The Ludwig Center for Molecular Oncology at MIT has been the driving force for establishing metastasis research into a major theme in the laboratory of Tyler Jacks. Having established a number of genetically-engineered mouse models (GEMMS) of human cancer over the past twenty six years, the Jacks lab has begun to focus increasingly on using these models to understand the molecular and cellular factors that control metastatic spread. The over-arching goals of this research is to translate the finding from mouse cancer models to human cancer with the hope of developing therapies aimed at inhibiting metastatic spread and designed to kill cancer cells at metastatic sites.
Small cell lung cancer (SCLC) is a highly aggressive neuroendocrine lung carcinoma that is characterized by its early and widespread metastasis. Despite decades of research, that has not been significant improvement in treatment outcomes for SCLC patients, and no targeted therapies have demonstrated efficacy in treating the disease. The Jacks Lab utilizes a murine model of SCLC, which involves the conditional deletion of tumor suppressor genes Trp53 and Rb1 in the lung epithelium. These animals develop neuroendocrine lung tumors that frequently metastasize to distant organs such as the liver, mimicking a key feature of the human disease. We have used this model, as well as cell lines derived from the model, to investigate various aspects of SCLC tumor progression and metastasis.
- We have successfully demonstrated the use of the CRISPR-Cas9 system to generate targeted mutations in SCLC tumors in vivo. This has allowed us to rapidly model and functionally validate candidate tumor suppressor genes in SCLC. We are currently using this system to test additional candidate genes that are found to be recurrently mutated in human SCLC tumors, in order to better understand the mechanisms and pathways that contribute to early tumorigenesis.
- In collaboration with the Manalis Lab and Shalek Lab at MIT, we have developed a microfluidics-based system to isolate circulating tumor cells (CTCs) from live tumor-bearing animals in a longitudinal fashion. We have demonstrated the use of this system to track CTC response to drug treatment within individual animals, which eliminates the variability associated with collecting samples from multiple different animals. We are currently using this system to study various aspects of CTC biology and metastatic spread, such as CTC dynamics within the bloodstream, as well as the relationship between primary tumors, CTCs and metastases.
- We have performed a CRISPR-based genetic screen to identify novel SCLC-specific vulnerabilities that may be potential therapeutic targets. The screen was performed in a panel of cell lines established from autochthonous SCLC tumors, including lymph node and liver metastases, using a custom sgRNA library targeting the druggable genome. By analyzing the results of our screen in parallel with other screens in lung adenocarcinoma and pancreatic ductal adenocarcinoma cell lines that were performed in our lab, we have identified a number of unique metabolic vulnerabilities in SCLC, which we are currently validating both in vitro as well as in vivo.
For many years, cancer research has been focused on understanding the mutational changes that occur within cancer cells. However, it is now recognized that tumors should be more appropriately viewed as heterogeneous collection of cells, admixtures of cancer cells and many other cell types that act both to inhibit and promote tumor progression. The considerable variability within tissue microenvironments as well as the multiclonality of cancers leads to intratumoral heterogeneity. This increases the probability of cellular states that promote resistance to therapy and eventually lead to reconstitution of the tumor by treatment-resistant cancer cells, which in some cases have properties of normal tissue stem cells. We have investigated how intratumoral heterogeneity contributes to SCLC metastasis. In particular, we are identifying the mechanism underlying functional collaborations between two different subpopulations of SCLC during different stages of the metastasis process. Abrogation of such collaboration during early stages of tumor development may thus decrease or even prevent metastasis formation.
In addition to exploring intratumoral heterogeneity, the Jacks laboratory is interested in the cancer cell-specific changes that contribute to metastasis. Despite the fact that the majority of lung cancer deaths are due to metastasis, the molecular mechanisms driving metastatic progression are poorly understood. We have previously developed a GEMM that models LUAD in which KRasLSL-G12D/+p53f/f (KP) mice are infected intratracheally with SPC-Cre to induce expression of KRasG12D and deletion of p53. KP mice develop pulmonary adenocarcinomas with substantial tumor burden in the lung and progression to advanced grade tumors and metastases. Approximately half of KP mice show local metastasis to mediastinal lymph nodes or the pleural cavity as early as 18-20 week post tumor induction by intratracheal Cre delivery. Distant metastases to liver or the kidneys are also observed around 20 weeks post-infection. Therefore the KP mouse model is highly physiologically relevant to study lung adenocarcinoma during primary tumor development and metastasis.
Sequencing studies have demonstrated that KP tumors do not acquire additional somatic mutations during tumor progression; therefore, we hypothesize that epigenetic dysregulation may be driving metastasis in this model. We have performed transposase-accessible chromatin using sequencing (ATAC-seq) in a panel of KP cell lines derived from individual non-metastatic and metastatic KP tumors in addition to primary lung tumors and metastases. ATAC-Seq utilizes a hyperactive transposase that incorporates adaptors into open regions of chromatin. We have found that KP metastatic cell lines and tumors have a more open chromatin state, and changes in chromatin state corresponds to dysregulation of key transcriptional programs. Furthermore, we have performed a technique termed HiChIP to compare the 3D chromatin state in the non-metastatic and metastatic KP tumors. We have found differential enhancer usage in the KP metastatic state, suggesting that changes in epigenetic state can drive tumor progression in KP-driven lung adenocarcinoma.
In recent years, the Jacks laboratory has moved into the burgeoning area of tumor immunology and utilizing GEMMs to understand the interactions between the immune system and cancer. We have utilized the before-mentioned autochthonous KP lung adenocarcinoma mouse model, which allows us to express KRASG12D and delete P53 upon intratracheal delivery of Cre recombinase and consequently leads to the formation of lung adenocarcinoma that mimics the genetic and histopathological features of the human disease. Natural killer (NK) cells are effector lymphocytes of the innate immune system and participate in early control of virus infection and tumor immunosurveillance both in humans and mice. In mouse, lack of major histocompatibility complex (MHC) class I expression or presence of NKG2D receptor ligands renders tumor cells susceptible to NK cell mediated cytotoxicity and results in rejection of transplanted tumors. Activating receptors on NK cells, such as NKG2D that recognizes stress-induced ligand expression on tumor cells or Ly49H that recognizes mouse cytomegalovirus (MCMV) encoded protein m157, can bolster NK cell cytotoxicity against target cells. In summary, the engagement of activating or inhibitory receptors upon interaction with the target cell results in a dynamic equilibrium that regulates NK cell activity and dictates whether NK cells kill their target cells or not. The Jacks lab is currently exploring how to utilize these naturally occurring interactions to develop targeted cancer therapeutics.
Lastly, the Jacks Lab is also developing mouse models of pancreas cancer. Pancreas cancer (PDAC) is the third leading cause of cancer-related deaths in the United States and the majority of patients with PDAC are found to have metastatic disease even at the time of diagnosis. We have adapted a technique to induce both pre-invasive (PanIN) and invasive/metastatic pancreatic adenocarcinoma in the autochthonous setting using retrograde pancreatic ductal delivery of lentiviral particles to GEM models. We are using this approach to functionally investigate multiple aspects of pancreatic cancer with a particular focus on metastasis. First, we are leveraging the power of CRISPR/Cas9 to explore the role of putative tumor suppressor genes in PDAC. Intriguingly, we have found that CRISPR-mediated deletion of SMAD4 cooperates with KRAS and TP53 mutations to induce rapid and widespread metastasis, consistent with prior clinical observations. We are collaborating with the Manalis Lab and Shalek Lab at MIT, to isolate and interrogate circulating tumor cells (CTCs) and to study how these relate to their primary tumor and macro-metastases. We are also using this surgical approach to initiate PDAC in the autochthonous setting harboring tumor-specific antigens, which provides a novel system in which we can model the interaction between the immune system and PDAC, both at early stages and at metastatic sites. In addition, we are utilizing murine and human pancreatic organoids to study the role of genetic perturbations on the metastatic cascade and to identify and validate actionable vulnerabilities in PDAC to reduce or prevent metastasis.