ABSTRACT: During human spermiogenesis, the elongated spermatids undergo a plasma membrane remodeling step that facilitates formation of the zona pellucida and hyaluronic acid (HA) binding sites. Various biochemical sperm markers indicated that human sperm bound to HA exhibit attributes similar to that of zona pellucida–bound sperm, including minimal DNA fragmentation, normal shape, and low frequency of chromosomal aneuploidies. In this work, we tested the hypothesis that HA-bound sperm would be enhanced in sperm of high DNA chain integrity and green acridine orange fluorescence (AOF) compared with the original sperm in semen. Sperm DNA integrity in semen and in their respective HA-bound sperm fractions was studied in 50 men tested for fertility. In the semen samples, the proportions of sperm with green AOF (high DNA integrity) and red AOF (DNA breaks) were 54.9% ± 2.0% and 45.0% ± 1.9%, whereas in the HA-bound sperm fraction, the respective proportions were 99% and 1.0%, respectively. The data indeed demonstrated that HA shows a high degree of selectivity for sperm with high DNA integrity. These findings are important from the points of view of human sperm DNA integrity, sperm function, and the potential efficacy of HA-mediated sperm selection for intracytoplasmic sperm injection.
Spermatozoa with dysmaturity with retained cytoplasm were deficient in zona pellucida binding. This was demonstrated in sperm hemizona (halved unfertilized human oocytes) complexes immunostained with creatine kinase, in which all bound spermatozoa were clearheaded without cytoplasmic retention (Huszar and Vigue, 1994; Huszar et al, 2000). This selectivity was attributed to a sperm plasma membrane remodeling step during spermiogenesis that facilitates the formation of the sperm zona pellucida binding site, as well as the binding sites for hyaluronic acid (HA; Huszar and Vigue, 1990, 1994; Huszar et al, 1997).
In earlier sperm DNA integrity studies with AO, chromatin structure of zona pellucida–bound spermatozoa was examined in sperm removed from the zona pellucida of unfertilized oocytes (Liu and Baker, 2007). The authors reported greater than 85% frequency of sperm with green AOF, reflecting high DNA integrity. In the present study, we examined the hypothesis that HA-bound sperm will be enriched in sperm with high DNA integrity (green AOF) compared with their respective semen fractions.
Materials and Methods
The subjects for this HA-bound sperm selection/AO study were husbands in infertile couples who were referred for semen analysis in the Sperm Physiology Laboratory, Department of Obstetrics, Gynecology, and Reproductive Sciences, Yale University School of Medicine. Semen samples were obtained by masturbation after 2–4 days abstinence. All studies were approved by the Yale Human Investigation Committee.
Sperm Preparation and HA Sperm Selection
We studied semen of 50 men (sperm concentration: 84.3 ± 9.3 × 106 sperm/mL [minimum to maximum (min–max), 11.2–284], sperm motility: 46.1% ± 3.4% [4%–86%]). The mean proportion of sperm with green and red AOF was 54.9% ± 2.0% and 45.1% ± 1.9%, respectively, in the 50 samples.
After liquefaction of the semen, a routine semen analysis was carried out by computer-assisted semen analysis, and the sperm were washed by diluting the semen 1:4 with human tubal fluid (HTF; Irwin Scientific, Santa Ana, California), and centrifugation at 600 ×g for 10 minutes at room temperature. Furthermore, the washed sperm were resuspended in HTF medium to a density of approximately 30 million sperm/mL HTF for the AO experiments.
HA Selection of Mature Sperm
For HA selection of mature sperm, we used 5-cm-diameter IVF Falcon petri dishes equipped with 4, 100-μm-diameter solid-state HA spots (PICSI Sperm Selection Device; Mid-Atlantic Scientific, Mount Laurel, New Jersey). An aliquot of the sperm suspension (10–20 μL) was layered over the HA spot of the petri dish. The sperm cells moved freely until the mature sperm bound to the HA head-first (similar to sperm–zona pellucida binding[Huszar et al, 2003]). The HA binding was observed by the lack of progressive motility and by the characteristic increase in tail cross-beat frequency of the bound sperm. Sperm lacking HA binding capacity or HA receptors, likely because of impaired or arrested spermiogenesis, have failed to “perceive” the HA and did not bind (Huszar et al, 1997, 2003). Nonmotile sperm and sperm without the characteristic higher tail beat frequency are not considered in the HA-bound sperm fraction.
After 15 minutes of incubation and sperm binding at room temperature, the unbound sperm were removed by slightly tilting the PICSI dish and applying HTF with a gentle drop-by-drop rinsing. As a control for the sperm selection process, after the rinsing step, another aliquot of the same sperm suspension (5–10 μL) was smeared on the peripherial area of uncoated plastic areas between the HA spots on same petri dishes, and the HA-selected and unselected sperm were subjected to fixation and AO staining.
Technical Aspects of Sperm Fixation for AO Staining
The usual fixative for sperm AO staining is the Carnoy solution, which is composed of a 3:1 ratio of methanol and glacial acetic acid. However, this solution is too corrosive for the HA spot used for selection of mature sperm. We have tried variations of the Carnoy solution that would fix the sperm and preserve the integrity of the HA selection spots. The fixative that was satisfactory on both accounts is a 9:1 ratio of methanol and glacial acetic acid, which was used in the experiments. To validate the fixation and AO staining results, we performed, in conjunction with each of the 50 experiments, a control study of unselected sperm suspensions fixed on glass slides with both the Carnoy and 9:1 ratio of methanol and glacial acetic acid solutions. The AO staining patterns after the 2 fixation methods showed excellent agreement (r = .93; Figure 1).
AO Staining of the Sperm Fractions
The fixed petri dish regions, including the HA spots with the selected sperm, and the areas between the spots with the unselected sperm suspension, as well as the glass slides of the fixation validation aliquots, were stained with acridine orange solution (Polysciences Inc, Warrington, Pennsylvania) for 10 minutes. Subsequently, petri dishes and glass slides were rinsed gently with distilled water, and samples were mounted and cover slips were applied. The AOF patterns with the Carnoy or 9:1 methanol and glacial acetic acid solutions were similar (proportion of sperm with green AOF, r = .97; red AOF, r = .89).
We also compared 2 AO solutions from Polysciences Inc. The first AO “lab-reconstituted” staining solution (04539) was prepared daily as follows: 1 ml of 1% AO stock solution was added to a mixture of 4 mL of 0.1 M citric acid and 0.25 mL of 0.3 M Na2HPO4, pH 2.5. The 1% AO stock solution was stored in the dark for up to 4 weeks. The second “ready-to-use” AO solution (24603), a much simpler and less work intensive process, was utilized according to the manufacturer's directions.
Scoring of AO Green and AO Red Sperm
The percentage of sperm with normal DNA was determined on each slide or regions of petri dish as follows: We selected 10 or more individual fields and evaluated a total of 200 sperm (40 000 sperm in 50 samples) under an Olympus BX51 fluorescence microscope with ×400 magnification and excitation of 450–490 nm. Observation of the fields did not exceed 40 seconds to prevent fading. The spermatozoa were recorded as “green” or “red.” Orange or yellow heads, as well as those displaying green and red color simultaneously, were also considered red, or sperm with fragmented DNA (Tejada et al, 1984). Each stained slide was evaluated immediately after staining.
A paired t test was used to compare the percentages of green sperm after treatment with the lab-made compared with ready-to-use Polysciences fixatives. Unpaired t tests were used to compare sperm concentrations and green and red sperm percentages, between the lowest 10 and highest 10 sperm concentration samples. Correlation analysis was performed by the Pearson test. Analyses were conducted with the Sigma Stat program (Jandel Scientific, San Rafael, California). A value of P < .05 was considered statistically significant. All data are expressed as x̄ ± SEM.
We have carried out 3 lines of experiments on each of the 50 samples. The first 2 studies addressed the validation of the methods regarding the fixative solution used and the type of the AO reagent, which is a very important element. Once we validated the methods, in the third study, we focused on the actual AO staining of unselected spermatozoa and HA-selected spermatozoa.
In Experiment 1, directed to the fixation method, sperm aliquots were fixed to glass slides with both Carnoy fixation solution (3:1 methanol: glacial acetic acid) and 9:1 methanol and glacial acetic acid solution, and the proportion of sperm with green or red AOF (acridine orange fluorescence) were assessed to assure that the 2 fixatives gave comparable results. In Experiment 2, the 2 Polysciences AO reagents were procured (lab-reconstituted AO and ready-to-use AO) and compared for further validation of the methods. In Experiment 3, Falcon petri dishes with HA spots were used for sperm selection. Furthermore, the proportions of sperm with green and red AOF, which were fixed and stained in an identical manner, were compared within the HA selection spots and the peripherial “control” sperm fractions.
Another important consideration was related to the study samples. The primary goal was to have sufficient sperm in the unselected and HA-selected fractions to provide populations for a convincing evaluation of green and red AOF patterns. Thus, considering the likely attrition attributable to handling during sperm-binding incubation, rinsing off unbound sperm, fixation and AO treatments, and washing steps, as indicated in “Materials and Methods,” the starting sperm suspensions were resuspended to concentrations of approximately 30 million sperm/mL. However, the studies are relevant for all constituent sperm because membrane remodeling and the HA binding properties of individual spermatozoa are independent of the other motile sperm within the samples (Huszar et al, 2007).
Development of the AO Fixative for the HA Studies (Experiment 1)
The first series of experiments were aimed at the development of a sperm fixative solution that is comparable to the Carnoy solution (3:1 methanol and glacial acetic acid) but does not affect the integrity of the solid-state HA coating on the petri dish. The use of a Carnoy-like solution with 9:1 methanol and glacial acetic acid satisfied these requirements. To validate the substitution, from each of the 50 study samples, 2 sperm aliquots were applied to glass slides, and after fixation with both the Carnoy and 9:1 solutions, sperm were stained with the AO reagents, and the fluorescence staining patterns were compared. The results were very similar (Carnoy vs 9:1 methanol: glacial acetic acid: green sperm, 45.7% ± 2.1% vs 45.0% ± 2.0%, r = .93, P < .01, Figure 1; red sperm, 46.1% ± 2.3% vs 41.6% ± 2.1%; r = .85, P < .01; n = 50 pairs).
Reproducibility of the AO Stains (Experiment 2)
In the second set of experiments, reproducibility of AOF patterns by the lab-reconstituted and ready-to-use AO solutions was tested. The 2 stains provided very similar results in the proportions of sperm with green and red AOF, whether using the Carnoy or the 9:1 methanol: glacial acetic acid solutions. The data also indicated close similarity between the ready-to-use and lab-made Polyscience reagents in the proportions of AO-stained sperm (green sperm with Carnoy: 54.7% ± 2.2% vs 54.7% ± 2.1%, lab-reconstituted vs ready-to-use; green sperm with 9:1 solution: 57.6% ± 2.1% vs 54.9% ± 2.0%, lab-reconstituted vs ready-to-use). Thus, we used the ready-made Polysciences reagent in the experiments, which facilitated the execution of studies yet was less labor intensive.
Proportion of Sperm with Green and Red AOF in the HA-Selected and Semen Sperm Fractions (Experiment 3)
The comparison of sperm with green and red AOF patterns indicated that the unselected semen sperm fractions placed on the “outside” (uncoated region) of the petri dishes were comparable to the AOF patterns in the original semen aliquots. However, in the regions of the HA spot, where the HA-selected sperm were bound (and the unbound sperm were gently rinsed off), the proportion of sperm with green AOF was in the range of greater than 99% (Table 1). This indicated that HA binding promotes the selection of sperm with high DNA chain integrity (Figure 2).
Table 1. . Summary of the 50 hyaluronic acid (HA) sperm selection experiments
x̄ ± SD
Green sperm in semen, %
54.9 ± 2.0
Green sperm, % (petri, outside control)
56.1 ± 1.9
Green sperm, % (petri, HA spot)
99.1 ± 0.2
Red sperm in semen, %
45.0 ± 1.9
Red sperm, % (petri, outside control)
43.9 ± 1.9
Red sperm, % (petri, HA spot)
0.9 ± 1.9
Considering the overall semen parameters, there was no correlation between proportion of sperm with green AOF and sperm concentrations. However, correlations between green AOF sperm and sperm motility or total motile sperm were moderate (r = .45, P < .01 and r = .36, P < .01, respectively).
Because of the known association in semen samples between oligozoospermia and sperm with arrested maturity, we expected that, in the low sperm concentration samples, there would be a higher representation of red AOF sperm. We addressed this question by comparing the proportion of green and red AOF sperm in samples with the 10 highest and 10 lowest sperm concentrations of the 50 men (Table 2). The sperm concentrations in the 2 groups showed an approximately 8-fold difference. The proportion of green fluorescing sperm was higher in the group with higher sperm concentration. However, the differences in green and red AOF proportions were not nearly as dramatic. Furthermore, it is of note that the proportion of sperm with green AOF was 99% in the HA-bound fraction of both concentration groups.
Table 2. . Comparison of the 10 samples with highest and lowest sperm concentrations
Lowest 10 Samples
Highest 10 Samples
Concentration, sperm × 106
22.8 ± 3.4
187.1 ± 17.1
Green sperm, %
52.1 ± 2.9
65.4 ± 3.6
Red sperm, %
47.9 ± 2.9
34.6 ± 3.6
Sperm DNA integrity is an important factor in the paternal contribution of human sperm to the fertilization process and the maintenance of pregnancy (Aitken et al, 2009). Regarding DNA integrity, zona pellucida–bound sperm contain a high proportion of sperm with green AOF (Liu and Baker, 2007). In the present study, we examined the hypothesis that the HA-bound sperm fraction will be enriched in sperm with high DNA integrity and green AOF. Indeed, in the 50 experiments, there was an almost exclusive presence (>99%) of sperm with green AOF and high DNA integrity in HA-selected bound sperm. (Figure 2).
In preparing for the studies, we had to address the vulnerability of the HA coating to the corrosive effects of the Carnoy fixative. An alternative fixative (the 9:1 methanol and glacial acetic acid solution) has satisfied the requirements of sperm fixation and preservation of the HA spot. Furthermore, the 9:1 methanol: glacial acetic acid solution was validated, with a greater than 90% correlation, by the proportion of sperm with green and red AOF when using both fixatives on glass slides (Figure 1). The staining methods were also validated by the comparable proportions of sperm with green and red AOF when using either of the 2 AO stains. Individual microscopic observation appropriately and efficiently facilitated the comparative studies of HA-selected and unselected spermatozoa of common origin, which were treated and examined identically.
With respect to semen parameters and AOF staining, there are 3 points of interest: 1) In semen samples with lower sperm concentrations, a higher proportion of sperm had DNA breaks, most likely because of the known relationship between reduced sperm production, sperm development, and increased DNA fragmentation. 2) Accordingly, a higher proportion of green AOF sperm appeared in samples with the 10 highest compared with the 10 lowest sperm concentrations (Table 2). However, in the HA-bound sperm fractions, both showed a greater than 99% frequency of green AOF sperm. Thus, the selection power of HA is related to the proportion of individual mature spermatozoa in the individual semen. In previous experiments, we have already demonstrated that the HA binding properties of individual spermatozoa are independent of other constituent sperm in the samples (Huszar et al, 2007). For instance, the HA-bound sperm fractions of normozoospermic and individually selected sperm from intracytoplasmic sperm injection (ICSI)–range oligozoospermic samples showed similar frequencies of chromosomal aneuploidy, regardless of the initial sperm concentrations or aneuploidy frequencies (Jakab et al, 2005).
3) Correlation was moderate (r = .45, P < .01) between the proportion of sperm with green AOF and sperm motility. In earlier sperm chromatin structure assay studies, a similar relationship was found (r = –.53; Giwercman et al, 2003).
The similarity in sperm selection power by the zona pellucida and HA may be explained by the related formation of their receptors via the plasma membrane remodeling in spermiogenesis (Huszar and Vigue, 1994; Huszar et al, 1998, 2007). Further confirmation of this relationship was supported by the similar degree of enrichment in sperm with Tygerberg normal morphology in the zona pellucida–and HA-bound sperm fractions compared with their respective semen sperm fraction (∼4.0 and 3.8 times; Kruger et al, 1986; Menkveld et al, 1996; Prinosilova et al, 2009). The focus of the present work was the DNA integrity of the HA-bound sperm fractions. Sperm morphology was not a goal because of the above, very recent study (Prinosilova et al, 2009). Another consideration was the harsh fixation method and the red/green fluorescence of AO staining that would have made sperm shape and morphology unreliable.
The present work independently confirms the notion that the sperm selection attributes of HA and of zona pellucida (as reported by Liu and Baker) are similar in selecting sperm with high DNA integrity. This is important information because the lack of zona pellucida binding is a major cause of IVF fertilization failure (Claassens et al, 1992; Liu and Baker, 1992, 2007; Menkveld et al, 1996; Angelopoulos et al, 1998).
We have established earlier that HA-selected spermatozoa do not exhibit attributes of dysmaturity, such as cytoplasmic retention, persistent histones, high levels of the apoptotic caspase-3, or DNA fragmentation (Huszar et al, 2003; Cayli et al, 2004). Moreover, individual sperm probed with multiple biochemical markers indicated an approximately 70% agreement between the various cytoplasmic and nuclear markers of normal and arrested development (Huszar et al, 2003; Sati et al, 2008). The present study, which is based on direct examination of single spermatozoa bound to HA, further extends the relationship between sperm cellular maturity, DNA integrity, and the selection role of HA for spermatozoa with attributes promoting fertilization potential and paternal contributions of sperm.
The HA-related diagnostic and sperm selection devices were invented by G.H. The ownership of the patent was assigned to Yale University. Yale licensed the technology to Biocoat Inc. G.H. is acting as a scientific advisor.
A.Y. has received support from the Research Fund of Afyon Kocatepe University, Turkey, and from an unrestricted gift from Vitrolife AB, Sweden.
Part of this research was presented at the 2008 annual meeting of European Society of Human Reproduction and Embryology (ESHRE), Barcelona, Spain.