Leukocyte detection in human semen using flow cytometry

Giuseppe Ricci1,3, Gianni Presani2, Secondo Guaschino1, Roberto Simeone1 and Sandra Perticarari2

1 Department of Obstetrics and Gynaecology, University of Trieste, Istituto per l'Infanzia `Burlo Garofolo', IRCCS and 2 Clinical Pathology Laboratory, Istituto per l'Infanzia `Burlo Garofolo', IRCCS, 34137 Trieste, Italy


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
This study set out to establish a new method, using flow cytometry, to evaluate leukocytes in semen. Ejaculates of 59 males, asymptomatic for genitourinary infections, were examined. Routine semen analyses were carried out as well as peroxidase and polymorphonuclear granulocyte–elastase detection. Leukocytes were detected combining flow cytometry and monoclonal antibodies (anti-CD45, anti-CD53). This technique reliably assessed the total number of leukocytes and differentiated subpopulations even at low concentrations. The peroxidase test and elastase determination showed good specificity, but only moderate sensitivity versus flow cytometry combined with monoclonal antibodies. No significant association was observed between semen parameters and leukocytospermia whether evaluated by conventional methods or flow cytometry except for a moderate correlation between spermatozoa and CD53-positive cell concentrations. A first comparison of data from patients grouped on the basis of leukocytospermia (>106 white blood cells, WBC/ml) or non-leukocytospermia revealed no significant differences in semen parameters; lowering the threshold value for leukocytospermia to 2x105 WBC/ml, sperm concentration was reduced in the group with a low number of WBC identified by monoclonal antibodies. Flow cytometry using monoclonal antibodies was seen to be a simple, reproducible method that enables leukocytes in semen to be accurately detected and to identify WBC subpopulations without preliminary purification procedures.

Key words: elastase/flow cytometry/human semen/leukocyte/peroxidase


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Elevated concentrations of leukocytes in semen have been associated with genital tract infection, poor semen quality and IVF and embryo transfer failure (Cohen et al., 1985Go; Talbert et al., 1987Go; Wolff, 1995Go). According to the World Health Organization (WHO), the threshold value for a normal seminal white blood cell (WBC) count is 1x106/ml (WHO, 1992), but there is no consensus in literature for this threshold value (Barratt et al., 1988Go; Wolff, 1995Go), possibly because there is no reliable method to detect and differentiate seminal WBC accurately. Leukocytes are usually counted manually and microscopically, but these methods are inappropriate for counting low numbers of WBC in semen. Moreover, direct counting of round cells is highly inaccurate because WBC cannot be distinguished from immature germ cells. WHO recommends the peroxidase test but this detects only granulocytes and not other WBC types (Endtz et al., 1974). Furthermore, to investigate the mechanisms by which WBC might cause sperm dysfunction, reliable differentiation of WBC subpopulations in semen is essential. Immunohistological staining employing monoclonal antibodies against all specific WBC subpopulations is now considered by many to be the gold standard of semen WBC tests, but it is expensive, time consuming and not standardized (El-Demiry et al., 1986Go; Wolff and Anderson, 1988; Schobel et al., 1989Go; Harrison et al., 1991Go; Eggert-Kruse et al., 1992Go; Kiessling et al., 1993Go). In the past 10 years, flow cytometry technology has been used for sperm analysis (Gledhill et al., 1990Go; Spanò and Evenson, 1993Go; Pasteur et al., 1994Go; Ferrara et al., 1997Go; Moilanen et al., 1998Go; Gandini et al., 1999Go). This technique, when used alongside monoclonal antibodies, allows a rapid, multiparameter analysis of particles and is suitable for counting and typing seminal leukocytes.

The purpose of this study was to establish a new method to evaluate low numbers of leukocytes in semen and to differentiate WBC subpopulations, combining flow cytometry and monoclonal antibodies. Moreover, the results obtained from employing two different monoclonal antibodies were compared with the results from a standard peroxidase technique and a polymorphonuclear (PMN) granulocyte–elastase determination method. The relationships between the number of WBC in semen and parameters of semen quality determined during routine laboratory tests were also investigated.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Patients
Fifty-nine randomly chosen males from couples attending the Reproductive Medicine Unit of the Department of Obstetrics and Gynaecology of Trieste University were evaluated. Ages ranged from 28 to 46 years with a median of 35 years. The duration of infertility ranged from 19 to 151 months with a median of 41 months. All patients were asymptomatic for genitourinary infections. Informed consent to use semen samples was obtained from all subjects.

Semen analysis
Semen samples were collected in the hospital by masturbation into a sterile container after 3–4 days of sexual abstinence. Routine semen analyses including sperm volume, pH, count, progressive motility and morphology were performed according to WHO recommendations (WHO, 1992).

Peroxidase test
The peroxidase test is recommended by WHO to detect PMN granulocytes (WHO, 1992). The protocol used was adapted from Endtz (1974). A stock solution was assayed by mixing 50 ml distilled water with 50 ml 96% ethanol and adding 125 mg benzidine (Sigma, Milan, Italy). The working solution was obtained by adding 5 µl 30% H2O2 to 4 ml of stock solution. Twenty µl of working solution were mixed with 20 µl of liquefied semen in a small test tube. After incubation for 5 min at room temperature, 20 µl of working solution was mixed with 20 µl of phosphate-buffered saline. Then, 10 µl were placed in a haemocytometer, and peroxidase-positive cells, i.e. dark brown round cells, were counted.

Determination of granulocyte elastase concentration
PMN granulocyte elastase concentration in seminal plasma was determined by means of a commercial immunoassay (Ecoline PMN-Elastase; Merck, Milan, Italy). Liquefied semen was centrifuged at 350 g for 15 min. Seminal plasma was aspirated and stored at -80°C until use. Latex particles were coated with antibody fragments F(ab')2 against human PMN elastase; changes in turbidity were measured photometrically. The extent of turbidity was proportional to PMN elastase concentration in the test sample. The reagent kit was used on an automatic analyser Cobas Fara (Roche, Milan, Italy) at 37°C and kinetic measurements performed at ~700 nm. The kit detected PMN elastase concentrations of >4 µg/l.

Preparation of semen samples for staining with monoclonal antibodies
An aliquot of semen specimen (0.5 ml of samples with 60–80x106/ml spermatozoa, 1.5 ml for lower spermatozoa concentration) was added to 1.5 ml of Hanks' balanced salt solution (HBSS; Sigma) containing 5% heat-inactivated fetal bovine serum (Sigma) and 2 U/ml of potassium heparin, 0.1% sodium azide, centrifuged at 350 g for 10 min. The supernatant was discarded and the pellet was resuspended in 4–5 ml of the same medium to reach a final concentration of ~5–10x106/ml spermatozoa. Direct immunofluorescence staining was performed in aliquots of 100 µl of sperm suspension, after vortex mixing to ensure monodispersion, with 20 µl of the following monoclonal antibodies: anti pan-leukocyte CD45 fluorescein isothiocyanate (FITC)-conjugated (Becton Dickinson, Milan, Italy) and CD53 phycoerithrin (PE)- conjugated (Pharmingen; Becton Dickinson, San José, CA, USA). As negative controls, samples were also stained with uncorrelated isotype monoclonal antibodies FITC or PE conjugated. Following incubation in the dark for 20 min at room temperature, cells were then prepared for analysis using a Lyse-no-Wash technique (ImmunoPrep©; Coulter Beckman, Fullerton, CA, USA) or washed with HBSS containing azide and re-suspended in 500 µl of 1% paraformaldehyde solution. In order to pinpoint leukocyte regions during flow cytometric analysis, a known amount of leukocyte suspension was added to further normal semen samples.

Gates were established using blood WBC because their cytometric properties were found to be comparable to semen leukocytes.

Peripheral leukocytes were obtained from heparinized blood of healthy donors by dextran sedimentation (Boyum, 1968Go) followed by hypotonic lysis with 0.87% ammonium chloride to eliminate red blood cells. Purified leukocytes were re-suspended in HBSS and counted on a haemocytometer. Since a commercial leukocyte-positive control in semen is not available, we needed to pinpoint leukocyte regions during cytometric analysis in order to identify leukocyte subpopulations and we added a known amount of peripheral blood leukocytes to semen samples without leukocytes; blood WBC were used to develop gates because we found their cytometric properties comparable to semen leukocytes. Having obtained different leukocyte regions, we confirmed these gates using positive controls.

Flow cytometry
Samples were acquired using a FACScan cytometer (Becton Dickinson) equipped with a 15 mW air-cooled 488 nm argon-ion laser. FL1 (FITC) signals were detected through a 530/30 nm band pass filter, FL2 (PE) signals were detected through a 585/42 nm band pass filter. 20 000 or 30 000 events were recorded in list mode and analysed using Lysys II software (Becton Dickinson). Data were displayed in a dot plot on the basis of the linear forward (FSC) and side scatter (SSC) properties of the cells; the events accumulated at the lower left corner, corresponding to debris, were excluded from the analysis. WBC populations were identified using the light scatter properties of the white cells together with the differential antigen density expression of the pan-leukocyte marker CD45, which had a low affinity with polymorphonuclear cells (CD45+), was more expressed on monocytes (CD45++) and had a high affinity with lymphocytes (CD45+++). The total number of leukocytes present was evaluated by multiplying the percentage of CD45- or CD53-positive cells by the total sperm count. Flow cytometry data were compared to data from simultaneous microscope evaluation of peroxidase tests on the same sample cells.

Statistical analysis
Non-parametric statistics were used to protect against non-normal distribution of data. A Spearman's rank correlation coefficient was used to assess correlations between seminal leukocyte detection methods, and between WBC concentrations and semen parameters. Sensitivity and specificity of the peroxidase test and PMN elastase determination were calculated using a contingency table and taking flow cytometry as the standard. Ejaculates were bracketed into leukocytospermic and non-leukocytospermic groups using different WBC concentration thresholds and statistically significant differences between the groups were examined by the Mann–Whitney U-test. P < 0.05 was considered statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Flow cytometry analysis
Cytometry analysis, acquiring 20 000 events or more, allowed a reliable enumeration of leukocytes even at low concentrations. In order to analyse all the sperm population the threshold value on the FSC was set sufficiently low, and a large gate was set so that only debris was excluded. On the sole basis of forward versus right-angle parameters it was impossible to distinguish WBC from sperm cells. The leukocyte window was identified by analysing a normal sample of spermatozoa to which a known number of isolated leukocytes had been added (10% of total spermatozoa), and stained with CD45 FITC; by analysing the CD45 fluorescence intensity and side scatter properties it was possible to define the regions R1, R2 and R3, representing lymphocytes, monocytes and granulocytes respectively. As shown in Figure 1Go, plot A represents the fluorescence versus side scatter of the sperm sample with leukocytes in which it was possible to distinguish the three leukocyte subpopulations in different colours corresponding to the three regions R1, R2, R3; plot B shows the scattergram of forward versus side scatter alone. Plots C and D show fluorescence and forward versus side scatter of the same sperm sample without the addition of leukocytes.



View larger version (54K):
[in this window]
[in a new window]
 
Figure 1. Cytometry analysis of a normal sperm sample with addition of known amount of isolated leukocytes, displayed in a dot plot on the basis of CD45 fluorescence versus side scatter (SSC; plot A) or of the scatter properties alone (forward scatter, FSC; plot B); regions R1 (red), R2 (green) and R3 (blue) represent lymphocytes, monocytes and granulocytes respectively. Plots C and D represent the same sample without addition of leukocytes. Events accumulated at the lower left corner, corresponding to debris, were excluded from the analysis.

 
The total number of leukocytes/ml and relative subset was calculated by applying the following formula:

Results were compared to the relative values of the sample to which leukocytes had been added, counted and typed by haemocytometer. Analysis of unknown samples was performed using the above-described criteria in order to identify the presence of leukocytes. Figure 2Go illustrates three examples of sperm samples: two pathological samples with white cells (plots A and B, C and D), and a normal sample without leukocytes (E and F). Parallel staining of samples with CD53 monoclonal antibody yields similar results.



View larger version (67K):
[in this window]
[in a new window]
 
Figure 2. Three examples of sperm samples: two pathological samples with presence of leukocytes (plots A and B, C and D) and a normal sample without leukocytes (E and F). For definition of regions see Figure 1Go legend.

 
Flow cytometry methods compared to the peroxidase test
White blood cell concentrations obtained via the peroxidase test and cytometry methods were significantly correlated. Spearman rank correlation coefficients (r) between peroxidase-positive cell concentrations and CD53-positive, CD45-positive cell concentrations were +0.692 (P < 0.0001) and +0.619 (P < 0.0001) respectively. There was a high degree of correlation between CD45- and CD53-positive cell concentrations (r = 0.818, P < 0.0001). When the threshold value for leukocytospermia recommended by WHO (106 WBC/ml) was applied, a number of samples were discordant. Compared to CD45, sensitivity of the peroxidase test was 58.8%, and specificity was 92.8%, whereas compared to CD53, sensitivity of the peroxidase test was 55.5% and specificity 94.1%.

Flow cytometry methods in relation to the determination of PMN granulocyte elastase concentration
Spearman rank correlation indicated a significant association of PMN elastase concentrations in seminal plasma with the numbers of WBC in semen counted by the CD45 monoclonal antibody cytometry method (r = 0.542, P < 0.0001); a similar correlation was observed using CD53 monoclonal antibody (r = 0.647, P < 0.0001). PMN granulocyte elastase concentration in semen ranged from 19 to 508 µg/l; this means that samples considered `positive' (>1000 µg/l), according to previous reports (Jochum et al., 1986Go, Wolff and Anderson, 1988), were not found within this population. Consequently, for statistical testing, the threshold value was arbitrarily lowered and a concentration of 250 µg/l was taken as the negative/positive threshold. In comparison with CD45 monoclonal antibody, sensitivity of PMN elastase was 78.8% and specificity 75%, whereas using CD53, sensitivity was 88.9% and specificity 100%.

Low and high WBC concentration ranges: conventional methods versus flow cytometry
To evaluate the relationships between conventional methods and flow cytometry methods in the low and high leukocyte concentration ranges respectively, semen samples were divided into two groups using 2x105 WBC/ml as a threshold value. When the WBC count was >2x105 WBC/ml, Spearman rank correlation revealed significant association for all the methods, even if a number of differences compared to results regarding all patients were detected (Table IGo). In samples with a low number of leukocytes the correlation between the peroxidase test and flow cytometry was significant, whereas no statistically significant association was found between PMN elastase concentrations and CD45 monoclonal antibody or CD53 monoclonal antibody-positive cell concentrations. There was only moderate concordance between the numbers of peroxidase-positive cells and the PMN granulocyte elastase concentrations in seminal plasma (r = 0.471, P < 0.001).


View this table:
[in this window]
[in a new window]
 
Table I. Correlations between different seminal leukocyte detection methods in total samples and in two subgroups of patients using 2x105 WBC/ml as a threshold value
 
Semen parameters in relation to seminal WBC
Spearman rank correlation between primary semen parameters (ejaculate volume, sperm concentration, progressive motility, morphology) and seminal WBC concentrations and PMN elastase concentrations was performed (Table IIGo). No significant association between semen parameters and leukocytospermia evaluated by conventional methods or flow cytometry with CD45 monoclonal antibody was observed. Using flow cytometry combined with CD53 monoclonal antibody, a moderate correlation was found only between sperm and CD53-positive cell concentrations. Data were also analysed after subgrouping patients according to leukocytospermia. No significant differences in any semen parameters were observed when leukocytospermic and non-leukocytospermic patients were compared (Table IIIGo). Lowering the threshold for leukocytospermia to 2x105/ml, sperm concentration was significantly reduced in the group with a low number of WBC identified by CD53 monoclonal antibody (P < 0.05), and a similar trend was observed for data obtained using CD45 monoclonal antibody (P < 0.05) (Table IVGo).


View this table:
[in this window]
[in a new window]
 
Table II. Correlations between semen parameters and three different seminal leukocyte detection methods in 59 samples of ejaculate
 

View this table:
[in this window]
[in a new window]
 
Table III. Comparison of semen parameters between leukocytospermic and non-leukocytospermic patients (threshold value = 106 WBC/ml) using three different seminal leukocyte detection methods
 

View this table:
[in this window]
[in a new window]
 
Table IV. Comparison of semen parameters between leukocytospermic and non-leukocytospermic patients (threshold value = 2x105 WBC/ml) using three different seminal leukocyte detection methods
 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Identification and quantification of the different types of leukocytes present in semen samples has long been difficult. Automated instrumentation based on image analysis is now entering the laboratory for semen analysis, but it is very expensive and of limited value in routine use (Knuth et al., 1987Go). Recently, analysis by flow cytometry, a well-accepted technique in cellular immunology, has entered the andrology laboratory and has been devised to detect antisperm antibodies (Haas and Cunningham, 1984Go; Rasanen et al., 1992Go; Ke et al., 1995Go), to evaluate sperm cell type, count and viability (Ferrara, 1997), and to study sperm chromatin structure (Spanò and Evenson, 1993Go; Pasteur et al., 1994Go). The advantage of flow cytometry is that many millions of cells can be analysed in a few seconds, giving a statistically more precise evaluation via a reproducible technique. Different results have been described when using flow cytometry to detect seminal WBC and conflicting data have been reported on the power of CD45 monoclonal antibody and CD53 monoclonal antibody to selectively identify leukocytes in semen. Previous immunohistological studies have shown monoclonal antibodies against the common leukocyte antigen CD45 to be reliable and accurate in detecting, in semen, granulocytes, lymphocytes, and macrophages at the same time (El-Demiry et al., 1986Go; Wolff and Anderson, 1988aGo; Barratt et al., 1990Go; Eggert-Kruse et al., 1992Go; Kiessling et al., 1993Go) whereas other monoclonal antibodies employed, such as CD 11 and CD14, do not recognize the entire PMN population (Schöbel et al., 1989Go). However, Wolff and Anderson (1988a) observed cross- reactivity of CD45 with immature germ cells in semen, and, since there was no reactivity of CD45 with epithelial or testicular germ cells, they suggested that immature germ cells in semen might absorb antigenic substances from secretions entering down-stream from the testis (Wolff and Anderson, 1988aGo). Barratt et al. (1991) reported cross-reactivity of CD45 with many anucleate bodies in semen and postulated that this cross-reactivity might be partly due to a cross-reactive antigen in seminal plasma which masks the other cells (Barratt et al., 1991Go). Diemer et al. (1994) were the first to attempt to evaluate seminal WBC using flow cytometry combined with Percoll density gradient centrifugation, but they failed to identify a leukocyte population, suggesting a possible cross-reaction between CD45 and other cells, such as spermatozoa or immature germ cells (Diemer et al., 1994Go). On the contrary, Denny et al. (1995) reported successfully using CD45 monoclonal antibody to identify a leukocyte gate to evaluate T-lymphocyte subsets present in semen (Denny et al., 1995Go). In the light of these controversial results, Ferrara (1997) employed monoclonal mouse anti-human pan-leukocyte antigen CD53 and identified the leukocyte plot without an overlapping signal in the sperm population (Ferrara, 1997); unfortunately, however, a comparison between CD45 and CD53 monoclonal antibody was not performed in this study. Moilanen et al. (1998) used CD16 (Leu-11c) monoclonal antibody (Moilanen et al., 1998Go); however, this reacts strongly with isoform NA1 but weakly with NA2 (Hibbs et al., 1994Go) and may underestimate granulocyte concentration in semen from NA2 homozygotes. Recently, Gandini et al. (1999) reported successfully using flow cytometry to identify leukocytes separated from immature germ cells using a discontinuous Percoll gradient (Gandini et al., 1999Go). This method required the simultaneous use of several monoclonal antibodies (CD14, CD45, CD11b, CD19, CD3) to evaluate leukocyte subpopulations. None of the studies published describes a correlation between flow cytometry and other WBC identification methods.

This study describes a simple method, able to identify leukocytes even when low in number, and to differentiate WBC subtypes, using only one monoclonal antibody and requiring no further steps, such as separation by Percoll density gradient centrifugation. The difficulty of detecting very low concentrations of WBC was overcome by acquiring high numbers of events, 20 000 or more, so that the number of positive cells ensures sound statistical significance. Moreover, using a single monoclonal antibody and scatter properties of cells, it was possible to detect the subtype of leukocytes present. By analysing the CD45 or CD53 fluorescence intensity and side scatter properties, the regions R1, R2 and R3 were defined representing lymphocytes, monocytes and granulocytes respectively. The results were reproducible using CD45 or CD53 monoclonal antibody. Unlike previous studies, CD45 monoclonal antibody showed no cross-reactivity with spermatozoa and other cells; this might be due to the greater sensitivity and specificity of flow cytometry compared with immunohistochemistry technique. Multiparameter measurement allowed not only staining with CD45 monoclonal antibody but also simultaneous staining with monoclonal antibody and consideration of scatter properties, such as size and granularity, which make it easier to distinguish leukocytes from other cells e.g. anucleate bodies.

The concordance between conventional methods and flow cytometry was also analysed. The peroxidase test and PMN elastase determination showed good specificity, but only moderate sensitivity versus flow cytometry combined with monoclonal antibodies. Critical comparisons between the results from the current study and those of other authors are difficult as this is the first study using flow cytometry analysis. However, a good correlation was found between WBC detected by the peroxidase test and CD45, confiming previous results obtained with an immunocytochemistry technique using the same monoclonal antibody (Politch et al., 1993Go). Spearman rank correlation revealed a significant but only moderate association between PMN elastase values and CD45-positive cells, consistent with immunocytochemistry findings by Eggert-Kruse et al. (Eggert-Kruse et al., 1995Go), but unlike another study that reported a highly significant relationship between PMN elastase concentrations and CD45 immunohistological technique (Wolff and Anderson, 1988bGo). On the contrary, when CD53 monoclonal antibody was used, the peroxidase test and PMN elastase determination both showed a good correlation with the flow cytometry results; similar comparisons using CD53 monoclonal antibody have not been made in previous studies. Correlation of methods was also tested in low and high seminal WBC ranges, taking 2x105 WBC/ml as a threshold value. For semen samples with a high WBC count, correlation between conventional methods and monoclonal antibodies was still significant, but lower, except for the correlation between elastase and CD53, which was higher. In the low range, the peroxidase test and monoclonal antibody methods were significantly correlated, while no significant association with elastase determination was noted. The reasons for these discrepancies might be that the peroxidase test and PMN elastase determination detect only granulocytes and that PMN elastase was inaccurate when few seminal leukocytes were present.

Correlation between results from PMN elastase determination and the peroxidase test was only moderate; this is consistent with other reports (Micic et al., 1989Go; Wolff et al., 1992Go; Reinhardt et al., 1997Go) and might be explained by different test principles. The PMN elastase method detects extracellular enzymes whereas the peroxidase test detects intracellular enzyme activities. When evaluating the accuracy of PMN elastase in detecting leukocytospermia in comparison with flow cytometry, no positive samples were observed (>1000 ng/ml), whereas when the threshold value was lowered to 250 ng/ml, specificity was 75%, and sensitivity was 78.8% versus CD45, whereas using CD53 monoclonal antibody specificity was 100% and sensitivity 88.9%. Other authors failed to find a significant number of positive samples using the threshold value suggested by earlier studies (Jochum et al., 1986Go; Wolff and Anderson, 1988bGo); in a series of 557 subjects, Wolff et al. (1992) detected only four positive samples (0.7%), and PMN elastase concentrations >1000 ng/ml were not found within 159 randomly chosen males by Eggert-Kruse et al. (1995). Therefore, a revised, lower threshold value could be used in clinical practice, in order to improve the accuracy of PMN elastase determination.

Concordance between CD45 and CD53-positive cells was high, but not absolute, and their correlation with seminal WBC detected by conventional methods showed a number of differences. It can only be concluded that some leukocyte subtypes are probably lacking either from significant CD45 or CD53 antigen expression. Furthermore, CD45 and CD53 antigens might identify a different activation of WBC and play a different role in leukocyte biology. For example, human neutrophils express high amounts of CD53, but this amount of antigen may change when cells are activated. Treatment of human neutrophils with their physiological activators, tumour necrosis factor alpha or platelet-activating factor, led to down-regulation of this antigen from the cell surface (Mollinedo et al., 1998Go). Moreover, from an analysis of the correlation between semen quality and WBC concentration, significant association of sperm concentration with CD53-positive cells but not with CD45-positive cells was observed.

No correlation was found between other semen parameters and WBC concentration. The effect of leukocytospermia on male fertility remains controversial: an increased leukocyte presence in semen has been associated with reduced sperm–oocyte fusion (Maruyama et al., 1985Go; Vogelpoel et al., 1991Go) and a significant correlation between leukocytospermia and impaired semen quality has been reported by numerous studies (Wolff et al., 1990Go; Eggert-Kruse et al., 1992Go; Gonzales et al., 1992Go; Yanushpolsky et al., 1996Go). Furthermore, it was observed that the concentration of WBC in semen was a strong predictive factor for IVF–embryo transfer success (Cohen et al., 1985Go), Talbert et al., 1987Go; Van der Ven, 1987) and, more recently, it was shown that the number of leukocytes expressing CD45 was negatively correlated with the fertilization rate of metaphase II oocytes (Moilanen, 1998). However, other studies reached opposite conclusions. It has been reported (Aitken et al., 1994Go) that leukocytospermia did not significantly influence any component of the semen profile, even if in-vitro contaminating leukocytes showed a powerful negative correlation with fertilization rates (Sukcharoen et al., 1995Go). It was shown that seminal leukocytes were not correlated with sperm fertilizing capacity and IVF outcome (Tomlinson et al., 1992aGo; De Geyter et al., 1994Go). Yet, when the number of WBC was >6x106/ml, IVF and embryo transfer success dropped dramatically (De Geyter et al., 1994Go). However, this last finding was not supported by the observation that significantly increased sperm ideal forms and motility in semen with high WBC concentration (mean 6.4 ± 1.5x106 WBC/ml) (Kiessling et al., 1995Go). The only prospective study published (Tomlinson et al., 1993Go) concluded that seminal leukocytes were not associated with either semen quality or conception in-vivo rates. These conflicting reports are probably due to different detection methods, different populations studied and to the fact that leukocyte subtypes in semen may have different functions. Most of these studies lacked an accurate, reliable methodology for detecting WBC, making it impossible to draw any firm conclusions from such results. Although immunocytochemistry is considered the gold standard for the detection of WBC in semen, it is however a manual, subjective method. The correlation between semen parameters and leukocyte concentration quantified using flow cytometry and monoclonal antibodies had never previously been performed. Data in this study show that correlation between flow cytometry combined with monoclonal antibodies and other methods is limited.

It was observed that seminal parameter mean values were not significantly different in leukocytospermic and non- leukocytospermic patients, grouped applying the standard WHO threshold value of 106 WBC/ml. However, a statistically significant difference in sperm concentration was detected at leukocytospermia threshold values of 2x105 WBC/ml in semen. Semen samples with very low WBC concentrations showed significantly lower sperm concentrations. These data were confirmed using CD45 or CD53 monoclonal antibody, but not with the peroxidase test. Thus, the results demonstrate that the seminal WBC detection method is of great importance. Furthermore, these findings suggest that while the presence of an excess of WBC in semen may adversely affect sperm function (De Geyter et al., 1994Go), on the contrary a very low seminal leukocyte concentration may also have detrimental effects and supports the hypothesis that, besides potentially negative effects, seminal leukocytes may play a positive role in seminal biology (Tomlinson et al., 1992aGo; Kiessling et al., 1995Go). The results of this study appear to agree with the report of Tomlinson et al. (1992a) which showed that oligozoospermic samples contained significantly fewer leukocytes. In contrast with a recent study (Thomas et al., 1997Go), no significant correlation between leukocytospermia and sperm morphology was observed. However, the results were not comparable because in Thomas's study WBC were detected using a conventional quantification method.

In conclusion, more studies are needed to clarify the biological significance of WBC in semen and their relationship with male fertility. The peroxidase test and PMN elastase determination appear to be useful screening methods to detect leukocytospermia in routine semen analysis. However, for the purposes of a careful evaluation of selected cases of leukocytospermia and for laboratory or clinical research, more sophisticated, reproducible techniques are required. In this study a simple, reproducible method is presented that enables leukocytes in semen to be accurately detected and to identify WBC subpopulations, using a single monoclonal antibody and with no preliminary purification procedure. This method could offer a new perspective on more precise evaluations of WBC in semen.


    Notes
 
3 To whom correspondence should be addressed. E-mail: ricci{at}burlo.trieste.it Back


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Aitken, R.J., West, K. and Buckingham, D. (1994) Leukocytic infiltration into the human ejaculate and its association with semen quality, oxidative stress, and sperm function. J. Androl., 15, 343–352.[Abstract/Free Full Text]

Barratt, C.L.R., Li, T.C. and Monterio, E.F. (1988) Diagnosis and detection of male accessory gland infection. In Barratt, C.L.R. and Cooke, I.D. (eds), Advances in Clinical Andrology. MTP Press, Lancaster, p. 91.

Barratt, C.L.R., Harrison, P.E., Robinson, A. et al. (1990) Antisperm antibodies and lymphocyte subsets in semen – not a simple relationship. Int. J. Androl., 13, 50–58.[ISI][Medline]

Barratt, C.L.R., Harrison, P.E., Robinson, A.J. et al. (1991) Seminal white blood cells in men with urethral tract infection. A monoclonal antibody study. Br. J. Urol., 68, 531–536.[ISI][Medline]

Boyum, A. (1968) Isolation of mononuclear cells and granulocytes from human blood. Scand. J. Clin. Lab. Invest., 97 (Suppl.), 77–89.

Cohen, J., Edwards, R., Fehilly, C. et al. (1985) In vitro fertilization: a treatment for male infertility. Fertil. Steril., 43, 422–432.[ISI][Medline]

De Geyter, C., De Geyter, M., Behre, H.M. et al. (1994) Peroxidase positive round cells and microorganisms in human semen together with antibiotic treatment adversely influence the outcome of in-vitro fertilization and embryo transfer. Int. J. Androl., 17, 127–134.[ISI][Medline]

Denny, T.N., Scolpino, A., Garcia, A. et al. (1995) Evaluation of T-lymphocyte subsets present in semen and peripheral blood of healthy donors: a report from the heterosexual transmission study. Cytometry, 20, 349–355.[ISI][Medline]

Diemer, T., Weidner, W., Michelmann, H.W. et al. (1994) Percoll density gradient centrifugation and consecutive flow cytometry do not identify leukocytes and leukocyte subtypes in ejaculate specimens. Andrologia, 26, 93–96.[ISI][Medline]

Eggert-Kruse, W., Bellmann, A., Rohr, G. et al. (1992) Differentiation of round cells in semen by means of monoclonal antibodies and relationship with male fertility. Fertil. Steril., 58, 1046–1055.[ISI][Medline]

Eggert-Kruse, W., Probst, S., Rohr, G. et al. (1995) Screening for subclinical inflammation in ejaculates. Fertil. Steril., 64, 1012–1022.[ISI][Medline]

El-Demiry, M.I.M., Young, H., Elton R.A. et al. (1986) Leucocytes in the ejaculate from fertile and infertile men. Br. J. Urol., 58, 715–720.[ISI][Medline]

Endtz, A.W. (1974) A rapid staining method for differentiating granulocytes from `germinal cells' in Papanicolaou-stained semen. Acta Cytol., 18, 2–7.[ISI][Medline]

Ferrara, F., Daverio, R., Mazzini, G. et al. (1997) Automation of human sperm cell analysis by flow cytometry. Clin. Chem., 43, 801–807.[Abstract/Free Full Text]

Gandini, L., Lenzi, A., Lombardo, F. et al. (1999) Immature germ cell separation using a modified discontinuous Percoll gradient technique in human semen. Hum. Reprod., 14, 1022–1027.[Abstract/Free Full Text]

Gledhill, B.L., Evenson, D.P. and Pinkel, D. (1990) Flow cytometry and sorting of male germ cells. In Melamed, M.R., Lindmo, T. and Mendelsohn, M.L. (eds), Flow Cytometry and Sorting, 2nd edn. Wiley–Liss, New York, p. 531.

Gonzales, G.F., Kortebani, G. and Mazzoli, A.B. (1992) Leukocytospermia and function of the seminal vesicles on seminal quality. Fertil. Steril., 57, 1058–1065.[ISI][Medline]

Haas, G.G. and Cunningham, M.E. (1984) Identification of antibody-laden sperm by cytofluorometry. Fertil. Steril., 42, 606–613.[ISI][Medline]

Harrison, P.E., Barrat, C.L.R., Robinson A.J. et al. (1991) Detection of white blood cell populations in the ejaculate of fertile men. J. Reprod. Immunol., 19, 95–98.[ISI][Medline]

Hibbs, M.L., Tolvanen, M. and Carpen, O. (1994) Membrane-proximal Ig-like domain of Fc gamma RIII (CD16) contains residues critical for ligand binding. J. Immunol., 152, 4466–4474.[Abstract/Free Full Text]

Jochum, M., Papst, W. and Schill, W.B. (1986) Granulocyte elastase as a sensitive diagnostic parameter of silent male genital tract inflammation. Andrologia, 18, 413–419.[ISI][Medline]

Ke, R.W., Dockter, M.E., Majumdar, G. et al. (1995) Flow cytometry provides rapid and highly accurate detection of antisperm antibodies. Fertil. Steril., 63, 902–906.[ISI][Medline]

Kiessling, A.A., Yin, H-Z., Purohit, A. et al. (1993) Formaldehyde-fixed semen is suitable and safer for leukocyte detection and DNA amplification. Fertil. Steril., 60, 576–581.[ISI][Medline]

Kiessling, A.A., Lamparelli, N., Yin, H-Z. et al. (1995) Semen leukocytes: friends or foes? Fertil. Steril., 64, 196–198.[ISI][Medline]

Knuth, U.A., Yeung, C.H. and Nieshlage, H. (1987) Computerized semen analysis: objective measurement of semen characteristics is biased by subjective parameter setting. Fertil. Steril., 48, 118–122.[ISI][Medline]

Maruyama, D.K., Hale, R.W. and Rogers, B.J. (1985) Effects of white blood cells on the in vitro penetration of zona-free hamster eggs by human spermatozoa. J. Androl., 6, 127–135.[Abstract/Free Full Text]

Micic, S., Macura, M., Lalic, N. et al. (1989) Elastase as an indicator of silent genital tract infection in infertile men. Int. J. Androl., 12, 423–429.[ISI][Medline]

Moilanen, J.M., Carpén, O. and Hovatta, O. (1998) Flow cytometric light scattering analysis, acrosome reaction, reactive oxygen species production and leukocyte contamination of semen preparation in prediction of fertilization rate in vitro. Hum. Reprod., 13, 2568–2574.[Abstract]

Mollinedo, F., Martin-Martin, B., Gajate, C. et al. (1998) Physiological activation of human neutrophils down-regulates CD53 cell surface antigen. J. Leukocyte Biol., 63, 699–706.[Abstract]

Pasteur, X., Metézéau, P., Maubon, I. et al. (1994) Identification of two human sperm populations using flow and image cytometry. Mol. Reprod. Dev., 38, 303–309.[ISI][Medline]

Politch, J.A., Wolff, H., Hill, J.A. et al. (1993) Comparison of methods to enumerate white blood cells in semen. Fertil. Steril., 60, 372–375.[ISI][Medline]

Rasanen, M.L., Hovatta, O.L., Penttila, I.M. et al. (1992) Detection and quantitation of sperm-bound antibodies by flow cytometry of human semen. J. Androl., 13, 55–64.[Abstract/Free Full Text]

Reinhardt, A., Haidl, G. and Schill, W.B. (1997) Granulocyte elastase indicates silent male genital tract inflammation and appropriate anti-inflammatory treatment. Andrologia, 29, 187–192.[ISI][Medline]

Schöbel, W.A., Schieferstein, G. and Uchanska-Ziegler, B. (1989) Immunocytochemical characterization of round cells in human semen using monoclonal antibodies and the APAAP-technique. Andrologia, 21, 370–376.[ISI][Medline]

Spanò, M. and Evenson, D.P. (1993) Flow cytometric analysis for reproductive biology. Biol. Cell., 78, 53–62.[ISI][Medline]

Sukcharoen, N., Keith, J. and Irvine, D.S. (1995) Predicting the fertilizing potential of human sperm suspensions in vitro: importance of sperm morphology and leukocyte contamination. Fertil. Steril., 63, 1293–1300.[ISI][Medline]

Talbert, L.M., Hammond, M.G., Halme, J. et al. (1987) Semen parameters and fertilization of human oocytes in vitro: a multivariable analysis. Fertil. Steril., 48, 270–277.[ISI][Medline]

Thomas, J., Fishel, S.B., Hall, J.A. et al. (1997) Increased polymorphonuclear granulocytes in seminal plasma in relation to sperm morphology. Hum. Reprod., 12, 2418–2421.[Abstract]

Tomlinson, M.J., White, A., Barratt, C.L.R. et al. (1992a) The removal of morphologically abnormal sperm forms by phagocytes: a positive role for seminal leukocytes? Hum. Reprod., 7, 517–522.[Abstract]

Tomlinson, M.J., Barratt, C.L.R., Bolton, A.E. et al. (1992b) Round cells and sperm fertilizing capacity: the presence of immature germ cells but not seminal leukocytes are associated with reduced success of in vitro fertilization. Fertil. Steril., 58, 1257–1259.[ISI][Medline]

Tomlinson, M.J., Barratt, C.L.R. and Cooke, I.D. (1993) Prospective study of leukocytes and leukocyte subpopulations in semen suggests they are not a cause of male infertility. Fertil. Steril., 60, 1069–1075.[ISI][Medline]

Van der Ven, H.H., Jeyendran, R.S., Perez-Pelaez, M. et al. (1987) Leukospermia and the fertilizing capacity of spermatozoa. Eur. J. Obstet. Gynecol. Reprod. Biol., 4, 49–52.

Vogelpoel, F.R., van Kooij, R.J. te Veld, E.R. et al. (1991) Influence of polymorphonuclear granulocytes on the zona-free hamster oocyte assay. Hum. Rep., 6, 1104–1107.[ISI]

Wolff, H. (1995) The biologic significance of white blood cells in semen. Fertil. Steril., 63, 1143–1157.[ISI][Medline]

Wolff, H. and Anderson, D.J. (1988a) Immunohistologic characterisation and quantification of leukocyte subpopulations in human semen. Fertil. Steril., 49, 497–504.[ISI][Medline]

Wolff, H. and Anderson, D.J. (1988b) Evaluation of granulocyte elastase as a seminal plasma for leukocytospermia. Fertil. Steril., 50, 129–132.[ISI][Medline]

Wolff, H., Politch, J.A., Martinez, A. et al. (1990) Leukocytospermia is associated with poor semen quality. Fertil. Steril., 53, 528–536.[ISI][Medline]

Wolff, H., Panhans, A., Zebhauser, M. et al. (1992) Comparison of three methods to detect white blood cells in semen: leukocyte esterase dipstick test, granulocyte elastase enzyme immunoassay, and peroxidase cytochemistry. Fertil. Steril., 58, 1260–1262.[ISI][Medline]

World Health Organization (1992) WHO Laboratory Manual for the Examination of Human Semen and Sperm–Cervical Mucus Interaction. Cambridge University Press, Cambridge.

Yanushpolsky, E.H., Politch, J.A., Hill, J.A. et al. (1996) Is leukocytospermia clinically relevant? Fertil. Steril., 66, 822–825.[ISI][Medline]

Submitted on October 25, 1999; accepted on February 18, 2000.