Intra-cytoplasmic sperm injection in a marsupial

REPRODUCTION RESEARCH

Intra-cytoplasmic sperm injection in a marsupial Nadine M Richings, Geoffrey Shaw, Peter D Temple-Smith and Marilyn B Renfree Department of Zoology, The University of Melbourne, Victoria, 3010, Australia Correspondence should be addressed to N M Richings; Email: [email protected]

Abstract Here we report the first use of intra-cytoplasmic sperm injection (ICSI) in a marsupial, the tammar wallaby (Macropus eugenii ), to achieve in vitro fertilization and cleavage. A single epididymal spermatozoon was injected into the cytoplasm of each mature oocyte collected from Graafian follicles or from the oviduct within hours of ovulation. The day after sperm injection, oocytes were assessed for the presence of pronuclei and polar body extrusion and in vitro development was monitored for up to 4 days. After ICSI, three of four (75%) follicular and four of eight (50%) tubal oocytes underwent cleavage. The cleavage pattern was similar to that previously reported for in vivo fertilized oocytes placed in culture, where development also halted at the 4- to 8-cell stage. One-third of injected oocytes completed the second cleavage division, but only a single embryo reached the 8-cell stage. The success of ICSI in the tammar wallaby provided an opportunity to examine the influence of the mucoid coat that is deposited around oocytes passing through the oviduct after fertilization. The presence of a mucoid coat in tubal oocytes did not prevent fertilization by ICSI and the oocytes cleaved in vitro to a similar stage as follicular oocytes lacking a mucoid coat. Cell– zona and cell – cell adhesion occurred in embryos from follicular oocytes, suggesting that the mucoid coat is not essential for these processes. However, blastomeres were more closely apposed in embryos from tubal oocytes and cell– cell adhesion was more pronounced, indicating that the mucoid coat may be involved in maintaining the integrity of the conceptus during cleavage. Reproduction (2004) 128 595–605

Introduction The only report of successful in vitro fertilization (IVF) in a marsupial is in a South American species, the grey shorttailed opossum (Monodelphis domestica), in which mature follicular oocytes were placed with preincubated epididymal spermatozoa in a complex culture medium and 67% of fertilized oocytes cleaved (Moore & Taggart 1993). Despite many efforts using standard insemination methods, fertilization has not been achieved in vitro in any Australian marsupial (Rodger 1994, Renfree & Lewis 1996), although sperm binding and penetration of the zona pellucida (Mate et al. 2000, Sidhu et al. 2003), sperm–oocyte fusion (Sidhu 2003) and decondensation of the sperm head (Sidhu 2003) have been observed in vitro. Intra-cytoplasmic sperm injection (ICSI) is a micromanipulation technique in which a single spermatozoon is inserted into the cytoplasm of an oocyte. ICSI has been used widely in eutherian mammals to achieve fertilization and embryo development (Palermo et al. 1992, Catt & Rhodes 1995, Burruel et al. 1996, Hewitson et al. 1996, Pope et al. 1998, Martin 2000, Deng & Yang 2001, Yamauchi et al. 2002). In humans, 67% of injected oocytes cleave and 82% of these embryos are deemed to q 2004 Society for Reproduction and Fertility ISSN 1470–1626 (paper) 1741–7899 (online)

be diploid (two pronuclei) (Richings et al. 1999). Recently, ICSI was used to examine the timing and ultrastructure of the events of fertilization in a marsupial (Magarey & Mate 2003b). Sperm collected from electro-ejaculated tammar wallabies (Macropus eugenii ) were injected into oocytes collected from hyperstimulated females. Oocyte activation, sperm head decondensation and pronuclear formation were similar to that of some eutherian species, however syngamy did not occur. While tammar oocytes from hyperstimulated cycles can support most of the events of fertilization, they may not be competent to complete fertilization. The ability of tammar oocytes from natural cycles to fertilize and cleave after ICSI has not been examined. There are no published reports of embryo development after fertilization in vitro in any Australian marsupial or after ICSI in any marsupial. The reproductive biology of the tammar has been extensively studied (Tyndale-Biscoe & Renfree 1987). It is a polyoestrous, monovular, seasonal breeder and has a postpartum oestrus with mating occurring about 1 h after birth (Rudd 1994). In the wild, tammars normally only produce one young a year (Tyndale-Biscoe & Renfree 1987). Ovulation is spontaneous and occurs about 40 h after mating and about 24 h after the luteinizing hormone surge that is DOI: 10.1530/rep.1.00270 Online version via www.reproduction-online.org

596

N M Richings and others

triggered by a rise in oestradiol produced by the Graafian follicle (Harder et al. 1984, Shaw & Renfree 1984, Tyndale-Biscoe & Renfree 1987, Renfree & Lewis 1996). As in all marsupials, fertilization occurs in the upper oviduct, the oocyte moves through the oviduct within 24 h and cleavage occurs in the uterus (Renfree & Lewis 1996). During passage through the oviduct and entrance to the uterus, the oocyte acquires two extra investments, a mucoid coat and a shell coat (Tyndale-Biscoe & Renfree 1987). Early cleavage in the tammar has been studied in vivo and in vitro (Renfree & Lewis 1996). In most marsupials, cell–zona adhesion precedes cell –cell adhesion (Selwood 2001). In the tammar, from the pronuclear stage the cells are attached to the zona pellucida with microvilli and the degree of attachment increases at the late 4-cell stage (Renfree & Lewis 1996). The degree of attachment between individual cells appears to increase at the late 4-cell or early 8-cell stage (Renfree & Lewis 1996). As in all marsupials, the cleavage divisions produce a unilaminar blastocyst with no inner cell mass that expands and develops to form a bilaminar blastocyst (Tyndale-Biscoe & Renfree 1987). In vivo fertilized tammar oocytes develop slower in culture than in vivo (Renfree & Lewis 1996). Most embryos collected from mated tammars at the 1-cell or 2-cell stage progressed to the 4-cell stage in culture. While a few of these embryos progressed to the 8-cell stage, none developed to blastocysts. In contrast, all embryos collected from mated tammars at the 4-cell or 8-cell stage developed to blastocysts (Renfree & Lewis 1996). This indicates that there may be a uterine signal to trigger cleavage past the third division in the tammar. The aim of this study was to establish methods for ICSI and to examine embryo development in vitro after the intra-cytoplasmic injection of sperm, in a marsupial, the tammar wallaby.

Materials and Methods Oocyte collection Oocytes were collected from females undergoing natural cycles at 2 –3 days post-partum, and therefore only a single oocyte could be collected from each animal. Female tammars (Macropus eugenii) (n ¼ 12) were killed by anesthetic overdose (100 mg/kg sodium pentobarbitone) and reproductive tracts were excised under sterile conditions. In females that had not yet ovulated (n ¼ 4), oocytes were dissected from Graafian follicles (4.83 ^ 0.24 mm) on the surface of the ovary into handling medium (HEPES-buffered Dulbecco’s modified Eagles’ medium (DMEM); L -glutamine, 2 mmol/ml; fetal bovine serum (FBS), 10%; penicillin G, 50 IU/ml; streptomycin, 50 mg/ml; all from Thermo Trace, Noble Park, Victoria, Australia). In females that had recently ovulated (n ¼ 8), oocytes were flushed from the oviduct with handling medium. When dissected from follicles, oocytes had few, Reproduction (2004) 128 595–605

if any, cumulus cells attached to the zona pellucida. These cells were easily removed before injection by mechanical manipulation with needles. Oocytes flushed from the oviduct were surrounded by a mucoid coat that ranged in thickness from 9 to 28 mm. All oocytes were examined using an inverted microscope with heated stage, differential interference contrast (D.I.C.) Nomarski optics and an attached video camera and monitor. Each oocyte was examined on the highest magnification at which the entire oocyte could be viewed on the monitor. The thickness of investments and the maximum diameter of the vitellus and the diameter perpendicular to it were recorded. The oocyte was then washed in fresh handling medium, transferred to equilibrated culture medium (DMEM; L -glutamine, 2 mmol/ml; FBS, 10%; penicillin G, 50 IU/ml; streptomycin, 50 mg/ml) and incubated at 37 8C in a humidified gas environment of 5% CO2 in air until the time of micro-injection.

Sperm collection and preparation Male tammars were anesthetized using 10 mg/kg Zoletil (Virbac, Peakhurst, NSW, Australia) and anesthesia was maintained using isoflurane (2% in O2; Abbot Australasia, Kurnell, NSW, Australia). Spermatozoa were collected under sterile conditions from the cauda epididymidis by either hemicastration or epididymal aspiration. In hemicastrated animals, the testis and epididymis were removed, the cauda epididymidis was dissected and washed in phosphate-buffered saline (PBS; Thermo Trace) to remove any blood. Small pieces of epididymis were placed into handling medium for very brief periods of time (2–10 s) to allow motile spermatozoa to be expelled from the tissue. Epididymal aspiration was performed by microsurgery or percutaneous aspiration (Bourne et al. 1995b). For microsurgical aspiration, a small incision was made through the tunica to expose the cauda epididymidis and motile spermatozoa were aspirated from this region using a 26 gauge needle attached to a 1 ml syringe containing 0.1 ml handling medium. For percutaneous aspiration, the cauda epididymidis was palpated and a 26 gauge needle attached to a 1 ml syringe containing handling medium was inserted through the skin into the duct and a small amount of fluid was collected. The epididymal sperm aspirate was resuspended in handling medium and kept at room temperature until injection. After sperm collection, incisions were sutured and males were allowed to recover and subsequently returned to the breeding colony.

ICSI All oocytes were injected with spermatozoa within 6 h of collection. Oocytes were placed individually into drops (15 ml) of handling medium surrounding a central drop of sperm suspension (0.5 –1 £ 106/ml) on a Petri dish (Becton-Dickinson-Falcon, Bedford, MA, USA) and covered with equilibrated mineral oil (Sigma, St Louis, MO, USA). www.reproduction-online.org

ICSI in the tammar wallaby

The micro-injection procedure was based on the method described by Palermo et al. (1992) with modifications as described by Bourne et al. (1995a). The sperm suspension did not contain polyvinylpyrrolidine (PVP) and motile spermatozoa were immobilized in the sperm suspension drop using the injection pipette (Cook Australia, Brisbane, Queensland, Australia) to crush the tail against the dish (Fig. 1a). A single immobilized spermatozoon was aspirated into the injection pipette and transferred to a drop containing an oocyte. The oocyte was held under suction on a holding pipette (Cook Australia) with the polar body 908 from the injection site (Fig. 1b). The spermatozoon was then injected into the cytoplasm of the oocyte with the injection pipette (Fig. 1c–h). The ease of injection was influenced by the investments surrounding the oocyte. The injection of follicular oocytes was straightforward (Fig. 1). The thickness of the mucoid coat in tubal oocytes varied (9–28 mm) and resisted penetration by the injection pipette. Although the method used to inject these oocytes was generally the same as that used to inject follicular oocytes, it required very precise and specific alignment of the holding and injection pipettes.

Oocyte and embryo culture After injection, the oocytes were transferred to 50 ml drops of equilibrated culture medium under mineral oil in four-well culture plates (Nunc, Roskilde, Denmark) and incubated at 37 8C in a humidified gas environment of 5% CO2 in air. The medium was replenished every second day and culture was stopped when no further changes were observed (up to 4 days). The injected oocytes were examined using an inverted microscope with D.I.C. Nomarski optics 16 –20 h after injection for the presence of pronuclei, extrusion of polar bodies and lysis or degeneration. Oocytes were evaluated twice daily for cleavage (cell number, retracting organelles, regularity of membranes), degree of fragmentation and signs of degeneration (retracting organelles, dark or granular cytoplasm, cell lysis). All oocytes and embryos were carefully examined in a number of planes by gently shaking the culture dish, which caused the oocyte/embryo to roll and therefore alter position. Four oocytes that did not cleave after injection were fixed in 4% glutaraldehyde/0.1 M cacodylate buffer overnight at 4 8C, washed and stored in fresh cacodylate buffer (0.1 M) at 4 8C until staining. Oocytes were rinsed thoroughly in PBS containing 0.3% bovine serum albumin (BSA; Sigma) and all debris was carefully removed from the investments. The oocytes were then placed in Hoechst 33258 (5 mg/ml; Sigma; in 0.3%BSA/PBS) for 30 –45 min at 37 8C in the dark. After this incubation, oocytes were rinsed thoroughly in three changes of wash solution (0.1% Tween 20; Sigma; 0.3% BSA in PBS) and mounted on microscope slides in 10% glycerol (Sigma) under coverslips. The slides were examined using a Leitz Dialux 20 microscope equipped with a u.v.-A filter block. After culture, four embryos were www.reproduction-online.org

597

processed by this method and the other embryos were used in another study.

Results Injection and fertilization Once the holding and injection pipettes were precisely aligned, there was no obvious difference in the ease of injection and compression of the vitellus during injection of follicular oocytes (no mucoid coat) or tubal oocytes (with mucoid coat). No oocytes showed signs of membrane lysis or cytoplasmic degeneration after injection. Signs of fertilization (pronucleus-like structures or polar body extrusion) were seen in all injected oocytes (Fig. 2). Two distinct polar bodies were seen in one-third of oocytes (four out of twelve) and one distinct polar body and cytoplasmic fragments that were probably the first polar body were evident in seven oocytes. The day after injection, spherical refractile structures resembling pronuclei were seen in most oocytes (ten out of twelve), but the structural detail of these pronucleus-like structures was difficult to discern because of the character of the cytoplasm (Figs 3c and d and 4a). Cleavage did not occur in oocytes which lacked these pronucleus-like structures. Hoechst staining was successful on three oocytes that were injected but did not cleave. Two oocytes had a nondecondensed sperm head, female chromosomes and a polar body and another oocyte had two pronuclei but no obvious DNA in the polar body.

Early cleavage in vitro Pronucleus-like structures were seen in all oocytes that cleaved. By 52 h post-injection (p.i.) half of the oocytes (six out of twelve) had completed at least the first cleavage division and 25% (three out of twelve) had completed the second cleavage division (Fig. 2). By 72 h p.i., 42% (five out of twelve) of the oocytes had not cleaved, 58% (seven out of twelve) had completed at least the first cleavage division, 33% (four out of twelve) had completed at least the second cleavage division and 17% (two out of twelve) had begun or completed the third cleavage division. No further cleavage occurred after 3 days of culture (Fig. 2). Hoechst staining was successful on three embryos. An oocyte that appeared to be a 2-cell embryo had only reached the pronuclear stage, so although early stages of fertilization had commenced, syngamy had not occurred. A 4-cell embryo had five nuclei and a cluster of dense chromatin (Fig. 4k). Four of these nuclei were possibly in two pairs which may each have been sister nuclei in blastomeres that had not yet undergone cytokinesis. The 8-cell embryo had seven nuclei and one dense cluster of chromatin (Fig. 4l). The cleavage rate varied between individual embryos (Fig. 2). Oocytes from the follicle and oocytes from the oviduct had many similarities in their subsequent development Reproduction (2004) 128 595–605

598

N M Richings and others

Figure 1 ICSI in the tammar wallaby. (a) Immobilization of sperm with the injection pipette. (b) Oocyte held on the holding pipette with the polar body (PB1) positioned 908 to the injection site and sperm visible in the injection pipette (arrow); Z ¼ zona pellucida. (c) The injection pipette inserted through the zona pellucida and oolemma and sperm visible in the injection pipette (arrow). (d) The injection pipette moved to the centre of the ooplasm. (e) Ooplasm has been aspirated up the injection pipette in the direction of the arrow and is visible as granular material in the lumen. (f) Ooplasm returned to the oocyte with sperm. (g) The injection pipette is then carefully and slowly withdrawn from the oocyte. (h) Sperm visible at the centre of oocyte (arrow). Scale bars ¼ 100 mm. Reproduction (2004) 128 595–605

www.reproduction-online.org

ICSI in the tammar wallaby

599

Figure 2 Dimensions of tammar oocytes at collection from either the follicle or the oviduct; timeline of observations after ICSI and treatment after culture. Site: F ¼ follicle, O ¼ oviduct; Vitellus ¼ dimension of oocyte without investments (mm); ZP ¼ thickness of zona pellucida (mm); MC ¼ thickness of mucoid coat (mm); n.a. ¼ not applicable; PBs ¼ number of polar bodies; 1 þ F ¼ 2nd PB and fragments of 1st PB; PN ¼ pronuclei, Y ¼ one or two spherical to ovoid refractile regions seen in cytoplasm; N ¼ no pronucleus-like structures seen in cytoplasm. Time of observations: F ¼ fertilization check (results in Day1 column); X ¼ culture stopped; PCP ¼ post-culture processing; GH ¼ glutaraldehyde fixation and Hoechst staining; G* ¼ glutaraldehyde fixation, but lost during processing; S ¼ cell spreading.

in vitro after ICSI. A developmental sequence from collection of the oocyte through early cleavage was constructed from observations of all oocytes that were injected and cleaved (Figs 3 and 4). After extrusion of the second polar body and formation of the pronuclei (Figs 3c and d and 4a), the vitellus shrank resulting in a larger perivitelline space (Fig. 4b and c). The organelles retracted away from the oolemma in some regions of the cytoplasm (Fig. 4b) and the oolemma became irregular and less distinct in these areas. One of these regions was near the polar bodies and another was opposite the polar bodies. When both regions were obvious the oocyte had a peach-like shape, with indentations on either side of the vitellus (Fig. 4d). This was clearly the cleavage furrow of the first division. At this time fragments were released near the cleavage furrow (Fig. 3e) and cytoplasmic outgrowths resembling ‘boxing-gloves’ or ‘palps’ were often present (Fig. 4e). These stages were also seen in oocytes that failed to complete the first cleavage division. At the 2-cell stage blastomeres were of similar size and a small cytoplasmic fragment, spherical to ovoid in shape and about 33– 50 mm in dimension, was usually present near the cleavage furrow (Fig. 4f). There was only a small cleavage cavity (i.e. perivitelline space) and the cells appeared to be pushed against the zona pellucida, though not attached. If the embryo failed to cleave further, by late on day 2 or day 3 the blastomeres had altered in shape being less ovoid, more irregular and attached to some degree to the zona pellucida (cell–zona adhesion). The cell–zona adhesion in these 2-cell embryos became more www.reproduction-online.org

pronounced during the culture period (Fig. 5a–c). At the 4-cell stage the blastomeres were less regular in size but did not follow an obvious pattern. Cell–zona adhesion was evident throughout most of the 4-cell stage but became more pronounced over time (Fig. 5d – f). Contact between blastomeres (cell–cell adhesion) only occurred in late 4-cell embryos. Only one embryo reached the 8-cell stage (Fig. 4i and j). The blastomeres differed in size and seemed to be in two groups, one small and one large. The two cell types seemed to be at opposite sides or poles of the embryo. Attachment of cells to the zona pellucida was apparent. Despite these similarities, several differences were also seen in embryos from follicular oocytes and embryos from tubal oocytes. At collection, oocytes from the ovary and oocytes from the oviduct differed in the number of investments, in vivo environment to which they had been exposed and post-ovulatory age. All follicular oocytes were surrounded by a zona pellucida while all tubal oocytes had a mucoid coat and a zona pellucida (Fig. 2). All tubal oocytes had been ovulated in vivo, then exposed to the oviductal environment for an unknown period of time. By 48 h p.i., cleavage had occurred in three out of four follicular oocytes and three out of eight tubal oocytes (Fig. 2). The proportion of blastomeres that cleaved at each cell stage differed between embryos from tubal or follicular oocytes (Figs 2 and 6). There was no obvious relationship between the thickness of the mucoid coat and subsequent development, since the oocyte with the thickest mucoid coat (28 mm) and the three oocytes with the Reproduction (2004) 128 595–605

600

N M Richings and others

Figure 3 A developmental sequence of a single oocyte, from collection to the 4-cell stage. (a) Tubal oocyte at collection with zona pellucida (Z), mucoid coat (M) and first polar body (PB1). (b) After ICSI with sperm visible near the centre of the ooplasm (arrow). (c) and (d) 16 h post-injection (p.i.) with pronucleus-like structures (arrows) and polar bodies (PB1, PB2) obvious. (e) 40 h p.i. after completion of the first cleavage division (cells indicated by arrows) and cytoplasmic fragments obvious between blastomeres. (f) At 48 h p.i. the conceptus remained at the 2-cell stage and undulations were apparent in the zona pellucida (UZ). (g) At 63 h p.i. the second cleavage division was complete, the four cells were obvious (arrows) and cell –zona adhesion (CZA) and zona undulations (UZ) were evident. (h) At 89 h p.i. the CZA was more prominent and cell–cell adhesion (CCA) was also evident. Scale bars ¼ 100 mm. Reproduction (2004) 128 595–605

www.reproduction-online.org

ICSI in the tammar wallaby

601

Figure 4 Typical developmental features during cleavage in vitro after ICSI of tammar oocytes. (a) Day 1 p.i.: follicular oocyte with two pronucleus-like structures (arrow heads) and two polar bodies (PBs). (b) Day 1 p.i.: follicular oocyte with retracted organelles (RO) and shrinkage (S) of the vitellus resulting in a large perivitelline space. (c) Day 1 p.i.: tubal oocyte with shrinkage (S) of the vitellus and cleavage furrow (arrow). (d) Day 1 p.i.: tubal oocyte with cleavage furrow (arrows) evident on opposite sides of the vitellus giving a ‘peach-like’ shape. (e) Day 1 p.i.: tubal oocyte with ‘palp-like’ structures that form in the cleavage furrow. (f ) Day 2 p.i.: tubal oocyte that has cleaved to a 2-cell embryo, with the cytoplasmic fragment (CF) often evident after this cleavage division. (g) Day 2 p.i.: follicular oocyte that has cleaved to a 4-cell embryo. Early cell–zona adhesion (CZA) is evident and a large cleavage cavity typical of embryos from follicular oocytes. (h) Day 2 p.i.: tubal oocyte that has cleaved to a 4-cell embryo (arrows indicate cells). CZA is evident. Minimal cleavage cavity and zona pellucida undulations (UZ) typical of embryos from tubal oocytes. (i) and (j) Day 3 p.i.: tubal oocyte that cleaved to an 8-cell embryo. Blastomeres are numbered (1–8) and the lines from each digit indicate the centre of the blastomere. Blastomeres 1 –4 are larger, blastomeres 5–8 are smaller. * ¼ above focal plane. (k) A 4-cell embryo had two polar bodies (arrow heads), a cluster of dense chromatin (1), and five nuclei, four appeared to be in two pairs (2, 3) and the fifth structure was on its own (not visible in this plane of focus). (l) An 8-cell embryo had seven nuclei (1–6, and one not visible in this plane of focus), one dense cluster of chromatin (7) and two polar bodies (arrow head, only one shown in this plane of focus). Z ¼ zona pellucida, M ¼ mucoid coat. Scale bars ¼ 100 mm. www.reproduction-online.org

Reproduction (2004) 128 595–605

602

N M Richings and others

Figure 5 Adhesion of blastomeres to the zona pellucida in 2-cell and 4-cell embryos became more prominent over time. (a –c) A single 2-cell embryo from a tubal oocyte: (a) 43 h p.i., (b) 63 h p.i. and (c) 87 h p.i. (d–f ) A single embryo from a follicular oocyte from the 2-cell to 4-cell stage: (d) 41 h p.i., (e) 54 h p.i. and (f ) 67 h p.i. CZA ¼ cell–zona adhesion, CCA ¼ cell–cell adhesion, CF ¼ cytoplasmic fragment, UZ ¼ undulating zona pellucida. Scale bars ¼ 100 mm. Reproduction (2004) 128 595–605

www.reproduction-online.org

ICSI in the tammar wallaby

603

Figure 6 The number and proportion of blastomeres at each cell stage that cleaved from follicular and tubal oocytes. Light grey, blastomeres did not cleave; dark grey, blastomeres cleaved.

thinnest mucoid coats (9, 16, 20 mm) all failed to cleave (Fig. 2). The cleavage cavity and intercellular spaces between blastomeres seemed to be more expansive in embryos from follicular oocytes (2-cell, see Fig. 5a compared with Fig. 5d; 4-cell, see Fig. 4g compared with 4h). Undulations developed in the zona pellucida of embryos from tubal oocytes after 2 or more days in culture (Figs 3 and 5a–c), but were never evident in embryos from follicular oocytes (Figs 4g and 5d –f).

Discussion This is the first published report of embryo development after ICSI in a marsupial. Mature oocytes collected from preovulatory follicles or flushed from oviducts and injected with epididymal spermatozoa could undergo cleavage as demonstrated by morphology and DNA (Hoechst) staining. At best, the injected oocytes developed to the same stage in vitro as oocytes fertilized in vivo before collection and culture (Renfree & Lewis 1996). We were unable to demonstrate any difference in developmental potential between embryos from tubal oocytes possessing a mucoid coat and follicular oocytes lacking a mucoid coat, although morphological differences were evident in culture. The number of pronuclei and extrusion of the second polar body are used to ascertain the ploidy of fertilized oocytes in vitro (Braude 1987, Vanderhyden & Armstrong 1989, Donoghue et al. 1992). The refractile structures observed in all oocytes that cleaved were assumed to be pronuclei. The presence of the pronuclei and extrusion of extra polar body material in the injected tammar oocytes were both good indicators of fertilization or oocyte activation. It was not possible to determine the number of www.reproduction-online.org

pronuclei and therefore the ploidy of the resultant embryos. Extrusion of the second polar body in the injected oocytes was a clear demonstration that the embryos were not triploid as a result of a retained polar body (digyny). Although the injection procedure can activate the oocyte of tammars, the injected spermatozoon is usually involved in fertilization (Magarey & Mate 2003a). This is also the case in eutherians (Tesarik et al. 1994, Catt & Rhodes 1995). It is therefore unlikely that more than a small proportion of cleaved oocytes were parthenogenotes as a result of oocyte activation by the injection procedure. Tubal oocytes are older than follicular oocytes in terms of post-ovulatory age and they have an extra investment, the mucoid coat (Tyndale-Biscoe & Renfree 1987). Possible functions of the mucoid coat are that it may be a barrier to polyspermy, provide nutrients to the embryo and act as an osmotic stabilizer during development (reviewed in Selwood 2000). Post-ovulatory aging is a continual process beginning within hours of ovulation (Mintz 1967, Kaufman 1983, Williams 2002) and, in eutherians, includes changes in organelles, cytoskeleton, cortical granule release, protein synthesis, spindle structure, plasma membrane and chromosomes (fragmentation, scattering, decondensation) (Tarin 1996). Tubal oocytes may have undergone post-ovulatory aging which is associated with an increased incidence of parthenogenesis in eutherian oocytes (Kaufman 1983). However, if thickness of the mucoid coat reflects time in the oviduct and therefore post-ovulatory age, the oldest oocyte clearly did not undergo parthenogenetic activation since it did not cleave. In tammars, fertilization occurs before the mucoid coat is deposited (Tyndale-Biscoe & Renfree 1987), but the oocytes retain the potential for fertilization and cleavage after this time since half of the tubal oocytes cleaved after Reproduction (2004) 128 595–605

604

N M Richings and others

ICSI. Embryos from tubal oocytes had a small cleavage cavity and closely apposed blastomeres and their morphology was similar to embryos from in vivo fertilization (Renfree & Lewis 1996). Embryos from follicular oocytes had a more obvious cleavage cavity and greater separation between blastomeres. Although cell– zona adhesion seemed to be unaffected in these embryos, the blastomeres were more spherical and there was reduced cell–cell adhesion. It is likely that these differences would affect further development, particularly blastocyst formation. Embryos of Monodelphis domestica generated from IVF of follicular oocytes were able to develop to the 16- to 32cell stage in vitro, but halted development at blastocyst formation (Moore & Taggart 1993). The observations in this study support previous suggestions that the mucoid coat is involved in providing an appropriate environment for the conceptus (Selwood 1989, 2000) by modulating the osmotic pressure in a manner that minimizes the volume of the cleavage cavity and promotes close contact between blastomeres. The morphology of tammar embryos generated from tubal oocytes after ICSI was similar to that of in vivo fertilized tammar embryos (Renfree & Lewis 1996). The cell–zona and cell–cell adhesions that are established in early cleavage of marsupial embryos (reviewed in Selwood 2001) were evident in embryos generated from both follicular and tubal oocytes indicating that the mucoid coat is not essential for the development of these contacts. Both cell –zona and cell–cell adhesions became more pronounced during the culture period, even in embryos that had ceased cleavage, suggesting that these processes are time dependent rather than cell number dependent. In vitro fertilized (injected) oocytes developed more slowly than was reported for in vivo fertilized oocytes (Renfree & Lewis 1996). Only 43% of the injected tammar oocytes in this study that underwent cleavage had reached the 4-cell stage within 24 h of fertilization, compared with 86% of comparable in vivo fertilized oocytes collected from mated tammars and placed in culture (Renfree & Lewis 1996). There was a broad range of cleavage times for each cell stage, reflecting the different cleavage rates of individual embryos. In humans, faster growing embryos have a significantly higher implantation rate than slower growing embryos (Edgar et al. 2000). In this study, the faster cleaving embryos developed further, possibly reflecting the better potential of these embryos. In this study, oocytes from natural cycles cleaved to the 8-cell stage after the injection of a single epididymal spermatozoon. However, oocytes collected from hyperstimulated tammars only develop to the pronuclear stage after injection of ejaculated spermatozoa (Magarey & Mate 2003a,b). The stimulation regimen used in those studies was probably not optimal since oocytes were unable to complete fertilization. Although spermatozoa were prepared by swim-up from the semen after electroejaculation (Magarey & Mate 2003a,b), they were Reproduction (2004) 128 595–605

exposed to seminal and prostatic components. Tammar epididymal spermatozoa have a relatively high rate of endogenous metabolism that can support motility and respiration for prolonged periods of time (Murdoch & Jones 1998). However, preparation of ejaculated spermatozoa by swim-up reduces or impairs their motility, ultrastructure, metabolism and rate of intracellular accumulation of sugars (Murdoch et al. 1999). PVP was used in that tammer ICSI study for sperm immobilization and injection (Magarey & Mate 2003b) and, since no cleavage was reported, it is possible that tammar oocytes or spermatozoa are sensitive to this polymer. PVP is not required to achieve fertilization with ICSI (Bourne et al. 1995a,b) and embryonic development, implantation and live birth rates are similar in embryos generated from ICSI or from standard IVF insemination (Richings et al. 1999). In the present study, cleavage was achieved after ICSI without the use of PVP in the injection procedure, so it is clearly not required. The embryos generated from ICSI were morphologically similar to those previously described from in vivo fertilized embryos (Renfree & Lewis 1996), but further research is needed to confirm the viability of embryos created from sperm injection in marsupials. This study demonstrated the potential for IVF in marsupial oocytes using ICSI, and has laid the foundation for developing techniques such as transgenesis, as well as assisted reproductive techniques for conservation of endangered species.

Acknowledgements We thank everyone in the Wallaby Research Group for help with animal handling, with special thanks to Sue Osborn and Michael Leihy for help with animals in surgery. We thank Dr Debra Gook for valuable advice and technical help with the Hoechst staining technique and for the use of her laboratory, materials and equipment. We thank David Paul for preparation of photographic plates. This work was supported by the Australian Research Council (DP0344941) and an Australian Postgraduate Award to N M R.

References Bourne H, Richings N, Harari O, Watkins W, Speirs AL, Johnston WIH & Baker HWG 1995a The use of intracytoplasmic sperm injection for the treatment of severe and extreme male infertility. Reproduction, Fertility and Development 7 237–245. Bourne H, Richings N, Liu DY, Clarke GN, Harari O & Baker HWG 1995b Sperm preparation for intracytoplasmic injection: methods and relationship to fertilization results. Reproduction, Fertility and Development 7 177–183. Braude PR 1987 Fertilization of human oocytes and culture of human pre-implantation embryos in vitro. In Mammalian Development: A Practical Approach, pp 281–306. Ed. M Monk. Oxford: IRL Press. Burruel VR, Yanagimachi R & Whitten WK 1996 Normal mice develop from oocytes injected with spermatozoa with grossly misshapen heads. Biology of Reproduction 55 709–714. Catt JW & Rhodes SL 1995 Comparative intracytoplasmic sperm injection (ICSI) in human and domestic species. Reproduction, Fertility and Development 7 161–167. www.reproduction-online.org

ICSI in the tammar wallaby Deng MQ & Yang XZJ 2001 Full term development of rabbit oocytes fertilized by intracytoplasmic sperm injection. Molecular Reproduction and Development 59 38–43. Donoghue AM, Howard JG, Byers AP, Goodrowe KL, Bush M, Blumer E, Lukas J, Stover J, Snodgrass K & Wildt DE 1992 Correlation of sperm viability with gamete interaction and fertilization in vitro in the cheetah (Acinonyx jubatus). Biology of Reproduction 46 1047–1056. Edgar DH, Bourne H, Speirs AL & McBain JC 2000 A quantitative analysis of the impact of cryopreservation on the implantation potential of human early cleavage stage embryos. Human Reproduction 15 175 –179. Harder JD, Hinds LA, Horn CA & Tyndale-Biscoe CH 1984 Estradiol in follicular fluid and in utero-ovarian venous and peripheral plasma during parturition and post-partum estrus in the tammar, Macropus eugenii. Journal of Reproduction and Fertility 72 551–558. Hewitson LC, Simerly CR, Tengowski MW, Sutovsky P, Navara CS, Haavisto AJ & Schatten G 1996 Microtubule and chromatin configurations during rhesus intracytoplasmic sperm injection: successes and failures. Biology of Reproduction 55 271–280. Kaufman MH 1983 Early Mammalian Development: Parthenogenetic Studies. London: Cambridge University Press. Magarey GM & Mate KE 2003a Fertilization following intracytoplasmic sperm injection of in vivo and in vitro matured oocytes from an Australian marsupial, the tammar wallaby (Macropus eugenii). Zygote 11 339–346. Magarey GM & Mate KE 2003b Timing and ultrastructure of events following intracytoplasmic sperm injection in a marsupial, the tammar wallaby (Macropus eugenii). Reproduction, Fertility and Development 15 397–406. Martin MJ 2000 Development of in vivo-matured porcine oocytes following intracytoplasmic sperm injection. Biology of Reproduction 63 109 –112. Mate KE, Sidhu KS, Molinia FC, Glazier AM & Rodger JC 2000 Sperm binding and penetration of the zona pellucida in vitro but not sperm-egg fusion in an Australian marsupial, the brushtail possum (Trichosurus vulpecula). Zygote 8 189 –196. Mintz B 1967 Mammalian embryo culture. In Methods in Developmental Biology, pp 379–400. Eds FH Wilt & NK Wessels. New York: Thomas Y Cromwell. Moore HDM & Taggart DA 1993 In vitro fertilization and embryo culture in the grey short-tailed opossum, Monodelphis domestica. Journal of Reproduction and Fertility 98 267–274. Murdoch RN & Jones RC 1998 The metabolic properties of spermatozoa from the epididymis of the tammar wallaby, Macropus eugenii. Molecular Reproduction and Development 49 92–99. Murdoch RN, Jones RC, Wade M & Lin M 1999 The ultrastructure and metabolism of ejaculated tammar wallaby sperm are impaired by swim-up procedures when compared with sperm from the cauda epididymidis. Reproduction, Fertility and Development 11 263–271. Palermo G, Joris H, Devroey P & Vansteirteghem AC 1992 Pregnancies after intracytoplasmic injection of single spermatozoon into an oocyte. Lancet 340 17–18. Pope CE, Johnson CA, McRae MA, Keller GL & Dresser BL 1998 Development of embryos produced by intracytoplasmic sperm

www.reproduction-online.org

605

injection of cat oocytes. Animal Reproduction Science 53 221 –236. Renfree MB & Lewis AM 1996 Cleavage in vivo and in vitro in the marsupial Macropus eugenii. Reproduction, Fertility and Development 8 725–742. Richings N, Bourne H & Baker H 1999 Comparison of embryological and clinical outcomes in embryos generated from conventional IVF and ICSI. In Proceedings of the 11th World Congress on In Vitro Fertilization and Human Reproductive Genetics, Sydney, Australia, Absract O-109 pp 116. Rodger JC 1994 Pre-fertilization gamete maturation events in marsupials. Reproduction, Fertility and Development 6 473–483. Rudd CD 1994 Sexual behaviour of male and female tammar wallabies (Macropus eugenii) at post-partum oestrus. Journal of Zoology 232 151–162. Selwood L 1989 Development in vitro of investment-free marsupial embryos during cleavage and early blastocyst formation. Gamete Research 23 399 –413. Selwood L 2000 Marsupial egg and embryo coats. Cells Tissues Organs 166 208– 219. Selwood L 2001 Mechanisms for pattern formation leading to axis formation and lineage allocation in mammals: a marsupial perspective. Reproduction 121 677–683. Shaw G & Renfree MB 1984 Concentrations of estradiol-17-beta in plasma and corpora lutea throughout pregnancy in the tammar, Macropus eugenii. Journal of Reproduction and Fertility 72 29–37. Sidhu KS, Mate KE, Molinia FC, Berg DK & Rodger JC 2003 Ionic calcium levels in oviduct explant-conditioned media from an Australian marsupial, the brushtail possum (Trichosurus vulpecula) and its relevance to in vitro fertilization. Zygote 11 285 –291. Tarin JJ 1996 Potential effects of age-associated oxidative stress on mammalian oocytes/embryos. Molecular Human Reproduction 2 717 –724. Tesarik J, Sousa M & Testart J 1994 Human oocyte activation after intracytoplasmic sperm injection. Human Reproduction 9 511 –518. Tyndale-Biscoe CH & Renfree MB 1987 Reproductive Physiology of Marsupials. Cambridge: Cambridge University Press. Vanderhyden BC & Armstrong DT 1989 Role of cumulus cells and serum on the in vitro maturation, fertilisation and subsequent development of rat oocytes. Biology of Reproduction 40 720–728. Williams CJ 2002 Signalling mechanisms of mammalian oocyte activation. Human Reproduction Update 8 313–321. Yamauchi Y, Yanagimachi R & Horiuchi T 2002 Full-term development of golden hamster oocytes following intracytoplasmic sperm head injection. Biology of Reproduction 67 534 –539.

Received 2 April 2004 First decision 9 July 2004 Revised manuscript received 30 July 2004 Accepted 9 August 2004

Reproduction (2004) 128 595–605