Genetically engineered mouse models in cancer research

Abstract

Mouse models of human cancer have played a vital role in understanding tumorigenesis and answering experimental questions that other systems cannot address. Advances continue to be made that allow better understanding of the mechanisms of tumor development, and therefore the identification of better therapeutic and diagnostic strategies. We review major advances that have been made in modeling cancer in the mouse and specific areas of research that have been explored with mouse models. For example, although there are differences between mice and humans, new models are able to more accurately model sporadic human cancers by specifically controlling timing and location of mutations, even within single cells. As hypotheses are developed in human and cell culture systems, engineered mice provide the most tractable and accurate test of their validity in vivo. For example, largely through the use of these models, the microenvironment has been established to play a critical role in tumorigenesis, since tumor development and the interaction with surrounding stroma can be studied as both evolve. These mouse models have specifically fueled our understanding of cancer initiation, immune system roles, tumor angiogenesis, invasion, and metastasis, and the relevance of molecular diversity observed among human cancers. Currently, these models are being designed to facilitate in vivo imaging to track both primary and metastatic tumor development from much earlier stages than previously possible. Finally, the approaches developed in this field to achieve basic understanding are emerging as effective tools to guide much needed development of treatment strategies, diagnostic strategies, and patient stratification strategies in clinical research.

Fig. 1

Genetically engineered mouse models of cancer. Technology is now available to control the overexpression and loss of gene expression, shown in gray, (A) throughout the mouse, as is often the case with classic knockouts, constitutive transgenics, or in some cases knockins; (B) in a specific organ (e.g., the heart) or at a specific time in development, using conditional knockouts or conditional overexpression transgenics or knockins; and (C) in single-cell knockouts through sporadic loss events.

Fig. 2

Modeling chromosomal rearrangements in mouse models of cancer. (A) Interchromosomal rearrangements are generated in translocator mice by inserting single flox recombination sites on different chromosomes. Recombination between the sites exchanges pieces of the two chromosomes. (B) Intrachromosomal rearrangements are generated by inserting two flox recombination sites on the same chromosome oriented in opposite directions. By using flox sites with two different mutations, the resulting flox sites on the inverted chromosome cannot recombine and the inversion becomes stable in the presence of Cre enzyme.

Fig. 3

Measuring mutagenesis levels using the Big Blue^® mouse. Big Blue^® mice have been engineered to carry many copies (30–40) of the bacterial lacI gene, encoding the LacI repressor protein. When Big Blue^® mice are subjected to mutagens or crossed to mice with a mutator phenotype, the lacI gene is mutated. Genomic DNA is isolated from the mice and infected into lacI⁻ bacteria via lambda phage. Bacteria carrying unmutated lacI continue to repress the lacZ gene, whereas bacteria carrying mutant lacI expressed lacZ and form blue colonies on X-gal indicator plates. The ratio of blue to clear bacterial colonies is a measure of mutagenesis.

Fig. 4

Mechanisms of retroviral mutagenesis. Because of the presence of enhancer and promoter sequences in the LTR regions of the retrovirus, there are four basic mechanisms by which retroviruses disrupt gene function. (A) Retroviruses can insert into gene promoters, providing a stronger promoter signal from the LTR and increasing the level of gene expression. (B) Retroviruses can insert into the 3′UTR of a gene, changing mRNA stability thus altering the amount of protein translated. (C) Retroviruses can act as enhancers of gene transcription when inserted upstream, or downstream, of the gene. (D) Retroviruses can mutate genes by inserting into an exon and causing premature truncation of the protein.

Fig. 5

Transposon insertional mutagenesis with Sleeping Beauty. Mice genetically engineered to carry arrays of transposon insertions are crossed to mice genetically engineered to express the transposase. (A) The Sleeping Beauty T2/Onc2 transposon structure is shown (Dupuy et al., 2005) in which the transposon carries two splice acceptors (SA) in opposite directions, a bidirectional poly(A) tail (pA), and the MCSV LTR (MCSV) with a splice donor (SD) that can act as a promoter/enhancer of gene expression. These sequences allow for both gain-of-function and loss-of-function mutations when inserted into a gene locus. The transposon is carried by transgenic mice in arrays of 150–350 transposon copies. Transposon-carrying mice are crossed to mice expressing the Sleeping Beauty (SB) transposase from a variety of different promoters to induce mutagenesis in different tissues. (B) The progeny of transposon and transposase mice express the tranposase and cause both intrachromosomal and interchromosomal hopping of the transposon to mutate the genome.

Fig. 6

Genomic tools for mapping cancer modifiers in mouse. (A) Classic mapping approaches have used two generations of breeding to generate recombinants in the genome, either through F2 intercrosses or backcrosses. Individual mice are then genotyped at polymorphic loci between the resistant and susceptible strain to correlate the genotype to the modifier phenotype. (B) Reference strain panels, such as chromosome substitution strains, recombinant inbred lines, and the Collaborative Cross, provide stable, genotyped recombinants and can be compared for their cancer phenotype to directly map modifiers, or can be crossed to genetically engineered mouse models of cancer to map modifiers in the first generation.

Fig. 7

The cancer stem cell model. In this model, tumors contain rare populations of cells with the ability to repopulate the tumor (cancer stem cells; CSC). These CSCs divide asymmetrically to give rise to more differentiated cells (light gray) and additional CSCs (dark gray). Only the CSCs can give rise to tumors. More differentiated cells can proliferate, but are not sufficient to repopulate the tumor.

Fig. 8

The complexity of the tumor microenvironment. This cartoon shows an example of the tumor cell types making up the tumor microenvironment, based on the current understanding of brain tumors. In addition to differentiated tumor cells that make up the bulk of the tumor, tumor stem cells reside in a specific niche along blood vessels. Normal cells, such as reactive astrocytes, can react to the changes in the local microenvironment to become altered in morphology or gene expression. Inflammatory cells, such as lymphocytes and microglia, invade the region of the tumor and can further change the local microenvironment.