Help from outside: cysteine to survive in AML

In this issue of Blood, Jones et al reveal that cysteine is critical for the survival of leukemic stem cells (LSCs) in patients with acute myeloid leukemia (AML) by aiding glutathione (GSH) synthesis and maintaining energy metabolism through the electron transport chain complex II (ETC II).¹ 

Because amino acid metabolism plays a crucial role in the sensitivity of the promising combination of venetoclax plus azacytidine,^2,3  the Jones et al study represents a major advance in the understanding of the mechanisms of resistance to these drugs. Specifically, the authors highlight the central role of the microenvironment and cysteine transport within the tumor and metabolism in drug resistance of AML cells.

Metabolic reprogramming is a hallmark of cancer and leukemia biology. Cell and mitochondrial metabolisms (especially oxidative phosphorylation) are key regulators in stem cell biology and therapeutic resistance. Therefore, knowledge of the metabolic pathways involved in the adaptive response of LSCs to stress and being able to identify available nutrients in the tumor microenvironment that fuel these metabolic processes have important implications for more effective therapies. Among these metabolic pathways, uptake and availability of amino acids contribute significantly to cancer development and drug resistance, in particular through maintenance of redox homeostasis.

To cope with oxidative stress, cancer cells must generate antioxidants that convert reactive oxygen species (ROS) into nontoxic molecules. GSH, the most abundant cellular antioxidant, is synthesized in a 2-step enzymatic process requiring 2 molecules of adenosine triphosphate, glutamate, glycine, and cysteine.²  Cysteine is derived from exogenous sources and also from several biosynthetic pathways and is involved in essential metabolic processes (see figure panel A). Interestingly, other groups have shown that endogenous production of cysteine is not sufficient to sustain GSH synthesis, and extracellular cysteine uptake is required through cystine-glutamate antiporter (xCT) or other membrane transporters such as the alanine-serine-cysteine system. In this context, direct disruption of extracellular cysteine-cystine pools by using cyst(e)inase (an engineered human cysteine-degrading enzyme) is a promising therapeutic option in breast cancer and chronic lymphocytic leukemia (CLL).³  Of particular interest, primary CLL cells have been reported to exhibit low expression of the xCT and thus have limited capability to import cysteine, whereas high expression levels of xCT in bone marrow stromal cells allows these cells to efficiently convert cystine to cysteine, which can be released into the microenvironment and used by CLL cells to maintain GSH synthesis.⁴ 

Jones et al explored the importance of cysteine availability in primary AML cells and in particular in LSCs. Previous studies from their laboratory demonstrated that LSCs were characterized by low levels of ROS (ROS-low LSCs⁵ ), which were dependent on amino acid uptake to support oxidative phosphorylation for survival⁶  (see figure panel B). Jones et al expanded their studies and reported that exogenous cysteine is exclusively metabolized into GSH, which is essential for maintaining the viability of ROS-low LSCs. Interestingly, they observed impairment of GSH synthesis but no changes in ROS levels upon cysteine depletion or treatment with cyst(e)inase.

Jones et al further demonstrated that the cysteine requirement for GSH synthesis modulates the activity of succinate dehydrogenase A (SDHA) (a subunit of ETC II) through glutathionylation protein modification in ROS-low LSCs (see figure panel C). Therefore, cysteine starvation leads to decreased SDHA glutathionylation and ETC II activity, resulting in reduced oxidative phosphorylation and selective LSC death, without toxic effects on normal hematopoietic stem/progenitor cells. In a previous study, disruption of ETC II activity through impaired SDHA glutathionylation in AML LSCs was proposed as the mechanism of action of the combination of venetoclax and azacitidine.⁷  Jones et al investigated whether this mechanism could be driven by cysteine availability. The effect of cysteine depletion was selectively observed in both de novo and relapsed/refractory ROS-low LSCs, with the latter being more resistant to the combination of venetoclax and azacitidine,⁸  suggesting differences in the capability of maintaining cysteine and GSH levels. Indeed, they showed that venetoclax with azacitidine decreased cysteine levels in LSCs from patients sensitive to this drug combination but not in patients who were resistant. Therefore, cysteine depletion might serve as an alternative therapy for patients who become resistant to venetoclax with azacitidine.

Protein glutathionylation seems to be crucial for defense against oxidative damage and redox balance. Thus, the posttranslational modification of mitochondrial proteins by glutathionylation might represent a more global protective response to oxidative damage by regulating protein function.⁹  Deciphering mitochondrial protein glutathionylation may also help determine whether SDHA glutathionylation is a specific physiological response to venetoclax plus azacytidine or to cyst(e)inase treatment.

The authors uncovered cysteine as a potential new biomarker of response to venetoclax plus azacitidine and/or resistance, but further clinical studies and analysis of additional primary specimens are needed to confirm this possibility. Intriguingly, residual LSCs escape amino acid depletion induced by venetoclax plus azacitidine by upregulating fatty acid oxidation (FAO), thus preventing oxidative phosphorylation from collapsing.⁶  A more comprehensive examination of the interplay between FAO, amino acid metabolism, and GSH synthesis is likely to provide new insights into oxidative phosphorylation regulation, particularly in the context of chemotherapy resistance in which oxidative phosphorylation plays a crucial role, which is an Achilles heel for treatment.¹⁰  Jones et al also observed the induction of a specific glutathione biosynthesis profile (supplementary Figure 2 in Jones et al) in response to cyst(e)inase treatment, underscoring the role of oxoproline in glutamate synthesis to support GSH production and putative adverse effects of its accumulation (eg, pyroglutamic acidosis).

Finally, the observations of Jones et al add new dimensions to the growing understanding of the role of microenvironmental nutrients available in this niche and those released from stromal cells, endothelial cells, and adipocytes to support survival and functions (such as resistance capacity) of cancer cells that are often exposed to a metabolically challenging environment. This metabolic stress leads to changes in the balance between the endogenous synthesis and exogenous uptake of carbohydrates, fatty acids, or amino acids, which are needed by cells for energy production; redox balance; histone, DNA, RNA, and protein modifications; and biomass. Identifying specific requirements of cancer cells to uptake, use, or degrade metabolic intermediates, in particular after conventional chemotherapies or treatments with novel agents, will help develop more effective treatment strategies and help us better understand AML biology.