Oocyte and zygote zona pellucida permeability to macromolecules

THE JOURNAL OF EXPERIMENTAL ZOOLOGY 271:145-150 (1995)

RAPID COMMUNICATION

Oocyte and Zygote Zona Pellucida Permeability to Macromolecules MICHAEL LEGGE Department of Biochemistry, University of Otago, Dunedin, New Zealand ABSTRACT Zona pellucida intact mouse pre- and post-ovulatory oocytes and zygotes were investigated for permeability to macromolecules using graded neutral fluorescein isothiocyanate (FITC) dextrans (molecular weight range 3.84-170 kDa) and a range of galactose-binding lectins (molecular weight range 40-247 kDa). The use of both FITC neutral dextrans and FITC galactose-binding lectins gave good agreement in the zona pellucida permeability studies. Zona pellucida intact pre- and post-ovulatory oocytes were permeable to both markers up to a molecular weight range of 170 kDa although movement across the zona pellucida was diminished between 100 to 150 kDa with no detectable permeability at molecular weight range of 170 kDa. Zygotes demonstrated a decreased zona pellucida permeability at around 110 kDa. Two galactose-containing oligosaccharide-rich zones in the zona pellucida were identified, and changes in zygote vitelline membrane galactose-bindinglectin affinity were detected. o 1995 Wiley-Lm, Inc.

Mammalian oocytes and pre-implantation embryos are surrounded by the zona pellucida, a n acellular glycoprotein matrix. Specific roles assigned to the zona pellucida are sperm attachment during fertilization, initiation of the acrosome reaction, and prevention of attachment of pre-implantation embryos to the reproductive tract (Wassarman, '87). It has also been proposed, based on earlier research, that the zona pellucida may act as a barrier to macromolecules or infections (Nichols and Gardner, '89). However, close examination of the literature does not present a complete picture in relation to consistency of the data. Previous reports of zona pellucida permeability to macromolecules cover a range of species, e.g., rabbits, rats, mice, and hamsters, variation in the stages of development investigated (primarily morula or blastocysts obtained either in vitro or in vivo), and a divergent range of macromolecules such as heparin, peroxidase, IgG, IgM, dyes, ferritin, and viruses (Austin and Lovelock, '58; Glass, '63; Gwatkin, '67; Hastings et al., '72; Sellens and Jenkinson, '75). In addition, Kapur and Johnson ('85, '86) reported the appearance of a n oviduct 215-kDa protein i n the perivitelline space of mouse oocytes and embryos which bound both a n IgG antibody and wheat germ agglutinin lectin (WGA). However, no other reports relating to this protein have been made, and the nature of the protein is still to be resolved. I n the present work the permeability of superovulated mouse oocyte and zygote zona pellu0 1995 WILEY-LISS, INC.

cida to macromolecules was determined under standardized experimental conditions. To determine the zona pellucida permeability two sets of molecular weight probes were used, graded molecular weight neutral dextrans and a series of galactosebinding lectins with a range of molecular weights, both conjugated with fluorescein isothiocyanate (FITC) for detection by fluorescence microscopy

MATERIALS AND METHODS Mice Oocytes and zygotes were collected from BALB/ cBy x C57BW6ByFl female mice. Zygotes were obtained following mating with C57BL/6By males. All female mice were superovulated using 12 IU pregnant mares' serum gonadotrophin (PMSG) followed 48 hours later by 20 IU human chorionic gonadotrophin (hCG). Both hormones were obtained from Sigma Chemical Co. (St. Louis, MO). Fertilization was confirmed both by the presence of a copulation plug and the confirmation of two pronuclei. Pre-ovulatory oocytes were obtained 48 hours after the administration of PMSG. Ovaries were dissected out, and large (2300-pm) follicles (Pedersen, '70) were dissected from the ovaries using a pair of watchmakers forceps and a 30gauge needle under a Olympus SZH141 stereo zoom dissecting microscope. The follicles were ruptured and the oocytes were recovered. All oocytes

Received July 14, 1994; revision accepted September 21, 1994.

146

M. LEGGE

and zygotes were collected into embryo culture medium (Quinn, '85). For the oviductal oocytes and zygotes, the cumulus oophorus was removed using bovine testicular hyaluronidase (Sigma Chemical Company Ltd., St Louis, MO) then washed three times in embryo culture medium before being placed in embryo culture medium at 37°C in a 5% carbon dioxide in air atmosphere until required for use. The pre-ovulatory oocytes were pre-incubated at 37°C in a modified embryo culture medium containing 50 mM sodium bicarbonate and 25 mM potassium chloride for 10 minutes. (This initiates cumulus oophorus expansion; M. Legge, unpublished observation.) The pre-ovulatory oocytes were then placed into bovine testicular hyaluronidase, and the cumulus oophorus was removed through enzymic action and gentle pipetting. The pre-ovulatory oocytes were then treated as described above. Following the removal of the cumulus oophorus, both oocyte groups a n d the zygotes were assessed for visual quality, i.e., blastomere fragmentation, refractibility, zona pellucida integrity, and thickness. Poor-quality oocytes and zygotes were discarded at this stage. When the removal of the zona pellucida was required, oocytes and zygotes were placed briefly into acidic Tyrodes solution (Nicolson et al., '75) and removed following dissolution of the zona pellucida. These were then washed three times in embryo culture medium before use. All animal experiments had the appropriate animal ethics committee approval.

Lectin reactivity

Six FITC-labeled lectins all with galactose-binding specificity and a range of molecular weights between 40 and 247 kDa were purchased from Sigma Chemical Co. (St Louis, MO) (Table 1).All lectins were prepared in embryo culture medium minus bovine serum albumin containing 0.2% sodium azide and at a final working concentration of 500 pg.rn1-l. For the negative control, prior to incubation with oocytes or zygotes, the lectin was pre-mixed with a n equal volume of the appropriate inhibitor (Table 1)then added to the oocytes or zygotes (Legge, '91). Each lectin experiment contained five t o ten oocytes or zygotes, and each lectin was tested a minimum of three times on different occasions. A final total of 30 oocytes or zygotes from each group was used. Both experimental and control groups were incubated and treated as described for the FITC dextrans. Three experimental groups were established for the FITC-lectin investigations: zona pellucida lectin binding, vitelline membrane lectin binding viewed through the zona pellucida, and zona pellucida free oocytes and zygotes. The latter group provided information on oocyte and zygote lectin binding unimpeded by the zona pellucida. To determine lectin binding to the vitelline membrane in zona pellucida intact oocytes and zygotes, the plane of focus of the microscope was adjusted and the fluorescence intensity scored. Fluorescence was scored using an arbitrary scoring system: 0 = negative; +- = marginal; + = weak; ++ = moderate; +++ = strong; ++++ = brilliant (1,egge and Sellens, '88; Legge, '91). All fluoresFITC dextrans cence measurements were made using a Leitz A range of FITC-labeled neutral dextrans was Dialux epifluorescence microscope with a BP450 purchased from Sigma Chemical Co. (St Louis, to 490 excitation filter, a n RKP beam-splitting MO) with molecular weights of 3.84,4.2,9.0,42.0, mirror, and a n LP515 suppression filter for FITC 71.2, 149.0, and 170.0 kDa, respectively Each was detection. prepared at a concentration of 1mg.ml-I then serially diluted to give a working concentration of 1 RESULTS pg.rn1-l of FITC dextrans in embryo culture meThe results for both t h e FITC dextrans a n d dium minus bovine serum albumin (Caulfield and the FITC lectins a r e shown i n Tables 2 a n d 3, Farquhar, '76). Oocytes and zygotes, five to ten respectively. per well, were placed into multiwell microtest plates (Sterilin, Feltham, UK) containing 10 ~1 FITC dextrans each of the FITC dextrans and incubated for 1 No fluorescence was detected in the zona pelluhour at 37°C i n a 5% C 0 2 i n air atmosphere. Following incubation, t h e oocytes a n d zygotes cida for both oocyte groups and the zygotes over were washed six times i n phosphate-buffered the entire molecular weight range. Both the presaline (pH 7.4) then transferred to phosphate- and post-ovulatory oocytes demonstrated identibuffered saline (pH 8.0) i n cavity slides and ex- cal perivitelline space fluorescence results, giving amined for fluorescence as described below moderate fluorescence over the molecular weight range of 3.84 t o 71.2 kDa, then weak fluorescence (Lectin Reactivity).

OOCYTE/ZYGOTEZONA PELLUCIDA PERMEABILITY

147

TABLE I . Panel ofgalactose-specific Lectins' MW Maculara pomifera

Abbreviation MPA

(x103)

Wisteria floribunda

WFA

68

Helix pomatia

HPA

79

Glycine m a x

SBA

110

Dolichos biflorus

DBA

140

Phaseolus limesis

LBL

247

Lectin

40

Carbohydrate specificity GalP13GalNAC > GalNAC > Gal GalPl-GGal> GalNAC > Gal GalNACal-3GalNAC > aGalNAC > aGal a-GalNAC = P-GalNAC > Gal GalNACal-3GalNAC > aGalNAC >iaGal GalNACa 1-3 [Fucal-2lGal> GalNCA > Gal

Inhibitor

Inhibitor concentration, M

P-galactose

0.5

J3-galactose

0.5

Galactosamine

0.3

Galactosamine

0.3

Galactosamine

0.3

Galactosamine

0.5

'The concentration that was required t o inhibit binding a t a lectin concentration of 500 pgml-l.

for the 149- and 170-kDa FITC dextrans. The zy- pellucida lectin-binding patterns were detected for gote group also gave identical results to both oo- both oocyte groups and the zygotes. When t h e vitelline membrane was viewed cyte groups up to 71.2 kDa, but for both the 149 and 170 kDa FITC dextrans only marginal fluo- through the zona pellucida both the pre- and postovulatory oocytes gave identical results for each rescence was detected in the perivitelline space. lectin tested. No lectin binding was observed for FITC lectins LBL (247 kDa); however, strong fluorescence was %o fluorescence zones could be distinguished observed for DBA (140 kDa). The zygote group in the zona pellucida with all the galactose-bind- had diminished lectin binding over the molecular ing lectins: a n outer or less intense fluorescence weight range 40 to 140 kDa. When compared with (moderate-to-strong fluorescence) and a n inner both oocyte groups, MPA (40 kDa) and WFA (68 zone of more intense fluorescence (strong-to-bril- kDa) demonstrated a reduction in fluorescence liant fluorescence). With the exception of WFA from strong to moderate, HPA (79 kDa) did not (GalP1-6 Gal) lectin binding more strongly to the change among the three groups, SBA (110 kDa) outer zone of the zygote than observed in both was reduced from moderate to weak. DBA (140 oocyte groups and HPA (GalNAc a 1 3 GalNac) kDa) changed from strong to marginal, and LBL binding more strongly to the inner zone in pre- (247 kDa) was negative. ovulatory oocytes zona pellucida than the oviducLectin binding for both zona pellucida free pretal oocytes a n d zygotes, no changes in zona and post-ovulatory oocytes was identical both be-

TABLE 2. Permeability of the zona pellucide to FITC dextrans of differing molecular weights i n pre- and post-ovulatory oocytes and zygotes' 3.84 Pre-ovulatory Zona' PVS3 Post-ovulatory Zona2 PVS3 Zygote Zma2 PVS3

4.20

FITC dextrans (molecular weight x lo3) 9.00 42.0 71.2

149.0

170.0

-

-

-

-

-

-

-

++

++

++

++

++

+

+

-

-

-

-

-

-

-

++

++

++

++

++

+

+

-

-

-

-

-

-

-

++

++

++

++

++

f

f

'Key: ++, moderate; +, weak; 2 ,marginal; -, negative fluorescence. 'Fluorescence within the matrix of the zona pellucida was scored. 3PVS, pen-vitelline space. Fluorescence associated with the space between the zona pellucida and the vitelline membrane was scored.

148

M. LEGGE TABLE 3. The binding of galactose-specific lectins with differing molecular weights to the zona pellucida a n d vitelline membrane of pre- a n d post-ouulatory oocytes and zygotes

Lectin Primary specificity Molecular weight

lo3

MPA GalP1-3 GalNAc 40-43

Zona pellucide Pre-ovulatory +++, ++ (inner, outer) Post-ovulatory +++, ++ (inner, outer) Zygote +++, ++ Vitelline membrane viewed through zona Pre-ovulatory +++ Post-ovulatory +++ Zygote ++ Zona free Pre-ovulatory +++ Post-ovulatory +++ Zygote ++ 'Key:

WFA Galpl-6Gal

HPA

LBL

68

GalNAcal-3 GalNAc 79

SBA aGalNAc = PGalNAc 110

DBA GalNAcal-3 GalNAc 140

GalNAcal-3 [Fucal-21 Gal 247

++++, +++

++++, ++

+++, ++

+++, ++

+++, ++

++++, +++

+++, ++

+++, ++

+++, ++

+++, ++

++++, ++++

+++, ++

+++, ++

+++, ++

+++, ++

+++ +++ ++

++ ++ ++

++ ++ +

+++ +++

-

*

-

+++ +++

++ ++ ++

++ ++ +++

+++ +++ +++

++ ++ ++

++

-

-, negative; f,marginal; +, weak; ++, moderate; +++, strong; ++++, brilliant fluorescence.

tween groups and with the zona pellucida intact oocyte groups. The exception was moderate LBL (247 kDa) binding in the zona pellucida free oocyte groups which was negative in the zona pellucida intact oocytes. The vitelline membrane of the zona pellucida free zygotes gave identical results to the zona pellucida intact zygotes for MPA (40 kDa), WFA (68 kDa), and HPA (79 kDa). However, the vitelline membrane of the zona pellucida free zygotes demonstrated significant binding for SBA (110 kDa) from weak t o strong, for DBA (9140 kDa) from marginal to strong, and for LBL (247 kDa) from negative to moderate fluorescence.

DISCUSSION The movement of macromolecules across the zona pellucida is a n unresolved issue, with variable results having been reported in the past. Previous research has used a variety of experimental techniques, and these data have been carried forward into t h e current literature. Austin a n d Lovelock ('58) used slightly compressed cumulus oophorus intact oocytes from rabbits, rats, and hamsters and monitored the movement of digitonin, alcian blue, toludine blue, and heparin (molecular weights 300, 1,200, 1,200, and 16,000 kDa, respectively); they concluded that oocyte membranes were not permeable to substances of molecular weights 16,000 kDa or greater. These experiments did not take into account the charge effect of the dyes and heparin or the nature of the compressed preparation. Glass ('63), using a

single intravenous injection of bovine plasma albumin prior to ovulation o r during oviductal development, demonstrated FITC-anti-bovine plasma albumin binding in the oocyte and embryo cytoplasm and concluded that the transmission of maternal macromolecules to the oocyte and embryo did take place. Later, Gwatkin ('67) reported the transmission of mengovirus through the zona pellucida of morulae cultured from the two-cell stage but discounted the possibility of passage through the zona pellucida, rather favouring access via channels left when follicle cell processes were withdrawn. Subsequent investigations by Hastings et al. ('72) who injected either peroxidase (molecular weight 40 kDa) or ferritin (molecular weight S O 0 kDa) into the uterine lumen of rats, mice, and rabbits before recovery of blastocysts, concluded that the zona pellucida did not exclude macromolecules. Similarly Sellens and Jenkinson ('75), using eight-cell embryos in culture for up to 72 hours, used complement-mediated lysis to demonstrate the penetration of IgG (molecular weight 150 kDa) and IgM (molecular weight 900 kDa) across the zona pellucida. They did not report the condition of the zona pellucidae in their experiments as a thinning of the zona pellucida occurs in vitro with advancing embryo development (M. Legge, unpublished observation). This may well influence the ability of macromolecules t o traverse the zona pellucida. Markers such as heparin, peroxidase, alcian blue, and ferritin in certain states are charged

OOCYTEEYGOTE ZONA PELLUCIDA PERMEABILITY

molecules, which may affect their ability t o move across membranes and other cellular structures, especially if the structure also carries a charge (Farquhar, '81). This limits the use of certain probes for determining the permeability of structures t o macromolecules. Previous research using the sialic acid-binding lectins of Limax f l a w s and LimuEus polyphenus has demonstrated that the zona pellucida of oocytes and zygotes is sialic acid rich (Legge and Sellens, '88; Legge, '91). The negative charge created by sialic acid in the zona pellucida may, therefore, influence the transit of charged molecules such as proteins and dyes. In the present study two approaches t o determine zona pellucida permeability t o macromolecules were employed, graded molecular weight neutral dextrans and galactose-binding lectins of differing molecular weights. Caulfield and Farquhar ('76) established that neutral dextrans over a molecular weight range of 40 t o 200 kDa behaved in the same way as comparable proteins and that they proved t o be reliablc markers for permeability studies for the structural elements of the glomerular capillary wall. This was subsequently confirmed by Bohrer et al. ('78). The inclusion of the galactose-binding lectins in the present work provided an additional opportunity to use a range of proteins with differing molecular weights and defined end point (galactose binding). Movement of the FITC dextrans across the zona pellucida was identified for both the pre- and postovulatory oocytes with a change of perivitelline space fluorescence occurring between the molecular weight range of 71.2 kDa and 149.0 kDa; however, weak fluorescence was still detectable in the perivitelline space for the two higher molecular weight dextrans, 149.0 and 170.0 kDa. This would indicate that similar-size proteins could access the zona pellucida intact oocyte. The zygote, however, demonstrated a more distinct cut off with a change from moderate t o marginal fluorescence for the 149.0- and 170.0-kDa dextrans, indicating that the zona pellucida becomes less permeable to high molecular weight compounds following fertilization. The use of the galactose-binding lectins provides a comparative measure of protein movement across the zona pellucida with the graded dextrans as well as information relating t o the comparative galactose lectin-binding patterns between the experimental groups. As no change in fluorescence was observed for the zona pellucida intact and free pre- and post-ovulatory oocytes for DBA lectin (molecular weight 150 kDa) it must be concluded that the zona pellucida is permeable t o proteins

149

up t o this molecular weight. However, the zona pellucida intact zygotes demonstrated a marked inhibition of vitelline membrane lectin binding for SBA (110 kDa) and DBA (140 kDa). This would indicate that fertilization induces changes t o the zona pellucida which renders it virtually impermeable to macromolecules in excess of 110 kDa. None of the experimental groups demonstrated zona pellucida permeability t o LBL lectin (247 kDa) although the vitelline membrane of the zona pellucida free oocytes and zygotes did bind LBL. These results are comparable with the FITC neutral dextran results (Table 2) which indicated that molecules with molecular weights up to 170 kDa could move across the zona pellucida of both pre- and post-ovulatory oocytes, but that movement is restricted following fertilization. These data are consistent with the earlier work of Glass ('63) who identified bovine plasma albumin in the perivitelline space of oocytes and zygotes using immunological techniques. However, a direct comparison with other previous research is not possible because of later embryo stages used (eight-cell to blastocyst). There were no major changes in the galactosebinding lectins between the two oocyte groups; however, all three experimental groups demonstrated two fluorescence zones in the zona pellucida, a n inner zone giving stronger brilliant fluorescence and an outer zone giving either moderate or strong fluorescence. As all these lectins specifically bind galactose-containing glycoconjugates, it must be concluded that there is a high concentration of these structures in the inner zone of the zona pellucida. Similar zone structures in the zona pellucida have been reported previously in other species by Boranska et al. ('75) using ruthenium red and electron microscopy and Dudkiewicz et al. ('76) using immunochemical techniques. Tesoriero ('84), using cytochemical techniques, also demonstrated gradients of complex carbohydrates in the mouse zona pellucida. The present galactose-binding studies confirm this heterogeneity of the structure of the zona pellucida, although its significance remains obscure. Fertilization did not alter this two-zone structure significantly; however, the vitelline membrane fluorescence did decrease following fertilization from strong t o moderate for MPA (GalP1-3 GalNac) and WFA (GalP1-6 Gal) and increased from moderate to strong for SBA (aGalNAc = PGalNAc). The significance of these oligosaccharide structural changes is not known at present.

M. LEGGE

150

LITERATURE CITED Austin, C.R., and J.E. Lovelock (1958) Permeability of rabbit, rat and hamster egg membranes. Exp. Cell. Res., l5:260-26 1. Bohrer, M.P., C. Baylis, H.D. Humes, R.J. Glassock, C.R. Robertson, and B.M. Brenner (1978) Permselectivity of the glomerular capillary wall: facilitated filtration of circulating polycations. J. Clin. Invest., 61~72-78. Boranska, W., M. Konwinski, and M. Kuyowa (1975) Fine structure of the zona pellucida of unfertilized eggs and embryos. J. Exp. Zool., 192~193-202. Caulfield, J.P, and M.G. Farquhar (1976) Distribution of anionic sites in glomerular basement membranes: their possible role in filtration and attachment. Proc. Natl. Acad. Sci. U.S.A., 73:1646-1650. Dudkiewicz, A.B., C.A. Shivers, and W.L. Williams (1976) U1trastructure of the hamster zona pellucida treated with zona precipitating antibody. Biol. Reprod., 14:175-185. Farquhar, M. (1981) The glomerular basement membrane: a selective macromolecular filter. In: Cell Biology of Extracellular Matrix. E.D. Hay, ed. Plenum Press, New York, pp. 335-378. Glass, L.E. 11963) Ransfer of native and foreign serum antigens to oviductal mouse eggs. Am. Zool., 3:135-156. Gwatkin, R.B.L. (1967) Passage of mengovirus through the zona pellucida of the mouse morula. J. Reprod. Fertil., 13:577-578. Hastings, R.A., H. Allen, C. Enders, and S. Schlafke (1972) Permeability of the zona pellucida to protein tracers. Biol. Reprod., 7:288-296. Kapur, R.P., and L.V. Johnson (1985) An oviductal fluid gly-

coprotein associated with ovulated mouse ova and early embryos. Dev Hiol., 122r89-93. Kapur, R.P., and L.V Johnson (1986) Selective sequestration of an oviductal fluid glycoprotein in the perivitelline space of mouse oocytes and embryos. J . Exp. Zool., 238:249-260. Legge, M. (1991) Oocyte and zygote zona pellucida lectin binding in BALBIcBy and C57BWGBy mice and their F 1 hybrids. J. Exp. Zool., 259~405-408. Legge, M., and M.H. Sellens (1988) Lectin reactivity of murine periovulatory oocytes and one cell embryos. In: Lectins, Biology, Biochemistry, Clinical Biochemistry. Vol. 6. T.C. Bog-Hansen and M. Freed, eds. Sigma Chemical Co., St Louis, pp. 651-658. Nichols, J., and R.L. Gardner (1989) Effect of damage t o the zona pellucida on development of pre-implantation embryos in the mouse. Hum. Reprod., 4:180-187. Nicolson, G.L., R. Yanagimachi, and H. Yanagimachi (1975) Ultrastructural studies of the mouse blastocyst substages. J. Embryol. Exp. Morphol., 32:675-695. Pedersen, T. (1970) Follicle kinetics in the ovary of the cycling mouse. Acta Endocrinol (Copenh.), 64:304-323. Quinn, F! (1985) Improved pregnancy rate in human in vitro fertilization with the use of a medium based on the composition of human tubular fluid. Fertil. Steril., 44r493-498. Sellens, M.H., and E.J. Jenkinson (1975) Permeability of the mouse zona pellucida t o immunoglobulins. J. Reprod. Fertil., 42~153-157. Teuoriero, J.V. (1984) Comparative cytochemistry of the developing ovarian follicles of the dog, rabbit and mouse: origin of the zona pellucida. Gamete Res., 10:301-318. Wassarman, PM. (1987) The zona pellucida: a coat of many colours. Bioessay, 6161-166.