Mouse Models of Gynecologic Pathology
http://www.100md.com
《新英格兰医药杂志》
Over the past two decades, the cataloguing of mutated oncogenes and tumor-suppressor genes in various types of human cancers has provided a rich knowledge base for the creation of genetic models. Although powerful in terms of their ability to elucidate gene function in a controlled environment in vivo, these models are prone to the perils of biologic reductionism. Often missed are the boilerplate goals of mimicking a particular human cancer sufficiently to render insight into the early natural history of the disease process and providing a relevant preclinical tool for testing therapies. Ovarian carcinoma has proved exceptionally refractory to genetic modeling; only during the past three years were the first two mouse models announced.1,2 Dinulescu et al.3 have recently described a new model that is a tour de force in genetic modeling of gynecologic abnormalities. They have engineered both a model of the highly prevalent, nonmalignant condition of endometriosis and a model of the endometrioid histologic subtype of invasive epithelial ovarian carcinoma, which sometimes arises within endometriotic lesions.
Mouse models of cancer are typically generated by one of two strategies: the engineering of transgenic animals to express, through the germ line, an activated oncogene, or the use of "knockout" animals that carry a mutationally inactivated tumor-suppressor gene. These relatively simplistic approaches often lead to unpredictable or undesirable outcomes, however. It is not unusual for either approach, but especially the use of genetic knockout animals, to result in embryonic lethality (which is not always uninformative) or adult mice that have a cancer other than that seen in humans who carry or acquire a defect in the orthologous gene.
The development of molecular techniques that allow the targeted manipulation of the mouse genome in a tissue- or time-specific fashion4 has provided a welcome means to overcome these obstacles. For example, one such strategy was used to generate invasive ovarian carcinomas in transgenic mice carrying the transforming oncogene from simian virus 40 DNA, which is under the transcriptional control of a promoter of a gene whose expression is specific to the ovarian epithelial cell.2
A powerful strategy for the conditional inactivation of tumor-suppressor genes typically involves the use of a recombination system composed of a recombinase enzyme and its target sequence. Although there are many variations of this concept, the basic approach is that the desired gene sequence, flanked by stretches of DNA that mediate recombination, is introduced into the germ line and the resultant mice are crossed with mice bearing the Cre recombinase gene, generally under the control of a tissue-specific or drug-inducible promoter. With the evolution of this technique, the Cre recombinase system can now be used to conditionally activate (gain of function) or inactivate (loss of function) gene expression in a time- or tissue-specific manner and thus allow highly accurate mouse modeling of human disease.4
In an effort to refine the existing models of mouse ovarian carcinoma, Dinulescu and colleagues3 set out to create a genetic model of endometrioid ovarian carcinoma, a compelling goal for two reasons. First, existing mouse models give rise to either undifferentiated ovarian cancer or the common serous histologic variant, and models producing more differentiated or less common histologic variants (that is, endometrioid, clear-cell, and mucinous ovarian cancer) would be of great value. Mutational activation of the K-RAS oncogene occurs in a small fraction of endometrioid ovarian carcinomas, and mutational inactivation of the PTEN tumor-suppressor gene is found in a substantial fraction of endometrioid ovarian cancers (but not in the other histologic types), so the candidate genetic targets were in hand. Second, approximately 30 percent of endometrioid ovarian carcinomas (and a smaller fraction of clear-cell carcinomas) are associated with endometriosis,5 but the molecular relationship of this common benign gynecologic entity to ovarian carcinoma remains obscure.
In an elegant application of Cre recombinase technology, Dinulescu et al. attained these goals in a two-phase study (Figure 1).3 First, they generated mice with a mutationally activated K-ras gene, whose expression was induced by the presence of Cre recombinase. After the delivery of a recombinant adenoviral vector expressing Cre recombinase to the ovarian bursa, which induced the expression of activated K-ras, benign endometrioid lesions involving the ovarian epithelium developed in all the mice (Figure 1). Remarkably, however, extensive peritoneal endometriosis, often extending into the abdomen, also developed in approximately half the mice. These lesions met all the established histopathological criteria for the human-disease counterpart, thus representing the first mouse model of spontaneous human endometriosis. Additional data informed a long-standing debate over the histologic origin of endometriosis. The coelomic-metaplasia hypothesis, as its name suggests, involves the genesis of endometrioid lesions within the peritoneal cavity. The data of Dinulescu et al. provide support for an alternative hypothesis, which states that the lesions are initiated by endometrium refluxed through the fallopian tubes into the peritoneal cavity. This model should prove invaluable for increasing our understanding of the natural history of endometriosis, its molecular and cell biology, and its susceptibility to various treatments. It is difficult to overstate the importance of this unique resource to the gynecology-research community, in the light of the high prevalence of this disease and its effect on women's health.
Figure 1. Making a Mouse Model of Gynecologic Disease.
Dinulescu et al.3 genetically engineered mice to carry latent alleles of active mutant K-ras (left side) or active mutant K-ras and inactive Pten (right side). Injection of an adenoviral Cre recombinase construct into the ovarian bursa led to tissue-specific expression of active mutant K-ras, resulting in pelvic endometriosis, or tissue-specific expression of active mutant K-ras and inactivation of Pten, resulting in metastatic endometrioid ovarian carcinoma.
In the second phase of the study, the mice in which K-ras could be activated were crossed with mice that had a Pten gene flanked by stretches of DNA targeted by recombinase. Injection of the adenoviral Cre recombinase construct into the ovarian bursal tissue resulted in mice that expressed mutant K-ras but lacked Pten. In these mice, invasive endometrioid carcinoma of the ovary developed within 7 to 12 weeks after the injection (Figure 1). Gross and histopathological examination at autopsy revealed frequent hemorrhagic ascites and lung metastases, an invariant origin of cancer in the ovarian surface epithelium, and all of the histologic hallmarks of human endometrioid ovarian carcinoma. Thus, this model rigorously recapitulates the corresponding human disease. Important clinical implications are evident from additional data showing that in tumor implants, the proto-oncogenic pathway phosphatidylinositol 3' kinase–Akt–mammalian target of rapamycin (mTOR), which is suppressed by PTEN, and p70 S6 kinase (a downstream target of mTOR) were activated, as was MAP kinase, a downstream effector of oncogenic K-ras, suggesting that this model will be helpful in testing the efficacy of specific inhibitors of these pathways (such as rapamycin and the small molecule PD 184352) in the treatment of this subtype of ovarian carcinoma.
These sophisticated genetic models of gynecologic disease illustrate the enormous power and promise of conditional manipulation of the mouse genome in advancing our understanding of the cause and course of human diseases, as well as providing essential tools for experimental therapeutics.
Source Information
From the Departments of Medicine and Surgery, Memorial Sloan-Kettering Cancer Center, New York.
References
Orsulic S, Li Y, Soslow RA, Vitale-Cross LA, Gutkind JS, Varmus HE. Induction of ovarian cancer by defined multiple genetic changes in a mouse model system. Cancer Cell 2002;1:53-62.
Connolly DC, Bao R, Nikitin AY, et al. Female mice chimeric for expression of the simian virus 40 TAg under control of the MISIIR promoter develop epithelial ovarian cancer. Cancer Res 2003;63:1389-1397.
Dinulescu DM, Ince TA, Quade BJ, Shafer SA, Crowley D, Jacks T. Role of K-ras and Pten in the development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005;11:63-70.
van der Weyden L, Adams DJ, Bradley A. Tools for targeted manipulation of the mouse genome. Physiol Genomics 2002;11:133-164.
Seidman JD, Russell P, Kurman RJ. Surface epithelial tumors of the ovary. In: Kurman RJ, ed. Blaustein's pathology of the female genital tract. 5th ed. New York: Springer-Verlag, 2002:791-904.(Jeff Boyd, Ph.D.)
Mouse models of cancer are typically generated by one of two strategies: the engineering of transgenic animals to express, through the germ line, an activated oncogene, or the use of "knockout" animals that carry a mutationally inactivated tumor-suppressor gene. These relatively simplistic approaches often lead to unpredictable or undesirable outcomes, however. It is not unusual for either approach, but especially the use of genetic knockout animals, to result in embryonic lethality (which is not always uninformative) or adult mice that have a cancer other than that seen in humans who carry or acquire a defect in the orthologous gene.
The development of molecular techniques that allow the targeted manipulation of the mouse genome in a tissue- or time-specific fashion4 has provided a welcome means to overcome these obstacles. For example, one such strategy was used to generate invasive ovarian carcinomas in transgenic mice carrying the transforming oncogene from simian virus 40 DNA, which is under the transcriptional control of a promoter of a gene whose expression is specific to the ovarian epithelial cell.2
A powerful strategy for the conditional inactivation of tumor-suppressor genes typically involves the use of a recombination system composed of a recombinase enzyme and its target sequence. Although there are many variations of this concept, the basic approach is that the desired gene sequence, flanked by stretches of DNA that mediate recombination, is introduced into the germ line and the resultant mice are crossed with mice bearing the Cre recombinase gene, generally under the control of a tissue-specific or drug-inducible promoter. With the evolution of this technique, the Cre recombinase system can now be used to conditionally activate (gain of function) or inactivate (loss of function) gene expression in a time- or tissue-specific manner and thus allow highly accurate mouse modeling of human disease.4
In an effort to refine the existing models of mouse ovarian carcinoma, Dinulescu and colleagues3 set out to create a genetic model of endometrioid ovarian carcinoma, a compelling goal for two reasons. First, existing mouse models give rise to either undifferentiated ovarian cancer or the common serous histologic variant, and models producing more differentiated or less common histologic variants (that is, endometrioid, clear-cell, and mucinous ovarian cancer) would be of great value. Mutational activation of the K-RAS oncogene occurs in a small fraction of endometrioid ovarian carcinomas, and mutational inactivation of the PTEN tumor-suppressor gene is found in a substantial fraction of endometrioid ovarian cancers (but not in the other histologic types), so the candidate genetic targets were in hand. Second, approximately 30 percent of endometrioid ovarian carcinomas (and a smaller fraction of clear-cell carcinomas) are associated with endometriosis,5 but the molecular relationship of this common benign gynecologic entity to ovarian carcinoma remains obscure.
In an elegant application of Cre recombinase technology, Dinulescu et al. attained these goals in a two-phase study (Figure 1).3 First, they generated mice with a mutationally activated K-ras gene, whose expression was induced by the presence of Cre recombinase. After the delivery of a recombinant adenoviral vector expressing Cre recombinase to the ovarian bursa, which induced the expression of activated K-ras, benign endometrioid lesions involving the ovarian epithelium developed in all the mice (Figure 1). Remarkably, however, extensive peritoneal endometriosis, often extending into the abdomen, also developed in approximately half the mice. These lesions met all the established histopathological criteria for the human-disease counterpart, thus representing the first mouse model of spontaneous human endometriosis. Additional data informed a long-standing debate over the histologic origin of endometriosis. The coelomic-metaplasia hypothesis, as its name suggests, involves the genesis of endometrioid lesions within the peritoneal cavity. The data of Dinulescu et al. provide support for an alternative hypothesis, which states that the lesions are initiated by endometrium refluxed through the fallopian tubes into the peritoneal cavity. This model should prove invaluable for increasing our understanding of the natural history of endometriosis, its molecular and cell biology, and its susceptibility to various treatments. It is difficult to overstate the importance of this unique resource to the gynecology-research community, in the light of the high prevalence of this disease and its effect on women's health.
Figure 1. Making a Mouse Model of Gynecologic Disease.
Dinulescu et al.3 genetically engineered mice to carry latent alleles of active mutant K-ras (left side) or active mutant K-ras and inactive Pten (right side). Injection of an adenoviral Cre recombinase construct into the ovarian bursa led to tissue-specific expression of active mutant K-ras, resulting in pelvic endometriosis, or tissue-specific expression of active mutant K-ras and inactivation of Pten, resulting in metastatic endometrioid ovarian carcinoma.
In the second phase of the study, the mice in which K-ras could be activated were crossed with mice that had a Pten gene flanked by stretches of DNA targeted by recombinase. Injection of the adenoviral Cre recombinase construct into the ovarian bursal tissue resulted in mice that expressed mutant K-ras but lacked Pten. In these mice, invasive endometrioid carcinoma of the ovary developed within 7 to 12 weeks after the injection (Figure 1). Gross and histopathological examination at autopsy revealed frequent hemorrhagic ascites and lung metastases, an invariant origin of cancer in the ovarian surface epithelium, and all of the histologic hallmarks of human endometrioid ovarian carcinoma. Thus, this model rigorously recapitulates the corresponding human disease. Important clinical implications are evident from additional data showing that in tumor implants, the proto-oncogenic pathway phosphatidylinositol 3' kinase–Akt–mammalian target of rapamycin (mTOR), which is suppressed by PTEN, and p70 S6 kinase (a downstream target of mTOR) were activated, as was MAP kinase, a downstream effector of oncogenic K-ras, suggesting that this model will be helpful in testing the efficacy of specific inhibitors of these pathways (such as rapamycin and the small molecule PD 184352) in the treatment of this subtype of ovarian carcinoma.
These sophisticated genetic models of gynecologic disease illustrate the enormous power and promise of conditional manipulation of the mouse genome in advancing our understanding of the cause and course of human diseases, as well as providing essential tools for experimental therapeutics.
Source Information
From the Departments of Medicine and Surgery, Memorial Sloan-Kettering Cancer Center, New York.
References
Orsulic S, Li Y, Soslow RA, Vitale-Cross LA, Gutkind JS, Varmus HE. Induction of ovarian cancer by defined multiple genetic changes in a mouse model system. Cancer Cell 2002;1:53-62.
Connolly DC, Bao R, Nikitin AY, et al. Female mice chimeric for expression of the simian virus 40 TAg under control of the MISIIR promoter develop epithelial ovarian cancer. Cancer Res 2003;63:1389-1397.
Dinulescu DM, Ince TA, Quade BJ, Shafer SA, Crowley D, Jacks T. Role of K-ras and Pten in the development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005;11:63-70.
van der Weyden L, Adams DJ, Bradley A. Tools for targeted manipulation of the mouse genome. Physiol Genomics 2002;11:133-164.
Seidman JD, Russell P, Kurman RJ. Surface epithelial tumors of the ovary. In: Kurman RJ, ed. Blaustein's pathology of the female genital tract. 5th ed. New York: Springer-Verlag, 2002:791-904.(Jeff Boyd, Ph.D.)