The past few years have witnessed an explosion in genome-wide association studies of cancer susceptibility in human populations. While these studies have revealed many new genetic variants that influence cancer risk, each variant is predicted to have a very small effect on susceptibility, and most heritable factors influencing risk remain to be discovered [26
]. Some risk is conferred by rare variants with large effects, such as the BRCA1
mutations that increase breast cancer susceptibility. Rare variations cannot be detected by genome-wide association studies, which analyze only common (typically >5% minor allele frequency) alleles. Epistatic interactions between common alleles may also contribute to cancer risk. The latter model is supported by studies of mouse models of cancer susceptibility, which have demonstrated that common alleles interact in a complex fashion to influence risk [27
]. However, even in mouse models that combine defined inbred strains with dramatically different tumor susceptibilities under well-controlled environmental conditions, classical mapping studies have not identified the majority of the risk factors [28
The realization that cancer susceptibility is an emergent property of the combinatorial effects of many genes necessitated the development of more complex network-based approaches that integrate classical genetics with gene expression analysis in normal and transformed tissues. We have previously used a systems genetics approach to analyze how gene expression networks in normal whole skin vary between animals that are susceptible or resistant to skin papilloma development. This approach led to identification of pathways controlling mitosis, inflammation and tissue remodeling in normal skin that affect individual susceptibility [8
]. In the present study we have focused on analysis of the rewiring of these normal gene expression networks during development of benign and malignant tumors from the same heterogeneous population of inter-specific backcross mice.
Our data illuminate the dynamic changes in cell populations, both tumor-derived and host-derived, that accompany the evolution of solid tumors. Genomic networks in squamous cell carcinomas are profoundly deregulated compared to normal epithelium and benign papillomas, reflecting major changes in gross tissue organization and signaling. Allelic variation continues to influence tumor gene expression, although this influence is reduced by the somatic alterations accompanying progression. The strongest reduction in tumors is seen in eQTL that act in trans
, possibly due to genomic instability leading to alterations in transcription factor-mediated control of gene expression and the tissue-specific nature of trans
-eQTL. eQTL under the control of cis
-acting elements in general have stronger effects than trans
-eQTL, and they may be more robust in the face of somatic genetic changes because the causal variant affects the gene directly. A recent study compared eQTL detected in hematopoietic cells at four stages of differentiation and demonstrated that many eQTL are unique to each state, and trans
-eQTL are less likely to be conserved between differentiation states than cis
-eQTL were detected in all four states.
We have also identified 'perturbation eQTL', which measure the degree to which changes in levels of gene expression between normal and transformed states are under genetic control. These eQTL reflect genetic control of the changes that occur in response to exogenous damage. In contrast to the steady state eQTL that are mainly cis
-acting, perturbation eQTL act primarily in trans
, similar to a scenario recently described for human lymphoblastoid cells subjected to ionizing radiation [25
]. The mechanistic basis for these observations remain to be determined by isolation and analysis of the trans
-acting factors responsible for these effects.
Genetic and gene expression analyses of tumors reveals features that cannot be detected by analysis of normal tissues, such as the cis
-eQTL controlling expression of Il18
, and Spry2
in tumors but not normal skin. Il18
has an important and complex role in inflammatory and immune responses; it has been reported to have both tumor-promoting and anti-tumor activities in different contexts [30
]. It remains to be determined whether the gain of germline influence over Il18
expression reflects a change in cell populations or a modification in cell-autonomous signaling. The presence of a tumor-specific eQTL for Il18
may reflect differences in the relative proportions of epithelial and inflammatory cells in the tumors, or may be due to rewiring of Il18
signaling during progression.
Unlike Il18, Gzme expression is not detectable in normal skin, and appears in papillomas and carcinomas concomitantly with the influx of innate immune cells. Mice with higher levels of Gzme within their papillomas were relatively resistant to papilloma development, in agreement with a protective role for Gzme, and possibly other granzymes within this gene cluster, in tumor development. In contrast, mice with high levels of Il18 in their papillomas were most susceptible to tumor development. These data suggest that innate immune cell responses against tumors are stronger in animals that carry the SPRET/Ei allele at the Gzme locus, due to a polymorphism resulting in higher Gzme expression. This analysis also suggests opposing roles in tumor susceptibility for Map2k4 and Spry2, genes that exert opposite effects on mitogen-activated protein kinase (MAPK) signaling.
Tumor signaling can be rewired due to oncogenic mutations or loss of tumor suppressor genes, possibly revealing activity of a germline polymorphism that is not evident in normal tissue. The identification of susceptibility genes by a combination of genetic and gene expression analysis of tumors highlights the power of this approach to elucidate the genetic architecture of cancer susceptibility. A combination of genetic and gene expression analysis of human tumors will complement genetic association methods and may identify additional susceptibility factors that cannot be detected using classical methods.