Reply to Flugge: the anti-metastatic potential of methionine restriction in melanoma

To the Editor,

We thank Dr Fluegge (1) for his interest and comments on our recently published study (2), in which we investigated the potential for methionine deprivation on primary and metastatic cancer using the mouse melanoma model.

The methionine auxotrophic nature of a cancer cell has been demonstrated for the first time almost half a century ago, and since then, it is one of the most interesting and, at the same time, one of the least understood phenomena of cancer biology (3). A number of studies investigating the specific mechanisms of methionine sensitivity have been performed and reported in the literature. The case of MeWo/MeWo-LC1 is one of those that indeed deserves close attention. The increased metastatic potential in the derived cell line MeWo-LC1 compared with the poorly metastatic parental MeWo cell line was linked to an impairment of cobalamin metabolism, as described by Liteplo (4). This metabolic change was eventually explained 20 years later with the finding that the promoter of the gene MMACHC is hypermethylated in MeWo-LC1, leading to loss of protein expression (5). The protein encoded by the gene MMACHC is responsible for the cleavage of the cobalt-carbon bond at the β-axial position of cobalamin (6). This same bond is involved in the N₂O poisoning described by Dr Fluegge.

However, the case of MeWo-LC1 seems to be an exception rather than the rule. MeWo-LC1 is a phenocopy of the cblC type of cobalamin disorder, which is in contrast to other methionine-dependent cell lines tested (7) Although disruption of cobalamin metabolism is one avenue leading to methionine dependence, it most likely is not the only one. For instance, a recent study by Strekalova et al. has convincingly demonstrated S-adenosylmethionine biosynthesis as ‘a targetable metabolic vulnerability of cancer stem cells’ both in vitro and in vivo using methionine restriction as an intervention strategy (8).

Manuscript by Thivat et al. (9), cited by Dr Fluegge, provides interesting findings in regards to the 1-day methionine deprivation effects in human participants, including those associated with the safety. However, based on our studies and those published by others, 1-day methionine restriction is not sufficient to substantially deplete the plasma methionine concentration. For instance, Durando et al. report that 1 day of dietary methionine restriction results in 58% depletion in plasma methionine concentrations (10). This, in turn, does not result in detrimental deprivation of tumor from methionine, as it has been reported that cancer cell competitively captures available in circulation methionine (11). Furthermore, as we have demonstrated in our study (2), tumor, in response to methionine deprivation, expedites metabolism of available methionine by upregulation of the Mat2a pathway, to compensate for the lack of endogenous levels of methionine. Therefore, it is clear that to achieve a clinically relevant effect, a much longer period of methionine deprivation will be required.

We emphasize that acquisition of methionine independence by the cancer cell does not indicate failure of this dietary intervention. It has been hypothesized that the cancer cell by selecting methionine independence may also be forced to select for the less malignant phenotype. For instance, Hoffman et al. have convincingly demonstrated that with the acquisition of methionine independence, cancer cells were found to have increased anchorage dependence and serum requirement for optimal growth, decreased cell density in medium with high serum and significant alterations in cell morphology (12). Similar to those findings, metastasis detected in the mouse fed with methionine-deficient diet combined with radiotherapy in our study, despite its considerable size, was confined to pulmonary vein leaving the endothelial vessel layer untouched (2).

Finally, our findings are not limited to animal melanoma models. Accumulating evidence indicates the success of methionine deprivation in other models, including aggressive breast cancer mouse models (combined with chemotherapy) (13,14) and rat model of rhabdomyosarcoma pulmonary metastasis (15).

The absence of a mechanism to explain methionine dependence has been a hindrance to the development of the field. We hope that modern analytical techniques, further development of experimental models coupled with successfully implemented clinical trials, will allow this promising avenue to occupy its niche in oncological practice.

Funding

This work was supported by National Institutes of Health (1P20GM109005), Arkansas Biosciences Institute and the Winthrop P. Rockefeller Cancer Institute.

Conflict of Interest Statement: None declared.