Human Sperm Membrane Research Articles

Carrier-mediated gamma-aminobutyric acid uptake in human spermatozoa indicating the presence of a high-affinity gamma-aminobutyric acid transport protein.

Human spermatozoa are capable of a carrier-mediated gamma-aminobutyric acid (GABA) uptake. The uptake is dependent on the concentration of Cl- and Na+ in the external medium, and the kinetics of the carrier resembles high-affinity GABA transport proteins. The time-dependent uptake of GABA displays large interindividual differences and is not correlated to motility parameters or morphology in the individual sample. Incubation of human spermatozoa with radiolabeled GABA was performed. Swim-up preparations of human spermatozoa were incubated with [3H]GABA, and subsequent GABA uptake was measured at various times by scintillation counting. GABA was accumulated intracellularly, and the uptake could be inhibited by preincubation of the samples in 200 microM nipecotic acid. Addition of aminooxyacetic acid in the medium did not alter the results, indicating that the internalized GABA remained unmetabolized intracellularly throughout the observation period. Kinetic analysis of GABA uptake was performed, and the Km for GABA transport was 14 microM. GABA uptake was reduced by equimolar substitution of NaCl in the capacitating medium by KCl, choline chloride, LiCl, N-methyl-D-glucamine (HCl) or D-glucuronic acid (sodium salt). Maximal reduction of [3H]GABA uptake was observed when the Na+ fraction of the medium was replaced with KCl. The results indicate the presence of a high-affinity GABA transport protein in the plasma membrane of human spermatozoa. GABA uptake was subsequently measured in 30 individual semen samples from men of barren couples. Large interindividual differences in GABA uptake was observed, but GABA uptake was not correlated to motility parameters or to morphology in the individual samples analyzed.

Laurdan fluorescence spectroscopy reveals a single liquid-crystalline lipid phase and lack of thermotropic phase transitions in the plasma membrane of living human sperm

Membrane lipid phase(s), phase coexistence, and thermotropic phase transitions have been investigated in viable human spermatozoa using Laurdan fluorescence spectroscopy. Generalized polarization (GP) values derived from Laurdan excitation and emission spectra confirm that the sperm plasma membrane is a low polar, highly rigid (liquid-ordered) structure, and give evidence that, in the range from 10°C to 42°C, membrane lipids are in a single liquid-crystalline phase. The absence of phase transitions in the same thermal range argues against the hypothesis that the lipid domains previously detected on the sperm surface are produced by lipid lateral phase separation. The above findings are likely accounted for by the high cholesterol to phospholipid molar ratio found in the human sperm membrane. This is the first time that membrane lipid phase and polarity have been detected and quantified in living mammalian spermatozoa.

Expression and characterization of a novel human sperm membrane protein.

A cDNA fragment (HSD-1) coding for part of a human sperm membrane protein (hSMP-1) was previously isolated from a human testis cDNA expression library, with the serum from an infertile patient used as a probe. By rescreening human testis cDNA libraries with the HSD-1 insert and using rapid amplification of cDNA ends, the complete cDNA of 2482 bp was identified and sequenced. An open reading frame of 1572 bp encodes 523 amino acid residues with a computed molecular mass of 55.08 kDa. This protein sequence does not match any other sequence in the databases, indicating that it represents a novel sperm antigen. Northern blot analysis of human and rat testis poly(A) mRNA detected a band of approximately 2.5 kb in both species. Reverse transcriptase polymerase chain reaction analysis showed that hSMP-1 mRNA was present in human testis but was not in either kidney or liver. When the cDNA was expressed in Escherichia coli under the control of the T7 promoter, the expressed protein accumulated to a level of about 50% of the total cellular protein. The expressed protein, which contained an N-terminal poly(his) sequence tag, was purified by chromatography on an nitrilo-tri-acetic acid affinity resin. Approximately 10 mg of pure protein was obtained from a 500-ml culture, purified, and used as antigen to generate a polyclonal antiserum in rabbits. Western blot analysis of human sperm extracts showed a single specific band at 55.5 kDa. Immunofluorescence data showed that hSMP-1 was localized to the head of human sperm. The fluorescent staining formed a cap-shaped pattern that was similar in morphology to the human sperm acrosome. The availability of large amounts of recombinant hSMP-1 and its antiserum will facilitate studies on the function and expression of the protein during spermatogenesis and the assessment of its potential value as a contraceptive immunogen.

Adhesion molecules in human sperm-oocyte interaction: relevance to infertility.

Fertilization involves cell-cell fusion of a sperm with the oocyte. This fusion restores the diploid genome, activates the oocyte, and initiates embryonic development. The identification of proteins mediating the fusion of sperm with oocyte plasma membrane (oolemma) is important to a deeper knowledge of fertilization. Defects in sperm-oocyte fusion may account for some form of human infertility. The hypothesis that sperm plasma membrane and oolemma carry complementary molecules involved in multistep fusion process has been validated by studies of cell adhesion molecules (integrins) in sperm-oocyte interaction in a number of animal models and human in vitro fertilization assays. Integrins or integrin-like molecules and complement proteins present on the surface of mammalian gametes, might be involved in the interaction between oocyte and sperm at fertilization. This review will provide an overview of the interaction of human sperm membrane with the oolemma, the nature of cell adhesion molecules, their expression profiles and their possible involvement in adhesive and fusogenic events in human fertilization. Unraveling the unique molecules involved in human sperm plasma membrane-oolemma fusion will be an important component for the development of a new set of contraceptive vaccines.

The presence of water channel proteins in ram and human sperm membranes.

Ram and human spermatozoa have a high coefficient of osmotic water permeability (Pf) with a low activation energy (Ea), suggesting the presence of water channels within their plasma membranes. Sperm membranes were examined for the presence of two known water channel proteins, CHIP28 and glucose transporters belonging to the GLUT family of proteins. The water permeability of ram spermatozoa was not inhibited by mercuric chloride to which the CHIP28 channel is sensitive. The CHIP28 protein was not located in western blots of ram sperm membrane preparations that used an anti-CHIP28 antibody. The water permeability of ram and human spermatozoa was inhibited in the presence of phloretin, an inhibitor of glucose transport. Rabbit spermatozoa, which have a low Pf and a high Ea value, suggesting a non-porous membrane, were unaffected by phloretin. These results indicate that the erythrocyte and proximal tubule water channel, CHIP28, is not present in sperm membranes but that sperm membrane glucose transporters may have a secondary water channel function.

Thymosin alpha-1 and FA-1 monoclonal antibody affect fertilizing capacity of human sperm by modulating protein phophorylation pattern

The present investigation was conducted to investigate the modulation of phosphorylation pattern of human sperm membrane proteins during capacitation by thymosin α-1 (Tα1) (which enhanced sperm penetration index) and anti-FA-1 monoclonal antibody (anti-FA-1 mAb) (which completely blocked sperm penetration) using 32P metabolic labeling, in vitro kinase assay and Western immunoblot analysis. In 32P metabolic labeling experiments, Tα1 (0.25 and 0.5 μg/100 μl) enhanced phosphorylation of 7 proteins in four molecular regions namely one protein (190 kDa) in 200-kDa, two proteins (112 and 104 kDa) in 97-kDa, two proteins (48 and 42 kDa) in 43-kDa and two proteins (31 and 25 kDa) in 29-kDa molecular regions, respectively. Anti-FA-1 mAb (10 μg/100 μl) resulted in a general decrease in the 32P labeling of these sperm proteins. In in vitro kinase assay using non-capacitated sperm extracts, Tα1 (0.5 μg/100 μl) enhanced autophosphorylation of 14 proteins in various molecular regions (122, 105, 95, 89, 73, 62, 48, 46, 40, 33, 30, 28, 25 and 22 kDa, respectively). The same concentration of Tα1 did not affect autophosphorylation of proteins in capacitated sperm extract. Anti-FA-1 mAb (10 μg/100 μl) inhibited autophosphorylation of a subset of 8 proteins (122, 104, 95, 89, 73, 62, 48 and 46 kDa, respectively) in non-capacitated sperm membrane extracts, and 12 proteins (112, 104, 95, 89, 73, 62, 48, 46, 33, 30, 28 and 25 kDa, respectively) in capacitated sperm membrane extracts. In the Western immunoblot analysis, Tα1 resulted in a concentration-dependent increase in tyrosine phosphorylation of two proteins (95 and 51 kDa) during capacitation of human sperm, whereas anti-FA-1 mAb inhibited tyrosine phosphorylation of both proteins. These results indicate that Tα1 and anti-FA-1 mAb affect the fertilizing capacity of human sperm by modulating phosphorylation of proteins especially tyrosine phosphorylation of 95- and 51-kDa proteins during capacitation. These findings also suggest that there may be a signal transduction pathway(s) involved in phosphorylation of membrane proteins during capacitation and that an exogenous stimulus affecting a single membrane protein component can modulate phosphorylation of all the relevant proteins involved in capacitation/acrosome reaction of human sperm.

High Water Permeability of Human Spermatozoa is Mercury-Resistant and not Mediated by CHIP281

A novel integral membrane protein with an apparent molecular mass of 28 kDa (CHIP28) was first isolated from human erythrocytes and is now recognized as a water channel protein. The expression of this protein has been found in several other cell types that all require high water permeability for their functions. Recent studies have shown that the water permeability (Lp) of human spermatozoa is among the highest reported for mammalian cells. Together with the low activation energy of human spermatozoa for Lp, this suggests that CHIP28 water channel may be present in the plasma membrane of human spermatozoa. However, our current studies do not support this hypothesis. Results from Western blot analysis on human sperm plasma membrane proteins, performed through use of an antibody against human erythrocyte CHIP28 protein, indicated that human spermatozoa do not express CHIP28 protein on their cell surface (n = 10). Consistent with the Western blot finding, mercuric chloride (HgCl2), a known water channel blocker, failed to reduce the osmotic water permeability of human spermatozoa. The calculated Lp values were 1.30 +/- 0.29 micron/min/atm (n = 16; mean +/- SEM) for the control group and 1.31 +/- 0.29 (n = 9; mean +/- SEM), 1.04 +/- 0.27 (n = 11; mean +/- SEM), and 1.34 +/- 0.19 (n = 6; mean +/- SEM), respectively, for the 10 microM, 30 microM, and 50 microM HgCl2-treated groups. These Lp values are not different (p > 0.05). In contrast, the same concentration of HgCl2 significantly blocked the osmotic water transport across the membrane of human erythrocytes.(ABSTRACT TRUNCATED AT 250 WORDS)

Electron microscopic study of human sperm membrane isolation

Our purpose was to isolated pure, homogeneous human sperm membranes, free of cellular contaminants. Donor semen samples collected after masturbation were stored at -70 degrees C and eventually pooled. Each attempt at sperm membrane isolation required 800 x 10(6) spermatozoa which were sonicated by ultrasound (40% output; Vibra Cell). The effect of sonication time (3 x 5, 3 x 15, and 180 sec) on membrane isolation was investigated. Sonicated samples were centrifuged (500 g, 5 min) and the supernatant was pipetted off. The supernatant of the centrifuged sample was layered on either a sucrose cushion (supernatant on 1.6 M sucrose) or a discontinuous sucrose gradient and centrifuged (100,000 g, 1 hr). Contents of supernatants of sonicated samples and fractions (sucrose interfaces) were then fixed in 1.0% tannic acid and 2.5% buffered glutaraldehyde and examined electron microscopically using standard procedures. (1) The optimal sonification time was found to be 3 x 15 sec. (2) Membrane isolation using a sucrose cushion was found to be inadequate, showing significant cellular contamination. (3) Sperm membrane isolation from the sucrose interface between 0.75 and 1.05 M sucrose was found to be most effective. The advantage of this method is its simplicity. The drawback of this method is the large number of spermatozoa required for membrane purification.

Steroid receptors in human spermatozoa.

The presence of steroid binding sites in (or on) human spermatozoa was first suggested in the late 1970s, by studies showing that some steroids were able to influence sperm function. Subsequently, several effects exerted on spermatozoa by biological fluids, such as follicular fluid, were found to be probably linked to the action of steroids, and among them progesterone. Since the effects of progesterone on spermatozoa were rapid, dose-dependent and not affected by progesterone conjugation with high molecular weight proteins unable to cross the plasma membrane, the existence of a novel class of non-genomic progesterone receptors was strongly suspected. This hypothesis was further supported by the observation that some of the effects of progesterone on human spermatozoa were not abolished by inhibitors of the classical progesterone nuclear receptors, nor mimicked by progesterone genomic receptor agonists. Recently, surface progesterone binding sites were directly identified on the membranes of human spermatozoa, and their mechanism of action partially characterized.

Osmotic Effects on Ram and Human Sperm Membranes in Relation to Thawing Injury

The effects of hyposmotic and hyperosmotic stresses on ram and human spermatozoa were examined. Human spermatozoa exhibited a precipitate decline in survival at osmolalities below 90 mOsm caused by cells swelling beyond their maximum volume-to-surface area ratio and lysing. Ram spermatozoa exhibited a progressive decline in cell survival at relatively small hyposmotic stresses before exceeding their maximum volume-to-surface area ratio; this prelytic cell loss could be prevented by decreasing the osmolality in a series of 25-mOsm steps. Repeated hyposmotic stress experiments indicated that cells sensitive to prelytic damage constitute a discrete subpopulation within the ram ejaculate. Spermatozoa of both species were apparently resistant to hyperosmotic stresses; human spermatozoa maintained membrane integrity when subject to stresses up to 2.5 Osm and ram spermatozoa up to 1 Osm. However, ram spermatozoa suffered an almost complete and irreversible loss of motility above 600 mOsm. Spermatozoa of both species exposed to hyperosmotic stress and returned to isosmotic conditions exhibited significant cell damage, although ram spermatozoa were the more vulnerable. These observations are related to cryopreservation of spermatozoa.

Inhibition by anti-sperm monoclonal antibodies of the penetration of zona-free hamster oocytes by human spermatozoa.

Fourteen mAb specific for human sperm membrane antigens were selected to investigate their inhibitory effect on fertilization. The antigens were not expressed in other human somatic tissues but were present in sperm from other species. The antibodies were purified from ascites fluid produced in mice. The zona-free hamster egg penetration assay was used for the evaluation of the blocking ability of these antibodies. The inhibition rate was generally related to a decrease in the number of adherent sperm. Three groups of antibodies were distinguished: (i) four mAb that have high inhibition at any concentration; (ii) four mAb with an intermediate inhibitory effect, that is more dependent on the antibody concentration tested; and (iii) six mAb with little or no effect at any concentration. The presence of antibodies leads to a lower penetration index, or number of penetrating sperm per oocyte. MAb specific for head antigens promote high inhibition of fertilization; these antibodies show a patchy staining on the sperm head. The antibodies localized in the midpiece have an intermediate inhibitory effect. No inhibition is detected with the equatorial region binding pattern. Sperm agglutination does not play any role in the inhibition caused by the antibodies described here.

Localization of wheat germ agglutinin lectin receptors on human sperm by fluorescence microscopy: utilization of different fixatives.

Wheat germ agglutinin (WGA) and Concanavalin A (Con A) seem to be the most widely studied of the lectins used to investigate sperm surface receptors. It has been reported that WGA receptors on human sperm membranes may be closely related to male fertility and that WGA receptor deficiency may cause human infertility. Although these claims may not be negated, there are many discrepancies in current literature regarding the localization and detection of human sperm WGA receptors. These discrepancies are probably the result of variations in the pretreatment of sperm before fixation and sperm-lectin interaction. This study investigated the effect of different fixation procedures on the localization of human sperm WGA receptors on normal spermatozoa. Comparative studies were performed on spermatozoa separated by wash-swim-up and fixed with 4% formaldehyde, 95% ethanol, or Bouin's solution, or nonfixed (air-dried) using fluorescein isothiocyanate-conjugated WGA. There were significant differences (p < .01) in the % WGA receptor localization following the different fixation procedures on samples receiving the same treatment previous to fixation. The different fixatives affect the plasma membrane to different degrees, resulting in an alteration of lectin receptor localization. To allow a better comparison and understanding of results, a standardized preparation procedure is recommended during studies using lectins as molecular probes. The results suggest than an aldehyde fixative may be the fixative of choice for sperm membrane studies.

Abnormal Physical Architecture of the Lipophilic Domains of Human Sperm Membrane in Oligospermia: A Logical Cause for Low Fertility Profiles

Three different domains of human sperm membrane, viz., the hydrophobic domains, the aqueous compartments and the surface proteins were probed to evaluate the molecular dynamics in these microenvironments in normal as well as oligospermic cases. Decreased rotational motion of the stearic spinlabels, designated rigid lipid matrix affording hindrance to the protein rotation in spermatozoa of oligospermic patients. The intrinsic tight anchoring of thiol containg proteins led to their increased ordering on the spermatozoa of such individuals. Fluorescence polarization studies revealed weaker hydrophobicity around these sperm proteins in oligospermia. Thus, this study identifies abnormal molecular mobility in the limiting membranes of oligospermic cells.

Fucoidin binding activity and its localization on human spermatozoa

We previously reported that fucoidin (a polymer of predominantly sulfated L-fucose) significantly inhibits: (1) tight binding of human sperm to human zona pellucida in vitro and (2) stimulation of the acrosome reaction by acid solubilized human zona pellucida. Here, we determined fucoidin binding activity on human spermatozoa and its localization on both live and permeabilized human sperm populations. A typical binding curve was demonstrated with biotinylated fucoidin. In competitive inhibition assays with unlabelled fucoidin or human sperm membrane extracts, IC, 50's were 4.0 μg/ml and 31.4 μg/ml, respectively. Fucoidin binding was localized over the acrosomal region of methanol-fixed human sperm and this pattern of binding significantly decreased from 92 ± 3% to 74 ± 6% with calcium ionophore A23187 treatment (p < 0.01). Binding of fucoidin-coated beads to live (non-permeabilized) human sperm was less than 1%. Addition of the detergent, Triton-X, to permeabilize sperm membranes resulted in a significant increase in binding (p = 0.001). These results provide evidence for the presence of a fucoidin binding compound in human spermatozoa that is localized to the membranes of the acrosomal region and can be extracted by a mild detergent extraction. Absence of binding by fucoidin to intact but not permeabilized spermatozoa suggests that the heteropolysaccharide

Adenosine stimulates human sperm motility via A2 receptors.

The effects of adenosine and its analogues on human sperm motility were studied using a transmembrane migration method. Specific binding sites for adenosine in human sperm were also investigated. Adenosine and 5'-N-ethylcarboxamidoadenosine (NECA) stimulated human sperm motility with similar efficacies and the maximal amplitudes of motility increases were both about 70%. 3,7-Dimethyl-1-propargylxanthine (DMPX), a potent A2 antagonist, competitively antagonized NECA-induced motility stimulation. Successively higher concentrations of DMPX shifted the dose-response curve of NECA to the right in a nearly parallel fashion. Dipyridamole, an inhibitor of adenosine uptake, does not reduce the ability of adenosine to stimulate human sperm motility. In radioligand-binding studies, adenosine A1 selective analogues, cyclopentyl-1,3-dipropylxanthine and 1-methyl-2-phenylethyl adenosine, have little competitive effect on [3H]NECA binding in human sperm membrane. These results provide evidence that adenosine enhances human sperm motility via adenosine A2 receptors on the surface of sperm membranes.

Identification and characterization of membrane cofactor protein of human spermatozoa.

Membrane cofactor protein (MCP) regulates C activation by serving as a cofactor for the cleavage of C3b and C4b by the serine protease factor I. An MCP-like molecule on the inner acrosomal membrane of human spermatozoa has been characterized. Three mAb and a rabbit polyclonal antibody against MCP recognized the sperm protein. On SDS-PAGE, it migrated as a single band with a molecular mass of 38,000 and 44,000 Da under nonreducing or reducing conditions, respectively. The molecular mass was 10,000 to 20,000 Da less than the two forms of MCP expressed on others cells. The electrophoretic pattern, by one- and two-dimensional gel analysis, and the isoelectric point profile (4.5 to 5.0) of the sperm protein were similar among multiple individuals. In contrast to MCP of other cells, digestion with endoglycosidases did not alter either the m.w. or the pI of the protein, suggesting that it is a poorly or nonglycosylated form of MCP. The solubilized sperm protein bound C3 with broken thioester bond to Sepharose and possessed cofactor activity for factor I-mediated cleavage of C3 with the broken bond. A mAb that blocks the regulatory function of MCP inhibited the cofactor activity of the sperm lysate. Thus, the sperm protein is an antigenic and functional homologue of MCP but has the distinct structural features of a lower m.w. and an apparent lack of glycosylation. MCP may play an essential role in the survival of the acrosome-reacted spermatozoa by modulating C activation in the female genital tract.

Reappraisal of the hypo-osmotic swelling test to improve assessment of seminal fertility status.

The hypo-osmotic swelling (HOS) test has been proposed as a useful assay for evaluation of the functional competence of the human sperm membranes. To assess this further, the HOS-test was evaluated in 187 semen samples collected from fertile men and from male patients consulting for infertility. These samples were classified as normal, oligo-, astheno- or oligoasthenozoospermic on the basis of their standard semen variables. The percentage of total sperm tail swelling and of sperm exhibiting different tail swelling patterns was recorded. In the fertile men and in the group of patients with normal semen variables, significantly more (P less than 0.001) HOS-reactive sperm were observed after hypo-osmotic treatment in comparison with those groups exhibiting abnormal semen parameters. Swelling of the sperm in a hypo-osmotic medium was highly correlated with both progressive motility (r = 0.62, P less than 0.001) and sperm viability (r = 0.65, P less than 0.001). A weak positive correlation was also observed between sperm swelling and sperm morphological features (r = 0.31, P less than 0.005) and between sperm swelling and sperm concentration (r = 0.31, P less than 0.005). No significant correlation was observed between sperm swelling and in-vitro sperm fertilizing capacity as assessed by the zona-free hamster oocyte penetration assay. However the majority of the semen samples (87.3%) showing a normal penetration rate (greater than or equal to 10%) also exhibited a 60% (or higher) reaction in the HOS-test.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of radioiodination, in the presence of iodogen, on human spermatozoa.

In order to study human sperm membrane proteins, a radioiodination technique using iodogen (Pierce, Interchim, Paris, France) has been tested. Iodogen was dissolved in redistilled chloroform and added in 50-μg quantities to nitric-acid-washed pyrex tubes. The chloroform was evaporated under a jet of nitrogen and the tubes rinsed with PBS shortly before use. The spermatozoa were obtained by a swim-up method in Locke buffer (0.15 M NaCl; 5.6 mM KC1; 24 mM NaHCO3; 0.25% dextrose; 2.25 mM CaCl2). Mobility of migrated sperm, evaluated in optic microscopy, was ≥85%. Migrated cells were then surface-labeled by adding 300 μCi carrier-free Na125 I (Amersham France, Les Ulis, France) to 1 ml cell suspension (50 × 107 cells) in the iodogen-coated tubes. After incubation at room temperature for 10 min, the reaction was terminated by diluting the spermatozoal suspensions with large volumes of PBS. The spermatozoa were then washed three times with PBS.

Staining of the inner acrosomal membrane of human spermatozoa with concanavalin A lectin as an indicator of potential egg penetration ability

To determine the sensitivity and functional significance of the fluorescein isothiocyanate concanavalin A (FITC-ConA) staining method of assessment of acrosomal status. Treatments were assessed for their ability to induce human sperm acrosomal loss. Penetration of zona-free human eggs by treated spermatozoa was subsequently determined. Zona-free human eggs were obtained from the Infertility Medical Centre, Richmond, Victoria, Australia. None. None. Acrosomal loss of human spermatozoa was determined at 4 hours and 10 hours of treatment incubation. Penetration of zona-free human eggs was assessed 16 hours after reinsemination. Human spermatozoa incubated in a strontium- or lanthanum-based medium, or T6 + 10% maternal human serum (HS) supplemented with 12 mM 8-bromo cyclic guanosine 3,5'-monophosphate and 10 mM imidazole for a 4-hour period before transfer to fresh T6 + 10% HS for a further 6 hours, demonstrated a significant increase (P less than 0.05) in acrosomal loss compared with T6 + 10% HS for a total 10-hour incubation. This increase in acrosomal loss with test treatments correlated with an increase in the development of pronuclei of zona-free human eggs (r = +0.98). The FITC-ConA staining procedure therefore reflects biological function as assessed by the penetration of zona-free human eggs and consequently provides a further research tool for the investigation of the human sperm acrosome reaction.

Hypoosmotic Swelling Test in Patients with Varicocele

The hypoosmotic swelling test is a simple test for measuring the functional competence of human sperm membrane. Thirty-five patients with varicocele were assessed by hypoosmotic swelling test and the results were compared with those in patients with idiopathic infertility. Percentage of swollen sperm in varicocele tended to be lower than that in idiopathic infertility, especially in sperm concentration between 30 x 10(6)/mL and 40 x 10(6)/mL. Varicocele with sperm motility of 40% or more revealed lower percentage of swollen sperm than idiopathic infertility did. The results indicate that varicocele may particularly injure sperm membrane function because hypoosmotic swelling test is a test for measuring functional ability of human sperm membrane. The percentage of swollen sperm recovered after the operation earlier than did the other sperm parameters such as sperm concentration or motility, which indicates that impaired sperm membrane injury by varicocele is reversible.