of sperm-zona pellucida penetration: relationship with results of

other sperm tests and fertilization in vitro. Human Reproduction 3,

489±496.

Liu, D. Y., Johnston, W. I. H., Clarke, G. N., Baker, H. W. G. &

Lopata, A. (1989) A sperm-zona pellucida binding test and in

vitro fertilization. Fertility and Sterility 52, 281±287.

Liu, D. Y., Lopata, A., Johnston, W. I. H. & Baker, H. W. G. (1988)

A human sperm-zona pellucida binding test using oocytes that

failed to fertilize in vitro. Fertility and Sterility 50, 782±788.

Marquant-Le Guienne, B., Humblot, P., Thibier, M. & Thibault,

C. (1992) In vitro fertilization as a tool to evaluate fertility in the

bovine. In: 12th International Congress on Animal Reproduction,

Vol. 2, pp. 662±664.

Oehninger, S. (1992) Diagnostic signi®cance of sperm-zona

pellucida interaction. Reproductive Medicine Review 1, 57±81.

Parrish, J. J., Susko-Parrish, J., Minner, M. A. & First, N. L. (1988)

Capacitation of bovine sperm by heparin. Biology of Reproduction

38, 1171±1180.

Rodriguez-Martinez, H. & Berrosteguieta, A. (1994) Viability of

frozen-thawed bull spermatozoa after swim-up through a

Hyaluronic acid solution. Biomedicine Research 2, 133±144.

Rodriguez-Martinez, H., Larsson, B., Zhang, B. R. & SoÈderquist,

L. (1996) Assessment of bull sperm fertility ability. Reproduction

in Domestic Animals 31, 515±517.

Rodriguez-Martinez, H., Larsson, B., Zhang, B. R. & SoÈderquist,

L. (1997) In vitro assessment of viability and fertilizing capacity

of bull spermatozoa. Journal of Reproduction and Development, 43,

1±11.

Roldan, E. R. S. & Gomendio, M. (1992) Morphological,

functional and biochemical changes underlying the preparation

and selection of fertilizing spermatozoa `in vivo'. Animal

Reproduction Science 28, 69±78.

Shamsuddin, M. & Larsson, B. (1993) In vitro development of

bovine embryos after fertilization using semen from different

donors. Reproduction in Domestic Animals 28, 77±84.

SoÈderquist, L. (1991) Sperm characteristics and fertility in dairy AI bulls.

PhD Thesis, Uppsala, p. 35.

Thomas, C. A., Garner, D. L., Burner, C. A., Senger, P. L.,

Johnson, M. S., Sasser, R. G. & Allen, C. H. (1996) Fluorescent

viability staining and the fertility of cryopreserved bovine sperm.

In: 13th International Congress on Animal Reproduction, Sydney,

Vol. 3 (eds G.M. Stone & G. Evans), pp. 22±24. Elsevier.

Woelders, H. (1990) Overview of in vitro methods for evaluation

of semen quality. Reproduction in domestic animals. Boar Semen

Preservation II Supplement 1, 145±164.

Wood, P. D. P., Foulkes, J. A., Shaw, R. C. & Melrose, D. R.

(1986) Semen assessment, fertility and the selection of Hereford

bulls for use in AI. Journal of Reproduction and Fertility 76, 783±

795.

World Health Organization (1992) WHO Laboratory Manual for the

Examination of Human Semen and Sperm-Cervical Mucus Interaction,

3rd edn, p 25. Cambridge University Press, Cambridge.

Zhang, B. R., Larsson, B. & Rodriguez-Martinez, H. (1995a)

In¯uence of batches of bovine oocytes on the outcome of a

sperm zona pellucida binding assay and in vitro fertilization.

International Journal of Andrology 18, 213±220.

Zhang, B. R., Larsson, B. & Rodriguez-Martinez, H. (1995b)

Results of an intact zona pellucida binding assay and in vitro

fertilization, using semen from bulls with high or low fertility

after AI. In: Proceedings of the 11th Annual Meeting on European

Embryo Transfer Association, Hannover, p. 262 (abstract).

Zhang, B. R., Larsson, B. & Rodriguez-Martinez, H. (1995c) In

vitro tests to predict fertility in A.I. bulls. In: Proceedings of the

14th Hokkaido Bovine Embryo Transfer Society Meeting, Sapporo

(ed. H. Kawa-Waga), p. 49 (abstract). Hokkaido University

Press, Sapporo.

Zhang, B. R., Larsson, B., Lundeheim, N. & Rodriguez-Martinez,

H. (1996) Ability of an intact zona pellucida binding assay to

discriminate bulls with different fertility after AI. Proceedings of

the 13th International Congress on Animal Reproduction, Sydney,

Vol. 3 (eds G.M. Stone & G. Evans), pp. 24±5. Elsevier.

Zhang, B. R., Larsson, B., Lundeheim, N. & Rodriguez-Martinez,

H. (1997) Relationship between embryo development in vitro

and 56-day nonreturn rates of cows inseminated with frozen-

thawed semen from dairy bulls. Theriogenology 48, 221±231.

Received 23 December 1997; accepted 28 February 1998



sperm characteristics and zona pellucida binding in relation to field fertility of frozen–thawed...

Documents