1327

BIOLOGY OF REPRODUCTION 67, 1327–1336 (2002)DOI 10.1095/biolreprod.102.003830

Progesterone Receptor as an Indicator of Sperm Function

Sushama Gadkar,2 C.A. Shah,2 Geetanjali Sachdeva,2 Urmila Samant,3 and C.P. Puri1,2

Institute for Research in Reproduction,2 Indian Council of Medical Research, Parel, Mumbai 400 012, IndiaTata Memorial Hospital,3 Parel, Mumbai 400 012, India

ABSTRACT

Expression of progesterone receptor (PR) localization onspermatozoa was determined in men with normal and abnormalspermiograms. Studies were also carried out to evaluate the po-tential of PR as a marker of sperm function. Progesterone re-ceptor expression on spermatozoa from men with normozoo-spermia (n 5 8), oligozoospermia (n 5 7), asthenozoospermia(n 5 8), oligoasthenozoospermia (n 5 7), and teratozoospermia(n 5 11) was analyzed using an immunocytochemical methodwith monoclonal antibodies against PR, and flow cytometry us-ing a cell-impermeable fluorescein-tagged progesterone coupledto BSA complex (P-FITC-BSA). Both methods revealed signifi-cantly fewer (P , 0.05) PR-positive spermatozoa in men witholigozoospermia, asthenozoospermia, oligoasthenozoospermia,and teratozoospermia compared with men with normozoosper-mia, thereby suggesting that down-regulation of PR expressionin spermatozoa may be one of the causes of male infertility.Spermatozoa from men with normozoospermia (n 5 12), oligo-zoospermia (n 5 12), asthenozoospermia (n 5 12), oligoasthen-ozoospermia (n 5 9), and teratozoospermia (n 5 10) were ex-posed to low osmotic conditions in the hypoosmotic swelling(HOS) test and then analyzed for PR expression using P-FITC-BSA complex. A significantly higher percentage (P , 0.05) ofspermatozoa with physiologically active plasma membrane(HOS1) lacked PR expression (HOS1PR2) in all categories ofmen with infertility, thereby suggesting that compared to theHOS test, PR expression is a better indicator of sperm function.Furthermore, PR expression in spermatozoa showed a strong (P, 0.05) positive correlation with their ability to undergo an invitro acrosome reaction. This was observed in all study groups(i.e., normozoospermia, r 5 0.8545; oligozoospermia, r 50.8711; asthenozoospermia, r 5 0.7645; oligoasthenozoosper-mia, r 5 0.9003; and teratozoospermia, r 5 0.8676). This sug-gests a potential role for PR in the events leading to the acro-some reaction in sperm.

acrosome reaction, progesterone receptor

INTRODUCTION

A reliable tool to assess the fertilizing potential of hu-man spermatozoa is not yet available. Therefore, it is notsurprising that the causes of infertility remain undetectablein a large percentage of men. Routine semen analyses thatinclude estimation of sperm concentration, motility, andmorphology have limitations as fertility indicators [1–3].These analyses lack the potential to ascertain the functionalcapacity of spermatozoa, and they cannot predict the pos-sible occurrence of both in vivo and in vitro conception [4–

1Correspondence. FAX: 91 22 4139412, 4964853;e-mail: vichin@vsnl.com and dirirr@vsnl.com

Received: 29 January 2002.First decision: 15 February 2002.Accepted: 28 May 2002.Q 2002 by the Society for the Study of Reproduction, Inc.ISSN: 0006-3363. http://www.biolreprod.org

6]. In the last few years, efforts have been directed towardthe development of various functional tests such as thezona-free hamster oocyte sperm penetration test [7, 8], thehypoosmotic swelling (HOS) test [9], and the acrosome re-action (AR) [10] to assess the sperm’s functional capacity.

The zona-free hamster oocyte sperm penetration test[11–13] has been developed as an indicator of the abilityof human spermatozoa to capacitate, acrosome react, andfuse with the vitelline membrane of the oocyte [14, 15].However, some reports have suggested that assessment ofsperm fertilizing capacity using this test does not have sig-nificant clinical benefit in diagnosing infertility [7, 16].Dead or immotile spermatozoa also penetrate zona-freehamster eggs under certain conditions [15, 17]. Further-more, this assay is time-consuming, complex, and subjectto many variables such as capacitation time [18], spermconcentration and incubation conditions [19], medium com-position [15], and the age of the ovum after ovulation [20].

The HOS test can predict the fertilizing capacity of hu-man spermatozoa [9, 21, 22]. A positive HOS test is anindicator of the functional integrity of the sperm membraneas judged by a swollen head and curled tail of spermatozoawhen exposed to hypo-osmotic conditions [9]. A decreasein the number of HOS-positive forms in the ejaculates ofmen with oligozoospermia, asthenozoospermia, and oli-goasthenozoospermia compared with men with normozoo-spermia has been reported [23]. Positive correlation wasalso observed between HOS and the fertilizing ability ofsperm [21, 22, 24–26], however, others have not found asignificant correlation between HOS and in vitro fertiliza-tion (IVF) outcome [27–30]. Thus, the clinical utility of theHOS test in assessment of sperm function is still debatable.

The AR, which is one of the prerequisites for successfulfertilization, seems to be a more relevant parameter for as-sessing sperm function [2, 7, 31–33]. AR is an irreversible,exocytotic, postcapacitational event that involves fusionand fenestration of the outer acrosomal membrane with theplasma membrane [10]. It has been demonstrated that pro-gesterone facilitates the AR [34–37]. Studies have alsobeen carried out to determine whether an assessment of asperm’s responsiveness to progesterone may predict its fer-tilizing ability in vitro [38–44]. A significant correlation hasbeen found between the outcome of IVF and progesterone-stimulated Ca21 influx [45, 46]. A physiological role forprogesterone in AR has also been suggested by other re-ports demonstrating a relationship between the ability ofsperm to respond in vitro to progesterone and male infer-tility [47–49].

The mechanism by which progesterone elicits an AR andsubsequent events probably involves its interaction with acell surface receptor on spermatozoa [50–52]. The presenceof membrane receptors that specifically bind to progester-one has been demonstrated on human spermatozoa [51, 53,

1328 GADKAR ET AL.

54], and there is evidence to suggest that blocking thesesurface receptors inhibits progesterone-induced AR [55,56]. This prompted us to investigate the possibility of anaberration in the expression of progesterone receptor (PR)on spermatozoa in men with an abnormal spermiogram andwhether it has any effect on sperm functions (i.e., the AR).Studies were also conducted to compare the predictive val-ue of PR with that of other tests such as HOS for spermfunction.

This is the first report to assess the characteristics ofHOS and positive PR in the same sperm samples. To ourknowledge, no such attempt has been made until now toanalyze the expression of a surface protein of potentialfunctional relevance on spermatozoa, which is categorizedon the basis of membrane activity. This is also the firststudy in which PR expression in spermatozoa from fertileand infertile groups of men was evaluated by both immu-nocytochemistry and flow cytometric analysis.

MATERIALS AND METHODS

Samples

Semen samples were collected by masturbation after 36–48 h of sexualabstinence from healthy and proven fertile (n 5 20) volunteers. Thesefertile men had fathered a child within the last 12 mo. Semen sampleswere collected from infertile (n 5 76) men attending the infertility clinicat Wadia Maternity Hospital, Mumbai. The wives of the volunteers hadnormal hormonal levels, menstrual cycle length, ovulation, and endome-trial function. Informed consent was obtained from each subject beforesamples were collected.

Samples were classified in accordance with World Health Organizationguidelines [6] by a trained andrologist. They were categorized either asnormozoospermic (n 5 20; count $20 million/ml, progressive motility$50%, morphology $30% normal forms), oligozoospermic (n 5 19;count ,20 million/ml, progressive motility $50%, morphology $30%normal forms), asthenozoospermic (n 5 20; count $20 million/ml, pro-gressive motility ,50%, morphology $30%), oligoasthenozoospermic (n5 16; count ,20 million/ml, progressive motility ,50%, morphology$30% normal forms), or teratozoospermic (n 5 21; count $20 million/ml, progressive motility $50%, morphology ,30% normal forms).

Spermatozoa from men with normozoospermia (n 5 8), oligozoosper-mia (n 5 7), asthenozoospermia (n 5 8), oligoasthenozoospermia (n 57), and teratozoospermia (n 5 11) were analyzed for PR expression byflow cytometric and immunocytochemical methods. Spermatozoa frommen with normozoospermia (n 5 12), oligozoospermia (n 5 12), asthen-ozoospermia (n 5 12), oligoasthenozoospermia (n 5 9), and teratozoo-spermia (n 5 10) were used in the HOS test followed by PR localizationand AR analysis.

Each ejaculate was allowed to liquefy at room temperature for 30 min,washed twice with 13 PBS pH 7.4, centrifuged at 980 3 g for 10 min,and assayed by all three techniques.

Collection of Follicular Fluid

Follicular fluid was obtained by aspiration of antral follicles fromwomen undergoing laparoscopy for IVF and embryo transfer. Follicularfluids were pooled and centrifuged at 2800 3 g at 48C to remove cellulardebris, and progesterone content was measured by radioimmunoassay [57].Follicular fluid was then aliquoted and stored at 2208C until used.

Preparation of Stock Solutions

Progesterone 3-(O carboxymethyl) oxime:BSA conjugate (5–10 mol ofprogesterone per mole of BSA) labeled with fluorescein isothiocyanate(FITC; P-FITC-BSA; Sigma, St. Louis, MO) was stripped of free proges-terone using dextran-coated charcoal [58]. FITC was conjugated with BSAaccording to the method described by Goding [59]. The conjugate solutionwas lyophilized and reconstituted in PBS to obtain a solution of 1 mg/ml.FITC-conjugated Pisum sativum lectin (Sigma) was dissolved in PBS at1 mg/ml and stored frozen (2208C) until used.

ImmunocytochemistryApproximately 1 3 106 spermatozoa in 10 ml of PBS were smeared

on clean, grease-free slides, air-dried, and fixed in chilled methanol for 30min. The slides were washed twice with 13 PBS and once with PBS-T(0.5% Tween 20 in 13 PBS) for 10 min at room temperature. After wash-ing, the sperm membrane was permeabilized with 0.1% sodium deoxy-cholate in 13 PBS at 48C for 30 min. Smears were washed with PBS-Tfor 5 min, and blocked in 1% BSA in 13 PBS at room temperature for20–25 min. Slides were incubated with mouse antisera (Affinity Biore-agents, Golden, CO) against the carboxyl terminus of PR (clone PR-AT4.14) diluted 1:100 in 13 PBS at 48C for 18–20 h. The slides werewashed with PBS-T for 5 min and then incubated with biotin-conjugatedgoat anti-mouse secondary antibody diluted 1:500 in 13 PBS for 1 h atroom temperature. The slides were washed twice with 13 PBS and in-cubated with avidin-biotin complex (Vectastain kit, Vector Laboratories,Burlingame, CA) for 30 min at room temperature. The slides were thenwashed twice with 13 PBS at room temperature for 30 min each, followedby treatment with 0.05% diaminobenzidine in 13 PBS with 0.06% H2O2for 10 min. The slides were counterstained with 1% hematoxylin for 2min, dehydrated in ascending grades of alcohol for 10 min each, and keptin xylene for 2 h followed by mounting in DPX. Specificity of the stainingwas evaluated by running a negative control (1% BSA without antibody).

The slides were examined using the 1003 objective of a brightfieldmicroscope (Olympus, Tokyo, Japan) with a double-blind technique. Atleast 200 spermatozoa were counted in 20 microscopic fields selected atrandom. Counterstained spermatozoa showing a brown color at the headregion were counted as PR-positive. Spermatozoa that did not react withthe antibodies against PR had no brown precipitates and appeared bluethroughout due to counterstaining and were counted as PR-negative.

Direct FluorescenceFour million sperm were incubated with 50 ml of 0.1% digitonin in

PBS at 48C for 30 min. These samples were washed twice with 13 PBSand centrifuged at 1000 3 g for 10 min. Washed pellets were incubatedwith either 0.1 mM P-FITC-BSA or FITC-BSA in 50 ml of PBS at 48Cfor 16–18 h. Unbound fluorescein-labeled compounds were removed bywashing with 13 PBS and centrifuged at 1000 3 g. The pellets weresuspended in 1 ml of 13 PBS. Ten microliters of each cell suspensionwas processed for direct fluorescence analysis and the remaining for flowcytometry.

Ten microliters of cell suspension was smeared on a clean, grease-freeslide, mounted, and visualized with a 1003 objective of the fluorescencemicroscope (Olympus) using a triple-band filter (Cube U-MNIBA, DM-505, BP-470-550, and BA-515-550). Positive staining following incuba-tion with P-FITC-BSA resulted in a green color at the acrosomal region,whereas negative staining as well as incubation of spermatozoa with FITC-BSA resulted in a red color.

Flow Cytometric AnalysisApproximately 4 million plain or digitonin-treated spermatozoa were

incubated with P-FITC-BSA or FITC-BSA as mentioned in the previoussection. The samples were washed and suspended in 1 ml of PBS. Thesesamples were analyzed using a FACScan cytometer (Becton Dickinson,San Jose, CA) equipped with a 15 mW air-cooled 488 nm argon-ion laser.FL1 (FITC) signals were detected through a 530/30 nm bandpass filter. Todistinguish spermatozoa from debris, a dot plot distribution of spermatozoaaccording to a forward angle light scatter (correlating with cell size) andright angle light scatter (correlating with cell density) were used to deter-mine a ‘‘sperm window’’ as described by Haas and Cunningham [60]. Thefluorescence intensity of 10 000 spermatozoa within the sperm windowwas computed twice in list mode and analyzed using Lysis II software(Becton Dickinson).

Intensity of the fluorescence of stained cells was presented in log scaleon the X-axis and number of cells showing fluorescence on the Y-axis inflow cytograms. Channels M1, M2, and M3 represented low, moderate,and high intensity staining, respectively. Flow cytometry data for PR pos-itivity were compared with the data for PR positivity and evaluated withthe immunocytochemical method.

Localization of PR on Sperm Subjected to the HOS TestThe HOS test was performed on 0.1 ml of the original ejaculate mixed

with 1 ml of hypo-osmotic solution of sodium citrate and fructose (150mOsm/L) [9]. After incubation for 40 min at 378C, 200 spermatozoa were

1329SPERM PROGESTERONE RECEPTOR IN FERTILE AND INFERTILE MEN

FIG. 1. Detection of PR using antibodiesagainst conventional PR in spermatozoafrom men with normozoospermia (b), oli-gozoospermia (c), asthenozoospermia (d),oligoasthenozoospermia (e), and terato-zoospermia (f). Negative control (no anti-body) is shown in a. Brown color indicatespositive staining for PR. Blue color appearsbecause of counterstaining with hematoxy-lin. Spermatozoa were visualized with an3100 oil immersion objective.

observed at a magnification of 4003 with a phase contrast microscope.The number of spermatozoa with head and tail changes typical of swelling(positive for hypoosmotic swelling) was counted [9]. Spermatozoa sub-jected to HOS were pelleted by centrifugation at 1000 3 g for 5 min andsmeared on clean, dry, grease-free slides. Smears were fixed in methanolfor 10 min at 48C. Smears were treated with 0.1% digitonin at 48C for 30min and washed three times with 13 PBS. The spermatozoa were incu-bated with 0.1mM P-FITC-BSA or FITC-BSA at 48C for 16 h and washedthree times with 13 PBS to remove unbound stain. The smears were thenmounted with glycerol in PBS (1:9). Stained slides were observed usinga 1003 objective of a fluorescence microscope (Olympus). Spermatozoasubjected to HOS showing green color at the acrosomal or equatorial re-gion were counted as PR-positive spermatozoa. PR staining was assessedin both HOS-positive and HOS-negative spermatozoa from fertile and in-fertile men, respectively. This evaluation of PR positivity was made bytwo different observers in a blinded manner. For quantitative assessmentof PR at least 200 spermatozoa were examined in randomly selected 20fields.

Acrosome ReactionSemen samples were suspended in Biggers Whitten Whittingham

(BWW) medium [61] and separated from seminal plasma by centrifugationat 1000 3 g at room temperature. Washed sperm (1–2 3 106) were over-laid with 1 ml of BWW medium supplemented with 3.5% human serumalbumin (HSA-BWW) and incubated at 378C in 5% CO2 and 95% air for6 h to allow capacitation [62]. Capacitated spermatozoa were divided intotwo equal aliquots and centrifuged at 1000 3 g. The acrosome reactionwas induced by incubating the capacitated spermatozoa in 20 ml of 20%follicular fluid in HSA-BWW for 20 min at 378C in 5% CO2 and 95% air.In another aliquot, 20 ml of HSA-BWW instead of follicular fluid wasused to take into account the number of spermatozoa undergoing a spon-taneous AR or an unstimulated AR. The sperm cells were then separatedfrom the medium by mild centrifugation at 1000 3 g to visualize the AR.

Visualization of Acrosomal StatusSeparated sperm cells were diluted (approximately 1 3 106/ml) in

BWW medium and smeared on clean, dry, grease-free slides. The smearswere fixed with methanol for 10 min, washed with PBS, and allowed toreact with FITC-labeled Pisum sativum lectin (1 mg/ml) for 16–18 h at48C. Excess stain was removed by washing with PBS. The smears werethen mounted with glycerol in PBS (1:9) and examined under a fluores-cence microscope (Olympus) using an oil immersion objective.

Two hundred spermatozoa per sample were counted to determinewhether they were intact (stained acrosomal cap) or acrosome reacted(patchy and equatorially stained and acrosome lost). The percentage of ARwas expressed as the difference between stimulated and unstimulated(spontaneous) acrosome reacted sperm population.

Statistical AnalysisAll statistical analyses were performed using the Statistical Program

for the Social Sciences (SPSS; Chicago, IL), version 9. Percentages of PR-positive sperm evaluated by immunocytochemistry and flow cytometrywere compared with one-way ANOVA.

Percentages of HOS1PR1, HOS1PR2, HOS2PR2 and acrosome-reacted sperm populations in different study groups were also comparedvia one-way ANOVA. The coefficients of correlation between %HOS,%HOS1PR1, %HOS1PR2, and %AR were determined with SPSS ver-sion 9.

RESULTS

PR Expression on Spermatozoa

Immunocytochemical localization using antibodiesagainst the conventional PR demonstrated localization ofPR, as evidenced by brown staining at the acrosomal region

1330 GADKAR ET AL.

FIG. 2. Localization of PR using P-FITC-BSA on spermatozoa from men with nor-mozoospermia (b), oligozoospermia (c),asthenozoospermia (d), oligoasthenozoos-permia (e), and teratozoospermia (f). Neg-ative control (FITC-BSA) is shown in a.Green color indicates positive PR and redcolor indicates negative staining. Sperma-tozoa were visualized with an 3100 oilimmersion objective.

TABLE 1. Percentage of PR-positive spermatozoa in fertile and infertilegroups assessed by immunocytochemical localization and flow cytome-try.a

GroupsImmunocytochemical

analysisFlow cytometric

analysis

Normozoospermic(n 5 8) 83.99 6 1.8 83.69 6 2.04

Oligozoospermic(n 5 7) 48.7 6 4.51* 58.11 6 4.38*

Asthenozoospermic(n 5 8) 33.47 6 7.47* 47.75 6 8.44*

Oligoasthenozoospermic(n 5 7) 38.82 6 10.31* 46.27 6 7.97*

Teratozoospermic(n 5 11) 61.69 6 4.35* 61.95 6 7.07*

a Values are mean 6 SEM.* P , 0.05, significance of difference when compared with normozoos-permic group.

(Fig. 1). Aliquots of the samples processed for flow cyto-metric analysis were also subjected to direct fluorescenceusing fluorescein-tagged ligand (P-FITC-BSA). This alsodemonstrated a similar pattern of PR localization (i.e., atthe acrosomal region) in spermatozoa from both fertile andinfertile groups (Fig. 2).

Immunocytochemical and flow cytometric analysis re-vealed a significant decrease (P , 0.05) (Table 1) in thenumber of PR-positive spermatozoa in infertile groupscompared with those that were fertile (Figs. 1 and 3).

PR was also localized to spermatozoa with abnormalmorphology (Fig. 4). It was interesting that the number ofPR-positive spermatozoa was higher in men with terato-zoospermia than it was in men with other types of infertil-ity. Very weak positive correlation was observed betweenPR expression and the traditional semen parameters suchas motility and morphology (data not shown).

HOS Positivity in Spermatozoa

A significant decrease in the percentage of HOS-positivespermatozoa (P , 0.05) was observed in all infertile

1331SPERM PROGESTERONE RECEPTOR IN FERTILE AND INFERTILE MEN

FIG. 3. Flow cytometric analysis to quantitate the number of PR-positivespermatozoa in samples from men with normozoospermia (a), oligozoos-permia (b), asthenozoospermia (c), oligoasthenozoospermia (d), and ter-atozoospermia (e). Respective controls (samples stained with FITC-BSA)are shown in the left panel. Comparison of PR positivity for each categorywas done by comparing the fluorescence in M2 channel.

groups, except in the teratozoospermic group, comparedwith those that were fertile (Table 2).

PR Expression and HOS Positivity in Spermatozoa

Spermatozoa subjected to hypoosmotic conditions re-vealed three patterns of PR expression (Fig. 5): 1)HOS1PR1: HOS positive spermatozoa with a functionallyactive plasma membrane demonstrating PR expression atthe equatorial or acrosomal region. 2) HOS1PR2: HOS-positive spermatozoa that lacked PR expression. 3)HOS2PR2: Spermatozoa that lacked a functionally activemembrane as well as PR expression.

A significant decrease (P , 0.05) in HOS1PR1 and asignificant increase (P , 0.05) in HOS1PR2 spermatozoawas observed in all infertile groups compared with nor-mozoospermic men (Table 3; Fig. 6). However, no signifi-cant change was observed in the number of HOS2PR2spermatozoa in infertile men except in those with oligoas-thenozoospermia compared with fertile men.

PR Expression and AR in Spermatozoa

Significantly fewer (P , 0.05) spermatozoa underwentan in vitro AR in all infertile groups compared with thosein the normozoospermia group (Table 2).

A weak positive correlation was found between the per-centage of HOS-positive forms and the percentage of ARspermatozoa in all groups (normozoospermia, r 5 0.4982;oligozoospermia, r 5 0.4958; asthenozoospermia, r 50.4680; oligoasthenozoospermia, r 5 0.5800), whereas anegative correlation (r 5 20.2204) was observed in menwith teratozoospermia (Table 4). In contrast, a very strongpositive correlation was observed between the percentageof HOS1PR1 and the percentage of AR spermatozoa inmen with normozoospermia (r 5 0.8545), oligozoospermia(r 5 0.8711), asthenozoospermia (r 5 0.7645), oligoas-thenozoospermia (r 5 0.9003), and teratozoospermia (r 50.8676) (Table 4). No strong positive correlation was foundbetween the percentage of HOS1PR2 and the percentageof AR spermatozoa in any group (Table 4).

DISCUSSION

Culmination of the sperm-egg interaction into an eventof successful fertilization requires sperm to undergo theAR, which is an irreversible, exocytotic, postcapacitationalevent. This consists of fusion and fenestration of the outeracrosomal membrane with the plasma membrane. AR dys-functions have been proposed to be one of the causes ofreduced fertility in men. The assessment of AR in vitro isthus expected to yield additional information on the fertil-ization potential of spermatozoa, which has not been de-rived from conventional semen analysis [63].

Progesterone, which is secreted by cumulus cells, hasbeen indicated as a physiological stimulus or costimulusfor initiating the AR in spermatozoa. It is speculated thatprogesterone-induced effects (i.e., an increase in calciuminflux, hyperactive motility, zona pellucida binding, andzona-free oocyte penetration) are mediated via distinct,nongenomic PR [55, 56]. It is also speculated that thesesperm functions probably involve three different types ofreceptors—a plasma membrane Ca21 ion channel, a GA-BAA receptor, and a membrane-associated protein tyrosinekinase [64]. Our previous studies on the biochemical andmolecular characterization of these membrane-bound pro-gesterone binding sites on human spermatozoa have re-

vealed similarities as well as dissimilarities of these recep-tors with the conventional intracellular PR [54, 65]. Thesestudies demonstrated that these progesterone binding siteson spermatozoa are membrane-bound, heat-labile, and lo-calized at the acrosomal region [54]. These sites were foundto be masked in spermatozoa of ejaculates from normalmen. It has been speculated that the reshuffling of mem-brane components during in vivo capacitation, the AR, orboth facilitates unmasking of these progesterone binding

1332 GADKAR ET AL.

FIG. 4. Immunolocalization of PR on human spermatozoa with abnormal morphology. Spermatozoa with head defects (c–f), midpiece defects (g andh), and tail defects (i and j) were stained with antibodies against the conventional PR. Brown staining indicates PR positivity and blue color appearsdue to counterstaining with hematoxylin. Spermatozoa with normal morphology are shown in b. Negative control (without antibody) is shown in a.Sperm were visualized with an 3100 oil immersion objective.

TABLE 2. Spermatozoa showing positivity for hypoosmotic swelling test and acrosome reaction in fertile and infertile men.a

Test

Group

Normozoospermicn 5 12

Oligozoospermicn 5 12

Asthenozoospermicn 5 12

Oligoasthenozoospermicn 5 09

Teratozoospermicn 5 10

Positive for thehypoosmoticswelling test

83.5%(79.44–87.55)

69.08%*(55.48–82.68)

67.08%*(58.16–76.0)

49.55%*(35.81–63.3)

72.1%(64.91–79.28)

Acrosome-reacted 24.0%(21.53–26.46)

11.08%*(7.83–14.32)

11.41%*(8.02–14.8)

9.22%*(4.96–13.48)

17.2%*(12.42–21.97)

a Values are mean; * P , 0.05, compared with normozoospermic group. Values in parenthesis indicate range.

sites. The present study was undertaken to investigatewhether PR expression on spermatozoa has any significancefor evaluating infertility in men.

In contrast to other reports demonstrating that only 11%of sperm in ejaculates from fertile men bear PR [47, 66],its expression was observed in 70%–80% of spermatozoafrom fertile men in the present study. Our results are inagreement with a report demonstrating significant respon-siveness to progesterone in 93% of spermatozoa from fer-tile donors [67]. It can be hypothesized that either the coat-ing of the sperm PR with proteins of testicular or epidid-

ymal origin or masking of these sites may stearically hinderthe binding of progesterone to its receptor. It is likely thattreatment of spermatozoa with mild detergents (i.e., digi-tonin) as carried out in this study unmasks these sites andthereby allows detection of a higher percentage of PR-pos-itive cells. In the present study, the functional assessments(acrosome reaction) of live spermatozoa were conducted inthe absence of detergent because these events led to re-shuffling of membrane components in spermatozoa andprobably subsequent unmasking of PR. However, for lo-calization of PR, fixed sperm smears were treated with mild

1333SPERM PROGESTERONE RECEPTOR IN FERTILE AND INFERTILE MEN

TABLE 3. Percentage of spermatozoa with different patterns of PR expression detected following HOST.a

PatternNormozoospermic

n 5 8Oligozoospermic

n 5 7Asthenozoospermic

n 5 8Oligoasthenozoospermic

n 5 7Teratozoospermic

n 5 11

HOST1PR1HOST1PR2

70.75 6 2.808.41 6 1.23

43.0 6 6.92*28.0 6 5.65*

39.5 6 3.56*27.2 6 4.0*

28.22 6 4.07*23.44 6 4.15*

49.20 6 4.79*23.0 6 5.82*

HOST2PR2 11.16 6 2.35 20.25 6 5.2 (NS) 15.33 6 3.85 (NS) 26.55 6 5.75* 14.30 6 3.21 (NS)a Values are mean 6 SEM. HOST1, Swollen head and coiled tail (cell with functionally active plasma membrane); HOST2, straight tail (damagedplasma membrane); PR1, stained equatorial or acrosomal region as indicated by yellow or green color; PR2, unstained equatorial or acrosomal regionas indicated by red color. NS, Not significant.* P , 0.05, compared with the normozoospermic group.

FIG. 5. PR localization on spermatozoa exposed to hypoosmotic con-ditions. a) Double positive, HOS1PR1, coiled tail (indicative of HOSpositivity), and green color (indicative of PR positivity); b) Single positive,HOS1PR2, coiled tail, and red color (PR negativity); c) double negative,HOS2PR2, uncoiled tail, and red color. Sperm were visualized with an3100 oil immersion objective.

detergent (either 0.1% digitonin or 0.01% sodium deoxy-cholate).

Evidence supporting a physiological role for progester-one in the AR comes from reports suggesting a relationshipbetween male infertility and the inability of spermatozoa torespond to progesterone in vitro [47–49]. However, thesestudies were conducted on spermatozoa from selectedgroups (either patients with unexplained infertility or oli-gozoospermia or teratozoospermia). The present study wasundertaken to analyze PR expression in spermatozoa frommen with various seminal abnormalities.

A significant decrease in the number of PR-positive sper-matozoa was detected in infertile men in the present study.To our knowledge, this is the first study in which two meth-ods (flow cytometry and immunocytochemistry) were usedto evaluate PR expression in spermatozoa from fertile andinfertile men. Percentages of PR-positive spermatozoa us-ing both immunocytochemical and flow cytometric meth-ods were found to be similar. These two techniques wereused to rule out the possibilities of measuring postacroso-mal fluorescence (due to diffusion of P-FITC-BSA intodead cells) and of manual bias in counting the positivecells, respectively. A slightly higher percentage of PR-pos-itive spermatozoa was recorded by flow cytometry in someinfertile groups compared with that assessed by immuno-cytochemistry. This may arise because of the cumulativefluorescence (postacrosomal, acrosomal, and equatorial)measured by the flow cytometer, whereas the data collectedusing immunocytochemistry took into account only thosecells that showed PR localization at the acrosomal region.The failure to express PR by spermatozoa from infertilemen may reflect an underlying pathological mechanism.

It was intriguing to find that only 24% of spermatozoafrom normal men undergo inducible AR. This may be at-tributed to the possibility of incomplete unmasking of PRon human spermatozoa or the requirement of additional in-ducers for AR. The other plausible reason for the observeddifference in the numbers of PR-positive spermatozoa andin vitro acrosome-reacted spermatozoa is that plain washed(not digitonin-treated) spermatozoa were subjected to theAR. It is also likely that the concentration of AR stimulator(follicular fluid in the present study) may not preciselymimic the effects of physiological inducers. In the presentstudy, the use of follicular fluid was preferred over chem-ical inducers (i.e., calcium ionophore), because these chem-ical inducers are known to be cytotoxic. Although follicularfluid has other AR-inducing components in addition to pro-gesterone, and it would have been more appropriate to eval-uate only progesterone-induced AR and rule out the con-tribution of other follicular fluid components to AR, follic-ular fluid is an AR inducer that closely mimics physiolog-ical conditions. Further, the same batch of follicular fluidwas used to stimulate the AR in spermatozoa from bothfertile and infertile groups.

In the present study, a significant decrease in the per-centage of acrosome-reacted spermatozoa was observed inmen with oligozoospermia, asthenozoospermia, oligoas-thenozoospermia, and teratozoospermia compared withthose with normozoospermia. These results are in agree-ment with the findings of Fuse et al. [40]. A decrease inthe percentage of spermatozoa with the potential to undergothe AR in infertile groups is suggestive of aberrations inthe cascade of events leading to the AR. This may arisebecause of the lower PR expression on spermatozoa in in-fertile men observed in this study.

The potential role of PR in the AR is also indirectlyevident by the observation of a higher percentage of bothacrosome-reacted as well as PR-positive spermatozoa inmen with teratozoospermia compared with those with oli-gozoospermia, asthenozoospermia, and oligoasthenozoo-spermia. It is interesting that no strong correlation was

1334 GADKAR ET AL.

FIG. 6. Detection of PR on spermatozoapreviously exposed to hypoosmotic condi-tions in men with normozoospermia (b),oligozoospermia (c), asthenozoospermia(d), oligoasthenozoospermia (e), and tera-tozoospermia (f). Negative control (FITC-BSA) is shown in a. Green-yellow colorindicates PR positivity and red indicatesnegative staining.

TABLE 4. Coefficient of correlation between % HOS1/% HOS1PR1/% HOS1PR2 and % AR spermatozoa in fertile and infertile men.

Coefficient ofcorrelation (r)

Group

Normozoospermicn 5 12

Oligozoospermicn 5 12

Asthenozoospermicn 5 12

Oligoasthenozoospermicn 5 09

Teratozoospermicn 5 10

% HOS1 and % AR 0.4982 0.4958 0.4680 0.5800 20.224%HOS1PR1 and % AR 0.8545 0.8711 0.7645 0.9003 0.8676%HOS1PR2 and % AR 20.2906 20.5246 0.1912 0.0902 20.8596

found between PR expression and other seminal parameterssuch as motility and morphology (data not shown). In lightof these observations, it is not surprising that spermatozoawith abnormal morphology retain their fertilization poten-tial as evident by in vitro fertilization studies [49, 68], andabnormal morphology is not necessarily accompanied by afailed AR [63].

Attempts have been made in the past to use the func-tional activity of the plasma membrane (i.e., the HOS test)as an indicator of sperm function [21, 22, 25, 26, 69]. How-ever, its diagnostic usefulness is still debatable [27–30]. Theability of the sperm tail to swell in the presence of a hy-poosmotic solution is an indicator of the membrane activity.In the present study, significantly fewer HOS1 spermato-zoa were found in men with oligozoospermia, astheno-

zoospermia, and oligoasthenozoospermia than in men withnormozoospermia, thereby indicating an impaired mem-brane integrity of spermatozoa in men with infertility. How-ever, HOS positivity alone failed to serve as an indicatorof sperm function in men with teratozoospermia. Further,when an attempt was made to find a correlation betweenHOS positivity and AR in spermatozoa, only a weak cor-relation was found. Although the HOS test and the ARrequire a functionally active sperm membrane, these seemto be independent events. In contrast, a strong positive cor-relation was found between the percentage of HOS1PR1and the percentage of AR sperm cells. Significantly fewerHOS positive spermatozoa from men with normozoosper-mia lacked PR expression compared with those in men withabnormal spermiograms. This probably suggests that not

1335SPERM PROGESTERONE RECEPTOR IN FERTILE AND INFERTILE MEN

only the integrity of the plasma membrane, but that PRexpression, are both required for the AR. This is the firststudy in which the expression of a surface antigen or mem-brane-bound molecule was analyzed in spermatozoa ex-posed to hypoosmotic conditions.

Evaluation of PR expression on spermatozoa may giveinsight into the functional competence of sperm becausethe PR expression on sperm bears a strong correlation withits ability to undergo in vitro AR. Further studies need tobe undertaken to understand the mechanism of progester-one-induced AR through PRs. Nevertheless, the presentstudy demonstrates that compared to HOS, PR expressionseems to be a better predictor of sperm function. In menwith an abnormal spermiogram, a higher percentage ofHOS-positive spermatozoa lacked PR expression. Further,the detection of PR on human spermatozoa is fast, easy toperform, and does not require expensive equipment. It ispossible to analyze the slides, and they can be permanentlypreserved and reassessed. Hence, analysis of PR expressionon spermatozoa is a useful method for evaluating spermfunction.

ACKNOWLEDGMENTS

The authors acknowledge Dr. Vandana Walvekar, the dean of WadiaMaternity Hospital, Mumbai, for providing the semen samples. We appre-ciate the suggestions by Dr. Geeta Vanage and Dr. Usha Natraj (assistantdirectors, Institute for Research in Reproduction, Parel, Mumbai) for im-munocytochemical studies. The authors also thank Dr. M. Hansotia of theFertility Clinic in Mumbai for providing human follicular fluid. The au-thors also thank Mr. D. Balaiah (assistant director, Institute for Researchin Reproduction, Parel, Mumbai) for assisting in statistical analysis.

REFERENCES

1. Steinberger E, Smith KD, Rodriguez-Rigan LJ (eds.). Relationshipbetween the results of semen analysis and the fertility potential of thecouples. In: Diagnostic Aspects in Andrology. Berlin: Grosse Verlag;1983: 10–22.

2. Blasco L. Clinical tests of sperm fertilizing ability. Fertil Steril 1984;41:177–192.

3. Polansky FF, Lamb EJ. Do the results of semen analysis predict futurefertility? A survival analysis study. Fertil Steril 1988; 49:1059–1065.

4. Hewitt J, Cohen J, Steptoe P (eds.), In: Male Infertility and In VitroFertilization. Edinburgh; 1987: 253–275.

5. Hinting A, Comhaire F, Vermeulen L, Dhont M, Vermeulen A, Van-dekerckhove D. Value of sperm characteristics of in vitro fertilizationfor predicting the outcome of assisted reproduction. Int J Androl 1990;13:59–66.

6. World Health Organization. WHO Laboratory Manual for the Exam-ination of Human Semen and Semen-Cervical Mucus Interaction, 3rded. Cambridge, U.K.: The Press Syndicate of the University of Cam-bridge; 1992: 3–26.

7. Yanagimachi R. Zona-free hamster eggs: their use in assessing fertil-izing capacity and examining chromosomes of human spermatozoa.Gamete Res 1984; 10:187–232.

8. Rogers BJ. The sperm penetration assay: its usefulness reevaluated.Fertil Steril 1985; 43:821–840.

9. Jeyendran RS, van der Ven HH, Perez-Palaez M, Crabo BG, ZaneveldLJ. Development of an assay to assess the functional integrity of thehuman sperm membrane and its relationship to other semen charac-teristics. J Reprod Fertil 1984; 70:219–228.

10. Yanagimachi R. Mammalian fertilization. In: Knobil E, Neill J (eds.),The Physiology of Reproduction. New York: Raven Press Ltd; 1988:135–185.

11. Yanagimachi R, Yanagamachi H, Rogers BJ. The use of zona freeanimal ova as a test system for the assessment of the fertilizing ca-pacity of human spermatozoa. Biol Reprod 1976; 15:471–476.

12. Borros C, Gonzales J, Herrera E, Bustos-Obregon E. Human spermpenetration into zona-free hamster oocytes as a test to evaluate thesperm fertilizing activity. Andrologia 1979; 11:197–210.

13. Rogers BJ, van Campen H, Uneo M, Lambert H, Bronson R, Hale R.

Analysis of human spermatozoal fertilizing ability using zona-freeova. Fertil Steril 1979; 32:664–670.

14. Yanagimachi R. Mechanisms of fertilization in mammals. In: Mas-troianni L Jr, Biggers JD (eds.), Fertilization and Embryonic Devel-opment In Vitro. New York: Plenum Press; 1981: 81–90.

15. Aitken RJ, Wang FU, Liu J, Best F, Richardson DW. The influenceof medium composition, osmolarity and albumin content on the ac-rosome reaction and fertilizing capacity of human spermatozoa: de-velopment of an improved zona-free hamster egg penetration test. IntJ Androl 1983; 6:180–193.

16. Wickings EJ, Freischem CW, Langer K, Nieschlag E. Heterologousovum penetration test and seminal parameters in fertile and infertilemen. J Androl 1983; 4:261–271.

17. Talbot P, Chacon RS. Ultrastructural observations on binding andmembrane fusion between human sperm and zona pellucida-free ham-ster oocytes. Fertil Steril 1982; 37:240–248.

18. Johnson AR, Syms MJ, Lipshultz LI, Smith RJ. Conditions influenc-ing human sperm capacitation and penetration of zona-free hamsterova. Fertil Steril 1984; 41:603–608.

19. Wolf DP, Sokoloski JE. Characterization of the sperm penetration bio-assay. J Androl 1982; 3:445–451.

20. Leibo SP, Giambernardi TA, Meyer TK, Bastias MC, Rogers BJ. Theefficacy of cryopreserved hamster ova in the sperm penetration assay.Fertil Steril 1990; 53:196–212.

21. Check JH, Nowroozi K, Wu CH, Bollendrof A. Correlation of semenanalysis and hypoosmotic swelling test with subsequent pregnancies.Arch Androl 1988; 20:257–260.

22. Okada A, Inomata K, Matsuhashi M, Fujio K, Miura K, Shiina K,Shirai M. Correlation of the hypo-osmotic swelling test, semen scoreand the zona free hamster egg penetration assay in humans. Int JAndrol 1990; 13:337–343.

23. Carreras A, Palama A, Mendoza C. Hypoosmotic swelling test in nor-mo-, oligo-, astheno- and oligoasthenozoospermic men before and af-ter swim-up separation of spermatozoa. Andrologia 1990; 22:313–317.

24. Cotezee K, Kruger TF, Menkveld RJ, Swanson RJ, Lombard CJ,Acosta AA. Usefulness of sperm penetration assay in fertility predic-tion. Arch Androl 1989; 23:207–212.

25. Jeyendran RS, van der Ven HH, Rachagan SP, Perez-Palaez M, Zan-aveld LJ. Semen quality and in vitro fertilization. Aust N Z J ObstetGynaecol 1989; 29:168–172.

26. Takahashi K, Uchida A, Kitao M. Hyoposmotic swelling test ofsperm. Arch Androl 1990; 25:225–242.

27. Chan SY, Fox EJ, Chan MM, Tsoi WL, Wang C, Tang LC, Tang GW,Ho PC. The relationship between the human sperm hypoosmotic test,routine semen analysis and the human sperm zona-free hamster ovumpenetration assay. Fertil Steril 1985; 44:668–672.

28. Barratt CLR, Osborn JC, Harrison PE, Monks N, Dunphy BC, LentonEA, Cooke ID. The hypoosmotic swelling test and the sperm mucuspenetration in determining fertilization of the human oocyte. HumReprod 1989; 4:430–434.

29. Sjoblom P, Coccia E. On the diagnostic value of the hypoosmoticsperm swelling test in an in vitro fertilization (IVF) program. J InVitro Fertil Embryo Transf 1989; 6:41–43.

30. Avery S, Bolton VN, Mason BA. An evaluation of the hypo-osmoticsperm swelling test as a predictor of fertilizing capacity in vitro. IntJ Androl 1990; 13:93–99.

31. Talbot P, Chacon RS. A triple-stain technique for the evaluation ofnormal acrosome reaction in human sperm. J Exp Zool 1981; 215:201–208.

32. Aitken RJ, Best FSM, Warnern P, Templeton A. A prospective studyof the relationship between semen quality and fertility in case of un-explained infertility. J Androl 1984; 5:297–303.

33. Prasad MRN. The in vitro sperm penetration test: a review. Int JAndrol 1984; 7:5–22.

34. Osman RA, Andria ML, Jones AD, Meizel S. Steroid induced exo-cytosis: the human sperm acrosome reaction. Biochem Biophys ResCommun 1989; 160:828–833.

35. Foresta C, Rossato M, Mioni R, Zorzi M. Progesterone induces ca-pacitation in human spermatozoa. Andrologia 1992; 24:17–21.

36. Parinaud J, Labal B, Vieitez G. High progesterone concentrations in-duce acrosome reaction with low cytotoxic effect. Fertil Steril 1992;58:599–602.

37. Bronson RA, Peresleni T, Golightly M. Progesterone promotes theacrosome reaction in capacitated human spermatozoa as judged byflow cytometry and CD46 staining. Mol Hum Reprod 1999; 5:507–512.

1336 GADKAR ET AL.

38. Cummins JM, Pember SM, Jequier AM, Yovich JL, Hartmann PE. Atest of the human sperm acrosome reaction following ionophore chal-lenge. J Androl 1991; 12:98–103.

39. Vigil P, Wohler E, Bustos-Obregon E, Comhaire F, Morales P. As-sessment of sperm function in fertile and infertile men. Andrologia1994; 26:55–60.

40. Fuse H, Okumura M, Sakamoto M, Kazama T, Katayama T. Acro-some-reacted sperm in infertile and fertile men using the triple-staintechnique. Arch Androl 1993; 30:41–45.

41. Tesarik J, Mendoza C. Insights into the functions of a sperm-surfaceprogesterone: evidence of ligand-induced receptor aggregation and theimplication of proteolysis. Exp Cell Res 1993; 205:111–117.

42. Calvo L, Dennison-Lagos L, Banks SM, Sherins RJ. Characterizationand frequency distribution of sperm acrosome reaction among normaland infertile men. Hum Reprod 1994; 9:1875–1879.

43. Luconi M, Krausz C, Forti G, Baldi E. Extracellular calcium nega-tively modulates tyrosine phosphorylation and tyrosine kinase activityduring capacitation of human spermatozoa. Biol Reprod 1996; 55:207–216.

44. Patrat C, Serres C, Jouannet P. The acrosome reaction in human sper-matozoa. Biol Cell 2000; 92:255–266.

45. Krausz C, Gervasi G, Forti G, Baldi E. Effect of platelet-activatingfactor on motility and acrosome reaction of human spermatozoa. HumReprod 1994; 9:471–476.

46. Krausz C, Bonaccorsi L, Luconi M, Fuzzi B, Criscuoli L, PellegriniS, Forti G, Baldi E. Intracellular calcium increase and acrosome re-action in response to progesterone in human spermatozoa are corre-lated with in vitro fertilization. Hum Reprod 1995; 10:122–124.

47. Tesarik J, Mendonza C, Moos J, Carreras A. Selective expression ofprogesterone receptor on the human sperm surface. Fertil Steril 1992;58:784–792.

48. Falsetti C, Baldi E, Krausz C, Casono R, Failli P, Forti G. Decreasedresponsiveness to progesterone of spermatozoa in oligozoospermic pa-tients. J Androl 1993; 14:17–22.

49. Oehninger S, Blackmore PF, Morshedi Ml, Sueldo C, Acosta A, Al-exander NJ. Defective calcium influx and acrosome reaction in sper-matozoa of infertile men with severe teratozoospermia. Fertil Steril1994; 61:349–354.

50. Thomas P, Meizel S. Phosphatidylinositol 4,5-biphosphate hydrolysisin human stimulated with follicular fluid or progesterone is dependentupon Ca21 influx. Biochem J 1989; 264:539–546.

51. Blackmore PF, Beebe SJ, Danforth DR, Alexander N. Progesteroneand 17a-hydroxyprogesterone: novel stimulators of calcium influx inhuman sperm. J Biol Chem 1990; 265:1376–1380.

52. Blackmore PF, Im WB, Bleasdale JE. The cell surface progesteronereceptor which simulates influx in human sperm is unlike the A ringreduced steroid site on the GABAA receptor/chloride channel. MolCell Endocrinol 1994; 104:237–243.

53. Meizel S, Turner KO. Progesterone acts at the plasma membrane ofhuman sperm. Mol Cell Endocrinol 1991; 11:R1–R5.

54. Ambhaikar MB, Puri CP. Cell surface binding sites for progesteroneon human spermatozoa. Mol Hum Reprod 1998; 4:413–421.

55. Sabeur K, Edwards DP, Meizel S. Human sperm plasma membraneprogesterone receptor(s) and the acrosome reaction. Biol Reprod1996; 54:993–1001.

56. Cheng FP, Gadella BM, Voorhout WF, Fazeli A, Bevers MM, Col-enbrander B. Progesterone-induced acrosome reaction in stallion sper-matozoa is mediated by a plasma membrane progesterone receptor.Biol Reprod 1998; 59:733–742.

57. Sufi SB, Donaldson A, Jeffcoate SL. WHO Matched Reagent Pro-gramme Method Manual, 18th ed. Geneva: World Health Organiza-tion; 1994.

58. Margalioth EJ, Bronson RA, Cooper GW, Rosenfeld DL. Luteal phasesera and progesterone enhance sperm penetration in the hamster eggassay. Fertil Steril 1988; 50:117–122.

59. Goding JS. Conjugation of antibodies with fluorochromes: modifica-tions to the standard methods. J Immunol Methods 1976; 13:215–219.

60. Haas CG Jr, Cunningham ME. Identification of antibody-laden spermby cytofluorometry. Fertil Steril 1984; 42:606–613.

61. Biggers JD, Whitten WK, Whittingham DG. The culture of mouseembryos in vitro. In: Daniel JC (ed.), Methods in Mammalian Em-bryology. San Francisco: Freeman Press; 1971: 86–116.

62. Singer SI, Lambert H, Cross NL, Overstreet JW. Alteration of humansperm surface during in vitro capacitation as assessed by lectin in-duced agglutination. Gamete Res 1985; 12:291–299.

63. Calvo L, Dennison-Lagos L, Banks SM, Dorfmann A, Thorsell LP,Bustillo M, Schulman JD, Sherins RJ. Acrosome reaction inducibilitypredicts fertilization success at in vitro fertilization. Hum Reprod1994; 9:1880–1886.

64. Tesarik J, Carreras A, Mendonza C. Single cell analysis of tyrosinekinase dependent and independent Ca21 fluxes in progesterone in-duced acrosome reaction. Hum Reprod 1996; 11:225–232.

65. Sachdeva G, Shah CA, Kholkute SD, Puri CP. Detection of proges-terone receptor transcript in human spermatozoa. Biol Reprod 2000;62:1610–1614.

66. Blackmore PF, Neulen J, Lattanzio F, Beeb SJ. Cell surface bindingsites for progesterone mediated calcium uptake in human sperm. JBiol Chem 1991; 266:18655–18659.

67. Plant A, McLaughlin EA, Ford WCL. Intracellular calcium measure-ments in individual human sperm demonstrates that the majority canrespond to progesterone. Fertil Steril 1995; 64:1213–1215.

68. Kiefer D, Check JH, Katsoff D. The value of motile density, strictmorphology and the hypoosmotic swelling test in in vitro fertilization-embryo transfer. Arch Androl 1996; 37:57–60.

69. Cotezee K, Kruger TF, Menkveld RJ, Lombard CJ, Sawnson RJ.Hypo-osmotic swelling test in the prediction of male infertility. ArchAndrol 1989; 23:131–138.