Accepting glowing cells as one’s own allows immune competent metastatic studies

From the Ghajar Lab, Public Health Sciences and Human Biology Divisions

Dr. Candice Grzelak began her postdoctoral research in the lab of Dr. Cyrus Ghajar, part of the Public Health Sciences and Human Biology Divisions, working to understand how breast cancer cells travel to the liver to either stay asleep (‘dormant)’ or grow out to form metastases. Wanting to study the role of the immune system in this process, she used mice with functional immune systems and injected them with breast tumor cells expressing a green fluorescent protein (GFP). GFP and other glowing proteins originate from organisms such as jellyfish or fireflies and are used as tools by biologists to easily track cancer cells’ movement and growth in mouse models. There was just one big problem: the injected GFP-expressing tumor cells were not forming GFP+ metastases in immune competent mice. The Ghajar lab realized this metastatic inhibition was due to the immune system attacking these cells - not because it was necessarily fighting tumor antigen(s) directly, but because it was attacking the cells due to the foreign GFP label! Dr. Grzelak explains this as a “massive roadblock”, not just for her research, but for any researcher broadly studying how cancer cells migrate and grow. Identifying this barrier to studying cancer metastases, Dr. Grzelak took a detour from her initial plans and set out on a journey to find a solution to this GFP immune tolerance problem. In their recent paper published in Cancer Cell, Dr. Grzelak and the Ghajar team overcame this problem by tricking the immune system into thinking that GFP was a natural mouse protein by utilizing a mouse centrally tolerized to GFP.

One of the first clues that the immune system was the culprit behind preventing GFP tumor cell growth, was when they observed that GFP-labeled cells formed metastases readily in mice with compromised immune systems. On the other hand, if metastases did form in mice with competent immune systems, tumor cells tended to have lost GFP expression. This finding supported an immune selection against GFP-expressing cells specifically, rather than against the tumor cells more generally. Consistent with this notion, the authors observed that injection of GFP-labeled tumor cells was accompanied by a robust GFP-specific T cell response. The Ghajar team wondered if they could simply mitigate the GFP-specific response by lowering GFP levels. Here they asked if reduced GFP expression was sufficient to allow for progression and metastasis. Unfortunately, the solution was not quite that easy as the authors found that tumor cells expressing low GFP levels still did not form metastases.

graphical abstract
Schematic of centrally tolerized GFP mouse model that permits GFP+ tumor metastases. Image taken from original article.

The Ghajar team then hypothesized that if they used a mouse engineered to normally express GFP in certain tissues, then the mouse should not have a GFP-specific response to the injected tumor cells. They used a previously developed “Glowing Head” mouse that expressed GFP and luciferase, the protein responsible for making fireflies glow, in a certain part of the head. Although this model had been designed to generate tolerance to GFP, the authors were surprised to find the same degree of tumor rejection encountered in Glowing Head mice. The Ghajar group began to suspect that unobstructed growth of GFP-expressing tumor cells would require a transgenic mouse that is centrally tolerized to GFP, meaning one that expresses GFP in antigen-presenting immune cells in the thymus, such as dendritic cells or medullary thymic epithelial cells. To their great satisfaction, mice with GFP-expressing dendritic cells were indeed centrally tolerized and permitted GFP-expressing tumor cells to now form tumors and metastasize! Because dendritic cells are important immune cells present in the tumor microenvironment, it makes it a bit tricky to tell single dormant tumor cells from host immune cells, without looking at specific cell markers. To further improve this model, Dr. Grzelak is currently working to develop a centrally tolerized mouse model where the host GFP-expressing cells will be “more tissue restricted and less abundant in the tissues I am interested in studying”. These GFP tolerized models will serve as valuable tools for Dr. Grzelak’s own research as well as for cancer research more broadly.

This work was supported by the Department of Defense (DoD) Breast Cancer Research Program, the National Breast Cancer Coalition’s Artemis Project (CMG), the National Institutes of Health and the Susan G. Komen Foundation.

UW/Fred Hutch Cancer Consortium member Cyrus Ghajar and Slobodan Beronja contributed to this research.

Grzelak CA, Goddard ET, Lederer EE, Rajaram K, Dai J, Shor RE, Lim AR, Kim J, Beronja S, Funnell APW, Ghajar CM. Elimination of fluorescent protein immunogenicity permits modeling of metastasis in immune-competent settings. Cancer Cell. 2022 Jan 10;40(1):1-2. doi: 10.1016/j.ccell.2021.11.004. Epub 2021 Dec 2. PMID: 34861158.