Cell Signaling and Tumorigenesis
Section Leader
James Robinson, Ph.D.
Assistant Professor

The Hormel Institute - James Robinson
The Robinson laboratory primarily is interested in the molecular mechanisms by which oncogenic signaling regulates tumorigenesis, with the ultimate goal of developing and improving existing therapeutic approaches to eliminate cancer. Our lab employs two experienced, full-time postdoctoral fellows Basak Celtikci, M.D. Ph.D.; and Hana Yang, Ph.D. This summer we also were joined by Celeste Underriner, whose Summer Undergraduate Research Experience (SURE) internship was generously funded with an Orville S. Privett Scholarship. As part of the University of Minnesota, The Hormel Institute has full access to the support service at the Masonic Cancer Center (MCC) and Genomics Center (UMGC), and we also have established collaborations with the MCC comparative pathology, bio-statistics and genomics center to assist in our research. We will continue to collaborate with worldwide experts in the fields of cell signaling, comparative pathology, and genetics.

Areas of investigation
Colon Cancer
Our work on colon cancer is funded by a National Institutes of Health (NIH) grant. Colorectal cancer (CRC) is one of the most common cancers worldwide and . after lung and prostate cancer . the leading cause of cancer deaths in the United States, with 132,770 new cases and 49,700 deaths anticipated in 2015. About 75 percent of these cases are sporadic, with no obvious evidence of an inherited disorder. The remaining 25 percent of patients have a family history of CRC that suggests a hereditary contribution; common exposures among family members; or combination of both.

Familial adenomatous polyposis (FAP) is one of the most clearly defined and well understood of the inherited colon cancer syndromes. The vast majority of FAP cases result either from dominantly inherited or de-novo mutations in the Adenomatous polyposis coli gene (APC). FAP is characterized by the development of numerous adenomatous polyps of the large intestine, and it is an excellent model of colorectal tumorigenesis as each adenoma is representative of the first step in CRC: APC loss. Individually, these polyps are histologically indistinguishable from sporadic non-familial colonic adenomatous polyps, and . although the proportion that progress to carcinoma is low . progression is inevitable due to the high number of polyps. For this reason, removal of the colon (a .colectomy.), typically is performed.

The study of FAP has led to advances in the understanding of the genetics of colon cancer and human malignancy. Precise elucidation of steps in FAP tumorigenesis, however, remains elusive. Both human and mice polyps develop without additional genetic alterations other than loss/inactivation of the normal copy of APC. Polyp multicity in both humans and mice can be attenuated greatly by COX-2 inhibition or with anti-inflammatory drugs. Diet, pregnancy, and exercise also might affect polyp numbers and the incidence of sporadic cancer. Our preliminary data has demonstrated that APC loss is insufficient for nuclear accumulation of .-catenin in intestinal epithelial cells. Additional growth signals or mutations also are required for nuclear accumulation of b-catenin and intestinal polyposis. Given that mouse models of FAP develop a multitude of intestinal polyps without additional genetic alterations, these additional signals are likely to arise from adjacent stromal cells. We currently are validating the role of key growth factors HGF and EGF while examining the role of IL17 in colorectal tumorigenesis by assessing if they can promote the nuclear accumulation of .-catenin. Aberrant stromal signaling following loss or haploinsufficiency of LKB1 or SMAD4 is known to drive polyposis in Peutz-Jeghers syndrome and juvenile polyposis syndrome. APC loss of heterozygosity (LOH) is detected in FAP polyp epithelium, and due to FAP not having a prominent stromal compartment, unlike the other syndromes, a proper assessment of stromal LOH was never attempted. If we can show that stromal signaling plays a driving role in tumorigenesis following or pre-empting epithelial LOH of APC, it should be possible to develop targeted therapeutics to block this signaling.

“Our studies will contribute to the development of novel therapies and
improve the outcome for patients with melanoma.”
Dr. James Robinson

Melanoma
Melanoma incidence is increasing at a greater rate than any other cancer. In 2015, it is estimated that 73,870 Americans will be diagnosed with melanoma and about 9,940 will die of the disease. Melanoma typically can be cured through surgery if detected early; however, the five-year survival rate for patients with metastatic disease is less than 15 percent. The MAPK signaling pathway (RAS>RAF>MEK>ERK) is constitutively activated in more than 85 percent of malignant melanomas. Recent advances in melanoma therapy have involved combinations of drugs that target this pathway. Vemurafenib treatment increases median survival by 6 months in approximately 50 percent of patients whose melanomas carry the BRAFV600E mutation. Although the initial response to BRAFV600E inhibition can be dramatic . sometimes causing complete tumor regression . melanomas eventually become resistant and reoccur. Combining MEK and BRAFV600E inhibition improves the response but the majority of patients still eventually experience disease progression (Figure 1). The U.S. Food & Drug Administration (FDA) recently approved humanized anti PD-1 antibodies (nivolumab, pembrolizumab, and opdivo) as a first line of treatment for melanoma, but most patients (about 80 percent) do not experience a clinical response and then are treated with BRAF and/or MEK inhibitors. Combining BRAF inhibition with PD-1 antibodies improves the response rate and overall survival; however, most patients still succumb to the disease. Other approved treatments for melanoma include dacarbazine interleukin-2 (IL-2) and ipilimumab (anti-CTLA-4). These agents, however, produce a response in only a small percentage of patients and their side effects can be pronounced. The increased incidence of melanoma, combined with the poor prognosis of patients with advanced disease, makes it imperative that we increase our understanding of the underlying causes of resistance to targeted therapies to enable the development of better therapeutic strategies.

The Hormel Institute - James Robinson Lab

(Left to right) Basak Celtikci, Celeste Underriner, James Robinson, Hana Yang
Not pictured: Shuxia (Susan) Jiang, Jaclyn Sweetapple

Mouse models that permit regulated expression of oncogenes are useful particularly for modeling the effects of targeted therapies because inhibition of the target. We have developed a novel, retroviral gene delivery mouse model of melanoma that permits control of oncogene expression using tetracycline. This model ideally is suited for testing the role(s) of specific genes in tumor initiation, progression, and maintenance. This model.s versatility eliminates the need to create a new transgenic mouse for testing each new gene. In our model, melanomas can be induced by mutant NRAS, BRAF or MEK in the context of Ink4a/Arf and/or Pten loss.

Importantly, tumors in our model evolve from developmentally normal somatic cells in an unaltered microenvironment. We have used this system to assess the efficacy of targeting NRASQ61R as a therapy for malignant melanoma. Most tumors respond to NRAS inhibition but reoccur after a prolonged latency. Analysis of the recurrent tumors has revealed the most common mechanism of resistance to be over-expression of receptor tyrosine kinases (RTK). In our ongoing research, we seek to define the common mechanisms of resistance to NRAS, BRAF, and MEK inhibition as well as test preemptive and coordinate targeted therapeutic approaches by targeting resistance mechanisms (Figure 2). Our studies will contribute to the development of novel therapies and improve the outcome for patients with melanoma.