Human Reproduction vol.15 no.3 pp.662–666, 2000 Bacterial contamination and sperm recovery after semen preparation by density gradient centrifugation using silane-coated silica particles at different g forces C.M.Nicholson1, L.Abramsson2, S.E.Holm3 and et al., 1984; Forman et al., 1987; Stovall et al., 1993; E.Bjurulf1,4 Liversedge et al., 1996; Bussen et al., 1997). It is not always clearly stated whether or not antibiotics were added to the 1Department of Obstetrics and Gynecology, 2Department of Urology and Andrology and 3Department of Clinical Bacteriology, culture media used in these studies. The only micro-organism Umeå University Hospital, S-90185 Umeå, Sweden reported to decrease pregnancy rates is Ureaplasma urealyticum (Montagut et al., 1991), where the authors speculate on an 4To whom correspondence should be addressed endometrial effect. The effects of density gradient centrifugation through The benefit of adding antibiotics to semen preparation silane-coated silica particles (PureSperm®) using 100, 200, solutions and IVF media has been questioned both by media 300 and 500 g on bacterial contamination of sperm samples producers and IVF clinics. The most widely used antibiotics and recovery of motile spermatozoa from sperm samples in IVF are a combination of penicillin and streptomycin. If were investigated with conventional culturing techniques these antibiotics are excluded, the embryos cleave faster (Magli and microscopic visual assessment. The recovery of motile et al., 1996). Penicillin G is degraded to 50% within 24 h spermatozoa was variable and was not improved using (Neftel et al., 1983), while streptomycin is stable (Kassem 500 g compared to the recommended 300 g. The bacterial et al., 1983) at 37°C. A prerequisite for the reduction or contamination was highly decreased by gradient centrifuga- exclusion of antibiotics from IVF media is the use of laboratory tion through PureSperm® and was almost abolished when methods which eliminate the bacterial contamination originat- strict aseptic techniques were used, with changes to sterile ing from the semen sample. Pasteur pipettes and tubes prior to washing procedures. In cases where the male has low sperm counts and intracyto- Key words: assisted reproduction/bacteria/density gradient plasmic sperm injection (ICSI) is necessary, it is important to centrifugation/semen preparation/spermatozoa have a sperm preparation method with high and reliable recovery. When performing other forms of assisted reproduc- tion, such as conventional IVF and intrauterine inseminations, a sperm preparation with good recovery could be crucial. Introduction Silane-coated silica particles have been reported both to provide Density gradient centrifugation of liquefied ejaculate through equal sperm recovery compared to PVP-coated particles polyvinylpyrrolidone (PVP)-coated silica particles (Percoll; (Claassens et al., 1998; Centola et al., 1998) and to have a Pharmacia, Uppsala, Sweden) was used for sperm preparation lower recovery (Chen and Bongso, 1999). (Bolton and Braude, 1984) until 1996. Percoll has since been The purpose of this study was to evaluate the effect of various withdrawn from use in clinically applied assisted reproduction centrifugation forces and the usage of aseptic techniques on and has been replaced by other products. Percoll is known bacterial contamination, recovery of spermatozoa and sperm selectively to clean sperm samples from bacterial contamina- motility under antibiotic-free culturing conditions. tion (Bolton et al., 1986), but the efficiency of new compounds, e.g. silane-coated silica particle solution (PureSperm®; Nidacon International AB, Gothenburg, Sweden), for removing bacterial Materials and methods contamination has not been proven. In this study, one sperm sample from each of 30 patients undergoing The clinical importance of a procedure that eliminates fertility investigation or IVF treatment were examined. The patients bacteria from the semen is evident as bacteriospermia is were between 23 and 47 years old (mean 35.9). The infertility common (Wille´n et al., 1996; Cottell et al., 1997). Moreover diagnosis of the couples was of a female origin in 21 cases and was there is a tendency of some bacteria to adhere to spermatozoa idiopathic in nine cases. No cases of diagnosed male infertility were 6 (Friberg and Fullan, 1983; Wolff et al., 1993; Diemer et al., included. For participation in the study a cut-off value of 80ϫ10 1996), and to impair both motility (Kaur et al., 1986) and the motile spermatozoa in the ejaculate was set. No patients were excluded inducibility of the acrosome reaction (el Mulla et al., 1996). because of previous infections. The ejaculate was collected by masturbation into a sterile container The clinical importance is further strengthened by another (Falcon No. 2070; Becton Dickinson Co., Franklin Lakes, NJ, USA) finding (Huyser et al., 1991) that, in bacterial infested cultures, and allowed to liquefy for 30 min on a rocking table. After the oocytes are degenerated. However, it is somewhat surprising semen analysis and microbiological examination, the remaining semen that the presence of bacteria in the original semen sample has sample was split into four aliquots and carefully placed on top of been reported to have no effect on fertilization, cleavage or PureSperm® gradients. Each gradient comprised two 1.0 ml layers of pregnancy rates in in-vitro fertilization (IVF) treatments (Riedel 90 and 45% (v/v) respectively, diluted in Gamete-100 medium 662 © European Society of Human Reproduction and Embryology Semen preparation and bacterial contamination (Scandinavian IVF Science, Gothenburg, Sweden) prepared in sterile at 200, 300 and 500 g. The percentage of motile spermatozoa conical tubes (Falcon No. 2095; Becton Dickinson Co.). The Pure- in the original specimen was 58.9 Ϯ 2.5%. Centrifugation ® Sperm gradients were run at 100, 200, 300 and 500 g for 20 min through PureSperm® increased the percentage of motile sper- at room temperature and without braking. The supernatants were matozoa significantly compared to semen. Even though the removed using sterile Pasteur pipettes, starting at the surface using a recovery of motile spermatozoa was increasing using higher circular motion around the inside of the tubes until approximately µ g force, the percentage of motile spermatozoa to total spermato- 100 l were left. The sperm pellets were carefully mixed with the Ͻ Ϯ remaining supernatants and resuspended in 2.5 ml Gamete-100. This zoa was significantly decreased (P 0.05) from 87.5 1.7% Ϯ Ϯ was performed in two different ways, either in the same tubes or (100 g) to 80.6 2.8 (200 g) and 77.5 2.6 (300 g) and after transfer of the pellets, with clean Pasteur pipettes, to clean tubes further decreased to 71.6 Ϯ 3.8% when centrifuged using 500 g. where the resuspensions were performed. The tubes were allowed to In 13 of 120 (11%) samples the motile sperm counts were stand for 5 min, after which they were centrifuged at 200 g for 10 under 2ϫ106 after preparation, irrespective of the centrifugal min. The supernatants were removed except for 200 µl in which the forces used (data not shown). sperm pellets were resuspended. This washing procedure was repeated When culturing from the original semen sample, only one once. Samples were taken from the sperm suspensions for microbio- lacked detectable bacteria, while none had detectable fungus. logical cultivation and sperm analysis. The most common bacterial type was coagulase negative Semen and sperm suspensions were analysed by microscopic staphylococci, followed by α-haemolytic streptococci (Table visual assessment using a Makler counting chamber (Sefi Medical I). Other bacterial species were found sporadically. Thirteen Instruments, Haifa, Israel) for the number of motile and immotile spermatozoa. semen samples had two or more types of bacteria. Ϯ Microbiological examinations of the ejaculates and sperm suspen- As seen in Figure 1, 30.83 7.91 colonies were formed/ sions were performed. The ejaculates and sperm suspensions were 10 µl semen in the original specimen. This number decreased plated with a standard 10 µl plastic loop onto blood agar (Columbia significantly to 3.23 Ϯ 0.96 colonies after preparation by II agar base, BBL; Becton Dickinson Co.), and saboraud agar (Lab density gradient centrifugation. When a strict aseptic laboratory M Ltd, Manchester, Lancashire, UK). A total of 50 µl samples were procedure was used, with change of pipettes and tubes before also inoculated in 1 ml of Todd Hewitt bouillon (BBL; Becton washing procedures, the number of colonies was further Dickinson Co.). This was performed within 1 h from the end of the decreased to 0.13 Ϯ 0.05 colonies. No significant differences liquidation or washing procedure. These bacteriological cultures were in bacterial numbers were obtained with different centrifugation all incubated at 37°C. The blood agar plates were cultured for 24 h forces (Table II). Even though the bacterial parameters in both and an additional 24 h if the initial reading was negative. The Todd Figure 1 and Table II were measurements of concentration, Hewitt bouillons were incubated for 48 h, then plated on blood agar dishes and incubated as previously mentioned. The fungal cultures the general picture and significant differences found were the were incubated at 30°C for 7 days. After the incubations, the numbers same when the numbers of bacteria loaded on the gradients of colonies were counted and the isolates identified using routine and in the prepared samples were calculated and compared. laboratory techniques. The number of samples without bacteria was significantly A control assay was performed to evaluate a possible bacteriostatic greater after preparation with strict aseptic techniques (40 of effect of PureSperm®. Isolates of three bacterial species were investi- 68 cases) both compared to original samples (one of 30 cases) gated: Escherichia coli, Staphylococcus epidermidis and Streptomycin and after preparation without changing pipettes and tubes prior sanguis. Four wells/strain were cut in blood agar dishes each filled to washing procedures (eight of 52 cases).