Date of Award

Winter 12-15-2022

Author's School

Graduate School of Arts and Sciences

Author's Department

Biology & Biomedical Sciences (Molecular Cell Biology)

Degree Name

Doctor of Philosophy (PhD)

Degree Type



The enzyme argininosuccinate synthetase 1 (ASS1) catalyzes the condensation of citrulline and aspartate into argininosuccinate as part of the urea cycle and citrulline-nitric oxide cycle. This reaction is essential for mammals to synthesize the amino acid arginine, which is required for all cells. Nearly all human tissues express at least some ASS1, but they import most of their arginine from the extracellular space after it is produced and released by the kidneys. Most solid tumors lack a functional level of ASS1, including over 85% of sarcomas, which are cancers of connective tissues. Published evidence suggests that this provides a proliferation advantage by reducing the consumption of aspartate by ASS1, resulting in a larger supply of aspartate to be used in the production of pyrimidines for nucleic acid synthesis. However, ASS1 deficiency causes these cancers to rely on extracellular arginine for survival and growth, which can be targeted through arginine deprivation therapy. PEGylated arginine deiminase (ADI-PEG20), an enzymatic drug that degrades extracellular arginine to citrulline, is the most widely used arginine deprivation therapy, currently being tested in many clinical trials.While often effective at slowing growth, many cancers, especially sarcomas, gain resistance to ADI-PEG20 in the long term by upregulating their expression of ASS1 and gaining the ability to synthesize arginine from the now abundant citrulline. This is so common because the ASS1 gene is almost always transcriptionally repressed rather than being deleted or mutated. However, it takes time to increase ASS1 expression sufficiently. In the short term, cells also upregulate autophagy to provide a temporary source of arginine for protein translation and survival. These canonical mechanisms have been studied extensively, and ASS1 re-expression is the only published pathway of long-term resistance to arginine deprivation therapy. Many other pathways could theoretically provide long-term resistance to ADI-PEG20, but none have yet been shown to do so. To determine whether ASS1 deficiency truly provides an advantage to sarcomas in vivo, a murine model of spontaneous sarcomas was developed with Ass1 knocked out (KO). Conditional Ass1 KO mice did not develop tumors sooner than control mice, nor did their tumors grow faster. In fact, tumors that expressed high levels of ASS1 initiated earlier and grew faster. These data cast doubt on the importance of previous findings explaining the advantages of ASS1 silencing; they suggest that the main reason for a lack of ASS1 in sarcomas may be inheritance from their tissues of origin. The apparent advantage conferred by ASS1 overexpression in these tumors remains unexplained but is a good target for future study. To characterize the kinetics and heterogeneity of the development of resistance to ADI-PEG20 in ASS1-deficient cancers, a sensor system was developed to monitor the availability of intracellular arginine for protein translation. The sensor consists principally of a genomically integrated gene encoding a reporter protein downstream of an arginine-rich region. Sensor expression is thereby regulated at the translational level, as ribosomes stall or move more slowly at the arginine-rich region, causing reporter protein expression to decrease when arginine supplies are low. Nuclear localization of the reporter and automated imaging allowed tracking of resistance to arginine deprivation in individual live cells. It was found that all ASS1-deficient cancer cells reduced their expression of the sensor when treated with ADI-PEG20 in vitro, followed by a period of heterogeneous recovery of expression. The timing and magnitude of resistance varied widely among individual cells. However, the sensor expression profile was quite different in vivo, as ADI-PEG20 unexpectedly had no impact on the expression of the arginine sensor even while tumor growth was slowed. Ass1 KO tumor cell lines generated from the mouse model described above also did not decrease their expression of the arginine sensor when grafted into syngeneic mice and treated with ADI-PEG20. Unexpectedly, these tumors grew robustly through arginine deprivation therapy in vivo, where they were expected to die as they do in vitro. This suggested that the tumor microenvironment lent strong growth support to the tumors by supplying arginine. This hypothesis was further supported with in vitro experiments showing that ASS1-competent fibroblasts could support growth of Ass1 KO tumor cells during ADI-PEG20 treatment. This growth support effect was found to likely be mediated through the uptake of fibroblast-derived extracellular vesicles (EVs) by macropinocytosis into cancer cells, followed by degradation and recycling of the EV components by autophagy/lysosomal degradation to yield free arginine for the cancer cells to use. Inhibition of this growth support phenomenon was shown to be possible by targeting either EV production and macropinocytosis or autophagy both in vitro and in vivo. These experiments uncovered a novel mechanism of resistance to arginine deprivation therapy, completely independent of intrinsic ASS1 expression, which was previously thought necessary. Further, these results highlight the importance and previously unknown magnitude of the ability of the tumor microenvironment to metabolically support tumors. Finally, this work has opened multiple promising avenues for future research that deserve to be explored.


English (en)

Chair and Committee

Brian A Van Tine

Committee Members

Ron Bose

Included in

Biology Commons