A prospective study was performed during 2017. Thirty men (24 healthy volunteers and six male partners from couples who underwent an infertility evaluation) participated and showed a total motile spermatozoa count (TMC) of at least 80 million. Informed consent was obtained from all participants. This study was approved by the Institutional Review Board of Bundang Seoul National University Hospital (IRB No. B-1705/395-307). None of the subjects had a history of genital inflammation or surgery, no subjective symptoms or self-reported medical risk factors were identified, and none of the subjects were taking any prescription medications. The participants ranged in age from 23 to 45 years, and three of them had biological children.

Semen samples were obtained by masturbation into a sterile container. Semen samples were liquefied for 30 minutes at room temperature and then divided equally into five aliquots, one of which was not processed and served as the control sample (raw semen). The remaining four aliquots underwent DGC procedures using PureSperm (Nidacon Laboratories, Gothenburg, Sweden), PureCeption (Sage Biopharma, Bedminster, NJ, USA), Sydney (William A. COOK, Queensland, Australia), and SpermGrad (Vitrolife, Englewood, CO, USA), respectively.

The raw semen sample and the four processed semen fractions were examined for TMC, motility, the mean percentage of normal morphology, DNA fragmentation level, and HA binding ability. Semen analysis was performed using motion analysis equipment (SAIS-PLUS 10.1; Medical Supply, Seoul, Korea). Strict criteria, as indicated in the World Health Organization (WHO) guidelines (fifth edition, 2010) were used to define normal spermatozoa in the morphological assessment.

The basic semen characteristics were as follows (median with interquartile range): semen volume, 3 mL (2–4 mL); sperm count, 111 million/mL (59–138 million/mL); total motility, 70.8% (60%–80%); TMC, 144 million (114–282 million); and mean percentage of morphologically normal spermatozoa, 6.9% (5.2%–8.3%).

DGC was performed as described in the WHO guidelines, but was modified appropriately according to the situation of our laboratory as described in our previous study [10]. The concentrations of the upper- and lower-phase gradient and centrifugation-specific criteria recommended by each manufacturer are summarized in Table 1. The recommended centrifugal force for PureCeption medium was 350–400×g, and we used 350×g to minimize the procedural differences across the medium types.

Each semen sample was mixed with 3 mL of Ham’s F-10 media (Gibco, Paisley, UK) with 10% synthetic serum substitute (SSS; Irvine Scientific, Santa Ana, CA, USA) and then centrifuged at 300×g for 5 minutes to remove seminal plasma. The precipitated pellet was suspended in 1.5 mL of Ham’s F-10 medium with 10% SSS. Two milliliters of lower-phase gradient medium was placed in a 15-mL conical tube and the same amount of upper-phase gradient medium was gently loaded. The suspended semen samples were placed onto the upper-phase gradient medium. Next, the conical tube was centrifuged at 300×g for 20 minutes. The pellet collected from the bottom layer was washed twice by resuspension in 4 mL of Ham’s F-10 medium with 10% SSS, followed by centrifugation at 300×g for 5 minutes. After two rounds of centrifugation, the supernatant was discarded and the final pellet was resuspended in 3 mL of Ham’s F-10 medium with 10% SSS.

The DNA fragmentation level was measured using the Halosperm assay kit according to the manufacturer’s instructions with minor modifications. In brief, each aliquot of semen sample was diluted to 10 million/mL in F-10 medium. Agarose was melted by heating in a water bath at 90°–100°C for 5 minutes and incubated in a water bath at 37°C for 5 minutes. After the temperature of agarose was equilibrated to 37°C, 25 μL of the semen sample was mixed with fused agarose. Twenty microliters of the mixture was pipetted onto a slide and covered with a 22×22 mm coverslip. The slide was cooled in the refrigerator at 4°C for 5 minutes to solidify the agarose. The coverslip was gently removed and the slide was treated with a denaturation solution (a mixture of 80 μL HCl and 10 mL of distilled water) for 7 minutes. The slide was treated with lysis buffer for 25 minutes and washed for 5 minutes in a tray containing distilled water. Stepwise dehydration was performed with increasing concentrations of ethanol (70%, 90%, and 100%; 2 minutes for each concentration), followed by air-drying. After Diff-Quick staining (Eosin G for 7 minutes, Azure B for 7 minutes), the slide was washed in tap water and air-dried.

Each slide was examined under a light microscope at ×200 magnification and at least 300 spermatozoa were scored. Each spermatozoon was assessed as having a large halo, a medium halo, a small halo, or no halo; degraded spermatozoa were also counted. A halo with a thickness equal to or greater than the minor diameter of the core was considered large, a halo with a thickness smaller than the minor diameter of the core and greater than one-third of the minor diameter of the core was considered medium, and a halo with a thickness equal to or smaller than one-third of the minor diameter of the core was considered small. Spermatozoa with a small halo or no halo, as well as degraded spermatozoa, were considered to have fragmented DNA. The DNA fragmentation level was calculated as the percentage of spermatozoa with DNA fragmentation out of all observed spermatozoa.

HA binding ability was assessed using the HBA Sperm Hyaluronan-Binding Assay kit (Origio, Måløv, Denmark) according to the manufacturer’s instructions. The semen samples were diluted with F-10 medium to a concentration of approximately 10 million/mL. A 10-μL drop of each semen sample was placed in the assay chamber, and a coverslip was carefully placed over the drop to avoid bubble formation. After 20 minutes, each slide was examined under a light microscope at ×400 magnification, and at least 300 spermatozoa were scored. The bound spermatozoa showed typical tail-beating without any progressive movement, which could be easily distinguished from the unbound spermatozoa. HA binding ability was calculated using the following equation: bound motile spermatozoa/total motile spermatozoa ×100 (%).

Statistical analysis was performed using IBM SPSS ver. 22.0 (IBM Corp., Armonk, NY, USA). All data were presented as median with interquartile range. The paired Wilcoxon signed-rank test was used to compare the semen parameters of the raw samples and each DGC fraction. The Kruskal-Wallis test was used to compare the semen parameters among the four DGC fractions. Spearman rank correlation analysis was used to analyze correlations between continuous variables.