J. Cell Set. 54, 161-171 (198a) 161Printed in Great Britain © Company of Biologists Limited 1982

THE COURSE OF THE ACROSOME REACTIONIN GUINEA-PIG SPERM

D. P. L. GREENDepartment of Pharmacology, University of Otago, Medical School,P.O. Box 913, Dunedin, New Zealand

SUMMARY

The acrosome reaction in guinea-pig sperm is accompanied by a marked cavitation of theacrosomal contents. Two divergent views are held as to whether this cavitation precedes orfollows the membrane fusion that occurs in the reaction. To distinguish between these 2 viewscavitation was induced in media containing a colloid, either Ficoll 70 or inulin, either by in-ducing a normal acrosome reaction using the calcium ionophore A2 3187 or by using the detergentTriton Xioo. Both Ficoll 70 and inulin, when incorporated into media of normal osmolality,were able to suppress various features of the cavitation. Complete retention of acrosomal shapewas achieved in sperm treated with detergent in 30 % (w/v) Ficoll 70 solution despite theabsence of the limiting acrosomal and plasma membranes. This evidence supports the sugges-tion that the cause of the cavitation is a colloid osmotic pressure within the acrosomal matrix.This in turn supports one of the 2 proposed mechanisms for the temporal sequence of eventsoccurring in the acrosome reaction.

INTRODUCTION

The acrosome reaction of mammalian sperm is a secretory discharge in whichouter acrosomal membrane fuses with overlying plasma membrane. Exocytosis of theacrosomal contents follows. The acrosome reaction in guinea-pig sperm is accom-panied by a dramatic increase in acrosomal volume.

Two mutually incompatible views are held as to the temporal relationship betweenthis volume increase and membrane fusion. Yanagimachi & Usui (1974) suggestedthat the volume increase precedes membrane fusion. Green (1978 a, b) on the otherhand, suggested that the volume increase follows membrane fusion.

It is possible, in principle, to distinguish between the 2 different mechanisms byinducing an acrosome reaction in media containing a suitable colloid. Experimentswere performed on guinea-pig sperm in which cavitation was produced either byinducing the acrosome reaction with the calcium ionophore A23187 (Green, 1976,1978a) or by using the detergent Triton Xioo (Green, 19786). Inulin and Ficoll 70were tested for their ability to suppress cavitation. Ficoll 70 (25%, w/v) producedcomplete retention of acrosomal shape despite the absence of intact acrosomal andplasma membranes. A more complex picture emerged at lower concentrations ofFicoll 70 and inulin. The results provide additional support for the mechanism inwhich membrane fusion precedes cavitation.

162 D. P. L. Green

MATERIALS AND METHODS

Ficoll 70 (lot nos. 5279 and 9524) was purchased from Pharmacia, Uppsala, Sweden: 25 %glutaraldehyde from Agar Aids, Bishop's Stortford, England; ethylene glycol tetra-acetic acidfrom Sigma, St Louis, Missouri; and inulin from Hopkins & Williams, Chadwell Heath,England. All other reagents were analytical grade. A23187 was a generous gift from Dr R. L.Hamill, Eli Lilly Laboratories, Indianapolis.

Stock solutions of Ficoll 70 and inulin of 50 % and 20 %, respectively, were made up indistilled water. Inulin is only sparingly soluble in water and was dissolved by heating thesolution to 90 °C in a water bath. The 20 % solution is stable for several hours after coolingalthough the inulin usually comes out overnight. Determinations of the osmolality of thesesolutions, and dilutions of them, were made by both freezing-point depression (model 3W,Advanced Instruments Inc., Needham Height, Mass.) and vapour pressure osmometry(5100 Series, Wescor Inc., Logan, Utah). Solutions of 20 % inulin and 25 % Ficoll 70 behavedas if their molecular weights were 2100 and 2600, respectively.

The suspending media were of 2 basic kinds: calcium medium contained 140 mM-NaCl,4 mM-KCl, 4 mM-HEPES, 10 mM-glucose and 2 mM-CaCl! (312 mosmol/1, theoretical;292 mosmol/kg, measured); magnesium medium was the same as calcium medium exceptthat 2 mM-MgCl| and 100 /*M-EGTA were substituted for the CaClt (293 mosmol/1, measured).The stock solutions of inulin and Ficoll 70 were diluted to form media with the concentrationsgiven in the text. Electrolyte concentrations were as for the simple calcium and magnesiummedia with the exception of (NaCl): this was reduced pro rata to compensate for the osmoticcontribution of the inulin and Ficoll 70, respectively. For example, a 15 % (w/v) solution ofFicoll 70 has an osmolality of 73 mosmol/kg (i.e. it behaves as if it were approx. 66 ITIM). TheNaCl concentration, in both calcium and magnesium media containing 15 % (w/v) Ficoll 70was, therefore, set at 74 mM. The final measured osmolalities of these 2 solutions were 29s and298 mosmol/kg, respectively. Similar compensations were also made for the cacodylate fixingbuffer (normally 0-15 M). However, no osmotic compensation was made for either the glutar-aldehyde or Triton Xioo where these were present. Hypertonic calcium and magnesium mediacontained 280 mM-NaCl, respectively; other concentrations remained as before (measuredosmolalities 559 and 554 mosmol/kg, respectively).

Experiments were of 2 kinds. Either sperm were suspended in a calcium medium and anacro8ome reaction was induced by A2 3187; or sperm suspended in a magnesium medium weretreated with Triton Xioo. In the first case, 10 fil of a stock solution of A23187 in dimethylsulphoxide (2 mg/ml) were added to 1 ml of sperm suspended in a calcium medium containingvarying amounts of inulin or Ficoll 70. After 1 min at 37 °C, 1 ml of 2 % glutaraldehyde incacodylate buffer (pH 7-2), containing the same amount of inulin or Ficoll 70 was added. After1 h, 8 ml of straight 0-15 M-cacodylate buffer (pH 7-2), was added and the suspension spunin a Beckman microfuge B for 6 min. In the second experiments, 1 ml of sperm suspended in amagnesium medium containing inulin or Ficoll 70 was mixed with 1 ml of the same solution

Fig. 1. A guinea-pig sperm head fixed 30 s after addition of a magnesium mediumcontaining 30 % (w/v) Ficoll 70 and 2 % (w/v) Triton Xioo to sperm suspended in amagnesium medium containing 30 % (w/v) Ficoll 70 (288 mosmol/kg). The plasmaand acrosomal membranes have been completely removed by the detergent. However,cavitation has not occurred and normal acrosomal shape is completely preserved.Note the integrity of the lighter area ' a' lying above the main sickle-shaped matrixof the acrosome. x 33 920.Fig. 2. A guinea-pig sperm head fixed 30 s after addition of a modified magnesiummedium containing 15% (w/v) Ficoll 70 and 2% (w/v) Triton Xioo to spermsuspended in a magnesium medium containing 15 % (w/v) Ficoll 70 (291 mosmol/kg).Again, the detergent has removed the plasma and acrosomal membranes. However,cavitation has now occurred in area ' a' of the acrosome shown in Fig. 1. This hasresulted in the complete loss of the contents of this area despite the absence of anyapparent break in the material immediately bounding it.

Course of acrosome reaction 1&3

\

D. P. L. Green

Course of acrosome reaction 165

containing 2 % Triton Xioo (w/v) and fixed after 1 min with 2 ml of 2 % cacodylate buffer(pH 7r2), containing the same amount of inulin or Ficoll. After 1 h, 6 ml of straight 0-15 M-cacodylate buffer (pH 7'2), was added and the suspension spun as before. AH pellets wereprocessed for electron microscopy as previously described (Green, 1978a). Silver to goldsections were cut on a Sorvall microtome and examined in a Philips 300 electron microscope.They were not post-stained.

RESULTS

The 2 colloids chosen for these experiments were both polysaccharides. Thereason for choosing these, and not proteins, was dictated by the need to be able tomake up fixative solutions in which the molecular weight characteristics of the colloidwere not shifted by any polymerizing action of the fixative. The disadvantages ofusing these 2 polysaccharides lies in the absence of any firm knowledge of the mol-ecular weight distributions at low molecular weights. For example, Ficoll 70 (lot no.5279) has an MT of 75 300 and an Mn of 37000, but despite this it behaves as if itsaverage molecular weight were about 2200. Clearly, the osmotic contribution of thelow-molecular-weight material completely dominates that of the higher.

The results themselves are consistent with the mechanism for acrosomal cavitationproposed by Green (1978 a) but not that proposed by Yanagimachi & Usui (1974).Fig. 1 shows that complete retention of acrosomal shape can be achieved in theabsence of the plasma and acrosomal membranes. Care was taken to ensure that theosmolality of the surrounding solution was similar to that of the simple supportingmedium. The conclusion is, therefore, that changing the composition of this mediumto contain colloid suppresses the cavitation. This suggests that an external colloidosmotic pressure is responsible, and that the driving force for cavitation is itself acolloid osmotic pressure.

Fig. 2 shows that, as the colloid component is reduced, selective cavitation isproduced. This selectivity is consistent with either a weaker matrix strength in area'a ' or a higher colloid osmotic pressure in this area or a combination of the two.Exactly the same result is obtained when the acrosome reaction is induced by A23187(Figs. 3, 4). This is important for 2 reasons. It suggests that the use of detergent toinduce cavitation is legitimate, and it helps to resolve the question of whether thereare 2 separate cavitation mechanisms operating side by side. It could be argued that,although in detergent solution cavitation occurred through a colloid osmotic pressure,in the normal acrosome reaction an alternative course was followed in which osmotic-ally active particles were generated (for example, by proteolysis). If the latter were

Fig. 3. A stack of guinea-pig sperm heads in sagittal section fixed 30 s after additionof A23187 to sperm suspended in a calcium medium containing 15 % (w/v) Ficoll 70(295 mosmol/kg). All sperm show induction of the acrosome reaction. The sperm Ashows complete retention of acrosomal content despite its partial expansion in theuppermost region. Sperm B also shows retention of much of its acrosomal materialalthough the pattern of expansion and cavitation is different from sperm A. The other4 sperm all show the complete loss of this material. All 6 sperm show no cavitation ofthe approximately sickle-shaped matrix in the lower part of each acrosome. x 25 560.

D. P. L. Green

Fig. 4. A close-up of sperms A and B in Fig. 3. showing the integrity of both plasmaand acrosomal membranes over the area in which cavitation has occurred. Note thatcavitation in sperm A has produced an amorphous granular array, whereas that in Bhas produced discrete cavities, x 57710.

Course of acrosome reaction 167

the case, however, the alteration of the composition of the external medium toinclude colloid would have been without effect. This is clearly not the case and this,therefore, suggests that there are not 2 mechanisms. The interesting feature of theA23i87-induced cavitation is that because membrane fusion can be arrested byfixation at various stages of development, a better understanding can be achieved ofthe morphological events occurring in area 'a'. This is of interest because of itssimilarity to granule content breakdown in other secretory granules (Jamieson &Palade, 1971; Schuel et al. 1972).

Fig. 5. shows the effect of A23i87-induced cavitation in sperm suspended inhypertonic media. The larger cavity, occurring in the sickle-shaped area, is verymuch larger than normal (cf. Green, 1978 a) and this appears to be a general featureof sperm treated in this way. One simple explanation is that the higher ionic strengthof the medium has produced a weakening of the structure. Two observations supportthis view. The first is that the matrix itself shows a number of smaller areas of cavita-tion not normally seen, suggesting a weaker structure. The second is that in hypertonicmagnesium medium, the matrix is completely dissolved. Inspection of the suspensionusing Nomarski optics (Zeiss IM 35, x 60 oil immersion, N.A. 1-4) shows no evidenceof intact but detached matrices. This suggests that a combination of high ionicstrength and detergent is enough to dissolve the matrix, and that, in turn, the matrixis held together by a combination of hydrophobic interactions and salt-bridges. Fig. 6shows that the ability to suppress cavitation is not restricted to Ficoll 70 but canbe mimicked equally by inulin. Both Ficoll 70 and inulin continue to suppresscavitation of the sickle-shaped area (shown occurring in Fig. 5) down to about the1 % (w/v) level.

The use of hypertonic media is an important test of the mechanism based on colloidosmotic pressure, since cavitation has to be independent of higher external osmoticpressures provided the pressure is due to components for which the matrix hasreflection coefficients close to zero. It could not be expected to eliminate unequivocallythe alfernative mechanism based on the generation of osmotically active particleswithin an intact acrosome, since there could reasonably be some margin of safety. Itwould raise the threshold to expansion but not so high that it was insurmountable.

DISCUSSION

One of the central events that occurs in the acrosome reaction of mammaliansperm is the fusion of plasma and acrosomal membrane. In guinea-pig sperm this isaccompanied by a substantial increase in acrosomal volume. Two mutually exclusiveproposals have been put forward for the temporal relationship between the membranefusion on the one hand and the increase in volume on the other. Yanagimachi &Usui (1974) proposed that the swelling preceded membrane fusion. To accommodatethe swelling, it was suggested either that the number of osmotically active particlesin the acrosome increased during the reaction or that the acrosomal membrane,formerly impermeable to water, underwent an increase in water permeability. Green(1978 a, c), however, proposed that the swelling followed the fusion of acrosomal and

Course of acrosome reaction 169

plasma membranes. Evidence was presented that the acrosomal contents exist as amatrix and that the formation of a large cavity within this matrix accounts for thebulk of the volume increase occurring in the acrosome reaction. The mechanismproposed by Green for cavity formation depends on the existence of a colloid osmoticpressure (c.o.p.) within the acrosomal matrix that the matrix is incapable of with-standing. This osmotic pressure would only become a net pressure on the matrixwhen equilibration of the readily diffusible components occurred between the acro-some and the extracellular fluid. This would happen in the acrosome reaction afterfusion of the plasma and acrosomal membranes, and also when these 2 membraneswere removed by detergent. The induction of cavitation when guinea-pig sperm aretreated with solutions containing Triton Xioo (Green, 19786) provides some of thebest evidence in favour of the c.o.p. model. Despite this, there is no direct evidencethat the matrix can act as a semi-permeable barrier with pores sufficiently small toprevent the movement of larger molecular weight material. Clearly, however, if thematrix is acting as a semi-permeable barrier, it should be possible to place a c.o.p.outside it by adding to the external medium a substance of suitable molecular weight.If present in sufficient quantity, this substance would counter any colloid osmoticpressure within the matrix and suppress cavitation.

The experiments described in this paper show that both inulin and Ficoll 70suppress cavitation when this is induced by either detergent in calcium-free mediumor A23187 in calcium-containing medium. This suggests that the proposed c.o.p.model provides an explanation for cavitation and its suppression. An additional testof this model is provided by the experiment in which the molality of the externalmedium is raised in the absence of Ficoll 70 or inulin. This solution fails to suppresscavitation, suggesting that it is not the external osmolality/w se that is important forthe suppression but whether the solution contains colloid. This is the result expectedwith the c.o.p. model. The experiment does not, however, completely eliminate thealternative model of Yanagimachi & Usui (1974), since it could be argued that raisingthe external molality merely necessitates the production of a greater number ofparticles within the acrosome before swelling occurs. This raises the question ofwhether any of the results described in this paper are inconsistent with this alternativemechanism. The key to the proposal of Yanagimachi & Usui lies in their choice of thesemi-permeable barrier upon which the osmotic pressure acts. Their mechanismdepends upon it being the acrosomal membrane. However, the occurrence of cavita-tion in the absence of the membrane (Green, 1978 c) effectively eliminates thispossibility. One argument that could be deployed to circumvent this objection would

Fig. 5. A guinea-pig sperm fixed 30 s after addition of A23187 to sperm suspended inhypertonic calcium medium (559 mosmol/kg). Cavitation has occurred in the area 'a'of Fig. 1 as well as in the sickle-shaped matrix. Note the very substantial volumeincrease in this second area, x 30100.

Fig. 6. Two acrosomes fixed 30 s after addition of A23187 to a suspension of guinea-pig sperm in a calcium medium containing 15 % (w/v) inulin (29s mosmol/kg). Thepattern of cavitation is similar to that shown by sperm in Fig. 3. x 37080.

170 D. P. L. Green

be to suggest that 2 different mechanisms are operating for cavitation, each dependingon the circumstances of its induction. According to this argument, when sperm aretreated with detergent cavitation is due to a c.o.p., but when an acrosome reaction isinduced normally it is due to an osmotic pressure acting on the intact acrosomalmembrane. If this were the case, however, then induction of the acrosome reactionby A23187 in medium of normal osmolality but containing Ficoll 70 would notsuppress cavitation. The evidence clearly shows that it does. These experiments,therefore, not only provide evidence that the matrix can act as a semi-permeablebarrier but also evidence against the alternative mechanism of Yanagimachi & Usui(1974). Once it is recognized that it is the acrosomal matrix against which the c.o.p.acts it is possible to place cavitation and the fusion of the plasma and acrosomalmembranes into a defined temporal sequence with the fusion coming first. The acro-some reaction therefore follows the following course. Intracellular free calcium risesin the anterior region of the head in response to an increase in the calcium permeabilityof the plasma membrane (Green, 1978a). The stimulus may be a catecholamine(Cornett & Meizel, 1978), possibly acting through stimulation of a guanyl cyclase(Santos-Sacchi & Gordon, 1980). The increased calcium level causes fusion of plasmaand acrosomal membranes by an unknown mechanism and this allows the colloidwithin the acrosomal matrix to develop a net osmotic pressure. This then causes thematrix to cavitate.

It is possible to make a very approximate estimate of the force needed to form theacrosomal cavity using an extension of the approach developed by Bell (1978). Cavi-tation can be regarded as the separation of 2 surfaces by an applied force. It is imposs-ible to measure the critical force needed to separate the 2 surfaces directly (Fo inBell's terminology) because the c.o.p., which exactly opposes that in the matrix,remains unknown. However, taking inulin as an example, the fall in c.o.p. from thehypothetical value, which just opposes that in the matrix to that of 1 % (w/v) inulin,must equal the pressure needed to separate the 2 surfaces. A further drop from 1 %(w/v) inulin to zero inulin allows this pressure to be exceeded. This fall in inulinconcentration from 1 % to zero represents a further pressure on the matrix, inaddition to the critical pressure, of about 12 kPa. This pressure is unlikely to representa small increment in that needed to separate the 2 surfaces. If one assumes, forargument's sake, that this increment is an additional 30 % in pressure over and abovethe critical pressure, the latter would be about 40 kPa. Bell calculated the criticalforce needed to separate a typical antigen-antibody interaction as about 40 pN perbond, taking the energy of the interaction as about 0-37 eV (357 kj/mole). Thecontacts in the matrix are unlikely to be as good. Taking a value of 8 pN/bond forthe critical force would give, with a separation pressure of 40 kPa, a distance betweeninteractions of about 14 nm. Although the assumptions in this calculation can bequalified, it shows that the c.o.p. estimated for the matrix is approximately commen-surate with that needed to split it.

The physiological function of acrosomal expansion is unknown but it may lie in theactivation of acrosin. This protease is normally held in the form of the inactivezymogen precursor, proacrosin, but it becomes activated during the acrosome reaction

Course of acrosome reaction 171

(Green, 19786). The mechanism by which activation is initiated remains unresolvedbut may depend on the influx of a specific ion, such as calcium, into the acrosome. Thefusion of plasma and acrosomal membranes that occurs in the acrosome reaction doesso in 2 specific but limited areas (Friend, Orci, Perrelet & Yanagimachi, 1977) andacrosomal cavitation begins as soon as these 2 areas fuse (Green, 1978 a). There is,therefore, a rapid acrosomal volume expansion before there is any substantial oppor-tunity for the diffusive loss of soluble acrosomal material. Since the acrosomal freecalcium concentration is Likely to be very low, i.e. approximately the same as thecytoplasmic free calcium concentration, and the external calcium concentration muchhigher, the expansion following membrane fusion is likely to produce a very rapidrise in the calcium concentration to which the acrosomal material is exposed. If thiswere to activate an enzyme, activation would take place in a confined space and, giventhe autocatalytic nature of proacrosin conversion, would be extremely rapid.

A number of secretory granules have condensed matrix cores that show cavitation.either when the granule is isolated and stressed (Jamieson & Palade, 1971; Schuelet al. 1972) or during secretion (Rohlich, Anderson & Uvnas, 1971; Lagunoff, 1973).The morphology of this cavitation is similar to that seen in the guinea-pig acrosomeduring the acrosome reaction (in particular, Figs, 3, 4). It is possible, therefore, thatthe mechanism for cavitation operating in the acrosome also operates in other secretorygranules and plays some role in secretion.

REFERENCES

BELL, G. I. (1978). Models for the specific adhesion of cells to cells. Science 200, 618-627.CORNETT, L. E. & MEIZEL, S. (1978). Stimulation of in vitro activation and the acrosome reac-

tion of hamster spermatozoa by catecholamines. Proc. natn. Acad. Sci. U.S.A. 75, 4954-4958.FRIEND, D. S., ORCI, L., PERRELET, A. & YANAGIMACHI, R. (1977). Membrane particle changes

attending the acrosome reaction in guinea-pig spermatozoa. J. Cell Biol. 74, 561-577.GREEN, D. P. L. (1976). Induction of the acrosome reaction in guinea-pig spermatozoa in vitro

by the Ca ionophore A23187. J. Pkysiol., Land. a6o, 18P-19P.GREEN, D. P. L. (1978a). The induction of the acrosome reaction in guinea-pig sperm by the

divalent metal cation ionophore A23187. J. Cell Sci. 33, 137-151.GREEN, D. P. L. (1978ft). The activation of proteolysis in the acrosome reaction of guinea-pig

sperm. .7. Cell Sci. 33, 153-164.GREEN, D. P. L. (1978 c). The osmotic properties of the acrosome in guinea-pig sperm. J. Cell

Sci. 33, 165-176.jAMffiSON, J. D. & PALADE, G. E. (1971). Condensing vacuole conversion and zymogen granule

discharge in pancreatic exocrine cells: metabolic studies. J. Cell Biol. 48, 503-522.LAGUNOFF, D. (1973). Membrane fusion during mast cell secretion. J. Cell Biol. 57, 252-259.ROHLICH, P., ANDERSON, P. & UVNAS, B. (1971). Electron microscope observations on com-

pound 48/80-induced degranulation in rat mast cells. J. Cell Biol. 51, 465-483.SANTOS-SACCHI, J. & GORDON, M. (1980). Induction of the acrosome reaction in guinea-pig

spermatozoa by cGMP analogues. J. Cell Biol. 85, 798-803.SCHUEL, H., WILSON, W. L., BRESSLER, R. S., KELLY, J. W. & WILSON, J. R. (1972). Purifica-

tion of cortical granules from unfertilized sea urchin egg homogenates by zonal centrifugation.Devi Biol. 39, 307-320.

YANAGIMACHI, R. & Usui, N. (1974). Calcium dependence of the acrosome reaction andactivation of guinea-pig spermatozoa. Expl Cell Res. 89, 161-174.

(Received 29 June 1981)