If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Microscopy, Laboratory of Cell Biology, Unit for Multidisciplinary Research in Biomedicine (UMIB), Institute of Biomedical Sciences Abel Salazar (ICBAS), University of Porto (UP), Porto, Portugal
Centre for Reproductive Genetics Prof. Alberto Barros (CGR-ABarros), 4100-009 Porto, PortugalDepartment of Genetics, Faculty of Medicine, UP, 4200-319 Porto, Portugal
Department of Microscopy, Laboratory of Cell Biology, Unit for Multidisciplinary Research in Biomedicine (UMIB), Institute of Biomedical Sciences Abel Salazar (ICBAS), University of Porto (UP), Porto, Portugal
In this prospective comparative study, sperm DNA fragmentation (sDNAfrag) was compared at each step of a sequential semen preparation, with semen parameters according to their degree of severity. At each step (fractions) of the sequential procedure, sDNAfrag was determined: fresh (Raw), after gradient centrifugation, washing, and swim-up (SU) for 70 infertile men enrolled in intracytoplasmic sperm injection cycles. sDNAfrag significantly (P = 0.04; P < 0.0001) decreased throughout the steps of semen preparation, with centrifugation and washing not increasing it. A negative correlation to sperm motility was observed in Raw and SU fractions, and a higher sDNAfrag was observed in samples with lower semen quality. Our results confirm that the steps of the sequential procedure do not compromise sperm DNA integrity and progressively decreased sDNAfrag regardless of the sperm abnormality and that semen parameters with lower quality present higher sDNAfrag. Four distinct patterns were observed, of which the entire sperm head staining was the pattern most expressed in all studied fractions. Additionally, the sperm head gene-rich region staining pattern was reduced by the procedure. This suggests that pattern quantification might be a useful adjunct when performing sDNAfrag testing for male infertility.
Spermatogenesis is the process responsible for male gamete production, which comprises the mitotic divisions of spermatogonia, the meiotic divisions of spermatocytes, the differentiation of spermatids (spermiogenesis) and, finally, spermiation (
). Chromatin compaction begins during spermiogenesis with replacement of the classic histones by protamines, whereas the final compaction occurs during sperm maturation in the epididymus where protamines are cross-linked by disulphide bonds. Compaction makes the sperm chromatin and epigenome resilient to exogenous damage during migration from the seminiferous tubules to the oviduct (
). In some men, however, this compact structure is not fully attained, leaving DNA more prone to damage as may be visualized through DNA fragmentation (
). Various putative causes of sperm DNA fragmentation (sDNAfrag) have been discussed: apoptosis, abortive spermatogenesis, genetic factors, meiosis and spermiogenesis defects, and oxidative stress (
Sperm DNA damage can be assessed in different ways. Direct assays, including TdT (terminal deoxynucleotidyl transferase)-mediated dUDP nick-end labelling (TUNEL) and single-cell gel electrophoresis (COMET) using neutral conditions, measure actual damage, i.e. the spontaneous or accidental breaking of DNA strands into nucleosomal fragments. Indirect assays measure ‘potential’ DNA damage since spermatozoa are exposed to denaturing conditions, thus relaxing and splitting double-stranded DNA into single-stranded strands through the breaking of the hydrophobic stacking between the bases. These indirect assays include the sperm chromatin structure assay, sperm chromatin dispersion test, and COMET using alkaline conditions (
). The predictive value of these tests can be influenced by several factors, such as single versus double-stranded DNAfrag; percentage of spermatozoa with DNA damage; whether the damage affects coding (exons) or non-coding (introns) regions; and the oocyte capacity to repair sperm DNA damage. The potential effect of the presence of sDNAfrag on pregnancy and embryo development will therefore depend on all these factors (
). The Practice Committee of American Society for Reproductive Medicine has published the threshold values used to define an abnormal presence of sDNAfrag and from which negative repercussions in fertility will occur (
The technological advances in assisted reproduction techniques, and especially the use of intracytoplasmic sperm injection (ICSI), mean that a single spermatozoon can be introduced into an oocyte (
). So, it became important to assess the quality of the paternal genetic material and to establish criteria by which to choose appropriate cohorts of spermatozoa. Various procedures for selection of spermatozoa have been implemented, the relative efficiency of which in decreasing or even removing spermatozoa with sDNAfrag have been reported (
The absence of concordant results makes urgent the need for prospective studies to select an appropriate sperm preparation technique during assisted reproduction technique procedures. The aim of the present study was to determine and compare the incidence of sDNAfrag at each step of a sequential semen preparation technique with regard to the degree of severity of semen pathologies, to determine if the steps of preparation influence the incidence; and whether a relationship could be found between semen parameters and sDNAfrag.
Materials and methods
Ethical considerations and biological material
Ethical guidelines were followed in the conduct of research, with written informed consent having been obtained before the beginning of the work. This work did not involve human or animal experiments. An approval by an Ethics Committee and the provisions of the Declaration of Helsinki, as revised in Tokyo 2004, on human experimentation does not apply to this kind of work. Therefore, according to the standards of National Law (PMA, Law 32/2006) and Council (CNPMA, 2008) criteria on Medically Assisted Procreation, 70 semen samples were used after patients had given informed and written consent. The patients, enrolled in ICSI cycles, suffered from primary infertility of at least 1 year, and presented a mean age of 36.3 ± 5.3 (27–52) years. Criteria for inclusion were absence of known pathologies and intake of medicines; normal physical examination, hormonal profiles and karyotypes; and semen analysis without agglutination, immature forms, leukocytes and microorganisms.
Sperm preparation
Semen samples were collected by masturbation into sterile cups after 2–4 days of sexual abstinence. Samples were left to liquefy for 30 min at 33°C in 5% CO2 in humidified air. Subsequently, a small fraction of each sample was taken (Raw fraction) and another aliquot was used for assessment of sperm parameters. The remainder was submitted to discontinuous gradient centrifugation (DGC) (Gradients PureSperm System 90–45%, Nidacon, Gothenburg, Sweden). After centrifugation at 450 g for 20 min at room temperature, the bottom gradient layer (purified populations of highly motile sperm) was recovered. An aliquot of this fraction was taken (Grad fraction). The pellet was thereafter washed twice (450 g for 10 min at room temperature) in sperm preparation medium with HEPES buffer (Medicult Origio, Jyllinge, Denmark) and at the end of this step another aliquot was taken (GW fraction). The pellet was then layered with 500 µl of swim-up solution (Universal IVF Medium, Medicult Origio), and incubated (1 h, 37°C, 5% CO2 in humidified air) to collect the swim-up fraction containing motile normal sperm that actively migrated from the pellet to the upper aqueous phase. An aliquot of this fraction was taken (SU fraction). The four fractions collected were spread onto a circle made by a diamond pen on adhesion glass slides, left to air-dry and used for detection of sDNAfrag.
Analysis of semen parameters
Conventional semen analysis was carried out according to the 2010 World Health Organization (WHO) guidelines (
). The original raw samples were examined for concentration, motility, vitality, by hypoosmotic swelling test and morphology. The latter was classified using Kruger's strict criteria (
The incidence of morphologically normal spermatozoa that had nuclear DNA strand breaks was identified by the TUNEL assay using the In-Situ Cell Death Detection kit, Fluorescein (Roche Diagnostics, Mannheim, Germany), according to the manufacturer's instructions as previously described (
). Briefly, each patient's cell fractions were fixed with 4% paraformaldehyde (Merck, Darmstadt, Germany) in phosphate buffered saline (PBS) (Sigma, Steinheim, Germany) for 1 h at room temperature and then permeabilized with 0.1% Triton X-100 (Sigma) in 0.1% sodium citrate (Merck) for 2 min at 4°C. After washing twice with PBS, sperm were incubated in 50 µl of labelling solution containing the terminal deoxynucleotidyl transferase (TdT) enzyme for 1 h at 37°C in a dark moist chamber. After two washes, slides were counterstained with Vectashield antifade medium containing 4′,6-diamino-2-phenylindole (DAPI) (Vector Laboratories, Burlingame, USA). At least 500 normal spermatozoa per slide were evaluated by fluorescence microscopy, using a Leitz DMBRE microscope fitted with a CCD camera and IM50 software (Leica Microsystems Ltd, Heerbrugg, Germany). As only morphologically normal sperm, without morphological anomalies in the head, mid-piece and tail are selected for use in ICSI treatments, only morphologically normal sperm were counted for sDNAfrag quantification. For each experiment, negative (TdT enzyme omitted and replaced by distilled water) and positive (deoxyribonuclease treatment) controls were included. A double-blind procedure was used. Each spermatozoon was assigned as apoptotic (if displaying intense green fluorescence) or normal (DAPI staining only). The incidence of TUNEL-positive sperm was taken to be the percentage of cells exhibiting sDNAfrag.
Statistical analysis
One ejaculate from each patient was independently assayed. The fractions obtained from each ejaculate were compared with sperm parameters, proportion of cells with sDNAfrag and staining patterns. Statistical analysis was carried out with the software STATISTICA (Version 11 Statsoft, USA). As we could not homogenize data, a non-parametric approach was taken. To compare the percentages of TUNEL-positive sperm between the four fractions as well as to analyse the morphological patterns in the fractions the Kruskal-Wallis ANOVA (comparisons between independent groups) and the Friedman analysis of variance (ANOVA) (comparisons between dependent groups) tests were used. 1/square root and Acos transformations were used to grant homogeneity of variances before analysis. Post-hoc analysis (comparisons between independent groups) was made with the multiple comparison test. Additionally, an analysis with Mann–Whitney U-test was used. After the Friedman ANOVA, the Sign test was applied to determine differences between groups. Spearman rank correlation was used to measure the association of the sDNAfrag, staining patterns and semen parameters. P < 0.05 was considered to be statistically significant.
Results
Semen evaluation
The characteristics of sperm concentration, normal morphology, RPM and total progressive motility (TPM) for patient semen analysis are shown in Table 1.
Table 1Characteristics of patient semen parameters.
In each patient, TUNEL assay was used to evaluate sDNAfrag on the four fractions (Raw, Grad, GW and SU). A progressive decrease was observed in sDNAfrag from the Raw toward SU fraction, with a significant decrease in the percentage of TUNEL-positive sperm after each procedure (P = 0.04 to P < 0.0001) (Raw [13.3%], Grad [5.6%], GW [4.1%] and SU [2.9%]). The Grad did not differ from the GW fraction using the Mann–Whitney U-test (Supplementary Table S1).
Comparisons of sDNAfrag within and between fractions according to severity of semen parameters
No significant differences were found in the percentages of sDNAfrag between subgroups analysed for sperm concentration in the Raw fraction, for morphology in each fraction and for RPM in the Raw, Grad and SU fractions (Supplementary Table S2). For all semen parameters considered, a significant (P range between <0.0001 and <0.05) decrease was found in sDNAfrag among fractions in the following order: Raw, Grad, GW and SU (Supplementary Table S2).
Comparisons of sDNAfrag within fractions according to sperm pathologies
An analysis of sDNAfrag within fractions was undertaken after subdividing patients according to different pathologies (oligozoospermia, asthenozoospermia, teratozoospermia, oligoasthenozoospermia, oligoteratozoospermia, oligoaasthenoteratozoospermia and asthenoteratozoospermia). Two types of control groups were used: absolute controls, derived from samples showing all spermiogram parameters simultaneously normal (normozoospermia: ≥20 × 106/ml sperm concentration + ≥15% normal morphology + ≥25% rapid progressive motility); and relative controls, specific for each spermiogram parameter (≥20 × 106/ml sperm concentration; or ≥15% normal morphology; or ≥25% rapid progressive motility) (
Compared with the Raw fraction, a significant (P < 0.001) decrease was observed in the percentage of TUNEL-positive sperm after each procedure toward the SU fraction, with no significant differences between the Grad and the GW fractions (data not shown).
Characterization of sDNAfrag staining patterns
Five hundred spermatozoa with normal morphology on each slide were examined and four staining patterns of sDNAfrag were observed (Figure 1). These were characterized by the differential binding of the FITC-dUTPs to fragmented DNA in the sperm head. The staining patterns were defined as follows: acrosome vesicle region (AVR): anterior 2/3 of the nucleus); equatorial region (ER): nuclear region under the equatorial acrosome segment); post-acrosome region (PAR): distal 1/3 of the nucleus); Head (nucleus whole staining). Although the sperm mid-piece contains mitochondrial DNA, no labelling was found at this region.
Figure 1DNA fragmentation patterns. (A) A normal spermatozoon showing the different patterns found; (B) The sperm head. In the acrosome vesicle region (AVR) region resides a higher number of coding genes. The equatorial region (ER) contains the highest concentration of protamines. The post-acrosome region (PAR) region is gene poor and is histone-rich; (C) results of TdT (terminal deoxynucleotidyl transferase)-mediated dUDP nick-end labelling assay on spermatozoa.
Comparisons of sDNAfrag staining patterns within fractions
In the Raw fraction, both Friedman and Kruskal–Wallis ANOVA analysis showed significant differences between patterns (P range between <0.0001 and <0.01), with the Head pattern being significantly higher than all others (P < 0.0001). The mean percentages of each labelling pattern decreased in the sequence: Head >AVR >PAR >ER (Table 2). In the Grad, GW and ER fractions, as in the Raw fraction, significant differences between patterns were seen (P range between <0.0001 and <0.02). The percentage of the AVR pattern was lower than that of the PAR pattern (P range between <0.0001 and <0.001) (Table 2). Thus, sDNAfrag decreased in the order: Head >PAR >AVR >ER (Table 2).
Table 2Comparisons of sperm DNA fragmentation staining patterns within and between fractions.
TUNEL positive sperm (%) within fractions (mean ± SD)
Fractions
Head
AVR
PAR
ER
P-value (intergroup)
Raw
60 ± 15
20 ± 10
16 ± 10
4 ± 5
a, b, c (<0.0001), d (<0.01), e, f (<0.0001)
Grad
63 ± 18
13 ± 11
20 ± 16
3 ± 5
a, b, c (<0.0001), d (<0.001), e, f (<0.0001)
GW
64 ± 19
12 ± 12
20 ± 13
2 ± 5
a, b, c (<0.0001), d (<0.001), e, f (<0.0001)
SU
55 ± 28
5 ± 8
31 ± 25
3 ± 0
a, b, c (<0.01), d (<0.0001), e (<0.02), f (<0.0001)
Comparisons of sperm DNA fragmentation staining patterns between fractions
P-value (intragroup)
Friedman–ANOVA
NS
A, B, C, D, E, F (<0.001)
A, B, C, D, E, F (<0.001)
A, B, C, D, E, F (<0.001)
Sign Test
NS
A, B, C, D, E, F (<0.01)
C, D, E, F (0.04)
B, C (<0.001), D, E (0.04)
Kruskal–Wallis ANOVA
NS
A, B, C, D, E, F (<0.0001)
A, B, C, D, E, F (<0.001)
A, B, C, D, E, F (<0.0001)
Multiple comparison
NS
A, B, C, E, F (<0.001)
C (<0.001)
B (<0.01), C (<0.001)
P < 0.05 was considered statistically significant. Each lowercase letter indicates a particular significantly different pair comparison within Raw, Grad, GW and SU fractions as follows: (a) Head ≠ AVR; (b) Head ≠ PAR; (c) Head ≠ ER; (d) AVR ≠ PAR; (e) AVR ≠ ER; (f) PAR ≠ ER. Comparisons within fractions were made by both Friedman and Kruskal–Wallis analysis of variance and, as similar results were obtained only one P-value (the least significant) is shown. Each uppercase letter indicates a particular significantly different pair comparison between fractions as follows: (A) Raw ≠ Grad; (B) Raw ≠ GW; (C) Raw ≠ SU; (D) Grad ≠ GW; (E) Grad ≠ SU; (F) GW ≠ SU. For comparisons between fractions the results of four statistical tests are presented as the P-values were different.
AVR = acrosome vesicle region; ER = equatorial region; Grad = gradient centrifugation; GW = gradient centrifugation plus wash; NS = not significant; PAR = postacrosome region; Raw = fresh liquefied semen; SU = gradient centrifugation, wash and swim-up.
Comparisons of sDNAfrag staining patterns between fractions
In order to compare each sDNAfrag staining pattern at different stages in the sperm preparation procedure, i.e. from the raw fraction to SU fraction, four statistical analyses were conducted (see intragroup P-value in Table 2). For the Head pattern, no significant differences between fractions were observed (Table 2). The percentage of labelling progressively decreased in the AVR pattern (P range between <0.0001 and <0.01). Using the more demanding Multiple Comparison Test, however, no significant difference between Grad and GW fractions was observed (Table 2). For the PAR pattern, the percentage of staining progressively increased towards the SU fraction (P range between <0.001 and 0.04). Using the more demanding Multiple Comparison Test, however, no difference was observed between the fractions except (P < 0.001) between the Raw and SU fractions (Table 2). For the ER pattern, the percentage of staining in the Raw fraction decreased compared with the GW fraction (P range between <0.0001 and 0.04); however, an increase occurred between the GW and SU fractions. Using the demanding Multiple Comparison Test, the only significant differences were between the Raw and GW fractions (P < 0.01) and the Raw and SU fractions (P < 0.001) (Table 2).
Correlation analysis among sDNAfrag staining patterns, semen parameters and fractions
Correlation analyses were not possible for the Grad and GW fractions, as well for patterns AVR, PAR and ER, because the dispersion of values among them rendered the group sizes too small. In the Raw fraction, significant negative correlations were observed between sperm RPM (r = −0.251; P = 0.036) and TPM (r = −0.243; P = 0.042), whereas in the SU fraction, sDNAfrag was only negatively correlated with TPM (r = 0.266; P = 0.026). For the Head pattern, negative correlations with RPM (r = −0.268; P = 0.025) and TPM (r = −0.275; P = 0.021) were only observed in the SU fraction. Positive correlations were found between semen parameters (concentration versus morphology [r = 0.364; P = 0.002], RPM [r = 0.311; P = 0.009] and TPM [r = 0.241; P = 0.044]; morphology versus RPM [r = 0.435; P < 0.001] and TPM [r = 0.378; P < 0.001] and RPM versus TPM [r = 0.916; P < 0.001]) (Table 3).
Table 3Correlation analysis between sperm DNA fragmentation staining patterns, semen parameters and fractions.
Variables
Raw fraction
SU fraction
TUNEL+
H-TUNEL+
TUNEL+
H-TUNEL+
Concentration
Sperm morphology
RPM
TPM
Raw fraction
TUNEL +
–
r = −0.005 NS
–
–
r = −0.075 NS
r = 0.050 NS
r = −0.251 P = 0.036
r = −0.243 P = 0.042
H-TUNEL +
r = −0.005 NS
–
–
–
r = 0.041 NS
r = 0.198 NS
r = 0.038 NS
r = 0.133 NS
SU fraction
TUNEL +
–
–
–
r = −0.058 NS
r = −0.192 NS
r = −0.183 NS
r = −0.179 NS
r = −0.266 P = 0.026
H-TUNEL+
–
–
r = −0.058 NS
−
r = −0.030 NS
r = −0.109 NS
r = −0.268 P = 0.025
r = −0.275 P = 0.021
Concentration
r = −0.075 NS
r = 0.042 NS
r = −0.192 NS
r = −0.030 NS
–
r = 0.364 P = 0.002
r = 0.311 P = 0.009
r = 0.241 P = 0.044
Sperm morphology
r = 0.050 NS
r = 0.198 NS
r = −0.183 NS
r = −0.109 NS
r = 0.364 P = 0.002
–
r = 0.435 P < 0.001
r = 0.378 P < 0.001
RPM
r = −0.251 P = 0.036
r = 0.038 NS
r = −0.179 NS
r = −0.268 P = 0.025
r = 0.311 P = 0.009
r = 0.435 P < 0.001
–
r = 0.916 P < 0.001
TPM
r = −0.243 P = 0.042
r = 0.133 NS
r = −0.266 P = 0.026
r = −0.275 P = 0.021
r = 0.241 P = 0.044
r = 0.378 P < 0.001
r = 0.916 P < 0.001
–
H = head pattern; NS = not significant; RPM = rapid progressive motility; SU = swim-up fraction; TPM = total progressive motility.
Incidence of sperm head staining patterns of sDNAfrag within fractions
In all four fractions analysed, it was determined that, for each patient, the contribution of each staining pattern for the percentage of TUNEL-positive sperm. The percentage of the staining patterns as the first or second most expressed ones was determined for the total of 70 patients analysed. It was also determined in which percentage of patients the staining patterns were absent or present (Table 4, Table 4).
Table 4Percentage of patients in whom each sperm DNA fragmentation staining pattern was absent and present, and when present in whom each staining pattern was the first or second more expressed one within fractions.
Fractions
Patterns
Head (%)
AVR (%)
PAR (%)
ER (%)
Raw
Prevalence
First
95.7
1.4
2.9
0
Second
4.3
67.1
30.0
0
Absence
0
0
0
34.3
Presence
100
100
100
65.7
Grad
Prevalence
First
94.3
1.4
4.3
0
second
5.7
38.6
45.7
1.4
Absence
0
20
11.4
60
Presence
100
80
88.6
40
GW
Prevalence
First
92.9
2.9
2.9
0
second
4.3
28.6
55.7
1.4
Absence
1.4
25.7
12.9
75.7
Presence
98.6
74.3
87.1
24.3
SU
Prevalence
First
75.7
0
17.1
0
second
14.3
8.6
52.9
1.4
Absence
10
61.4
22.9
81.4
Presence
90
38.6
77.1
18.6
AVR = acrosome vesicle region; ER = equatorial region; Grad = gradient centrifugation; GW = gradient centrifugation plus wash; PAR = postacrosome region; Raw = fresh liquefied semen; SU = gradient centrifugation, wash and swim-up; first: most expressed pattern; second: second most expressed pattern.
In all cases of TUNEL-positive sperm, most of the sperm showed the Head pattern as the most prevalent one, with a decrease towards the SU fraction (95.7–75.7%). The AVR and PAR patterns were the second most expressed. Although the AVR pattern decreased towards the SU fraction (67.1–8.6%), the PAR pattern increased towards the SU fraction (30.0–52.9%). The ER pattern was the least present pattern.
Along the steps of sperm preparation, all patterns of sDNAfrag decreased in its presence in the samples analysed. This decrease was especially noted in the AVR region (from 100% of the 70 samples analysed in the Raw fraction to 38.6% of the 70 samples analysed in the SU fraction) and the ER region (from 65.7% of the 70 samples analysed in the Raw fraction to 18.6% of the 70 samples analysed in the SU fraction).
The most expressed sequences, from the most to the least expressed patterns, were Head >AVR >PAR >ER and Head >PAR >AVR >ER. The frequency of these predominant sequences was determined for the total number of patients (70) analysed in this study. In the Raw fraction, the sequence Head–AVR–PAR–ER (50%) was more frequent than Head–PAR–AVR–ER (27.1%). In the Grad fraction, although the sequence Head–AVR–PAR–ER was also predominantly found (35.7%), the sequence Head–PAR–AVR–ER was similarly frequent (34.3%). On the contrary in the GW fraction, the sequence Head–PAR–AVR–ER was predominantly found (38.6%), although at a similar frequency to Head–AVR–PAR–ER (28.6%). This inversion in the predominance of the sequences was complete in the SU fraction, with the Head–PAR–AVR–ER sequence being found in 24.3% against the Head–AVR–PAR–ER sequence found in 7.1% of all patients analysed.
Discussion
Spermatozoa must be separated from the seminal plasma, as prolonged exposure to it has been associated with reduced viability (
). This separation can be achieved simply by washing the semen sample. Density gradient centrifugation was then established to separate a subpopulation of sperm with the best motility, morphology (
). A further possible source of sperm DNA damage could be iatrogenic owing to the use of assisted reproduction techniques. Patients treated using intrauterine insemination, IVF and ICSI require sperm processing to select sperm with better chances of embryological and clinical outcomes. It is therefore important to use a sperm-processing technique that enables better selection of functional sperm while causing the minimal damage. This is further important in the ICSI technique, as the single spermatozoon injected, which is only selected based on morphology and motility, may carry a high risk of having damaged DNA (
Sperm deoxyribonucleic acid fragmentation is increased in poor-quality semen samples and correlates with failed fertilization in intracytoplasmic sperm injection.
Research outcomes on the potential adverse effect of semen processing on human sDNAfrag remain controversial. Several investigators have found no improvement in sDNAfrag rates when using DGC alone (
determined that DGC can select spermatozoa with lower nuclear damage. In the present study, we observed a significant (P = 0.04 to <0.0001) decrease in the percentage of sDNAfrag between the Raw and the SU fractions with a progressive decrease at each sequential step in the sequence: Raw (13.3%), Grad (5.6%), Wash (4.1%) and SU (2.9%). Therefore, our results indicate that the discontinuous double-layered DGC and the wash steps used routinely in IVF laboratories do not increase sDNAfrag, and that higher levels of sperm DNA integrity are observed after the use of a total sequential technique.
The relationship between sDNAfrag and raw semen abnormalities also seems to be controversial. Several investigators observed lower sDNAfrag in men with higher sperm concentration, total motility, progressive motility and normal morphology (
). However, after use of sperm-preparation techniques, results appear more consistent, several reports finding that after swim-up the percentage of sDNAfrag was negatively correlated with motility and morphology (
Sperm deoxyribonucleic acid fragmentation is increased in poor-quality semen samples and correlates with failed fertilization in intracytoplasmic sperm injection.
Relationship of sperm DNA fragmentation, apoptosis and dysfunction of mitochondrial membrane potential with semen parameters and ART outcome after intracytoplasmic sperm injection.
). In relation to sperm pathologies, using DGC followed by wash and swim-up, a significant correlation between the degree of sDNAfrag and asthenozoospermia was not found, but a significantly higher percentage of sDNAfrag was reported in men with teratozoospermia (
In the present work, use of within-fractions analysis, and regarding sperm concentration, results showed no relationship between sperm DNA damage and different degrees of sperm abnormality in the Raw fraction; however, a significantly higher sperm DNA damage was seen in cases with >1 to ≤5 × 106 spermatozoa/ml in the Grad and GW fractions, and in cases with >5 to <20 × 106 spermatozoa/ml in the Grad and SU fractions compared with cases with a normal sperm concentration. In relation to morphology and RPM, no significant differences were observed except in the GW fraction where ejaculates with more than 25% of spermatozoa with RPM presented significantly lower percentages of sDNAfrag than samples with ≥10 to <25% of spermatozoa with RPM. In contrast, for TPM, sDNAfrag significantly decreased in samples with more than 25% of spermatozoa with TPM when compared with samples with ≥10 to <25% of spermatozoa with TPM in all fractions and to samples with ≤5% and >5 to <10% of spermatozoa with TPM in the Grad, GW and SU fractions.
In the between-fractions analysis, and regarding concentration, morphology, RPM and TPM, a significant decrease in sDNAfrag occurred among fractions in the following order: Raw, Grad, GW and SU, with the Raw fraction presenting the highest sDNAfrag. Our results therefore confirm that higher sperm damage occurs in semen samples with lower semen parameters, and that sDNAfrag is progressively decreased during the washing procedure, which does not therefore contribute to DNA fragmentation.
To our knowledge different sperm head staining patterns of sDNAfrag have not been identified, described previously, or both. Comparisons among washed fractions revealed that patterns differed significantly, progressively decreasing in the following order: Head, AVR, PAR and ER patterns. Our results provide unequivocal evidence that different patterns of DNAfrag are present in human morphologically normal spermatozoa and that the resulting fluorescence pattern does not depend on the sperm preparation technique. Indeed, comparisons between fractions showed that the Head pattern did not differ between the washing steps; the AVR and ER patterns differed between all steps with a significant decrease towards the SU fraction; and the PAR pattern significantly increased from the Raw to the SU fraction. Further, the fact that we have only counted spermatozoa with DNA fragmentation in 500 morphologically normal spermatozoa and not randomly in 200 as most published studies on sDNAfrag, adds a new significance to the method used in this study as well as to the results obtained.
The sperm nucleus shows specific organized chromosome territories, with centromeres collected into a compact chromocentre and telomeres localized at the periphery (
). Chromosomes are concentrated at the nuclear centre in defined regions, apical (gene rich with a higher density and size of genes, and higher density of protein coding sequences) (AVR region) and basal (gene poor region) (PAR region). This organization indicates a functional link between chromosome distribution and regulation of gene expression. This organization is thought to facilitate the sequential activation of the paternal genome after fertilization (
). Sperm histones also have a non-random positioning in the sperm nucleus. Protamine 1 is primarily located at the centre (ER region), where chromatin is more highly compacted, whereas the PAR region and the nuclear periphery (gene poor regions) have a lower protamine content but a higher histone content, rendering them more susceptible to oxidative damage (
). In the sperm nucleus, histones are associated with specific genes, largely at gene promoter regions where sperm mRNAs are also concentrated. Some of these sperm histones play specific roles both in the zygote and during embryo development (
Taking into account these findings, the lower frequency of sDNAfrag at ER might be explained by the predominance of a more compact chromatin at this region, whereas the higher frequency of sDNAfrag at PAR can be explained by the lower content of protamines. Our sequential sperm preparation procedure also selectively decreased the AVR pattern (gene-rich region), thus indicating the efficiency of the sperm preparation procedure in decreasing the number of spermatozoa with a higher risk of sDNAfrag (Figure 1). As apoptosis is an inherent, rapid and continuous process that can take place in a few hours or even minutes (
), the different sDNAfrag staining patterns could alternatively reflect distinct stages of apoptosis dependent of the observation time. Another hypothesis is that sDNAfrag originates during spermatogenesis and that, because of the abortive apoptosis mechanism, apoptotic sperm escape full elimination during spermiogenesis (
). It will therefore be of fundamental importance in the future to apply the protocol of sperm preparation used in the present study, in combination with magnetic or fluorescence activated cell sorting (
), to determine if this association can efficiently decrease or eliminate any sDNAfrag staining pattern to substantially reduce the risk of using spermatozoa with DNA fragmentation in assisted reproduction techniques. Specifically, reduce or eliminate the patterns attaining the sperm gene-rich regions (AVR; Head). To test the biological impact of each pattern it will be necessary to sort spermatozoa with each pattern individually and study their fertilizing capacity.
In conclusion, our results confirm that the use of DGC and the following wash do not compromise sperm DNA integrity. We also show that the use of a sequential procedure, in which DGC is followed by swim-up, is most effective at decreasing sDNAfrag, and we confirm that semen with lower concentration, motility and morphology have higher levels of sDNA fragmentation. Finally, we describe sperm head staining patterns that are correlated with the degree of sDNAfrag, and show that the pattern marking the gene-rich region was reduced by our procedure. This last result suggests that pattern quantification might be a useful adjunct when performing sDNAfrag testing for male infertility. Prospective clinical trials are now required to clarify the relevance of the sDNAfrag staining patterns.
Acknowledgements
The authors appreciate the contribution of Paulo Viana, BSc, Clinical Embryologist-ESHRE and Ana Gonçalves, Cláudia Osório and Nuno Barros, BSc (CGR-ABarros) in sample preparation; of Joaquina Silva, MD, Senior Clinical Embryologist-ESHRE (CGR-ABarros) for IVF and Andrology Laboratory supervision and of Carolina Almeida, PhD in study design.
This work was supported in part by UMIB that is funded by National Funds through FCT-Foundation for Science and Technology, under the Pest-OE/SAU/UI0215/2014.
Appendix. Supplementary material
The following is the supplementary data to this article:
Sperm deoxyribonucleic acid fragmentation is increased in poor-quality semen samples and correlates with failed fertilization in intracytoplasmic sperm injection.
Relationship of sperm DNA fragmentation, apoptosis and dysfunction of mitochondrial membrane potential with semen parameters and ART outcome after intracytoplasmic sperm injection.
Rosália Sá graduated in 2004 from University of Aveiro and obtained her PhD degree in Biomedical Sciences in 2012 from the Institute of Biomedical Sciences Abel Salazar (ICBAS), University of Porto (UP), Porto, Portugal. Currently, she is Assistant Professor of the Department of Microscopy, Laboratory of Cell Biology and Investigator of the Biology and Genetics of Reproduction Group of the Multidisciplinary Unit for Biomedical Research, ICBAS-UP. Research in her laboratory focuses on the genetic causes of human infertility and biological mechanisms associated to human gametogenesis and embryogenesis. Her current research interests are male fertility preservation, in-vitro spermatogenesis and oocyte dysmorphisms.
Article info
Publication history
Published online: July 07, 2015
Accepted:
June 23,
2015
Received in revised form:
June 16,
2015
Received:
January 13,
2015
Declaration: The authors declare no conflicts of interest that could be perceived as prejudicing the impartiality of the research reported. The authors alone account for the content and writing of the manuscript.
00315-6&id=rbmo1395-fig-0001.jpg){kind=link}