A. In Vitro Fertilization in Eutherian Mammals.

In the early 1950s it was recognized that mammalian spermatozoa must undergo physiological and structural changes as a prerequisite to fertilization. These changes in the spermatozoon were termed capacitation. Sperm capacitation normally occurs in the female genital tract but it could be reproduced in cell-free systems containing body fluids from several sources (e.g., follicular fluid, blood serum). For this reason, during the 1960s it was thought that a factor or factors were involved in inducing this phenomenon. In the early 1970s, our research was directed at demonstrating that sperm capacitation and fertilization could be achieved in vitro in a defined physiological solution free of macromolecules

To test these hypotheses, we used a heterologous in vitro fertilization system composed of guinea pig spermatozoa and zona-free hamster oocytes. Guinea pig spermatozoa underwent the acrosome reaction and other physiological changes consistent with capacitation after incubation for a few hours in a defined saline solution. Furthermore, these capacitated spermatozoa fertilized zona-free hamster oocytes both in a defined saline solution or in a saline solution supplemented with bovine serum albumin.

Fig. 1. A zona-free hamster oocyte penetrated by several guinea pig spermatozoa. Fig. 2. Male pronucleus development within a zona-free hamster oocyte.

Fig. 3. Time-lapse dark field micrographs captured on a single photographic plate (top) and corresponding drawing renditions (bottom) of a not activated guinea pig spermatozoon (left) and an activated (capacitated) guinea pig spermatozoon (right).

At the resolution of the phase contrast microscope, fertilizing spermatozoa, fused with the oolemma, decondensed their chromatin and formed structures consistent with male pronuclei (Figs. 1, 2). These observations were corroborated later by others using transmission electron microscopy. Similarly, work by us and others demonstrated that the acrosome reaction and an increase/change in sperm motility (termed activation) obtained under these in vitro conditions were similar to that obtained during in vivo sperm capacitation (Fig. 3.). This work was conducted in the laboratory of my first mentor, Dr. Claudio Barros, while I was an undergraduate student at the Catholic University, Santiago.

B. Ultrastructure of the Acrosome of the Octodon degus Spermatozoon.

The acrosome of mammalian spermatozoa has been the subject of intense research. In the mid 1970s we became interested in investigating at what stage of spermatogenesis the acrosome conferred spermatozoa the ability to fertilize eggs. To conduct these studies we selected the spermatozoon of a small hystrichomorph rodent O. degus. Mature spermatozoa from O. degus have several small protrusions in their acrosome (Fig. 4). The appearance of these protrusions may be used as markers for sperm maturation. Our first task was to capture the sequence of morphological events leading to a fully mature acrosome and thus a mature spermatozoon. Electron micrographs of late spermatids were assembled together to give a complete picture of the stages leading to the formation of a mature acrosome in this organism. This work was conducted in the laboratory of Dr. Claudio Barros while I was a graduate student at the Catholic University, Santiago.

Fig. 4. O. degus late spermatids. Electron micrograph showing the final stages in the assembly of their acrosome.

C. Nuclear Assembly: Post-Fertilization Events in Primary Rabbit Oocytes.

Primary oocytes displaying a germinal vesicle (GV) cannot be fertilized normally; their ability to be fertilized is acquired during the resumption of meiosis just before ovulation. In the mid 1970s phase contrast microscopy showed evidence that seemed to confirm that spermatozoa reaching the perivitelline space could not fuse with and were not incorporated into the ooplasm of rabbit GV oocytes. In contrast, ultrastructural analyses of similar experiments in hamster GV oocytes revealed that they were able to incorporate fertilizing spermatozoa in a normal fashion. We became interested in evaluating, at the ultrastructural level, whether rabbit GV oocytes were indeed penetrated by spermatozoa and if post-fusion events like cortical granule breakdown, chromatin decondensation and pronuclear formation were also

Fig. 5. Transmission electron micrographs of ultrathin sections through the nucleus of rabbit spermatozoa that have penetrated primary oocytes.

occurring normally. We conducted these experiments in vivo by first artificially inseminating the fallopian tubes of surrogate female rabbits with ejaculated spermatozoa. A few hours later, GV oocytes collected from the ovaries of non-stimulated donor female rabbits were also placed into the already inseminated fallopian tubes. After incubation in the upper female genital tract, oocytes were recovered from the fallopian tubes, fixed and prepared for transmission electron microscopy (Fig. 5). Ultrastructural analyses revealed that, spermatozoa do fuse with the oolemma of rabbit GV oocytes, but in contrast to what is observed in mature rabbit oocytes (metaphase II oocytes), cortical granule breakdown was only partial, the nuclear envelope of fertilizing spermatozoa remained intact and chromatin decondensation was delayed implying that resumption of meiosis may bring about functional changes to ooplasmic components. This work was conducted in the laboratory of my second mentor, Dr. Michael Bedford, while I was a research trainee at Weil Cornell Medical College, New York.

D. Nuclear Assembly in a Drosophila Cell-Free System.

During the early 1980s several groups developed cell-free nuclear assembly systems using extracts from activated Xenopus oocytes. These cell-free systems provided important new insights into the structure and function of the nucleus. In the late 1980’s my laboratory began to develop similar in vitro nuclear assembly systems in the fruit fly Drosophila melanogaster.

Fig. 6. Phase contrast micrographs showing the decondensation of demembranated Xenopus spermatozoa and the assembly of nuclei in cell-free extracts from Drosophila embryos.

Specifically, we developed a heterologous nuclear assembly cell-free system that used demembranated Xenopus spermatozoa as a source of chromatin and low speed supernatants (cell-free extracts) from early Drosophila embryos as a source of ooplasmic nuclear components. This cell-free system was capable of decondensing demembranated spermatozoa from various species and assembled nuclei (male pronuclei) starting from demembranated frog spermatozoa (Fig. 6). Using indirect immunofluorescence microscopy, we demonstrated that these in vitro assembled nuclei were assembled in part from Drosophila embryo components originally stockpiled in oocytes. This work was conducted in my laboratory at Stony Brook.

E. Myosin-like ATPases and the Nuclear Pore Complex.

Since the 1960s several groups attempted to identify the protein(s) responsible for an ATPase activity histochemically localized to nuclear pore complex-enriched fractions isolated from vertebrate cells. Although many attempts were made, definitive identification and characterization of the enzyme or enzymes associated with nuclear pore complex-enriched fractions were precluded by the inability to solubilize the ATPase in active form.

Fig. 7. Effect of “cold” dATP and dGTP on ATPase activity (A,B) and direct UV photoaffinity labeling (C,D) of a 188-kD polypeptide band associated with a Drosophila nuclear pore complex-enriched fraction.

As in vertebrate systems, in the early 1980s our attempts to purify the ATPase or ATPases associated with Drosophila nuclear pore complex-enriched fractions were frustrated by enzyme insolubility. To circumvent this problem, we used direct UV-photoaffinity labeling in tandem with protein purification under SDS-denaturation conditions. After UV-dependent cross-linking with [a³²P]ATP, proteins from nuclear pore complex-enriched fractions were solubilized in SDS and subjected to SDS-PAGE and autoradiographic analyses.

Fig. 8. Transmission electron micrographs showing immunogold labeling of Drosophila nuclear pore complexes (arrows) by antibodies directed against a 188-kD myosin heavy chain-like polypeptide associated with a Drosophila nuclear pore complex-enriched fraction. Bar: 50 nm.

Using this strategy, a single radiolableld polypeptide band migrating at approximately 188-kD on SDS-PAGE gels was identified by autoradiography (Fig. 7). Specific antibodies were raised against the photolabeled polypeptide and used to biochemically characterize and localize the ATPase in situ by immunofluorescence and immunogold transmission electron microscopy. Results from these experiments revealed that the enzyme shared properties with myosin heavy chains and it was localized to nuclear pore complexes by both immunofluorescence and immunoelectron microscopy (Fig. 8). Similar studies also revealed the presence of a myosin light chain-like polypeptide associated with Drosophila nuclear pore complex-enriched fractions.

Fig. 9. Transmission electron micrographs showing immunogold labeling of Drosophila nuclear pore complexes (arrows) by antibodies directed against a myosin light chain-like polypeptide. N: nucleoplasm. Bar: 300 nm.

Antibodies directed against the myosin light chain-like polypeptide localized this subunit to Drosophila nuclear pore complexes (Fig. 9). This work was started in the laboratory of my doctoral thesis advisor, Dr. Gunter Blobel and was later continued in my own laboratory at Stony Brook.