Hum. Reprod.

Advance Access published May 24, 2013

Human Reproduction, Vol.00, No.0 pp. 19, 2013
doi:10.1093/humrep/deq098

ORIGINAL ARTICLE Infertility

rm

Y chromosome microdeletions, spe

DNA fragmentation and sperm
oxidative stress as causes of recurr

ent
spontaneous abortion of
unknown

etiology

2
J. Bellver 1, , M. Meseguer1, L. Muriel
, S. Garca-Herrero 1,

M.A.M. Barreto 1, A.L. Garda 3, J. Remoh 1, A. Pellicer 1,

Correspondence address. Tel:
and
N. Garrido 1 +34-963050900; Fax: +34-963050999; E-mail:
jbellver@ivi.es
1

Instituto Valenciano de Infertilidad (IVI), University of Valencia, Plaza de la Polica Local, 3, 46015 Valencia, Spain IVI
Vigo, Plaza Francisco
3
Fernandez del Riego, 7, 36203 Vigo, Spain IVI Murcia, Navegante Macas del Poyo, 5, 30007 Murcia, Spain

byguest
onMarc
h17,201
6http://h
umrep.o
xfordjo
urnals.o
rg/Dow
nloaded
from

Submitted on December 4, 2012; resubmitted on February 24, 2013; accepted on March 23, 2013

background: The aim of the present study was to evaluate the implication of male factor, in terms of sperm DN
A oxidation and
fragmentation, and Y chromosome microdeletions in recurrent spontaneous abortion (RSA) of unknown origin in a strictly sel
ected cohort.

methods: A prospective cohort study was carried out in a private university-afliated setting. Three groups, each
comprised of 30 males,
were compared. The rst was formed by healthy and fertile sperm donors (SD) with normal sperm parameters (control gro
up), the second
by men presenting severe oligozoospermia (SO) without RSA history, and the third by men from couples who had experien
ced idiopathic
RSA. Frequency of Y chromosome microdeletions and mean sperm DNA fragmentation and oxidation were determined.

results: Y chromosome microdeletions were not detected in any of the males enrolled in the study. Moreover, sperm
DNA oxidation
measurements were not demonstrated to be relevant to RSA. Interestingly, sperm DNA fragmentation was higher in the SO
group than in
the RSA and the SD groups, and also higher in the RSA group compared with the SD group, but lacked an adequate predicti
ve power to be
employed as a discriminative test of RSA condition.

c o n c l u s i o n s : Sperm DNA features and Y chromosome microdeletions do not seem to be related to RSA of unknown
origin. Other
molecular features of sperm should be studied to determine their possible inuence on RSA.
Clinicaltrials.gov reference: nCT00447395.
Ke y words: recurrent spontaneous abortion / semen / Y chromosome microdeletions / sperm DNA fragmentation / sperm o
xidative stress

Introduction

2005), some inherited thrombophilias, such as Factor V Leiden

and proRecurrent spontaneous abortion (RSA), dened as 3 clinic thrombin G20210A gene mutation (Rey et al., 2003; Kovalevsk
y et al.,
al preg2004; Krabbendam et al., 2005), congenital or acquired
nancy losses before the fetus has reached viability (Rai an uterine
d Regan,
anomalies (Devi Wold et al., 2006), endocrine, autoimmune or
2006), is a frustrating condition that affects 1% of couples of challoimildbearmune disturbances (Arredondo and Noble, 2006; Christiansen
ing age (Porter and Scott, 2005). Although some causes h et al.,
ave been
2006) and perhaps unhealthy lifestyle habits (smoking, obesity,
identied, others remain mere speculation. Thus, 50% of psychocouples
logical stress; Lashen et al., 2004; Pandey et al., 2005). Howev
are still classied as having unexplained RSA since the u er, most of
nderlying
these conditions are related to the woman, with the males cont
mechanism(s) is never found (Porter and Scott, 2005).
ribution
Known causes include paternal or de novo chromosomal aberremaining relatively underexplored.
rations
The alteration of sperm parameters evaluated by classic
(Rai and Regan, 2006), antiphospholipid syndrome (Empsoncriteria
et al.,
(concentration, motility, morphology; World Health Organi
zation,
byguest
onMarc
Bellver et al. h17,201
6http://h
umrep.o
xfordjo
not clearly associated with the risk of either sp
Classic parameters of sperm evaluation cannot be emplournals.o
rg/Dow
abortion (Carrell et al., 2003b; Bhattacharya yed to
nloaded
identify structural alterations in the sperm chromatin (Erfrom

& The Author 2013. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved.
For Permissions, please email: journals.permissions@oxfordjournals.org

1999) is
oradic or
recurrent
, 2008),
although some of these sperm deciencies could represent the
manifestation of an underlying related cause. This is the case for e
mbryos
with chromosomally abnormalities originating in meiotic segr
egation
errors in the spermatozoa of men with severe oligoasthen
oteratozoospermia (Bernardini et al., 2005; Pang et al., 2005). In fa
ct, some
authors have proposed performing uorescent in situ hybri
dization
analysis in severely altered sperm samples to detect cases
that may
benet from preimplantation genetic screening, thus reduc
ing the
abortion rate (Rubio et al., 2001). Nevertheless, there is no co
nsensus
in the current medical literature about this topic (Harper et al.,
2008).
In addition, many reported cases of male partners of RSA
subjects
show normal parameters in their ejaculates (Al-Hassan et al.,
2005).
Moreover, as sperm physiology is ascertained, an increasing n
umber
of molecular factors are implicated in male fertility (Garrid
o et al.,
2008a, b), such as oxidative stress-related molecules
(Meseguer
et al., 2004), DNA oxidation and fragmentation (Aguilar
et al.,
2009) and mRNA expression proles (Garrido et al., 2008a,
b), but
their implication in RSA has received little or no attention.

enpreiss
et al., 2006; Erenpreiss et al., 2008; Avendano et al., 200
9; CohenBacrie et al., 2009). Sperm DNA fragmentation has been re
lated to
male subfertility, sporadic abortion and poorer repr
oductive
outcome, especially after assisted conception technologies (
LarsonCook et al., 2003; Borini et al., 2006; Lewis et al., 2008;
Lin et al.,
2008; Zini et al., 2008), but with contradictory results (Collin
s et al.,
2008; Nicopoullos et al., 2008; Tavalaee et al., 2009). In
addition,
different methods of analysis have been employed to gat
her data
[Comet Assay, Sperm Chromatin Structure Assay (SCSA),
TUNEL
Assay; Practice Committee of American Society for Repro
ductive
Medicine, 2008; Zini et al., 2008], and scarce information is c
urrently
available about their implication in RSA (Carrell et al.,
2003a).
Similarly, some authors have reported increased DNA da
mage in
sperm caused by oxidative stress in infertile men (Mesegu
er et al.,
2008; Aguilar et al., 2009), but nothing is known about this in
relation
to RSA.
Two reports have also described a notably higher prevalenc
e of Y
chromosome microdeletions in the male partners of coupl
es with

The control group included 30 healthy SD of 1835 years of

RSA (Dewan et al., 2006; Karaer et al., 2008), but this has age,
all
not been
Caucasian, with normal karyotype, no family history of diseases,
conrmed (Kaare et al., 2008; Lu et al., 2008).
normal
The present study was designed to prospectively evalu sperm analysis result (more than 80 million total motile sper
matozoa
ate the
relationship of Y chromosome microdeletions, sperm DNA frag per ejaculate) and previous term pregnancies without complications
conmentation and sperm DNA oxidation caused by oxidative stress wit ceived with their sperm samples.
The oligozoospermic group (SO) consisted of 30 Caucasian men
h RSA
with
and to compare their values among healthy and fertile sperm
.1 year infertility, ,45 years of age, absence of autoimmune
donors
or endo(SD) with normal sperm parameters, men with severe oligozo crine disorders, normal karyotype, no history of RSA, normal
ospergenital
mia (SO) and men from couples with RSA.
examination, normal FSH, LH, testosterone and prolactin values
and at
least two sperm analysis results showing SO ( 5 million sperm
atozoa/
ml, but .1 million of total sperm per ejaculate in order to make f
easible
the sperm tests). These men were considered to represent
Study design
cases of
This is a prospective casecontrol study, in which men were severe idiopathic oligozoospermia. No assisted reproduction treat
enrolled
ment
between 1 January 2010 and 1 February 2012. Subjects were had been performed for these males at the time they entered in th
assigned
e study.
to one of the three groups according to the following inclusion crit The RSA group was composed of the male partners of 30 Ca
eria.
ucasian
couples who had previously experienced 3 clinical rst t
rimester
(514 weeks) spontaneous abortions, with normal karyotypes o
f both
male and female and no autoimmune or endocrine disorders.
These
couples had not attempted assisted reproduction treatments at the
time
they were accepted for the study. The female partners of these s
ubjects
were ,38 years old, presented normal ovarian function, a
normal
uterus conrmed by vaginal ultrasound and/or hysterosalpingog
raphy/
hysteroscopy and absence of acquired or inherited thrombophilia
(antiphospholipid syndrome, activated protein C resistance, serum
fasting
homocystine, protein C, protein S, antithrombin III, proth
rombin
G20210A gene mutation, factor V Leiden). All the men were u
nder 45
years old and previous sperm analysis had given a normal o
r only a
mildly altered result (. 10 million/ml, .35% A + B motile sperm
atozoa,
.10% of normal spermatozoa following Krugers strict criteria).
To ensure the reliability of the results obtained, subjects compl
eted a
clinical questionnaire about previous consumption of alcohol
and/or
drugs, recent episodes of fever, exposure to gonadotoxins (e.g.
during
chemotherapy or radiotherapy), pesticide or heavy metals (professi
onally)
and vitamin intake in the 3 months prior to the enrolment in th
e study.
Andrological examinations were performed to rule out the prese
nce of
varicoceles, testicular torsion, traumatisms and other alterations
in the

Materials and Methods

genital tract, such as genital tract inammation or recurrent i Applied Science, Indianapolis, Ind.) according to the manufa
nfections
cturers
(Viloria et al., 2009). None of the men presented any of these con instructions. Screening for Yq microdeletions was carried out in co
ditions.
ntrols
The study was approved by the Institutional Research Board and and patients using polymerase chain reaction (PCR) techniques by a
Ethics
mplifyCommittee. The study purposes and procedures were carefully expl ing 20 different sequence-tagged sites (STSs) on the long arm o
ained,
f the Y
and the informed consent was obtained from those willing to particchromosome, corresponding to four AZF loci spread over inter
ipate.
vals 5
The study was also registered in Clinicaltrials.gov, with the r and 6. These included SY14 from SRY, SY81, SY84s and SY
eference
86 from
NCT00447395.
AZFa; SY182 from KALY, SY121, SYPR3, SY124, SY127, SY128,
SY130,
SY133 and SY134 from AZFb; SY145 and SY152 from proximal
Y chromosome microdeletions: isolation
AZFc/
of DNA and sequence-tagged site analysis d locus; SY242, SY208, SY254 and SY255 from AZFc; and SY157 (
Genomic DNA was extracted from the peripheral blood leukocytes heterochromatic distal Yq region).
using
Multiplex PCR was performed using the Y chromosome AZF A
the MagNA Pure Compact instrument (Roche Diagnostics Corpor
nalysis
ation)
and the MagNA Pure Compact Nucleic Acid Isolation I method ( System (Promega, USA) under the PCR conditions recommended
by the
Roche
Male factor and idiopathic recurrent miscarriage

byguest
onMarc
h17,201
6http://h
umrep.o
supernatant.
No ejaculate displayed signicant (.1 million/ml) levels of lexfordjo
urnals.o
ukocytes
rg/Dow
(a known and relevant source of ROS), as demonstrated by Quic nloaded
from
k Panop-

manufacturer. For analysis, the microdeletions PCR products were

run by
electrophoresis on a 2% agarose gel impregnated with ethidium bro
mide for
visualization under UV light. Failure of amplication for a given STS
was contic (QCA, Barcelona, Spain) staining of semen smears. The se
rmed twice. DNA from a fertile male and water served as negative
nsitivity of
Y chromothe test is 10 000 leukocytes/ml.
some microdeletion controls, whereas DNA from man with a Y chro
mosome
DNA fragmentation (SCD)
microdeletion served as a positive control for multiplex reactions.

Sperm analysis
Semen parameters of every ejaculate were evaluated by two inde
pendent
observers. After 10 min liquefaction of the semen at 378C and 5
% CO 2,
the samples were examined for concentration and motility in a
Mackler
chamber (Se Laboratories, Tel Aviv, Israel), according to th
e WHO
guidelines (12). Results were only accepted when the d
ifferences
between two observers were less than 5%.
Two semen aliquots were taken from each semen sample to
be processed for ow cytometry and sperm chromatin dispersion (SCD
) tests.
The samples were directly frozen by immersion in liquid nitrogen
(SCD)
or in cold ethanol 70% (for ow cytometry experiments). These pr
otocols
have been validated previously and have been demonstrated to be
safe in
regard to the maintenance of sperm characteristics, allowing the an
alysis of
all the samples in the same experiment.
The total number of spermatozoa collected for the experiments r
anged
between 1 and 10 million/ml in order to provide a sufcient quan
tity for
the analysis after centrifugation for 10 min at 400g and eliminatio
n of the

Eppendorf tubes with gelled aliquots of low-melting point agar

ose were
placed for 5 min in a water bath at 901008C to fuse the
agarose, and subsequently in another water bath at 378C. After 5 min incubation at
378C for
temperature equilibration, 60 ml of the diluted semen sample were
added to
the Eppendorf tube and mixed with the fused agarose. 20
ml of the
semen-agarose mix were pipetted onto slides precoated with
agarose and
covered with a 22 22 mm coverslip. The slides were placed
on a cold
plate in the refrigerator (48C) for 5 min to allow the agarose to p
roduce a
microgel with the sperm cells embedded within. The coverslips we
re gently
removed and the slides immediately immersed horizontally in an a
cid solution,
previously prepared by mixing 80 ml of HCl from an Eppendorf t
ube with
10 ml of distilled water and incubated for 7 min. The slides were h
orizontally
immersed in 10 ml of the lysing solution for 25 min. After 5 min w
ashing in a
tray with abundant distilled water, the slides were dehydrated in i
ncreasing
concentrations of ethanol (7090100%) for 2 min each and then
air-dried.
For bright-eld microscopy in the improved SCD test (Halo
spermw

kit), slides were horizontally covered with a mix of Wrights staini quantied using ow cytometry (the OxyDNA Assay is perform
ed and
ng solobserved directly in the cells). Sperm cells were washed twice in
ution (Merck, Germany) and phosphate buffer solution
PBS and
(Merck,
xed-permeabilized with ice-cold ethanol 70% at 2208C. EthanolGermany) (1:1) for 510 min with continuous airow. Slid
treated
es were
cells were pelleted at 230 g for 5 min, and the supernata
briey washed in tap water and allowed to dry. Strong staining is p
nts were
referred
removed and pellets were washed twice with PBS. Sperm cells wer
to easily visualize the periphery of the dispersed DNA loop halos.
e resusFor this study, a minimum of 500 spermatozoa per sample were
pended in 3 ml of Wash Solution [Tris-buffered Saline/TWEE
scored
Nw 20
under the 100 objective of the microscope.
Detergent (TBST), containing Thimerosal], and centrifuged at 2
30g for
DNA oxidation (8-oxoguanine DNA sperm 5 min, after which the supernatant was removed and non-specic
binding
measurement by ow cytometry)
sites were blocked with 50 ml freshly prepared Blocking Solution
The OxyDNA Assay (OxyDNA assay kit, Calbiochem, Barcelona, S
for 1 h
pain) is
at 378C. Then 3 ml of Wash Solution was added and washes were
based on the direct binding of a uorescent probe to 8-oxoguanine repeated
moieties
twice, and then incubated overnight with 100 ml FITC-Conjugate
in the DNA of xed cells (Aguilar et al., 2009). Fluorescence can at 48C.
then be
Samples were washed twice (Wash Solution) and resuspended
in 2 ml
PBS. Cells were maintained in the dark on ice until their resuspe
nsion in
FACS uid and were read in a ow cytometer with FITC lters.
Flow cytometry analysis was performed on an Epics Elite Fl
ow Cytometer (Coulter Cytometry, Hialeah, FL, USA) using an argonion laser
tuned at 488 nm and 15 mW. Debris was excluded by the a
nalysis of
the scatter properties. At least 10 000 events per sample wer
e stored
in list-mode les. Data were expressed as percentage of stai
ned cells
(compared with negative controls) and uorescence intensity m
easured
as uorescence arbitrary units.

Statistical analysis
Statistical analysis was performed using the Statistical Package f
or Social
Science version 17.0 (SPSS Inc., Chicago, IL, USA). Categor
ical data
were expressed as number and percentage, and numerical
data as
means with 95% condence intervals (95% CI). When a quantitativ
e analysis of the data was performed, groups were compared by analysis
of variance, with Bonferroni and Scheffes post hoc analysis if more
than two
groups were compared and a Students t-test when only two
groups
were compared. Crosstabs and chi-square tests were applied for
the qualitative analysis of the data. The predictive power of sperm DNA ox
idation
and fragmentation to forecast miscarriage were determined by
receiver
operator curve (ROC) analysis. Signicance was dened as P , 0
.05.

Results
Demographics

The baseline comparisons of the three groups are presented in T million/ml in sperm concentration, from 0.09 to 7.42 million in
total
able I.
motile sperm cells and from 3 to 53% in A + B forms), as ex
In brief, age was signicantly lower among SD males (ranging
pected
from
20 to 34 years) than among men in the SO (range 2644 considering the inclusion criteria.
When sperm quality of the SD males was compared with that
years)
of
the
and RSA (range 3042 years) groups. Sperm quality was also
RSA
group (concentration ranging from 11 to 110 million/ml),
signia
sigcantly higher in the rst group, in which the values of sperm c
nicant difference was observed in sperm concentration, but
oncentration (ranging from 29 to 250 million/ml), sperm motility (r not in
other parameters. However, in both SD and RSA, sperm
anging
from 29% to 67% of A + B forms) and the total number of concentration was still normal (.20 million/ml).
motile
The percentages of smokers were statistically comparable
sperm cells (ranging from 80 to 278 million) were superior, partic
in
the
ularly
three
groups, and in both males and females, thus ruling
in comparison with the selected group of SO (ranging from 0.3
out
any
to 5
confounding inuence of this parameter on the results. Nevert
heless,

Bellver et al.

byguest
onMarc
h17,201
6http://h
Table I Demographics of the study population.
umrep.o
xfordjo
Control donors
Severe oligospermia
Recurrent miscarriage P-value
........................................................................................................................................................ urnals.o
rg/Dow
.....................................
nloaded
Male
from
a,b
a
b
Age (years)
26.50 (24.5828.52)
34.00 (32.3835.32)
35.24 (33.8436.64)
,0.001
Presence of Y chromosome microdeletions

0/30 (0%)

Smoker (yes/total and %)

10/30 [33.33% (16.5650.20)]

(8.2838.46)]

3/30 [10% (020.74)]

0/30 (0%)

0/30 (0%) ns
7/30 [23.33%

ns
a,b

Mean number of cigarettes per day

0.15 (00.4)

5.21 (2.068.35)

4.28 (0.877.69)

Ejaculate volume (ml)

4.02 (3.224.83)

4.38 (3.565.19)

3.39 (3.003.78)

Sperm concentration (million/ml)

74.70 (55.1394.28)

% of A + B sperm

a,b
a

ns

51.75 (40.1763.33)
a,b

b,c

,0.001

,0.001

26.38 (20.6632.10)

43.88 (38.5549.21)

119.49 (92.76146.218)
,0.001

3.31 (1.854.77)

83.91 (55.04

34.67 (33.6135.72)

31.94 (30.4733.40)

ns

Smoker (Yes/Total and %)

13/30 (43.33 (25.6061.06))8/30 (26.67 (10.8442.50))

Mean number of cigarettes per day

3.60 (1.505.70)

3.21 (0.555.86)

ns

3.76 (3.39

7.42 (6.937.41)

Total motile sperm (million)

b
112.79)

46.62 (42.8650.38)

a,c

3.15 (2.044.25)

0.019

a,b

Female
Age (years)

Mean number of miscarriages per patient

4.14)
Mean week at pregnancy termination

ns

Data are expressed as means or proportions with 95% condence intervals. Superscript letters (a, b, and c) denote statistical signicance between groups within th
e same row sharing the
superindex, and ns denotes non-signicant differences. P-values of ,0.05 were considered statistically signicant. A + B sperm are dened as sperm cells with
progressive motility either
rapid and straight (A) or less rapid or presenting changing direction (B). Normal ranges are considered above 50% of A + B forms.

a trend toward a lower mean number of cigarettes smoked per day

was observed in the SD group (ranging from 0 to 3 cigarettes/day)
in comparison with the SO (ranging from 0 to 20 cigarettes/day)
and RSA (ranging from 0 to 30 cigarettes/day) groups.
The mean number and mean gestational age of miscarriages are also
presented in Table I. The number of pregnancy losses among the
couples of the RSA group ranged from 3 to 7, whereas the mean
gestational week in which pregnancy was lost ranged from 5 to 14.

Males genetic contribution to RSA

We were unable to nd any Y chromosome microdeletions in any of

the males included in the SD, SO or RSA groups.
Regarding sperm DNA quality, two parameters were assessed:
sperm DNA fragmentation and sperm DNA oxidation. The results
can be found in Figs 1 and 2.
In short, the mean number of sperm cells with fragmented DNA was
24.06% (95% CI 20.8727.23) in the SD group, in comparison with
46.01% (95% CI 40.4051.62) in the SO group and 33.48%
Figure 1 Sperm DNA fragmentation analysis results depending
(95% CI
on
28.2438.72) in the RSA group, with statistically signicant
the origin of the sperm sample. The horizontal lines represent
differences
the
between the SD and RSA and between the SD and SO groups. medians, the box width the interquartile range, the lower li
The results of the analysis of sperm DNA oxidation in th ne of
the box the rst quartile, the upper line of the box the third qua
e three
rtile,
groups were comparable in terms of percentage of oxidize
and the whiskers are maximum and minimum.
d cells,
with values of 32.14% (95% CI 24.6239.65) in the S
D group,
24.01% (95% CI 16.8531.17) in the SO group and 32.43 The predictive value of sperm DNA fragmentation and DNA
oxi% (95%
dation for forecasting recurrent miscarriage are shown in Fig.
CI 24.6240.24) in the RSA group.
When the DNA oxidation staining intensity was compare 3. The
byguest
low values of the areas under the curve indicate the absence ofonMarc
d, the
h17,201
mean sperm cell DNA oxidation was 87.25 (95% CI 61.16 a pre6http://h
dictive power of sperm DNA fragmentation or oxidation anal umrep.o
113.34)
xfordjo
relative uorescence units (RFU) in the SD group, 87. ysis in
urnals.o
spermatozoa with respect to RSA.
58 RFU
rg/Dow
If only males with non-severe sperm alterations (i.e. mem nloaded
(95% CI 63.69111.48) in the SO group and 68.04 RFU
bers of
from
(95% CI
53.5582.53) in the RSA group. The differences betw the SD and RSA groups) are considered, the capacity of the
sperm
een the
groups were not signicant.
Male factor and idiopathic recurrent miscarriage

5
affected by this condition is not advisable since their alterat

ions are
not currently demonstrated to present a causative role.
DNA oxidative damage is indicative of a number of conditions
involving the generation of oxygen-free radicals. This damage could
be the
primary source of DNA fragmentation (Aguilar et al., 2009). In
fact, a
defective antioxidant system activity could be related to an imp
airment
of chromatin packaging (Tarozzi et al., 2009). However, this is
not the
only the mechanism of DNA fragmentation (Erenpreiss et al.
, 2006;
Practice Committee of American Society for Reproductive Med
icine,
2008). This is why we considered DNA fragmentation and DN
A oxidation independently because, despite the association tha
t exists
between the two, they are not synonymous.
We analyzed DNA oxidation by ruling out all the external f
actors
that have been related to increased free radical damage
(see the
Materials and Methods section; Practice Committee of A
merican

Society for Reproductive Medicine, 2008; Viloria et al., 2009

). Similarly, a smoking habit, which is also associated with increas
ed DNA
damage in animal models and humans (Kunzle et al., 2003; Pas
qualotto
et al., 2008), was identied in a similar proportion in the thre
e study
groups. Despite the fact that almost 27% of women include
d in the
RSA group were smokers (Table I), the mean number of ci
garettes
smoked per day was low (3.2). Still, the number of
cigarettes
smoked daily in the RSA and the SO groups was signicantl
y higher
than in the SD group and this may contribute to the higher
rate of
DNA fragmentation. However, a smoking has not been
clearly
related

to

unexplained

recurrent

miscarriage

in

recent

studies
(Zhang et al., 2010). The correlation between male age an
d sperm
DNA fragmentation is controversial, with some reports show
ing an
increase in sperm DNA damage with age (Plastira et a
l., 2007;
Vagnini et al., 2007) and others nding no relationship
(Winkle
et al., 2009), even after 40 years. In our study, we aimed
to avoid
this possible confounding factor by limiting the age of the study
population to 45 years old, although the SD group was signicantly y
ounger
Figure 2 Sperm DNA oxidation analysis results depending on
the
origin of the sample. Results expressed as the percentage of oxi
dized
cells (positively stained, top) and mean staining levels (bottom).
The
horizontal lines represent the medians, the box width the interq
uartile range, the lower line of the box the rst quartile, the upper l
ine of
the box the third quartile, and the whiskers are maximu
m and
minimum. The dots outside the box are abnormally high
values.
Asterisks here denote extreme values, but not statistical differ
ences
among groups.

DNA fragmentation and oxidation measurements to discri

minate
between males from couples with or without recurrent misc
arriages
can be seen in Fig. 4. Only the percentage of oxidized cells pr
esented
a mild capacity to adequately predict RSA occurrence, though
it was
not sufcient for it to be applied as a diagnostic tool (
where an

optimal area under the curve should be .0.8).

Discussion
In the present study, Y chromosome microdeletions and sper
m DNA
fragmentation and oxidation did not prove to be related to
RSA of
unknown origin, at least in men with normal or mildly altere
d sperm
samples. Therefore, the assessment of these parameter
s in men

compared with the SD and the RSA groups. This also may hav Increased levels of impairment to DNA integrity in severe olig
e conospertributed to the higher DNA fragmentation rates in the latt mic males have been reported previously in the literature an
er two
d could
groups.
explain the poor results described in assisted reproduction treat
According to our results, sperm DNA quality, in terms of ments
DNA
by some authors (Meseguer et al., 2008). However, in
fragmentation and oxidation, two of the main symptoms of r a recent
eactive
meta-analysis of 13 studies involving infertile couples, despite
oxygen species attack (Garrido et al., 2004; Aguilar et al., a small
2009), is
but statistically signicant association between sperm DNA i
not related to recurrent idiopathic spontaneous abortions. Alth ntegrity
ough
test results and pregnancy outcome in IVF and ICSI cycles
statistical differences were conrmed among the groups, th , sperm
ese do
DNA integrity did not prove to be predictive of pregnancy out
not point to an increase in sperm DNA fragmentation or oxi come
dation
and provided no clinical usefulness in the routine practice
in men from RSA couples with respect to the other groups of pat (Collins
ients.
et al., 2008).
Interestingly, the group with the highest sperm DNA fragmen The inuence of sperm DNA damage on sporadic pregnanc
tation
y loss
was that formed by males yielding severely oligospermic s has been mainly evaluated after assisted reproductive tecbyguest
onMarc
amples.
hnology.

Bellver et al.

h17,201
6http://h
umrep.o
xfordjo
urnals.o
rg/Dow
nloaded
from

Figure 3 ROC curve analysis of the predictive value of sperm DNA fragmentation and oxidation to forecast RSA.

Though controversial, a recent systematic review and metaanalysis of

11 studies concluded that the risk of pregnancy loss after IVF a
nd ICSI
was 2.5 higher in men with DNA damage, but differed depend
ing on

the assay (SCSA or TUNEL) employed (Zini et al., 2008). Ho

wever,
only two previous reports have considered couples with unexp
lained

RSA. In both studies (Carrell et al., 2003a; Qiu et al., 2008), D Similar levels of DNA oxidation, in terms of both the percent
age of
NA fragcells with oxidized DNA and mean intensity of oxidation
mentation was signicantly higher in the RSA group tha , were
n among
found in all the three groups assessed (SD, SO and mal
donors of known fertility and men from the general po es from
pulation.
couples with RSA). Only the percentage of oxidized cells ex
Classic parameters of sperm quality correlated with sperm DN hibited
A fraga mild capacity to adequately predict RSA occurrence (
mentation in one of the studies (Qiu et al., 2008), but not in t Fig. 4),
he other
although this predictive power does not appear to be sufci
(Carrell et al., 2003b). Our results are similar, since sperm sa ent for
mples of
its implementation as a diagnostic tool (where an optim
the RSA group were normal or only mildly altered, and DNA fra al area
gmenunder the curve should be .0.8).
tation in the same group was signicantly higher than amo One of the most signicant ndings of this work is that no
ng fertile
differdonor controls. However, when we assessed the clinical pr ences in the prevalence of Y chromosome microdeletions
edictive
were
value of this nding by means of ROC curves, it was abse found between the groups. In fact, no presence of Y microde
nt due to
letions
the overlap of values between men belonging to both was detected in any of the 90 subjects included. These res
byguest
groups
ults are
(Fig. 4). The ndings were the same when oligozooper in complete contrast to what was initially published by Dewa onMarc
h17,201
mic men
6http://h
n et al.
were taken into account in the ROC curves (Fig. 3). T (2006). In their work, 14 out of 17 males (82%) from RSA umrep.o
xfordjo
herefore,
couples
urnals.o
despite the fact that DNA fragmentation seems to be increpresented microdeletions in one or more of the four segm rg/Dow
ased in
nloaded
ents of
the sperm of men with unexplained RSA, the clinical value o the proximal AZFc region studied, whereas this percentage drofrom
f sperm
pped
DNA integrity analysis as a predictive tool is negligible. Simil to 20% in infertile males with severely impaired spermat
arly, the
ogenesis
causative role of DNA fragmentation in RSA cannot be demons and was non-existent among men with proven fertility. R
trated.
ecently,
To the best of our knowledge, this is the rst study in which Karaer et al. (2008) also detected Y chromosome microdeleti
sperm
ons in
oxidative stress has been evaluated in relation to idiopat one or more of the four STSs assessed in the AZFb an
hic RSA.
d AZFd
regions in 16% of men from couples with RSA and repor
ted no
case among fertile men. However, the percentage of microde
letions
detected differed hugely between these two studies (82 versus
16%)
and the regions affected were mostly different.
Male factor and idiopathic recurrent miscarriage

Figure 4 ROC curve analysis of the predictive value of sperm DNA fragmentation and oxidation to forecast RSA considering t
hose males with
normal sperm parameters.

byguest
onMarc
In our work, the increased sample size should have allowed chromosome microdeletions in our SO group, since these alterh17,201
ations
clari6http://h
cation of the considerable differences reported previously by have been described in higher proportions in men with vumrep.o
xfordjo
ery low
Dewan
urnals.o
et al. (2006). At least, some Y chromosome microdeletions sperm counts (,1 million/ml; Martinez et al., 2000). In rg/Dow
addition,
should
nloaded
have been observed given that we have considered simil despite the strict inclusion criteria employed, the sample sifrom

ze conar populations. In addition, we assessed 20 Y-linked STSs belonging sidered in the study (90 men) may also represent a limit
ation for
to four
AZF loci regions (AZFa, AZFb, AZFc and AZFd) rather th the results obtained.
Similarly to our study, other recent reports have not iden
an only
four STSs in one or two AZF regions (Dewan et al., 2006 tied Y
chromosome microdeletions in male partners of RSA couples.
; Karaer
Kaare
et al., 2008).
Furthermore, Dewan et al. (2006) argue that their patien et al. (2008) evaluated 40 men from couples with at least thr
ee mists were
selected from a tertiary referral center that receives a dispro carriages and analyzed 33 STS loci spanning the whole Y chrom
osome
portionate number of RSA patients and that these patients were likely plus the 4 STS loci used by Dewan et al. (2006). No microdelet
ion was
to have
failed evaluations and treatments prior to referral, thus ske found in any case. A study of a Chinese population by Lu et al.
(2008)
wing the
results. In our study, patients were selected following strict i also failed to detect any case of Y chromosome microdeletions
in 26
nclusion
criteria. The couples described as infertile in the Dew chorionic villous samples of abortive male embryos or in 5
1 blood
an et al.
report exhibited similar conditions of sperm quality as those samples of men whose wives had experienced RSA. They
studied
in our
study, with the exception of those males presenting azoo 12 STSs in the AZFa, AZFb and AZFc regions of Yq11.2.
Differences in the presence of Y chromosome microdeletions
spermia,
who were not included in our population due to the impossibi could
be due to variations in the study populations (race and e
lity of
performing the described DNA fragmentation and oxidatio thnicity),
subject selection criteria, the STSs considered and method
n tests
with their samples. This fact may partly explain the l ological
ack of Y

8
aspects of the analysis. Nevertheless, no clear evidence
currently

Bellver et al.

exists regarding the relationship between idiopathic RSA

and Y
chromosome microdeletions.

outcome of ovum donation cycles. Fertil Steril 2009; June 1

9 [Epub
ahead of print].
Al-Hassan S, Hellani A, Al-Shahrani A, Al-Deery M, Jaroudi K,
Coskun S.
Sperm chromosomal abnormalities in patients with une
xplained
recurrent abortions. Arch Androl 2005;51:6976.
Arredondo F, Noble LS. Endocrinology of recurrent pregnancy loss.
Semin
Reprod Med 2006;24:3339.
Avendano C, Franchi A, Duran H, Oehninger S. DNA fragmenta
tion of
normal spermatozoa negatively impacts embryo quality and
intracytoplasmic sperm injection outcome. Fertil Steril 2009; M
arch 30
[Epub ahead of print].
Bernardini LM, Calogero AE, Bottazzi C, Lanteri S, Vent
urini PL,
Burrello N, De Palma A, Conte N, Ragni N. Low total normal
motile
count values are associated with increased sperm disomy and d
iploidy
rates in infertile patients. Int J Androl 2005;28:328336.
Bhattacharya SM. Association of various sperm paramet
ers with
J.B. designed the study, selected the patients and wrote th
unexplained repeated early pregnancy losswhich is most imp
ortant?
e paper;
Int Urol Nephrol 2008;40:391395.

To conclude, there seems to be no clinical use for the asse

ssment
of Y chromosome microdeletions and DNA fragmentation an
d oxidation when dealing with couples with idiopathic RSA, at least
when
the sperm samples of the male partner are not severely i
mpaired.
Until such a clinical use is demonstrated, this assessment
can only
serve research purposes. Given the continuous description of
sperm
molecular factors related to male infertility, further resea
rch into
RSA using the newly discovered molecules as markers should
be conducted. Recent advances in microarray technology applied to
sperm
will undoubtedly constitute powerful tools in the characteriza
tion of
these factors.

Authors roles

M.M. helped in the design of the study and performed the DN

A oxidation studies; L.M. performed the DNA fragmentation
studies;
S.G.-H. collaborated in the DNA oxidation studies and
in the
writing of the paper; M.A.M.B. included many of the patients a
nalyzed
and helped in the design of the paper; A.L.G. performe
d the Y
helped in the
desi
chromosome microdeletion analysis; J.R.
gn and
monitored the study; A.P. helped in the statistical approach and
monitored the study and the paper; N.G. designed the study, perf
ormed
the statistical analysis, coordinate the sperm and blood anal
ysis and
helped in the writing of the paper.

Acknowledgements
The authors would like to acknowledge the Andrology lab tech
nicians
Yolanda Marquez, Carolina Rico, Paloma Rodrguez, Alicia M
encas,
Alicia Zurilla, y Vicente Centelles for their help in the samples
management and storage.

References
Aguilar C, Meseguer M, Garcia-Herrero S, Gil-Salom M, OCon
nor JE,
Garrido N. Relevance of testicular sperm DNA oxidation f
or the

Borini A, Tarozzi N, Bizzaro D, Bonu MA, Fava L, Flamigni C, Coticc

21:20612064.
hio G.
Erenpreiss J, Elzanaty S, Giwercman A. Sperm DNA damage in me
Sperm DNA fragmentation: paternal effect on early postn from
implantation
infertile couples. Asian J Androl 2008;10:786790.
embryo development in ART. Hum Reprod 2006;21:28762881.
Garrido N, Meseguer M, Simon C, Pellicer A, Remohi J. Pro-oxidative
Carrell DT, Liu L, Peterson CM, Jones KP, Hatasaka HH, Eric
and
kson L,
anti-oxidative imbalance in human semen and its relation wit
Campbell B. Sperm DNA fragmentation is increased in
h male
couples
fertility. Asian J Androl 2004;6:5965.
with unexplained recurrent pregnancy loss. Arch Androl
Garrido N, Martinez-Conejero JA, Jauregui J, Horcajadas JA, Simo
2003a;
n C,
49:4955.
Remohi J, Meseguer M. Microarray analysis in sperm from ferti
Carrell DT, Wilcox AL, Lowy L, Peterson CM, Jones KP, Erick
le and
son L,
infertile men without basic sperm analysis abnormalities re
Campbell B, Branch DW, Hatasaka HH. Elevated sperm chromo
veals a
some
signicantly different transcriptome. Fertil Steril 2008a;
aneuploidy and apoptosis in patients with unexplained re
91(Suppl
current
4):13071310.
pregnancy loss. Obstet Gynecol 2003b;101:12291235.
Garrido N, Remohi J, Martinez-Conejero JA, Garcia-Herrero S,
Christiansen OB, Nielsen HS, Kolte AM. Future directions of
Pellicer A,
failed
Meseguer M. Contribution of sperm molecular features to e
implantation and recurrent miscarriage research. Reprod Biomed
mbryo
Online
quality and assisted reproduction success. Reprod Biomed
2006;13:7183.
byguest
Online
Cohen-Bacrie P, Belloc S, Menezo YJ, Clement P, Hamidi J,
onMarc
2008b;17:855865.
Benkhalifa M.
h17,201
Harper J, Sermon K, Geraedts J, Vesela K, Harton G, Thor
6http://h
Correlation between DNA damage and sperm paramet
nhill A,
umrep.o
ers: a
Pehlivan T, Fiorentino F, SenGupta S, de Die-Smulders C et al.xfordjo
prospective study of 1,633 patients. Fertil Steril 2009;91:1801
urnals.o
What
1805.
next for preimplantation genetic screening? Hum Reprod rg/Dow
Collins JA, Barnhart KT, Schlegel PN. Do sperm DNA integrit
nloaded
2008;
from
y tests
23:478480.
predict pregnancy with in vitro fertilization? Fertil Steri
Kaare M, Painter JN, Ulander VM, Kaaja R, Aittomaki K. Sex chromo
l 2008;
some
89:823831.
characteristics
and
recurrent
miscarriage.
Fertil
Steril
Devi Wold AS, Pham N, Arici A. Anatomic factors in recurrent preg
2008;
nancy
90:23282333.
loss. Semin Reprod Med 2006;24:2532.
Karaer A, Karaer K, Ozaksit G, Ceylaner S, Percin EF. Y chro
Dewan S, Puscheck EE, Coulam CB, Wilcox AJ, Jeyend
mosome
ran RS.
azoospermia factor region microdeletions and recurrent preg
Y-chromosome microdeletions and recurrent pregnancy loss.
nancy
Fertil
loss. Am J Obstet Gynecol 2008;199:662.e1662.e5.
Steril 2006;85:441445.
Kovalevsky G, Gracia CR, Berlin JA, Sammel MD, Barnhart KT. Eval
Empson M, Lassere M, Craig J, Scott J. Prevention of recurrent misc
uation
arriage
of the association between hereditary thrombophilias and rec
for women with antiphospholipid antibody or lupus anticoa
urrent
gulant.
pregnancy loss: a meta-analysis. Arch Intern Med 2004;164:558
Cochrane Database Syst Rev 2005;CD002859.
563.
Erenpreiss J, Bungum M, Spano M, Elzanaty S, Orbidans J, Giwerc
Krabbendam I, Franx A, Bots ML, Fijnheer R, Bruinse HW. Thrombop
man A.
hilias
Intra-individual variation in sperm chromatin structure assay
and recurrent pregnancy loss: a critical appraisal of the literatur
parameters
e. Eur J
in men from infertile couples: clinical implications. Hum Reprod
Obstet Gynecol Reprod Biol 2005;118:143153.
2006;
Male factor and idiopathic recurrent miscarriage
9
Kunzle R, Mueller MD, Hanggi W, Birkhauser MH, Dres
cher H,
Bersinger NA. Semen quality of male smokers and nonsm
okers in
infertile couples. Fertil Steril 2003;79:287291.
Larson-Cook KL, Brannian JD, Hansen KA, Kasperson KM, Aam
old ET,
Evenson DP. Relationship between the outcomes of
assisted
reproductive techniques and sperm DNA fragmentation as me
asured
by the sperm chromatin structure assay. Fertil Steril 2003;80:89
5902.

Lashen H, Fear K, Sturdee DW. Obesity is associated with increa

sed risk
of rst trimester and recurrent miscarriage: matched cas
econtrol
study. Hum Reprod 2004;19:16441646.
Lewis SE, Agbaje I, Alvarez J. Sperm DNA tests as useful a
djuncts to
semen analysis. Syst Biol Reprod Med 2008;54:111125.
Lin MH, Kuo-Kuang Lee R, Li SH, Lu CH, Sun FJ, Hwu Y
M. Sperm
chromatin structure assay parameters are not related to fe
rtilization
rates, embryo quality, and pregnancy rates in in vitro fertiliz
ation and

intracytoplasmic sperm injection, but might be related to spont Plastira K, Msaouel P, Angelopoulou R, Zanioti K, Plastiras A, Pot
hos A,
aneous
Bolaris S, Paparisteidis N, Mantas D. The effects of age
abortion rates. Fertil Steril 2008;90:352359.
on DNA
Lu HY, Cui YX, Xia XY, Shi YC, Yang B, Shao Y, Ge YF, Yao
fragmentation,
chromatin
packaging
and
conventional
B, Li XJ,
semen
Huang YF. AZF microdeletions are not related with r
parameters in spermatozoa of oligoasthenoteratozoospermic pa
ecurrent
tients.
spontaneous abortion. Zhonghua Nan Ke Xue 2008;14:1099
J Assist Reprod Genet 2007;24:437443.
1102.
Porter TF, Scott JR. Evidence-based care of recurrent miscarria
Martinez MC, Bernabe MJ, Gomez E, Ballesteros A, Landeras J, Glo
ge. Best
ver G,
Pract Res Clin Obstet Gynaecol 2005;19:85101.
Gil-Salom M, Remohi J, Pellicer A. Screening for AZF deletion in
Practice Committee of American Society for Reproductive Medicin
a large
e. The
series of severely impaired spermatogenesis patients. J Andr
clinical utility of sperm DNA integrity testing. Fertil Ste
ol 2000;
ril 2008;
21:651655.
90:S178S180.
Meseguer M, Garrido N, Simon C, Pellicer A, Remohi J. Concentrat
Qiu Y, Wang L, Zhang L, Yang D, Zhang A, Yu J. Analysis o
ion of
f sperm
glutathione and expression of glutathione peroxidases 1 and 4 i
chromosomal abnormalities and sperm DNA fragmentation in inf
n fresh
ertile
sperm provide a forecast of the outcome of cryopreserv
males. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 2008;25:681
ation of
685.
human spermatozoa. J Androl 2004;25:773780.
Rai R, Regan L. Recurrent miscarriage. Lancet 2006;368:601611.
Meseguer M, Martinez-Conejero JA, OConnor JE, Pellicer A, Re
Rey E, Kahn SR, David M, Shrier I. Thrombophilic disorders and fet
mohi J,
al loss:
Garrido N. The signicance of sperm DNA oxidation in
a meta-analysis. Lancet 2003;361:901908.
embryo
Rubio C, Gil-Salom M, Simon C, Vidal F, Rodrigo L, Minguez Y,
development and reproductive outcome in an oocyte
Remohi J,
donation
Pellicer A. Incidence of sperm chromosomal abnormalities in
program: a new model to study a male infertility prognosti
a risk
c factor.
population: relationship with sperm quality and ICSI outcom
Fertil Steril 2008;89:11911199.
e. Hum
Nicopoullos JD, Gilling-Smith C, Almeida PA, Homa S, Norman-Taylor
Reprod 2001;16:20842092.
JQ,
Tarozzi N, Nadalini M, Stronati A, Bizzaro D, Dal Prato L, Coticc
Ramsay JW. Sperm DNA fragmentation in subfertile men: the eff
hio G,
ect on
Borini A. Anomalies in sperm chromatin packaging: implicati
the outcome of intracytoplasmic sperm injection and correlati
ons for
on with
assisted reproduction techniques. Reprod Biomed Online
sperm variables. BJU Int 2008;101:15531560.
2009;
Pandey MK, Rani R, Agrawal S. An update in recurrent spo
18:486495.
ntaneous
Tavalaee M, Razavi S, Nasr-Esfahani MH. Inuence of sperm ch
abortion. Arch Gynecol Obstet 2005;272:95108.
romatin
Pang MG, Kim YJ, Lee SH, Kim CK. The high incidence of meiotic
anomalies on assisted reproductive technology outcome. Fert
errors
il Steril
increases with decreased sperm count in severe male factor inf
2009;91:11191126.
ertilities.
Vagnini L, Baruf RL, Mauri AL, Petersen CG, Massaro FC, Po
Hum Reprod 2005;20:16881694.
ntes A,
Pasqualotto FF, Umezu FM, Salvador M, Borges E Jr, Sobr
Oliveira JB, Franco JG Jr. The effects of male age on sperm DNA d
eiro BP,
amage
Pasqualotto EB. Effect of cigarette smoking on antioxidant le
in an infertile population. Reprod Biomed Online 2007;15:514519.
vels and
Viloria T, Meseguer M, Martinez-Conejero JA, OConnor JE, Rem
presence of leukocytospermia in infertile men: a prospectiv
ohi J,
e study.
Pellicer A, Garrido N. Cigarette smoking affects specic
Fertil Steril 2008;90:278283.
sperm
oxidative defense but does not cause oxidative DNA da
mage in
infertile men. Fertil Steril 2009; April 24 [Epub ahead of print].
Winkle T, Rosenbusch B, Gagsteiger F, Paiss T, Zoller N. The corr
elation
between male age, sperm quality and sperm DNA fragment
ation in
320 men attending a fertility center. J Assist Reprod Genet 2009;2
6:4146.
World Health Organization. Laboratory Manual of the WHO
for the
Examination of Human Semen and Sperm-Cervical Mucus In
teraction.
Annali dellIstituto Superiore di Sanita, 1999.
Zhang BY, Wei YS, Niu JM, Li Y, Miao ZL, Wang ZN. Risk fact
ors for

unexplained recurrent spontaneous abortion in a population with an increased risk of pregnancy loss after IVF and ICSI: sy
from
stematic
southern China. Int J Gynaecol Obstet 2010;108:135138.
review and meta-analysis. Hum Reprod 2008;23:26632668.
Zini A, Boman JM, Belzile E, Ciampi A. Sperm DNA damage is asso
ciated