Manipulating the genome to study reproduction

Mice with `humanized' zonae pellucidae

Philip.E. Castle1,1 and Jurrien Dean

1 Laboratory of Cellular and Developmental Biology, NIDDK, Building 6, Room B 1–26,National Institutes of Health, Bethesda, MD 20892, USA

The current revolution in biology has provided new insights into mechanisms of mammalianprocreation. However, detailed knowledge of the molecular basis of human reproduction remainslimited because of ethical, regulatory, and societal considerations. Although non-human primates are closely related and can serve as valuable models of human biology, cost and care of primate colonies are formidable and beyond the means of many investigators. Thus, development of small animal models that more closely approximate human reproduction is highly desirable.

Human genes can be introduced into mice using a variety of methodologies (Cohen-Tannoundji and Babinet, 1999Go). One approach is to establish homozygous `knockout' mouse lines from embryonic stem cells in which the gene of interest has been inactivated by insertional mutagenesis. This line can be crossed with a second transgenic line that expresses the homologous human protein. The flexibility of this approach lies in the ability to rescue the `null' phenotype with either normal or mutated forms of a human protein. A potential drawback derives from varying positional effects of the randomlyinserted transgene. A second approach is to directly introduce the transgene into the endogenous gene locus by homologous recombination using `hit and run' or `knock-in' strategies. Although technically more demanding and potentially less flexible, these latter approaches have the advantage that only a single mouse line need be established.

The mouse zona pellucida is composed of three major sulphated glycoproteins (moZP1, moZP2, moZP3), one of which (moZPl) is present as a dimer. Human homologues of ZP2 and ZP3 are 61 and 67% identical to the corresponding mouse roteins (Chamberlin and Dean, 1990Go;Liang and Dean, 1993Go) and a conceptual human protein that is 69% identical to moZP1 is encoded on human chromosome 1 lql2.2 (GenBank: AC004126; D.Hughes, unpublished observations). The presence of an additional human protein ZPB, that is 46% identical to moZP1 (Harriset al., 1994Go) and distinct from the aforementioned conceptual protein (as well as from huZP2 and huZP3) raises the possibility that the formation of the human (and perhaps other) zona pellucida involves four polypeptides. The functional correspondence of the mouse and human zona proteins has been inferred from sequence similarity, but not established experimentally.

Using transgenesis, mouse lines have been established in which moZP3 has been replaced byhuZP3 (Rankinet al., 1998Go). These mice were generated by first establishing genetically altered `null' mice that carry an insertional mutation inZp3 and express moZP1, moZP2, but not moZP3. In the absence of moZP3, no zona pellucida matrix is formed and `null' females are sterile in vivo. The infertile phenotype can be reversed by crossing the `null' line with mice carrying a humanZP3 transgene. These mice have eggs with `humanized' zonae consisting of moZP1, moZP2, and huZP3. Females are as fertile as normal litter mates and `humanized' zonae bind mouse, but not human spermatozoa,in vitro. These mice provide useful models for investigating the molecular basis of fertilization, developing strategies for immunocontraception, understanding the pathogenesis of autoimmune oophoritis and targeting pharmaceutical agents to the ovary.

Molecular basis of mammalian fertilization

In-vitro biochemical studies have suggested functions for individual zona pellucida proteins in mice. ZP3 is considered the primary sperm receptor and induces the acrosome reaction after sperm binding to the zona pellucida. ZP2 is a secondary sperm receptor involved in post-acrosomal induction binding, and ZP1 dimers cross-link ZP2/ZP3 heterodimeric filaments to provide structural integrity to the zona matrix (Wassarman, 1999Go). Based on amino acid homology to moZP3, it has been suggested that huZP3 acts as a primary human sperm receptor and recombinant huZP3 has been reported to induce the acrosome reactionin human spermatozoa (van Duinet al., 1994Go;Brewiset al., 1998Go). Thus, it is surprising that capacitated human spermatozoa do not bind to `humanized' mousezonae of rescue females containing huZP3 (Rankinet al., 1998Go). Even more striking is the continued fertility of the rescue femalesin vivo and the ability of mouse spermatozoa to bind to zonae in the absence of moZP3. These results suggest that either ZP3 by itself is insufficient to dictate order-specific sperm binding or that post-translational modifications of ZP3 within the mouse oocytes functionally convert huZP3 to moZP3.

To sort through these possibilities, mouse lines lacking each of the other zona proteins have been created using embryonic stem cell technology and transgenesis. Additional mouse lines expressing each human zona gene are being crossed into the `null' mice to establish lines in which huZP1 and huZP2 replace the corresponding endogenous mouse zona proteins. Using in-vivo fertility and mouse/human sperm bindingin vitro, it should be possible to determine whether a single zona protein is sufficient to determine the specificity of sperm binding. If not, mouse lines expressing pairs and eventually all of the human zona proteins can be generated by simple crosses. Parenthetically, establishment of mouse lines to which human spermatozoa bind could provide useful reagents for the diagnosis of male infertility.

Immunocontraceptive development

The continued fertility of the huZP3 rescue mice also suggests their utility in the rational design of zona-based immunocontraceptive for humans. Parenteral administration of monoclonal antibodies to huZP3 (but not moZP3) results in long-term, reversible contraception in mice with `humanized' zonae (Greenhouseet al., 1999Go). The advantages of this animal model are low cost, high fertility, short reproductive cycles, and ease of use. Although `passive' immunization has proven effective in rodent models, translation of these results into a peptide vaccine targeting particular epitopes within the zona remains an elusive goal. The huZP3 rescue mice should be useful in evaluating candidate epitopes on huZP3 for immunogenicity, contraceptive efficacy as well as detecting any adverse effect of antibodies on ovarian function and fetal development.

Vaccination of mice with the entire ZP3 protein can result in reversible contraception, but often with an associated autoimmune ovarian dysfunction (Jacksonet al., 1998Go). Thus, there has been considerable interest in devising more targeted peptide vaccines that would elicit a humoral response to specific ZP3 epitopes without inducing the T-cell mediated oophoritis. For example, vaccination with the mouse ZP3 epitope recognized by the monoclonal antibody described above can reversibly inhibit fertility (Millaret al., 1989Go). However, even this 15 amino acid peptide causes autoimmune oophoritis in some inbred mouse lines, although insightful molecular design can obviate this adverse side effect (Rhimet al., 1992Go;Louet al., 1995Go;Tunget al., 1997Go).

If the ultimate goal is to create a human vaccine, it seems advantageous to test epitopes from human zona proteins in a rodent model such as the huZP3 rescue line. Only those peptide epitopes that result in effective, but reversible, contraception in the absence of oophoritis would be considered for further evaluation. Such an approach should result in considerable savings and only the most promising vaccine candidates would progress to further testing of safety and biological efficacy in primates. Whether experimental success in laboratory models will translate into effective, safe immunocontraceptive vaccines for humans remains an open question.

Autoimmune oophoritis and premature ovarian failure

Premature ovarian failure (amenorrhoea, infertility, sex-steroid deficiency and elevated gonadotrophins prior to the age of 40) occurs in ~1% of otherwise healthy women. The causes are multiple, but at least some have been ascribed an autoimmune aetiology (Hoeket al., 1997Go). Certain inbred strains of mice [(C57Bl/6xA/J)Fl, (SWR/JxA/J)F1, BALB/cBy] are susceptible to experimental post-thymectomy autoimmune oophoritis (Taguchi and Nishizuka, 1980Go). Although affected females have antibodies to oocyte-specific antigens including the zonapellucida, the disease has been ascribed to T-cell mediated immunity. Similar T-cell mediatedovarian dysfunction can be provoked in these same mouse lines by vaccinating thymus-intactfemales with ZP3 peptides (Rhimet al., 1992Go). As in the thymectomized mice, adoptive transfer of T-cells from the ZP3 vaccinated mice to naivelitter mates causes oophoritis, but without any observable antibodies to the zona pellucida peptide.Unlike other antigens, immunizing neonatal female mice with the ZP3 peptide does not result intolerance to the disease, although male mice with ovarian grafts develop a state of tolerance (Garzaet al., 1997Go).

The huZP3 rescue mice, in which endogenous mouse ZP3 has been replaced by human ZP3, have unremarkable ovarian histology and are as fertile as normal litter mates. By crossing the huZP3 rescue line into the susceptible mouse background, female mice lacking the endogenous mouse ZP3 peptide but otherwise genetically identical can be created. These susceptible lines with huZP3 would eliminate gender-based differences observed above and should prove a potent model for investigating experimental autoimmune oophoritis. Additionally, vaccination of these mice with human ZP3 peptides may provoke T-cell mediated autoimmune ovarian dysfunction. This could provide a useful model to examine the molecular basis of autoimmune oophoritis in humans and establish a rodent system in which diagnostic reagents and therapeutic modalities for autoimmune premature ovarian failure could be evaluated.

Targeting pharmacological agents to the ovary

The oocyte-specific expression of the zona pellucida raises the possibility of utilizing monoclonal antibodies directed against zona proteins to target the ovary. A 125I-labelled monoclonal antibody specific to a mouse zona protein has been successfully used to image murine ovaries by external scintigraphy (Eastet al., 1984Go). In an analogous manner, it should be possible to target the ovaries in huZP3 rescue mice by using monoclonal antibodies specific to human ZP3. By coupling reagents, including isotopes, to the monoclonal antibody, strategies for directing a wide range of pharmacological agents to the human ovary can be developed. These reagents should be useful in radiological imaging of the ovary, the delivery of ionizing isotopes for ablative therapy (voluntary sterilization, cancer treatment) or for the possible delivery of cancer therapeutic reagents (Bookman, 1998Go).

Currently, only mouse monoclonal antibodies to the human zona proteins have been produced. However, if results in the huZP3 rescue mice are promising, technologies now exist to create `humanized' antibodies to avoid adverse complications arising from the administration of mouse antibodies to humans. Such approaches might be further refined by creating single chain `humanized' immunodiagnostic or immunotherapeutic reagents (Vaughanet al., 1998Go). Although the use of anti-zona antibodies would be limited to pre-menopausal women whose ovaries contain zona pellucida-encased oocytes, they could prove useful for diagnosis and treatment of ovarian dysfunction.

Future directions

The advent of transgenesis has remarkably broadened the range of animal model systems available for the study of human reproductive biology. As evidenced by the huZP3 rescue mice, we now have the ability to establish mouse lines in which each of the endogenous mouse zona proteins is replaced by the human homologue. Theoretically, by simple breeding strategies, these lines can be used to derive mice with `humanized' zona pellucida containing any combination of the human zona proteins. Such mouse lines should provide potent models for investigations into normal and abnormal reproduction, diagnostic testing of human spermatozoa, evaluation of contraceptive strategies and the delivery of imaging or therapeutic reagents to the ovary. Similar transgenic strategies may also be useful in the evaluation of sperm surface proteins involved in order-specific binding of spermatozoa to the zona pellucida. Although these model systems remain imperfect abstractions of human biology, they represent a significant advance in experimental testing and should complement in-vitro and tissue culture techniques.

Notes

1 To whom correspondence should be addressed Back

This debate was previously published on Webtrack 67, May 25, 1999.

References

Bookman, M.A. (1998) Biological therapy of ovarian cancer: Current directions.Semin. Oncol.,25,381–396.[ISI][Medline]

Brewis, I.A., Clayton, R., Browes, C.E.et al. (1998) Tyrosine phosphorylation of a 95 kDa protein and induction of the acrosome reaction in humanspermatozoa by recombinant human zona pellucida glycoprotein 3.Mol. Hum. Reprod.,4,1136–1144.[Abstract]

Chamberlin, M.E. and Dean, J. (1990) Human homolog of the mouse sperm receptor.Proc. Natl Acad. Sci.USA,87,6014–6018.[Abstract]

Cohen-Tannoundji, M. and Babinet, C. (1999) Beyond `knock-out' mice: New perspectives for the programmed modification of the mammaliangenome.Mol. Hum. Reprod.,4,929–938.[Abstract]

East, I.J., Keenan, A.M., Larson, S.M. and Dean, J. (1984) Scintigraphy of normal mouse ovaries with monoclonal antibodies to ZP-2, the major zonapellucida protein.Science,225,938–941.[ISI][Medline]

Garza, K.M., Griggs, N.D. and Tung, K.S. (1997) Neonatal injection of an ovarian peptide induces autoimmune ovarian disease in female mice:requirement of endogenous neonatal ovaries.Immunity,6,89–96.[ISI][Medline]

Greenhouse, S., Castle, P.E. and Dean, J. (1999) Antibodies to human ZP3 induce reversible contraception in transgenic mice with `humanized'zonae pellucidae.Hum. Reprod.,14,593–600.[Abstract/Free Full Text]

Harris, J.D., Hibler, D.W., Fontenot, G.K.et al. (1994) Cloning and characterization of zona pellucida genes and cDNAs from a variety of mammalianspecies:ZPA,ZPB, andZPC gene families.DNA Seq,4,361–393.[ISI][Medline]

Hoek, A., Schoemaker, J. and Drexhage, H.A. (1997) Premature ovarian failure and ovarian autoimmunity.Endocr. Rev.,18,107–134.[Abstract/Free Full Text]

Jackson, R.J., Maguire, D.J., Hinds, L.A. and Ramshaw, I.A. (1998) Infertility in mice induced by a recombinant ectromelia virus expressing mouse zona pellucidaglycoprotein 3.Biol. Reprod.,58,152–159.[Abstract]

Liang, L. and Dean, J. (1993) Conservation of the mammalian secondary sperm receptor genes results in promoter function ofthe human homologue in heterologous mouse oocytes.Dev. Biol.,156,399–408.[ISI][Medline]

Lou, Y., Ang, J., Thai, H.et al. (1995) A zona pellucida 3 peptide vaccine induces antibodies and reversible infertility without ovarianpathology.J. Immunol.,155,2715–2720.[Abstract]

Millar, S.E., Chamow, S.M., Baur, A.W.et al. (1989) Vaccination with a synthetic zona pellucida peptide produces long-term contraception in femalemice.Science,246,935–938.[ISI][Medline]

Rankin, T., Tong, Z., Castle, P.E.et al. (1998) HumanZP3 restores fertility inZp3 null mice without affecting order-specific sperm binding.Development,125,2415–2424.[Abstract/Free Full Text]

Rhim, S.H., Millar, S.E., Robey, F.et al. (1992) Autoimmune disease of the ovary induced by a ZP3 peptide from the mouse zona pellucida.J. Clin. Invest.,89,28–35.[ISI][Medline]

Taguchi, O. and Nishizuka, Y. (1980) Autoimmune oophoritis in thymectomized mice: T cell requirement in adoptive cell transfer.Clin. Exp. Immunol.,42,324–331.[ISI][Medline]

Tung, K.S., Lou, Y.H., Garza, K.M. and Teuscher, C. (1997) Autoimmune ovarian disease: mechanism of disease induction and prevention.Curr. Opin. Immunol.,9,839–845.[ISI][Medline]

van Duin, M., Polman, J.E.M., De Breet, I.T.M.et al. (1994) Recombinant human zona pellucida protein ZP3 produced by Chinese hamster ovary cells inducesthe human sperm acrosome reaction and promotes sperm–egg fusion.Biol. Reprod.,51,607–617.[Abstract]

Vaughan, T.J., Osbourn, J.K. and Tempest, P.R. (1998) Human antibodies by design.Nat. Biotechnol.,16,535–539.[ISI][Medline]

Wassarman, P.M. (1999) Mammalian fertilization: Molecular aspects of gamete adhesion, exocytosis, and fusion.Cell,96,175–183.[ISI][Medline]