The Origins of Oncomice: a History of the First Transgenic Mice Genetically Engineered to Develop Cancer
Total Page:16
File Type:pdf, Size:1020Kb
Downloaded from genesdev.cshlp.org on October 2, 2021 - Published by Cold Spring Harbor Laboratory Press HISTORICAL PERSPECTIVE The origins of oncomice: a history of the first transgenic mice genetically engineered to develop cancer Douglas Hanahan,1,4 Erwin F. Wagner,2 and Richard D. Palmiter3 1Department of Biochemistry and Biophysics, Diabetes Center, and Comprehensive Cancer Center, University of California at San Francisco, San Francisco, California 94143, USA; 2Research Institute for Molecular Pathology (IMP), Vienna A-1030, Austria; 3Department of Biochemistry, University of Washington, Seattle, Washington 98195, USA This perspective describes the concurrent development cells within normal tissues of a mammalian organism in the 1980s of the first transgenic mice genetically en- could lead to tumor development. In 1992, gene knock- gineered to express dominant oncogenes, involving inde- out technology converged in a similar fashion with tu- pendent researchers who were largely unaware of each mor suppressor genetics in the generation of mice that other’s strategies and progress. We relate the experimen- developed cancers by virtue of lacking tumor suppressor tal designs, the pitfalls and challenges encountered, and gene function (for review, see Jacks 1996). the eventual success in developing distinctive mouse The significance of these early technological innova- models of cancer, wherein tumors arose heritably in vari- tions, which lead to genetically engineered mice en- ous organs. These early oncomice have produced a dowed to heritably develop particular forms of cancer, wealth of new knowledge, become topics of intellectual can be appreciated by considering the scientific/techni- property, and spawned a vibrant field of cancer research cal landscape then, and now. Before these developments, that is revealing mechanisms of tumorigenesis and sug- cancer was largely modeled by tissue culture of cell lines gesting new therapeutic strategies for treating the hu- established from human and animal tumors, and by the man disease. inoculation (transplantation) of such cell lines under the skin of immunodeficient mice, where lump-like solid Supplemental material is available at http://www.genesdev.org. tumors would form. While of clear utility in studying parameters of tumor growth, such models did not nec- In the early 1980s, a technology for generating lines of essarily recapitulate the subtleties observed in human “transgenic mice” carrying cloned genes integrated into tumors arising in different organs, in terms of polymor- the mouse genome was demonstrated to be a tractable phic genetic susceptibility, histological characteristics, and reproducible method (Gordon et el. 1980; Brinster et and progression from benign premalignant lesions to tu- al. 1981; Costantini and Lacy 1981; Wagner et al. 1981; mors of increasing aggressiveness. Moreover, tumor for review, see Palmiter and Brinster 1986). Concur- transplant models, while frequently used as a bench- rently, there was considerable excitement in cancer re- mark to document activity of anti-cancer drugs in pre- search, with the continuing discoveries and molecular clinical therapeutic trials, did not necessarily predict the cloning of viral and then cellular oncogenes. These genes limited benefits or failures seen when some such drugs were causally implicated in particular natural cancers were subsequently tested in clinical trials (Kerbel 2003; and demonstrably capable of inducing transformation of Sharpless and DePinho 2006). In an insightful alternative cultured cells that would form tumors when trans- approach to tumor transplant models, Rudolf Jaenisch planted in appropriate host animals. The two areas of and Beatrice Mintz (Jaenisch and Mintz 1974) sought to research came together when transgenic mice carrying introduce the genome of the SV40 DNA tumor virus into cloned oncogenes were generated and found in some mice via viral infection of early embryos; although the cases to have heritable predispositions to the develop- resultant SV40-containing mice did not transmit the ment of cancer. The first reports appeared in 1984, with SV40 genome to progeny or evidence tumors, the ap- others following in 1985–1987, collectively substantiat- proach reflects on the subsequent development of tu- ing the hypothesis that oncogene expression in normal mor-prone oncomice. In the current era, in addition to such transplant tumor models, the biomedical research community has at its [Keywords: Transgenic mice; mouse models of human cancer; onco- disposal an ever-expanding set of research tools, consist- genes; tumors; oncomouse patents] ing of hundreds of mouse models of organ-specific can- 4Corresponding author. E-MAIL [email protected]; FAX (415) 731-3612. cer, wherein tumors arise out of normal cells resident in Article is online at http://www.genesdev.org/cgi/doi/10.1101/gad.1583307. their natural tissue microenvironments in the context of 2258 GENES & DEVELOPMENT 21:2258–2270 © 2007 by Cold Spring Harbor Laboratory Press ISSN 0890-9369/07; www.genesdev.org Downloaded from genesdev.cshlp.org on October 2, 2021 - Published by Cold Spring Harbor Laboratory Press The origins of oncomice intact immune systems and other tissue-intrinsic barri- adult mice that developed from the injected eggs. This ers to tumorigenesis. These models in some cases reca- experiment led to their first transgenic mice (Brinster et pitulate the genetic mutations, the polymorphic genetic al. 1981). They found that HSV-TK enzyme activity was susceptibility and resistance, the histological subtleties, induced in the liver of the transgenic mice by metals and/or the progressions that are characteristic of differ- known to activate the endogenous Mt1 gene, and that ent types and subtypes of human cancer. These research the transgenes could be transmitted to progency and re- tools provide new insights into the mechanisms of can- tain expression (Palmiter et al. 1982b). cer development and progression, and present platforms Even before the report of their first transgenic mice for testing experimental therapeutics in conditions that was published, Palmiter and Brinster began talking about may more accurately mimic different human cancers. other genes they might express from the Mt1 gene pro- Scientific conferences regularly focus on these engi- moter. Among the candidates were growth hormone and neered mouse models of cancer and draw thousands of oncogenes. They first met in November 1981, when Ri- scientists every year to exchange new results and ideas. chard Palmiter visited Ralph Brinster in Philadelphia The breadth of involvement by the cancer research com- (Fig. 1A). Their transcontinental collaboration was main- munity, the excitement of the new knowledge forthcom- tained by 2- to 3-h telephone calls every Saturday, a tra- ing, and the future promise of this burgeoning field are dition that continued for 15 years. apparent, arguably unimaginable when these mouse Palmiter and Brinster’s development of tumor-prone models were first being developed. At the time it was not transgenic mice came from a convergence of sorts, in- self-evident that such early models could be built or that volving both a deliberate aim (initially unsuccessful) to they would be so extraordinarily useful, in and of them- produce transgenic mice expressing cellular oncogenes selves, as prototypes for a new field of cancer research. and a serendipitous approach (unexpectedly successful) In this historical reflection, we focus on the concur- originally intended to amplify the levels of transgene ex- rent development of the first transgenic mice carrying pression using elements from the DNA tumor virus dominant oncogenes, termed “oncomice,” involving SV40. The convergent approaches illustrate an interest- four independent research groups who were largely un- ing dynamic of scientific discovery. Logical, hypothesis- aware of each other’s strategies, efforts, and progress. We present a perspective on how experimental designs evolved in each of the groups, on the pitfalls and chal- lenges they encountered, and on their eventual success in developing transgenic mouse models of cancer. This treatise does not seek to cover the equally important development of gene knockout technology or the sophis- ticated second- and third-generation models that have been engineered to further refine mouse models as re- flective of human cancers to better enable investigation of mechanisms and testing of new therapies. Rather, our focus is on the early days of oncomice and on the lessons forthcoming from their engineering and initial analysis. Richard Palmiter and Ralph Brinster: transgenic mice developing brain and lymphoid tumors The collaboration between Richard Palmiter and Ralph Brinster began in the fall of 1980, initiated by a series of phone conversations. Their entrée into transgenic mice began with a hybrid gene that Allen Senear, a post-doc- toral fellow in Palmiter’s laboratory, had constructed. The recombinant gene, called MK, consisted of the mouse metallothionein-1 gene (Mt1) promoter fused to transcribe the thymidine kinase (TK) gene from herpes simplex virus (HSV). At that time, the HSV-TK gene was being used both as a selectable gene in TK-null cells and as a convenient reporter gene. Palmiter was interested in asking whether regulation of metallothionein promoter Figure 1. Richard Palmiter, Ralph Brinster, and transgenic mice developing brain tumors. (Top) Palmiter and Brinster are activity by metals or hormones could be conferred