Inability of mature oocytes to create functional haploid genomes from somatic cell nuclei

Abstract

Several investigators have reported that nuclei of adult somatic cells undergo meiosis-like divisions after incorporation into maturing and mature oocytes and suggested that the nuclei thus produced could substitute for male or female gamete nuclei for treating human infertility (1Kaneko M. Takeuchi T. Veeck L.L. Rosenwaks Z. Palermo G.D. Haploidization enhancement to manufacture human oocytes.Hum Reprod. 2001; 16: 4-5Crossref Google Scholar, 2Lacham-Kaplan O. Daniels R. Trounson A. Fertilization of mouse oocytes using somatic cells as male germ cells.Reprod Biomed Online. 2001; 3: 205-211Abstract Full Text PDF PubMed Scopus (46) Google Scholar, 3Takeuchi T. Kaneko M. Veeck L.L. Rosenwaks Z. Palermo G.D. Creation of viable human oocytes using diploid somatic nuclei. Are we there yet?.Hum Reprod. 2001; 16: 5Google Scholar, 4Tesarik J. Nagy Z.P. Sousa M. Mendoza C. Abdelmassih R. Fertilizable oocytes reconstructed from patient’s somatic cell nuclei and donor ooplasts.Reprod Biomed Online. 2001; 2: 160-164Abstract Full Text PDF PubMed Scopus (52) Google Scholar, 5Palermo G.D. Takeuchi T. Rosenwaks Z. Oocyte-induced haploidization.Reprod Biomed Online. 2002; 4: 237-242Abstract Full Text PDF PubMed Scopus (43) Google Scholar). The segregation of somatic chromosomes into two groups within maturing oocytes was reported previously by other investigators (6Kubelka M. Moor R.M. The behaviour of mitotic nuclei after transplantation to early meiotic ooplasts or mitotic cytoplasts.Zygote. 1997; 5: 219-227Crossref PubMed Scopus (27) Google Scholar), but correct assortment of chromosome homologues was not shown. Mature oocytes were collected from oviducts of superovulated (B6D2)F1 females. Oocytes were enucleated and injected with single nuclei from mouse (2n = 40) or Chinese hamster (2n = 22) cumulus cells as described elsewhere (7Wakayama T. Perry A.C.F. Zuccotti M. Johnson R.K. Yanagimachi R. Full-term development of mice from enucleated oocytes injected with cumulus cell nuclei.Nature. 1998; 394: 369-374Crossref PubMed Scopus (1925) Google Scholar). Two to 3 hours after nucleus injection, oocytes were activated by 2-hour treatment with 5 mM SrCl2 in Ca2+-free Chatot-Ziomek-Bavister (CZB) medium, followed by 4-hour incubation in regular CZB medium without Sr2+ at 37°C under 5% CO2 in air. The oocytes that emitted a pseudo polar body were cultured in CZB medium containing 0.01 μg/mL vinblastine for 13–14 hours to enforce arrest at the metaphase of the first mitotic division and then were fixed and air-dried for chromosome examination (8Mikamo K. Kamiguchi Y. A new assessment system for chromosomal mutagenicity using oocytes and early zygotes of the Chinese hamster.in: Ishihara T. Sasaki M.S. Radiation-induced chromosome damage in man. Alan R. Liss, New York1983: 411-432Google Scholar). Some oocytes were immunocytochemically stained to observe morphology of spindle before the emission of a pseudo polar body. The number of mouse cumulus chromosomes that remained in activated oocytes after pseudo polar body extrusion was less than or greater than the expected haploid number of 20 chromosomes in 153 of 168 cases (91.1%; Fig. 1A). In 4.2% of the constructs, all 40 chromosomes remained in the oocyte. In 1.2% of the constructs, no chromosomes remained in the oocyte. The number of chromosomes segregated to the oocyte showed a mode of 20 and a mean of 20 ± 8 SD. Immunofluorescence staining demonstrated that the single chromatids of the somatic nucleus were randomly arranged on the spindle before pseudo polar body extrusion (Fig. 1B). To determine whether 8.9% of oocytes could segregate homologous chromosomes from cumulus cell nuclei, we produced additional constructs by injecting Chinese hamster cumulus cell nuclei. Unlike mouse chromosomes, individual Chinese hamster chromosomes are readily identifiable (9Ray M. Mohandas T. Report of the committee on chromosome markers proposal banding nomenclature for the Chinese hamster chromosomes (Cricetulus griseus).Cytogenet Cell Genet. 1976; 16: 83-91Crossref PubMed Scopus (103) Google Scholar). The modal number of Chinese hamster chromosomes retained in the mouse oocyte was 10, with a mean of 12 ± 3 SD. Only 18 of the 128 hamster-mouse constructs (14.1%) retained 11 chromosomes—the expected haploid number of chromosomes in the hamster—and on karyotyping, none had a normal haploid complement of chromosomes. The above results clearly show that authentic haploid nuclei are not generated in activated oocytes following somatic cell nuclear transfer. This result is not surprising because, in the absence of normal prophase events of meiosis (i.e., synapsis and recombination between homologous chromosomes), there is no physical means of ensuring the orderly segregation of homologous chromosomes. Assuming random segregation of homologous chromosomes, the probability of achieving correct chromosome assortment can be calculated by a statistical formula, 2r/nCr, where n and r are diploid and haploid chromosome numbers, respectively. If the chance of obtaining oocytes with a haploid number of chromosomes is taken as 8.9% in the mouse (2n = 40) and as 14.1% in the Chinese hamster (2n = 22), the chance of correct haploidization is approximately 6.79 × 10−7 in the former and 4.09 × 10−4 in the latter. Because the human diploid chromosome number is 46, it is certain that the chance of correct haploidization in humans is less than that in the mouse. Further, even if haploidization were to occur, it is unlikely that the resulting group of chromosomes would provide a functionally normal haploid complement. Normal embryo development requires appropriate genomic imprinting of both parental genomes (10Latham K.E. Epigenetic modification and imprinting of the mammalian genome during development.Curr Top Dev Biol. 1999; 43: 1-49Crossref PubMed Scopus (67) Google Scholar). Even if a normal haploid number of chromosomes were to be retained within the oocyte, it is extremely unlikely that such a chromosome complement would, by simple assortment, contain exclusively maternally or paternally imprinted chromosomes, as would be needed to substitute for authentic gamete genomes. One might propose that imprints could be erased and reestablished in this system. However, given the stage specificity of the imprinting process, it is unlikely that metaphase II oocytes would contain factors needed to erase imprints and reestablish maternal imprints. There is clearly no reason to suppose that an oocyte of any stage could establish a paternal system of imprints. In short, the proposed use of somatic cell nuclear transfer into mature oocytes for the purpose of creating functional haploid genomes appears untenable.