Metformin: Metabolic Rewiring Faces Tumor Heterogeneity

Abstract

Tumor heterogeneity impinges on all the aspects of tumor history, from onset to metastasis and relapse. It is growingly recognized as a propelling force for tumor adaptation to environmental and micro-environmental cues. Metabolic heterogeneity perfectly falls into this process. It strongly contributes to the metabolic plasticity which characterizes cancer cell subpopulations-capable of adaptive switching under stress conditions, between aerobic glycolysis and oxidative phosphorylation-in both a convergent and divergent modality. The mitochondria appear at center-stage in this adaptive process and thus, targeting mitochondria in cancer may prove of therapeutic value. Metformin is the oldest and most used anti-diabetic medication and its relationship with cancer has witnessed rises and falls in the last 30 years. We believe it is useful to revisit the main mechanisms of action of metformin in light of the emerging views on tumor heterogeneity. We first analyze the most consolidated view of its mitochondrial mechanism of action and then we frame the latter in the context of tumor adaptive strategies, cancer stem cell selection, metabolic zonation of tumors and the tumor microenvironment. This may provide a more critical point of view and, to some extent, may help to shed light on some of the controversial evidence for metformin's anticancer action.

Keywords: ETC; NFkB; OXPHOS; STAT3; autophagy; cancer stem cells; metabolic heterogeneity; metformin; mitochondria; therapeutic target.

Figure 1

Schematic model of metformin action. Systemically, metformin lowers the glucose levels and, indirectly, reduces insulin levels. Metformin crosses the plasma membrane at least partially through organic cation transporters (OCTs and related) and enters mitochondria where it affects complex 1 coupling and causes altered redox status and increased AMP/ATP ratio. This latter activates, allosterically, the kinase LKB1 which phosphorylates AMPK. mTORC1 functions as an environmental sensor and is activated by insulin signaling and growth factors to promote anabolic metabolism and to inhibit autophagy. Metformin-stimulated AMPK reverses mTORC1 actions. mTORC2 is activated by AMPK and stimulates increased glucose entry into muscles and reduced glucose production in the liver. ACC inhibition reduces fatty acid synthesis and may increase collective protein acetylation (increased acetyl-CoA) thus exerting transcriptional modulation. Metformin-stimulated-AMPK controls transcriptionally hepatic gluconeogenesis. Metformin interferes with the increase in HIF-1α and the control of glycolytic genes and GLUT transporters, including the expression of PFK2 which increases the levels of the metabolite fructose-2,6-P₂, allosterically activating PFK-1. Downregulated expression of ACC, FASN, ACLY and SCD1 was promoted by metformin, through interference with SREBP-1c and TR4.

Figure 2

A schematic representation of the main factors modulating the anticancer effects of metformin. The genetic background of the tumor may primarily influence the inter-tumor heterogeneity. The extent of hypoxia and fibrosis and, ultimately, the nutrient availability and the release of oncometabolites may greatly vary among patients as a function of tumor size and location, thus creating spatially defined metabolic zonation phenomena inside the tumor. The composition of the tumor microenvironment (TME) influences and is influenced by therapy-induced changes within the tumor which may trigger epigenetic reprogramming. Altogether, these factors promote the emergence of stress-adapting cancer cell subpopulations, with features of cancer stem cells (CSCs). The tumor stage-dependent change in mitochondrial fitness may contribute to the selective pressure driving the oligo-clonal emergence of CSCs. Those cell subpopulations may ultimately drive tumor metastasis and relapse and may resist metformin treatment as a single agent by acquiring dynamic metabolic states. Please note that the circular arrows indicate that both inter- and intra-tumor heterogeneity are self-fueling processes which strongly influence each other.