Investigators at KU Leuven have discovered that although mTOR signaling was important in primary breast tumors and lung metastases alike, the signals that activated mTOR were different between the two, and mTOR signaling could be inhibited through different mechanisms.

Specifically, serine biosynthesis supported mTOR signaling in metastases, but not in primary tumors.

The team reported its results in the December 18, 2020, online issue of Molecular Cell.

Senior author Sarah-Maria Fendt, who is a biochemist at the VIB Center for Cancer Biology and KU Leuven, told BioWorld Science that the work offers both new scientific insights and new targeting possibilities for metastases, which are what ultimately kills most cancer patients.

Scientifically, "our paper demonstrates that common growth signals, such as the mTOR pathway... are differently important depending on the environment -- in this case, primary tumor versus metastases," she said.

The mTOR complex integrates nutrient sensing with growth decisions, making it a key player in normal and tumor metabolism alike. And nutrient availability differs between primary tumors and metastases, prompting Fendt and her team to take a look at whether mTOR activation differed as well.

The team first compared cells from primary tumors and metastases in a mouse model of breast cancer. They showed that while mTORc1 activity was increased in both cell types compared to normal cells, metastasis-derived cells had stronger mTOR activity than those from primary tumors.

The team next looked at the expression of several different nutrient transporters into cells, to see what might be driving mTOR signaling in metastases.

They found that the pyruvate transporter MCT2 was expressed at high levels specifically in metastasis-derived cells. Treating animals with an MCT2 inhibitor decreased mTOR activity in metastatic cells, but not primary tumor cells.

The team next showed that in metastases, pyruvate uptake activated mTOR by increasing the biosynthesis of the amino acid serine. Silencing the enzyme phosphoglycerate dehydrogenase (PHGDH), which is important in serine synthesis, reduced the growth of metastatic cells but not primary breast tumors.

Although serine is the most abundant amino acid in human proteins and thus important to have around in its own right, serine levels did not appear to be the critical factor in mTOR activation. Instead, activation depended on the levels of a metabolic byproduct of serine biosynthesis, alpha-ketoglutarate.

The work is in line with a paper published earlier in 2020 in Cancer Discovery. In that paper, researchers from NYU Langone Health Center and Weill Cornell Medical College reported that PHGDH inhibition slowed the growth of metastases in the brain, where serine is in short supply, but not of primary tumors.

Therapeutic implications

The work now published by Fendt and her team demonstrates that there is a broader set of circumstances where a therapeutic approach can differ in its effectiveness at fighting primary tumors and metastases. Such differential effectiveness can happen when metastases evolve mutations that set them apart from the primary tumor, but the new studies show that it can be due to metabolic circumstances as well.

Fendt said that her team's work offers a way "to stratify whether a person with a lung metastasis would respond to an mTOR inhibitor. If they are not expressing PHGDH, then they are not sensitive."

Additionally, she said "we can take this one step further" to identify new targets that exploit the metabolic vulnerabilities of metastases.

Pyruvate uptake, she added, is "a very attractive target... [that] could be interesting from multiple perspectives."

Inhibiting pyruvate uptake by blocking its transporter, MCT2, could inhibit both serine biosynthesis and mTOR activation, which "could maybe prevent metastases and treat metastases with one drug," she said.

Scientifically, the findings give new insights into the metastatic process. They increase the understanding of why certain tumor types are more likely to metastasize to certain sites, and may ultimately enable the prediction of where a tumor is likely to metastasize to, based on its metabolic characteristics, that could help clinicians understand who needs to be monitored closely for possible recurrences, and where.

The team also plans to look at dormant cells, to see whether their metabolic state is similar to metastatic cells. Breast cancer is unusual because patients relapse at a constant low rate for decades after the original tumor has been removed. For most other tumor types, if patients do not relapse within 5 years, their risk of a renewed diagnosis is no higher than that of the general population.

Dormant cells, Fendt said, are "complicated to study -- in vitro, it is very hard to put these cells into a dormant state, and in vivo, there are very few of them."

But "for breast cancer, this would really make a big difference, finding something that can kill dormant cells." Her prediction was that dormant cells "will have a very particular metabolism."

Unlike primary tumor cells, their metabolism is not optimized for cycling through division at a breakneck pace.

But unlike cells in a fully formed metastasis they "need to withstand a harsh environment, because they lack the extracellular support system" of the tumor microenvironment, Fendt explained: "They need particular metabolic pathways to support that." (Rinaldi, G. et al. Mol Cell 2020, Advanced publication).