Metabolic sensing by p53: Keeping the balance between life and death

There is no life without stress, and there is no life without mechanisms for adaptation to stress. Stress perturbs the balance of a living system, thereby scrutinizing the plasticity of its regulatory networks. These networks shape the composition and behavior of living systems. Biological networks serve the adaptability of the whole system, wherein multiple sensors and effectors cooperate to achieve maximal efficiency. Such interlacing cooperation needs to be controlled in relays (effectors) that integrate incoming signals and direct outgoing signals. The list of such powerful cellular effectors is short. In multicellular organisms, proteins of the p53 family (1) are well-studied examples of such effectors; as transcriptional factors, they balance the regulation of cell fate between proliferation, differentiation, and death. p53 is a relatively short-lived protein whose stability is regulated by multiple counteracting mechanisms targeting p53 to, or preventing it from, ubiquitin-dependent (2) or -independent (3) proteasomal degradation. The article by Khutornenko et al. (4) in PNAS reports a surprising connection between p53 stabilization (activation) in cancer cells and mitochondrial function.

The mitochondrion is a versatile enzymatic engine that senses the metabolic requirements of cells and utilizes biochemical pathways in accordance with the cells’ proliferative or differentiation status or with a normal or transformed state. Regulation of mitochondrial activity in transformed cells also requires adaptation to changing aerobic and anaerobic conditions depending on energy supply through blood vessels and metabolic symbiosis in tumor cells (5). Because many metabolic pathways intersect in mitochondria, and because some of them directly rely on electron flow through the electron transport chain (ETC) (e.g., de novo pyrimidine synthesis), the communication between this organelle and the cell nucleus is essential. The cross-talk between these organelles has been best studied in yeast. During retrograde response (RTG) (6), for example, three proteins, Rtg2 (sensor) and Rtg1/Rtg3 (transcription factor heterocomplex), cooperate to switch ATP production from respiration to glycolysis and glutaminolysis. The signal flow from mitochondria to the nucleus is mediated by the activation and translocation of transcription factors from the cytoplasm into the nucleus and aims at adjusting gene expression according to the metabolic state of the cell. In this pathway, Rtg2 plays a remarkable function by keeping the balance between adequate activation of the responsive genes and protection of cells against deleterious genomic instability, which accompanies the RTG (7). In human cells, exploration of the regulatory network-coordinating metabolism with gene expression is the focus of current research (8). The emerging evidence for p53 being a transcriptional regulator of metabolic integration (9, 10) came as a surprise at first glance.

The study by Khutornenko et al. is an important contribution to understanding the metabolic function of p53 (4); they analyzed protein levels of p53 in cancer cells with functionally suppressed mitochondria. These investigators found that of various pharmacological interventions disturbing mitochondrial activity, only interruption of the electron transfer to complex III elicited accumulation of p53, followed by apoptotic death (Fig. 1). The causative chain of events starts with the uncoupling of complex III and inner mitochondrial membrane-bound dihydroorotate dehydrogenase (DHODH), which converts dihydroorotate to orotate using ubiquinone as a direct electron acceptor. By stalling de novo pyrimidine synthesis at this step (which can also be simulated by a specific DHODH inhibitor), two cellular enzymes, NQO1 and NQO2, become involved in the process of stabilization and nuclear accumulation of p53. NQO1 and NQO2 are basically responsible for the protection of cells against chemical poisons, but they also regulate p53’s lifetime (3, 11), thereby establishing a link between the metabolic state of a cell and the “guardian” activity of p53. How pyrimidine deficiency is sensed by NQO1 and NQO2 remains to be elucidated. A plausible explanation might be deduced from the link between NADH, NQO1, and p53. Interaction of NQO1 with p53 is regulated by NADH, whose levels decide between ubiquitin-independent 20S proteasomal degradation (at low NADH) and protection from degradation (at high NADH) (3). Interfering with electron transfer will uncouple the NADH shuttle system, and cytoplasmic NADH, produced during glycolysis, will accumulate and induce p53 stabilization. Addition of orotate and uridine to the culture medium, in turn, abrogates p53 stabilization, probably attributable to “normalization” of cellular physiology after resumption of RNA synthesis, resulting in increased NADH consumption.

Fig. 1.

The mitochondrial step of de novo pyrimidine biosynthesis is linked to p53 function (details provided in the text). Under physiological conditions, metabolic stress (e.g., pyrimidine deficiency) is balanced by an antiapoptotic response of p53. The proapoptotic p53 response is an extreme outcome attributable to nonphysiological conditions. UQ, ubiquinone; III, complex III; DHO, dihydroorotate; NQO1, NAD(P)H:quinone oxidoreductase 1; NQO2, NRH:quinone oxidoreductase 2; 20S, 20S proteasome.

Drug-induced pyrimidine deficiency activates the p53 pathway, leading to apoptotic cell death, which fits into the dominating picture of p53 as the executioner of a stress response (Fig. 1). However, the link of p53 with RTG provides another explanation in which p53-induced apoptosis is not the primary goal of the p53 response but, instead, an extreme outcome, because the level of induced metabolic deficiency exceeded a physiological, and thus tolerable, threshold. Transcriptional regulation by p53 is mostly considered within the framework of a response to a stressor and leads to the activation (enhancement) and repression (dampening) of multiple target genes, controlling their activity on demand (12). Because of the importance of the p53 network in preventing cell transformation, the p53 gene is frequently mutated (or deleted) in tumor cells as a means to escape from p53’s guardian function (13). It is thus generally assumed that in tumors retaining WT p53 expression, its physiological function is more or less compromised by different mechanisms. However, loss of p53 is not necessarily always advantageous in terms of cell transformation and tumor growth. Otherwise, it is difficult to conceive why expression of WT p53, which can also be at least partially activated in tumor cells and always acts in an antiproliferative manner on activation, is retained in some tumors. In this regard, one has to consider that p53 not only controls the expression of genes whose products regulate cell cycle progression and apoptosis but the expression of genes coding for components of the metabolic machinery (14). Transcriptional activation by p53 of genes encoding the negative regulator of glycolysis (TIGAR), the component of complex IV of the ETC (SCO2), and the recently identified mitochondrial phosphate-activated glutaminase (GLS2) (14) shifts the metabolic energy production from the glycolytic pathway to mitochondrial respiration and glutaminolysis. As a consequence, loss of p53 activity favors the use of glycolysis for energy production, thereby contributing to metabolic reprogramming and creation of a transformed phenotype (8, 15). On the other hand, retention of a WT p53 in transformed cells that acts as an intelligent metabolic regulator and adapts their proliferation rate according to their metabolic state would be advantageous.

The inevitable consequence of mitochondrial respiration is the production of reactive oxygen species (ROS), including hydrogen peroxide, superoxide, hydroxyl radicals, and singlet oxygen. ROS can act as signaling molecules, but they largely exert a damaging effect. While promoting mitochondrial activity, p53 simultaneously balances the deleterious effect of side products. Genes coding for antioxidant defense enzymes [e.g., aldehyde dehydrogenase (ALDH4), glutathione

p53 might play an important role as a “guardian of metabolic balance.”

peroxidase (GPX1), Mn-superoxide dismutase (SOD2), sestrins (SESN1 and SESN2), and glutaminase (GLS2; responsible for generation of the major endogenous antioxidant reduced glutathione through glutaminolysis)] are transcriptional targets of p53 (9, 14). Thus, the picture emerges that the primary function of p53 is to mediate the balance between interwoven opposing pathways, compensating for the deleterious consequences of one pathway by an “antidote” function of the complementary pathway, thereby ensuring survival of the cell, rather than to promote growth arrest and death (9). However, p53 will induce cellular collapse when effector actions cross a threshold that exceeds a tolerable maximum and results in deleterious effects. In the case of massive cellular damage, cell death will occur, initiated by the proapoptotic function of p53, among others. Under still physiological conditions, however, p53 can antagonize apoptosis by activating, for example, the expression of the TRIAP1 gene (coding for TP53-regulated inhibitor of apoptosis 1) (16), and thereby maintain a balance between cell life and death.

In many aspects, the physiology of tumor cells and normal cells is inherently similar because it relies on the same principles. Indeed, the heterogeneity of differentiation and metabolic states observed in tumors mimics the respective normal tissues and their complexity. How complex is tumor tissue? Is it more complex than the normal tissue? At least at the genetic and epigenetic levels, tumor cells display a higher complexity and variability of successful “scenarios,” promoting and maintaining the growth and performance of a tumor as a tissue-like system. The interactions among tumor cells themselves and between tumor cells and their microenvironment composed of stromal, endothelial, and immune cells provide a basis to consider a tumor as an “ecosystem” and cancer as a systems biology disease (17). In such a tumor cell “ecosystem”, p53 might play an important role as a “guardian of metabolic balance” by adapting the physiological state of each single tumor cell within the system to a changing environment, thereby defining a long-sought after role for p53 under physiological conditions.

Balancing the metabolic state of normal and cancer cells thus emerges as an important effector function of p53 rather than simply another node in the p53-centered regulatory network. As a clue to p53’s physiological function, the challenge is to explore more deeply the interconnections between cellular metabolism and gene regulation as a network, adjusting the state of individual cells and the behavior of cell communities and p53’s role in these processes.