Is the sperm bacterial ratio a determining factor in impairment of sperm motility: an in-vitro study in man with Escherichia coli

Transcription

1 International Journal of Andrology, 1991, 14, pages Is the sperm bacterial ratio a determining factor in impairment of sperm motility: an in-vitro study in man with Escherichia coli M. R. AUROUX*, L. JACQUES?, D. MATHIEU? and J. AUERt *Biologic de la Reproduction et du Developpement and f Laboratoire de Microbiologie, CHU Bicgtre, Le Kremlin Bicetre, Cedex, France Summary The effects of urogenital infection on male fertility are controversial. The object of this study was to assess whether contact between E. coli, one of the bacteria encountered most frequently in semen cultures, and sperm was involved in decreasing motility of the sperm. Sperm from healthy donors were therefore incubated at two concentrations (1.10' and 4.10' ml-') with bacteria (lo4 and lo6 bacteria ml-' respectively). Sperm motility was assessed as a function of time. The endotoxin effect was also evaluated. Aliquots of the sperm were used as controls. The motility of a population of lo6 sperm ml-' was reduced significantly more by the presence of 10' ml-' E. coli than a sperm population four times more numerous. Since the endotoxin had no effect on sperm motility, it is possible this phenomenon is due to bacterial adherence to the sperm. From this study, it is therefore probable that the presence of E. coli in semen decreases sperm motility, but that this depends on the sperm:bacterial ratio ml semen-'. Keywords: bacterial adherence, Escherichia coli, sperm motility, sperm-bacteria ratio. Introduction Do urogenital infections influence male fertility? The answer has long been a matter for debate with some authors convinced that it does (Getzoff, 1958; Gnarpe & Friberg, 1973; Diqueiou et al., 1986), whilst others do not (Riley & Masters, 1956; Gump et al., 1984; Cintron et al., 1981; Battin et al., 1984; Hellstrom et al., 1986). If urogenital infections do play a role in male infertility, they may (a) directly affect the sperm, or (b) indirectly result in either obstruction or non-destructive lesions of the male excretory ducts, in lesions of the accessory sex glands, or, finally, in the production of anti-sperm antibodies. The present study concerns the mode of action of Escherichia coli on human sperm. A previous study has suggested already that some bacteria such as Chlamydia frachomatis (CT) may affect sperm by contact (Auroux ef al., 1987). Indeed, examination of semen smears infected experimentally with CT revealed that distribution of the bacteria corresponded to the sperm distribution, and that areas Correspondence: Dr M. R. Auroux. Biologie de la Reproduction et du Developpement, CHU de BicPtre Le Kremlin Bicetre. Cedex. France. 264

2 SpermatozoalBacteria Ratio 265 without sperm exhibited none or very few bacteria. It was therefore assumed that some bacteria might affect sperm only when they are in contact with them, which would suggest that a phenomenon of adherence is involved. In this study, we have tried to determine whether such a phenomenon is implicated when E. coli, one of the bacteria encountered most frequently in semen cultures, is brought into contact with sperm. In this respect it is worth noting that E. coli has been shown already to decrease sperm motility (Bisson & Czyglik, 1974). This action is not due to the release of a toxin (Taegue et ul., 1971), and some strains show adherence properties in certain pathologies (Archambaud et al., 1988). Patients and methods General principles If bacteria affect sperm only when they are in contact with them, the decrease in motility of a given sperm population due to the presence of E. coli should be all the more important if the bacterial population is numerous or the sperm count is low. Two sperm concentrations from healthy donors were therefore incubated with two different concentrations of bacteria. Sperm motility was assessed as a function of time. Aliquots of the sperm were used as controls. Semen collection Semen samples were obtained by masturbation under aseptic conditions. Patients were requested to wash their hands and genital region carefully with bactericidal and antifungal soap and to rinse the soap away with isotonic solution. A total of nine ejaculates from seven different donors were examined. E. coli struins The bacterium used was isolated from the semen culture of a subfertile man presenting with chronic prostatitis. The strain was stored on Le Minor gelose (Diagnostics Pasteur, France), then transferred to solid medium 24 h prior to experimentation. It was isolated for purity assessment. The strain used showed two operons, designated as 'pap' and 'sfa', and known to be responsible for adherence to uro-epithelial cells when they are expressed (Archambaud et al., 1988). After incubation in soja-trypticase for 18 h, the bacterial culture was centrifuged at 500 g for 10 min, washed twice then resuspended in sterile Tyrode's solution. Concentrations were adjusted finally to lo" and lo7 bacteria ml-'. Incubation of sperm with E. coli Immediately after semen liquefaction and mixing using a pipette, the number of sperm was adjusted to two different concentrations: 1 X lo7 and 4 X lo7 sperm ml-' in sterile Tyrode's solution. An aliquot of 450 yl from each sperm concentration was then incubated with 50 yl from each bacterial culture (at concentrations of 10' and lo7 bacteria ml-') to give final ditutions of 10' and 10' bacteria ml-'. In control samples, the bacterial culture was replaced by 50 PI sterile Tyrode's solution. All samples were maintained at 37 C.

3 266 M. R. Auroux et al. ph, sperm motility and vitality assessment For all samples, the ph was measured, and a blind count of motile sperm was performed under the light microscope (X 40), immediately before incubation with bacteria (To) and then again 1, 2, 3, 4 and 5 h later. After routine sperm analysis, sperm vitality was assessed, using eosin-nigrosin staining, at To, T4 and T5. Due to lack of time, the two concentrations of one sperm sample tested were only examined from To to T4 and with only the 10' bacteria concentration (Table 1). Table 1. Sperm viability (%) as a function of time and of the ratio of sperm to E. coli Sperm + E. coli number mi-' Toh T4h 10' lo lo ' 407 4n7 + loh f f f f 8.98" Results are the mean f SD. "hcp<o.ol, h'ns (analysis of variance) f f i f 17' f f f f f i f f 13.92' f f f i Effect of endotoxin To establish that the effect of E. coli on sperm motility was not due to the release of bacterial endotoxin, aliquots of lo7 sperm suspended in 0.9 ml Tyrode's solution provided by the same donor were incubated with 0.1 ml Tyrode's solution containing 1 pg ml-' Endotoxin (Sigma L2880,7600 U/pg). The control sample, from the same sperm preparation, was a suspension of 10' sperm in 1 ml Tyrode's solution. Assessment of sperm motility was performed blind for all samples immediately prior to the addition of endotoxin, and then at 1, 2, 3, 4 and 5 h later. Three ejaculates from three different donors were examined in this way. Effect of trypticase-soja solution To eliminate the effect of Trypticase-soja traces (which could remain on the surface of bacteria) on sperm motility and vitality, lo7 sperm in 0.45 ml sterile Tyrode's solution were incubated with 0.05 ml Trypticase soja mixture. The control sample, from the same sperm preparation. was a suspension of lo7 sperm in 0.5 ml sterile Tyrode's solution. Sperm motility and vitality were evaluated at To and Ts. Three ejaculates from three different donors were examined in this way. Statistical analysis Variations in sperm motility as a function of time were evaluated using coefficient of regression. Adjustment of each curve to a straight line was made using the linear regression test, and the slopes obtained were then compared (Schwartz, 1963). Analysis of variance was used to compare means results.

4 SpermatozoalBacteria Ratio 261 Results The sperm motility of the control samples decreased significantly with time of incubation, irrespective of the sperm count: thus, control samples with lo7 sperm ml-i (Fig. 1; curve A) gave a coefficient of regression = -0.62, P<O.OOl; control samples with 4 x lo7 sperm ml-' (Fig. 2; curve A') gave a coefficient of regression of = -0.57, P<O.OOl Time (Hours) Fig. 1. Regression of sperm motility as a function of time. A: lo7 sperm ml-'; B: lo7 sperm + lo4 bacteria ml-'; C: lo7 sperm + lo6 bacteria m1-i zn Q A' + 11' * C'., I Time (Hours) Fig. 2. Regression of sperm motility as a function of time. A': 40' sperm m1-i; B': 407 sperm + lo4 bacteria ml-'; C': 40' sperm + 10' bacteria ml-'. Effect of lo4 E. coli ml-' on sperm motility As illustrated by curves B (Fig. 1) and B' (Fig. 2) corresponding respectively to sperm concentrations of lo7 ml-' and 4 x lo7 ml-', sperm motility was decreased significantly as a function of time (B: r = -0.55, P<O.OOl; B': r = -0.62, P<O.OOl). However, comparison of the slopes of the A and B curves on the one hand, and of the A' and B' curves on the other hand did not reveal any significant

5 268 M. R. Aurouxet al. differences (P>0.05). The presence of lo4 E. coli ml-i did not therefore influence sperm motility. Effect of lo6 E. coli ml-' on sperm motility The presence of 10' E. coli ml-i in a suspension of lo7 sperm ml-' (curve C, Fig. 1) resulted in a significant decrease in sperm motility as a function of time (r = -0.90, P<O.OOl). The decrease was also significant at a sperm concentration of 4 X lo7 m1-l (curve C', Fig. 2, r = -0.84, P<O.OOl). Comparison of the slopes of the A and C curves on the one hand, and of the A' and C' curves on the other hand revealed a significant difference between the control and contaminated sperm (A/C: P<O.OOl; A'/C': P<O.OOl). Furthermore, comparison between slopes of the C and C' curves also showed a significant difference, P<0.02). ph assessment For all samples, the ph remained between 7.4 and 7.6 from To to TS, except at Ts, when samples contained 10' bacteriah-' and either of the two concentrations of sperm, when the ph was 7.2. Effect of Endotoxin on sperm motility The decrease in sperm motility as a function of time was significant in the control and endotoxin samples. Control (n = 3). To (mean %): 55 t- 5; TI: 53.3 k 5.8; Tz: 53.3 k 5.8; T3: 51.7 t- 2.9; T4: 51.7 f 2.9; TS: 46.7 k 5.8 (r = -0.90, PC0.02). Endotoxin (n = 3). To (mean %): 55 f 5; TI: ; T2: 51.7 k 2.9; T3: 50 f 0.0; T4: 46.7 f 5.8; Ts: 43.3 k 2.9 (r = -0.98, P<O.Ol). Comparison of the slopes of the corresponding curves did not reveal any significant difference between the control and endotoxin samples. Effect of E. coli on sperm viability The only significant difference observed between sperm viability at To, on the one hand, and those at T4 or TS, on the other hand, concerned the lowest sperm concentration associated with the highest bacterial concentration (Table 1, abc). The mean viability of the other sperm samples, including the controls at To, T4 and TS, were not significantly different (Table 1). Effect of Trypticase-soja mixture on sperm motility and viability The Trypticase soja (TS) mixture did not have any significant effect on either sperm motility or viability: Motility (%). To. Control (n = 3): 51.7 k 2.9; TS -mixture (n = 3): 51.7 k 2.9. TS. Control (n = 3): 45 k 8.7; TS -mixture (n = 3): 43.3 f 5.8. Viability (%). To. Control (n = 3): 82 k 1.7; TS - mixture (n = 3): TS. Control (n = 3): 76.3? 6.5; TS - mixture (n = 3): 77.7 k 5.9. Discussion Our results suggest a certain number of points. First, it is interesting to note that a concentration of lo4 E. coli ml-' does not seem to impair the motility of sperm, whatever their number. Since a significant decrease in sperm motility was found at a bacterial concentration of lo6 E. cofi ml-', the presence of E. coli may begin to

6 SpermatozoalBacteria Ratio 269 be damaging at concentrations ranging from lo4 to lo bacteria ml-i. These figures are above those reported by other authors, i.e., lo3 (Fari et al., 1977) or lo4 bacteria ml- (Comhaire et al., 1981). It is noticeable however, that the estimates reported in the literature were made on the basis of clinical studies performed in retrospect, and hence did not result from the study of experimentally infected sperm, as is the case here. Our results also suggest that, with the ph remaining the same, the motility of a population of lo7 sperm ml-i is reduced more significantly by the presence of 10 E. coli ml- than is a sperm population of four times this value. As our data, in accord with those of Taegue et al. (1971), showed that endotoxin had no effect on sperm motility, it is suggested that bacterial adherence to sperm is the most likely cause of this phenomenon. In this respect, the probability that the presence of E. coli in semen may decrease sperm motility would be dependent not only on the duration of incubation of both populations, but would also fluctuate according to the ratio of the sperm bacterial concentration ml- semen. Consequently, the chances of semen contaminated by E. coli fertilizing an oocyte would not necessarily be zero, but might be reduced due to the number of unaffected sperm. Infection by E. coli could therefore be damaging or not, depending on both intra- and inter-individual variations in the number of sperm among ejaculates (Heuchel et al., 1982) and intra-individual variations in the composition of the sperm cell-coat (Courtot et al., 1981). If this were also the case for bacteria other than E. coli, this would explain the contradictory data reported in the literature regarding the role of infection in subfertility of the male. However, the adherence of E. coli to sperm remains to be demonstrated. Finally, concerning the nature of the effect of E. coli on sperm, the question remains as to whether bacteria affect sperm motility by impairing the quality of their movement or by killing them. In this respect, it is noticeable that at T4 and Ts the lowest motility values corresponded with the poorest sperm viability, and that only in these cases was the decrease in viability significant. We can therefore assume that the sperm:bacterial ratio was most unfavourable in this situation and led to lethal conditions for sperm, whereas, in the other cases, only their movement was altered. The value of this ratio might therefore determine the lethal power of E. coli at a given time. Acknowledgments We thank Mrs Labigne and Professor Le Minor (Institut Pasteur, Paris) for genotypic characterization of the bacteria used and Miss Malanguerray for her secretarial assistance. References Archambaud, M., Couroux, P., Ouin, V., Chabanon, G. & Labigne-Roussel, A. (1988) Phenotypic and genotypic assays for the detection and identification of adhesins from pyelonephritic Escherichia coli. Annales de I lnstitut Pasteur, microbiologie, 139, Auroux, M., Cortesse, A,, Feneux, D., Auer, J., Mathieu, D. & Jacques, L. (1987) L infection du sperme: modes d action des bactkries et relativitk des consequences. Contraception, Fertilitt?, Sexualiti, 15, Battin, D. A., Barnes, R. B.. Hoffman, D. I., Schachter, J., Dizerega, G. S. & Yonekura, M. L. (1984)

7 270 M. R. Auroux et al. Chlamydia trachomatis is not an important cause of abnormal post-coital tests in ovulating patients. Fertility and Sterility, 42, Bisson, J. P. & Czyglik, F. (1974) Retentissement de I infection gknito-urinaire sur les spermatozoides. Journal d Urologie et de Mkphrologie, 7-8, Cintron, R. D., Wortham, J. W. E. & Acosta, A. (1981) The association of semen factors with the recovery of ureaplasma urealyticum. Fertility and Sterility, 36, Comhaire, F., Verschraegen, G., Vermeulen, L. (1981) Diagnosis of accessory gland infection and its possible role in male infertility. In: Oligozoospermia: Recenf Progress in Andrology (eds G. Frajese, E. S. E. Hafez, C. Conti and A. Fabbrini) pp Raven Press, New York. Courtot, A. M., Auroux, M., Obrenovitch, A. & Monsigny, M. (1981) Intraindividual variability of concanavalin A binding to the plasmalemma of human spermatozoa. Archives of Andrology, 7, Diquelou, J. Y., Pastorini, E. & Lanoe, N. (1986) Rdle de Chlamydia trachomatis dans la survenue d anomalies de la mobilitt des spermatozoides. 2eme CongrPs mondial sur les Maladies Sexuellement Transmissibles, Paris 25-28, juin, Fari, F., Trevoux, R., Verges, J., Belaisch, J. & Lafarge. J. (1977) Incidences des 6tats inflammatoires ou infectieux des glandes gtnitales annexes sur le sperme. In: Infection et fkconditk. (eds J. Henri-Suchet, A. Steg and A. Constantin) pp Masson, Paris. Getzoff, P. (1958) The effects of chronic prostato-vesiculates on sperm mobility. International Journal of Fertility, 3, Gnarpe, H. & Friberg, J. (1973) T-mycoplasma on spermatozoa and infertility. Nature, 245, Gump, D. W., Gibson, M. & Ashikaga, T. (1984) Lack of association between genital mycoplasmas and infertility. New England Journal of Medicine, 310, , Hellstrom, W. G., Schachter, J., Sweet, R. L. & McClure, R. D. (1986) Is there a role for Chlamydia trachomatis and genital mycoplasma in male infertility? Congress of Urology 81st meeting of AVA. New York, May (Abstract). Heuchel, V., Schwartz, D., Czyglik, F. (1982) Between and within subject correlations and variances for certain semen characteristics in fertile men. Andrologia, 2, Riley, F. J. & Masters, W. H. (1956) Problems of male infertility 111. Bacteriology of Human semen. Fertility and Sterility, 7, Schwartz, D. (1963) Mkthodes statistiques a l usage des mkdecins er des biologistes. p Flammarion, Paris. Teague, N. S., Bogarsky, S. & Glenn, J. F. (1971) Interference of human spermatozoa motility by Escherichia coli. Fertility and Sterility, 22, Received 5 November 1990; accepted 22 March 1991 (after revision)