Cytoplasmic droplets are normal structures of human sperm but are not well preserved by routine procedures for assessing sperm morphology

Trevor G. Cooper1,3, Ching-Hei Yeung1, Sabina Fetic1, Aligholi Sobhani2 and Eberhard Nieschlag1

1 Institute of Reproductive Medicine of the University Clinic, Domagkstrasse 11, D-48129 Münster, Germany and 2 Department of Anatomy, Tehran University of Medical Sciences, Tehran, Iran

3 To whom correspondence should be addressed. Email: cooper{at}uni-muenster.de


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
BACKGROUND: There is a discrepancy between the use of terminology employed by clinicians and basic scientists concerning the cytoplasmic droplets of sperm. Most clinicians consider their presence on sperm to be indicative of abnormal sperm, whereas basic scientists consider them to be attributes of normal sperm. METHODS: The presence of cytoplasmic droplets on human sperm was examined using conventional air-dried, fixed and stained sperm smears and in living and fixed wet preparations. RESULTS: Cytoplasmic droplets were found on the majority of motile sperm and in fixed preparations but only half of them were found in air-dried smears. There was no relationship between the presence of abnormally large cytoplasmic droplets, indicative of abnormal sperm, and the droplets found on living cells. CONCLUSION: The term ‘cytoplasmic droplet’ is confusingly used to describe two different sperm structures: large amounts of retained, excessive cytoplasmic remnants, that survive the air-drying procedure and are observed on abnormal sperm in conventionally stained sperm smears, and osmotically sensitive vesicles that are present on normal living sperm. A plea is made to retain the term ‘cytoplasmic droplet’ for the latter structure of normal sperm and to use the term ‘excess residual cytoplasm’ to describe the abnormally retained cytoplasm observed on abnormal sperm in smears.

Key words: artefacts/cytoplasmic droplets/human sperm/morphology/nomenclature


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Cytoplasmic droplets are a normal component of mammalian sperm, found at the neck of immature caput sperm and at the end of the midpiece in mature cauda sperm. Their movement along the midpiece during migration through the caput epididymidis is a characteristic feature of sperm maturation and they are retained on the majority of mammalian sperm stored in the cauda epididymidis (Cooper and Yeung, 2003Go). There are fewer sperm with droplets in the bovine ampulla (Branton and Salisbury, 1947Go; Rao and Hart, 1948Go) and few on ejaculated sperm from bulls (O'Donnell, 1969Go), rams (White et al., 1959Go) and boars (Lasley and Bogart, 1944aGo,bGo). Recent studies suggest that droplets may play an important role in fertility, since sperm are subjected to a hypotonic challenge upon ejaculation into the female tract, and water would initially enter the sperm's cytoplasm, the bulk of which is in this organelle. In transgenic mice, sperm that cannot maintain their volume upon osmotic challenge exhibit flagellar angulation occurring at the site of the cytoplasmic droplet, and the mice are infertile (Cooper et al., 2004Go).

Observations on human ejaculated sperm that have been swollen by the channel blocker quinine and that display poor penetration of surrogate mucus suggest that volume regulation may play a role in human fertility (Yeung and Cooper, 2001Go; Yeung et al., 2003Go). The vast majority of literature on ‘cytoplasmic droplets’ of human sperm considers them to be indicative of abnormality with sperm being described as of ‘diminished maturity’ (Gergely et al., 1999Go) or ‘immature sperm’ (Ollero et al., 2000). The different terminology used to describe the cytoplasmic structure of human sperm (‘cytoplasmic droplets’, Rago et al., 2003), ‘cytoplasmic residues’ (Keating et al., 1997Go), ‘residual sperm cytoplasm’ (Aitken et al., 1994Go), ‘abnormal retention of cytoplasmic droplets’ (Zini et al., 2000Go), ‘retention of cytoplasm’ (Mak et al., 2000Go) attests to the confusion surrounding this organelle.

Such disagreement in terminology is not helped by the different descriptions given by semen analysis manuals. According to the World Health Organization (1999)Go, cytoplasmic droplets observed in air-dried semen smears should only be considered morphological defects when large (greater than one-third or one-half the sperm head size), implying that smaller droplets are not abnormal. The ESHRE/NAFA handbook (2002)Go confirms that sperm with retained cytoplasm less than one-third the sperm head size are normal but adds to the confusion by stating that residues larger than this are abnormal ‘and classified as cytoplasmic droplets’, associating that name with an abnormal structure.

Recent observations on osmotically sensitive ‘midpiece vesicles’ have prompted the view that such vesicles may be a normal component of human sperm (Abraham-Peskir et al., 2002Go; Chantler and Abraham-Peskir, 2004Go), although these authors consider them distinct from ‘cytoplasmic droplets’ by accepting that such ‘droplets’ survive air-drying and are characteristic of immature sperm.

The recommended clinical procedure for evaluating human seminal sperm (the production of air-dried smears before fixation: World Health Organization, 1999Go) can cause morphological artefacts. For instance, it produces severely swollen sperm heads when applied to immature epididymal sperm (Yeung et al., 1997Go; Soler et al., 2000Go), yet no swollen sperm heads are observed if the cells are fixed before making the smears. Such a drastic procedure as air-drying has been shown to disrupt fragile, osmotically sensitive midpiece vesicles (Abraham-Peskir et al., 2002Go). Thus the apparent absence of droplets from a normal spermiogram may be an artefact of semen preparation for sperm morphological analysis.

This report describes experiments performed in order to assess the presence of true (not abnormal) cytoplasmic droplets on human sperm in living, in fixed and wet, and in air-dried preparations.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Preliminary experiments
Three observations on semen from one donor, a healthy father, revealed the presence of cytoplasmic droplets on human sperm. (i) An ejaculate was produced directly in a vessel containing 5% (v/v) glutaraldehyde in phosphate-buffered saline (PBS) (Dulbecco's; Sigma, Germany) and aliquots of the material were compressed under a cover slip for microscopic evaluation. (ii) Within 30 s of production an ejaculate was incubated at 37°C and within 1 min, and every 5 min for 30 min before and after liquefaction, 50 µl was transferred to 500 µl 5% (v/v) glutaraldehyde fixative in Biggers–Whitten–Whittingham (BWW) medium (Biggers et al., 1971Go) and the percentage of sperm bearing droplets was ascertained microscopically. (iii) An ejaculate was brought to the laboratory within 75 s and incubated under 2 ml BWW medium [osmotic pressure (OP) 328 mmol/kg] containing 4 mg/ml bovine serum albumin. Sperm emerging from the liquefying ejaculate were evaluated in 5 µl aliquots for motility and presence of droplets in unfixed, wet preparations. After liquefaction, routine semen smears for morphology were made for evaluation of cytoplasmic droplets.

Ejaculates
Human ejaculates of widely differing quality were obtained with informed consent from 37 patients attending the Institute of Reproductive Medicine and from 12 student volunteers. Ten normozoospermic samples were included and the characteristics of the semen provided in this study, as analysed according to the World Health Organization manual (1999)Go, are presented in Table I.


View this table:
[in this window]
[in a new window]
 
Table I. Characteristics of semen used in this study

 
Measurement of osmotic pressure
The osmotic pressure of 10 µl semen was measured after liquefaction with a vapour pressure osmometer (Wescor Vapro 5510; Kreienfeld Scientific Measuring Systems, Germany), calibrated daily with a 290 mmol/kg standard solution. As the viscosity of semen retards saturation of the detection chamber, a timed delay of 2 min was employed before reading to ensure accuracy, as recommended by the manufacturer. BWW medium with osmotic pressure of 230 mmol/kg (BWW230) was made by reducing the amount of NaCl.

Microscopical observations
After liquefaction, semen samples were incubated for 15 min at 37°C and 3 µl were examined under a 22 x 22 mm cover slip by phase contrast microscopy (Olympus BH-20, Japan) with a x40 objective and x10 ocular on a heated stage (Mini-Tüb, Germany) at 37°C. The percentage of motile sperm (WHO grades a + b + c) was determined and immotile and motile sperm were separately assessed for the presence of (I) cytoplasmic droplets (small, regular distensions at the neck or midpiece: Figure 1a, c, e), (II) abnormal residual cytoplasm (large, irregular material along the mid-piece: Figure 1d, h, iFigure 1d, h, i), (III) coiled or looped tails (Figure 1g, j, k) or (IV) none of the above categories. To 10–20 µl of these samples was added an equal volume of 7% (v/v) glutaraldehyde and after 60 min at room temperature the fixed cells were washed by addition of 1 ml PBS, and centrifugation at 500 g for 5 min. The pellet was examined as a wet preparation at x400 magnification for the presence of categories I–IV above. Similar preparations were examined in which semen was mixed with appropriate volumes of BWW230 to a final osmolality of 290 mmol/kg, the osmotic pressure of cervical mucus (Rossato et al., 1996Go).



View larger version (129K):
[in this window]
[in a new window]
 
Figure 1. Examples of human ejaculated sperm observed in fixed wet preparations (phase contrast optics: a, b, h, j), live wet preparations (differential interference contrast optics: d, e, f, i, k) and Papanicolaou-stained, air-dried smears (c, g). Examples of true cytoplasmic droplets (a, b, c, e, f), abnormal cytoplasmic residues (d, h, i) and coiled tails (g, j, k). Bars = 5 µm.

 
The routine air-dried, Papanicolaou-stained semen smears of the same ejaculates were examined at both x400 and x1000 magnification by the same observer of the wet, fixed preparations and scored according to the same criteria for defining droplets. These stained smears were also examined for the presence of normal cytoplasmic droplets, defined as being smaller than one-third to one-half the sperm head size, and also for the presence of abnormal cytoplasmic residues by experienced andrology technicians.

Statistics
Differences between populations were assessed by paired or unpaired t-tests, and relationships by linear regression. P<0.05 was accepted as statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Preliminary observations
In order to determine if cytoplasmic droplets, as seen in other mammalian species and considered as normal structures, were present on human sperm, one ejaculate was fixed immediately upon ejaculation and two were fixed within 75 s of sample production. Observations on the sample delivered into fixative revealed the presence of droplets on the majority of sperm within the unliquefied coagulum. When aliquots of an ejaculate were fixed at intervals before and after liquefaction, 47–65% of sperm had visible droplets over the 30 min examined. When sperm taken from the medium surrounding a liquefying ejaculate were examined in wet preparations, 49–57% of immotile sperm bore droplets, whereas 68–92% of motile sperm did. The mean percentages of droplets observed in conventional Papanicolaou-stained smears of the two liquefied samples (14.5, 10.0%) were far lower than those found in the glutaraldehyde-fixed samples examined at 1 min (65%) or 20 min (67%).

Sperm morphology and motility
The majority of live sperm viewed in wet preparations in semen (mean 336 mmol/kg at the time of processing) and in media of female tract tonicity (290 mmol/kg) were observed to have cytoplasmic droplets at the neck regions, sometimes extending along the length of the midpiece (Figure 1b, f). The percentage of motile sperm with droplets significantly exceeded that of immotile cells in both semen and at the osmolality of cervical mucus (290 mmol/kg) and there was no difference in motility of sperm in these fluids (Table II). The same sperm suspensions fixed in glutaraldehyde also revealed cytoplasmic droplets on the majority of them. In order to compare this value with that obtained from the live wet preparations, a weighted mean was obtained by multiplying the percentage of motile and immotile sperm by their respective percentages of droplets. This percentage (40.5±1.5, n=50, mean±SEM) was not different (paired t-test) from that of the glutaraldehyde-fixed sperm (38.4±1.7).


View this table:
[in this window]
[in a new window]
 
Table II. Sperm motility in semen and Biggers–Whitten–Whittingham (BWW290) and the percentage of motile and immotile sperm each displaying cytoplasmic droplets

 
Sperm morphology in fixed and smeared preparations
The percentage of conventionally air-dried and Papanicolaou-stained sperm revealed significantly fewer droplets than observed in fixed wet preparations, whether observed at x400 (magnification used for the fixed and wet preparations: 52%) or x1000 (magnification used for routine semen analysis: 51%), although there was a significant correlation between the percentages of droplets observed in the dried and wet preparations by the same observer [Figure 2: r=0.500 (x40), r=0.432 (x100)]. When abnormal cytoplasmic residues (defined as greater than one-third to one-half the sperm head size: World Health Organization, 1999Go) were assessed by experienced andrology technicians, the percentage was low (~9.6±0.6) and bore no relation to the percentage of true droplets found in wet and fixed preparations (Figure 2). There was a statistically significant linear relationship between the percentage of what the andrology technicians assessed as abnormal ‘cytoplasmic droplets’ and what were assessed as residual cytoplasm (category II) in 47 wet preparations at x100 (r=0.584; data not shown).


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
The results of this study indicate that ejaculated sperm from men possess true cytoplasmic droplets, which are normal structures of sperm in all other mammalian species, as judged from wet preparations on living gametes and in rapidly fixed preparations. The percentage of motile sperm with droplets in both semen and medium of 290 mmol/kg being higher than that of immotile cells suggests that they are not deleterious to vitality or motility and are clearly not a marker for poor sperm quality. That the percentage of sperm bearing droplets in live preparations agreed with estimates obtained from glutaraldehyde-fixed preparations confirmed that such droplets were not artefacts of cell fixation. Thus human sperm resemble other mammalian sperm in having a midpiece cytoplasmic droplet; what differs is their position, at the neck rather than the end of the annulus, and that they remain attached to the spermatozoon in the ejaculate. Ultrastructural micrographs of well-fixed, human ejaculated sperm also demonstrate a cytoplasmic droplet at the neck (Holstein and Roosen-Runge, 1981Go; Johnson, 1982Go; Neugebauer et al., 1990Go) as they do in the epididymis (Ånberg, 1957Go).

A related and important observation for the morphological assessment of human ejaculates was the far lower percentage of droplets on sperm observed in air-dried, Papanicolaou-stained smears. Indeed, the same observer counted far fewer droplets in stained smears than in the live or fixed wet preparations of the same ejaculates. This attests to the general inadequacy of the routine method for preserving structures sensitive to the stresses accompanying air-drying before fixing and staining. As even human sperm heads may expand under these conditions (Yeung et al., 1997Go; Soler et al., 2000Go), it should not be surprising that far less rigid and osmotically sensitive vesicles would collapse during preparation.

These findings confirm and extend the observations of osmotically sensitive ‘midpiece vesicles’ (MPV) extending along the length of the midpiece of human sperm in semen and cervical mucus (Abraham-Peskir et al., 2002Go; Chantler and Abraham-Peskir, 2004Go), which were also considered not detrimental to sperm and were not found in air-dried preparations. These authors considered the so-called MPV to be distinguishable from ‘cytoplasmic droplets’ by the absence of visible content and their higher incidence than conventional ‘cytoplasmic droplets’. MPV were found to have no content as rendered visible by X-ray and differential contrast microscopy of wet preparations, whereas ‘cytoplasmic droplets’ assessed in air-dried preparations were expected to contain a proteinaceous content which is stained green with Papanicolaou dye. Others have considered swollen cytoplasmic droplets to be artefacts of Percoll preparations (Arcidiacono et al., 1983Go), although they clearly can be observed in the absence of Percoll, and their detection by confocal microscopy (Sofikitis et al., 1994Go) was also considered abnormal, presumably also because they resembled large cytoplasmic remnants observed in air-dried preparations.

We consider that MPV are ‘true’ cytoplasmic droplets as observed here (see Figure 1b, f), as they are present on living gametes and neither survive air-drying well. Thus we disagree with Chantler and Abraham-Peskir (2004)Go that the two organelles are distinct and believe that this difference reflects one of terminology: the abnormal (large, protein-rich, green-stained) ‘cytoplasmic droplets’ that survive the air-drying procedure are considered by many to be the equivalent of cytoplasmic droplets found on well-fixed sperm, whereas we argue that they are excess residual cytoplasm retained by abnormal sperm produced by imperfect spermiogenesis. This view is supported by electron micrographs of well-fixed semen that reveal the membranous components typical of a true cytoplasmic droplet within a vesicle extending the entire length of the human sperm midpiece (see Figure 3 in Smith et al., 1988). The presence of droplets on living sperm within cervical mucus (Abraham-Peskir et al., 2002Go) and on fixed sperm recovered from the Fallopian tube after artificial insemination by husband (Mortimer et al., 1982Go) suggest that they may be markers for functionally superior cells. Whether the presence of true cytoplasmic droplets observed in semen smears are related to fertility deserves examination.

The present observations on the cytoplasmic droplet on the midpiece of the majority of living human sperm highlights the inconsistent and confusing nomenclature whereby ‘cytoplasmic droplets’ are considered normal components of healthy mammalian sperm by basic scientists, but to be a structure associated with abnormal sperm by clinicians. This difference may well have to do with the preparation methods employed by each; generally well-fixed epididymal sperm versus air-dried smears of ejaculated sperm, respectively. Midpiece structures surviving the latter, drastic treatment would include abnormally large amounts of excess cytoplasm not removed at spermiation, adhering to the midpiece and staining green with Papanicolaou. As such excess residual cytoplasm has been associated with sperm from smokers (Mak et al., 2000Go) and men with varicocele (Zini et al., 2000Go) and with deficiencies in sperm DNA (Fischer et al., 2003Go) and phospholipid-bound docosahexanoic acid (Zini et al., 2000Go), its presence is indeed indicative of abnormal spermiogenesis. Such sperm should not be described as of ‘diminished maturity’ (Gergely et al., 1999Go) or as immature (Ollero et al., 2000Go), but merely abnormal, since they cannot and do not undergo maturation in the epididymis.

A plea is made to adhere to a common nomenclature and reserve the term ‘cytoplasmic droplet’ of human sperm to that defined for sperm from all other Eutherian species, namely an anatomically normal, osmotically sensitive component of a well-formed spermatozoon, produced by a functional testicular tubule, and which does not survive well the air-drying of semen smears. Conversely, the large, irregular ‘abnormal cytoplasmic droplet’ (World Health Organization, 1999Go), that does survive air-drying and is stained in routine semen smears, should not be termed a cytoplasmic droplet; the term ‘excess residual cytoplasm’ (Aitken et al., 1994Go) is suggested to indicate the nature of this organelle, found on abnormally formed sperm liberated from a testis displaying damage to the seminiferous epithelium. As both true cytoplasmic droplets and residual cytoplasm could be said to be forms of midpiece vesicles, this term is rejected in favour of the long-existing term cytoplasmic droplet. Acceptance of this terminology should obviate the confusion that has surrounded this topic in the past and reduce confusion arising when the same term is used for two completely different sperm structures.


Figure 2. Relationship between the percentage of human ejaculated sperm bearing cytoplasmic droplets assessed wet after glutaraldehyde fixation (abscissa) (x400 magnification, phase contrast optics; abscissa) and in air-dried, fixed, Papanicolaou-stained smears of the same ejaculates (ordinate), observed at magnification x400 (
, phase contrast) and x1000 (
, oil immersion). The percentage of sperm with excess residual cytoplasm (greater than one-third to one-half the size of the sperm head, ) showed no relationship with that of cytoplasmic droplets. Regression analysis revealed correlation coefficients for ‘true’ droplets in stained smears against droplets in wet, fixed samples (n=48) of 0.500 (x400) and 0.432 (x1000).


    Acknowledgements
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
We thank Ele Kürten, Sabine Rehr, Külli Nurmik and Daniele Hanke for performing the routine semen analysis.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 Acknowledgements
 References
 
Abraham-Peskir JA, Chantler E, Uggerhoj R and Fedder J (2002) Response of midpiece vesicles on human sperm to osmotic stress. Hum Reprod 17, 375–382.[Abstract/Free Full Text]

Aitken R, Krausz C and Buckingham D (1994) Relationships between biochemical markers for residual sperm cytoplasm, reactive oxygen species generation, and the presence of leukocytes and precursor germ cells in human sperm suspensions. Mol Reprod Dev 39, 268–279.[ISI][Medline]

Ånberg Å. (1957) The ultrastructure of the spermatozoon. An electronmicroscopic study of the spermatozoa from sperm samples and the epididymis including some observations of the spermatid. Acta Obstet Gynecol Scand 36 (Suppl 2), 1–133.

Arcidiacono A, Walt H, Campana A and Balerna M (1983) The use of Percoll gradients for the preparation of subpopulations of human spermatozoa. Int J Androl 6, 433–445.[ISI][Medline]

Biggers JD, Whittem WK and Whittingham D (1971) The culture of mouse embryos in vitro. In Daniel Jr, JC (ed.) Methods in Mammalian Embryology. Freeman, San Francisco, pp. 86–116.

Branton C and Salisbury GW (1947) Morphology of spermatozoa from different levels of the reproductive tract of the bull. J Anim Sci 6, 154–160.[ISI]

Chantler E and Abraham-Peskir JV (2004) Significance of midpiece vesicles and functional integrity of the membranes of human spermatozoa after osmotic stress. Andrologia 36, 87–93.[CrossRef][ISI][Medline]

Cooper TG and Yeung CH (2003) Acquisition of volume regulatory response of sperm upon maturation in the epididymis and the role of the cytoplasmic droplet. In Toshimori K (ed.) The Biology of Spermatozoa Maturation. Wiley-Liss Inc, New York, pp. 28–38.

Cooper TG, Yeung C-H, Wagenfeld A, Nieschlag E, Poutanen M, Huhtaniemi I and Sipilä P (2004) Mouse models of infertility due to swollen spermatozoa. Mol Cell Endocrinol 216, 55–63.[CrossRef][ISI][Medline]

ESHRE/NAFA (2002) Manual on basic semen analysis. http://www.ki.se/org/nafa/manual.

Fischer MA, Willis J and Zini A (2003) Human sperm DNA integrity: correlation with sperm cytoplasmic droplets. Urology 61, 207–211.[CrossRef][ISI][Medline]

Gergely A, Kovanci E, Senturk L, Cosmi E, Vigue L and Huszar G (1999) Morphometric assessment of mature and diminished-maturity human spermatozoa: sperm regions that reflect differences in maturity. Hum Reprod 14, 2007–2014.[Abstract/Free Full Text]

Holstein AF and Roosen-Runge EC (1981) Atlas of Human Spermatogenesis. Grosse Verlag, Berlin.

Johnson L (1982) A re-evaluation of daily sperm output of men. Fertil Steril 37, 811–816.[ISI][Medline]

Keating J, Grundy CE, Fivey PS, Elliott M and Robinson J (1997) Investigation of the association between the presence of cytoplasmic residues on the human sperm midpiece and defective sperm function. J Reprod Fertil 110, 71–77.[ISI][Medline]

Lasley JF and Bogart RA (1944a) Comparative study of epididymal and ejaculated spermatozoa of the boar. J Anim Sci 3, 360–370.

Lasley JF and Bogart R (1944b) Some factors affecting the resistance of ejaculated and epididymal spermatozoa of the boar to different environmental conditions. Am J Physiol 141, 619–624.[Free Full Text]

Mak V, Jarvi K, Buckspan M, Freeman M, Hechter S and Zini A (2000) Smoking is associated with the retention of cytoplasm by human spermatozoa. Urology 56, 463–466.[CrossRef][ISI][Medline]

Mortimer D, Leslie EE, Kelly RW and Templeton AA (1982) Morphological selection of human spermatozoa in vivo and in vitro. J Reprod Fertil 64, 391–399.[ISI][Medline]

Neugebauer DC, Neuwinger J, Jockenhoevel F and Nieschlag E (1990) ‘9 + 0’ axoneme in spermatozoa and some nasal cilia of a patient with totally immotile spermatozoa associated with thickened sheath and short mid-piece. Hum Reprod 5, 981–986.[Abstract]

O'Donnell JM (1969) Electrical counting and sizing of mammalian spermatozoa and cytoplasmic droplets. J Reprod Fertil 19, 263–272.[ISI][Medline]

Ollero M, Powers RD and Alvarez JG (2000) Variation of docosahexanoic acid content in subsets of human spermatozoa at different stages of maturation: implications for sperm lipoperoxidative damage. Mol Reprod Dev 55, 326–334.[CrossRef][ISI][Medline]

Rao CK and Hart GH (1948) Morphology of bovine spermatozoa. Am J Vet Res 9, 117–124.[ISI]

Rossato M, Di Virgillio F and Foresta C (1996) Involvement of osmo-sensitive calcium influx in human sperm activation. Mol Hum Reprod 2, 903–909.[Abstract]

Smith AC, Anderson CW, Barratt CLR and Williams MA (1988) Ultrastructural morphometric data on human spermatozoa. Andrologia 20, 396–403.[ISI][Medline]

Sofikitis NV, Miyagawa I, Zavos PM, Toda T, Iino A and Terakawa N (1994) Confocal scanning laser microscopy of morphometric human sperm parameters: correlation with acrosin profiles and fertilising capacity. Fertil Steril 62, 376–386.[ISI][Medline]

Soler C, Perez-Sanchez F, Schulze H, Bergmann M, Oberpenning F, Yeung C and Cooper TG (2000) Objective evaluation of the morphology of human epididymal sperm heads. Int J Androl 23, 77–84.[Medline]

White IG, Larsen LH and Wales RG (1959) Method for the in vivo collection of epididymal spermatozoa and for their comparison with ejaculated cells. Fertil Steril 10, 571–577.[ISI]

World Health Organization (1999) Laboratory Manual for the Examination of Human Semen and Sperm-Cervical Mucus Interaction. 4th edn., Cambridge University Press, Cambridge.

Yeung CH and Cooper TG (2001) Effects of the ion-channel blocker quinine on human sperm volume, kinematics and mucus penetration, and the involvement of potassium channels. Mol Hum Reprod 7, 819–828.[Abstract/Free Full Text]

Yeung CH, Perez-Sanchez F, Soler C, Poser D, Kliesch S and Cooper TG (1997) Maturation of human spermatozoa (from selected epididymides of prostatic carcinoma patients) with respect to their morphology and ability to undergo the acrosome reaction. Hum Reprod Update 3, 205–213.[Abstract/Free Full Text]

Yeung CH, Anapolski M, Depenbusch M, Zitzmann M and Cooper TG (2003) Human sperm volume regulation. Response to physiological changes in osmolality, potential sperm osmolytes and channel blockers. Hum Reprod 18, 1029–1036.

Zini A, Defreitas G, Freeman M, Hechter S and Jarvi K (2000) Varicoele is associated with abnormal retention of cytoplasmic droplets by human spermatozoa. Fertil Steril 74, 461–464.[CrossRef][ISI][Medline]

Submitted on May 5, 2004; accepted on June 16, 2004.