Proteomics and metabolomics in cancer drug ...

4 downloads 0 Views 344KB Size Report
of these observations, Warburg's hypothesis was recently revived. 55 and expanded, as it became to be considered one key target for therapeutic treatments [5].


Proteomics and metabolomics in cancer drug development Expert Rev. Proteomics 10(5), 000–000 (2013)

Angelo D’Alessandro and Lello Zolla* Department of Ecological and Biological Sciences, University of Tuscia, Largo dell’Universita`, snc, 01100 Viterbo, Italy *Author for correspondence: Tel.: +39 0761 357 100 Fax: +39 0761 357 630 [email protected]

In this review article, the main recent advancements in the field of proteomics and metabolomics and their application in cancer research are described. In the second part of the review the main metabolic alterations observed in cancer cells are thoroughly dissected, especially those involving anabolic pathways and NADPH-generating pathways, which indirectly affect anabolic reactions, other than the maintenance of the redox poise. Alterations to mitochondrial pathways and thereby deriving oncometabolites are also detailed. The third section of the review is a discussion of how and to what extent (mutations to) tumor suppressors and oncogenes end up influencing cancer cell metabolism and cell fate, either promoting survival and proliferation or autophagy and apoptosis. In the last section of the review, an overview is provided of therapeutic strategies that make use of metabolic reprogramming approaches.




KEYWORDS: cancer • drug development • mass spectrometry • metabolism • proteomics

Metabolomics, the new/old hallmark of cancer

Human tumor pathogenesis is characterized by the progressive accumulation of changes to normal cells, changes that make cells evolve to a neoplastic state through the gradual acquisition of a series of hallmark capabilities. This multistep process utterly enables normal cells to become tumorigenic and, ultimately malignant [1]. While metabolic reprogramming has only recently been included in the list of the socalled ‘hallmarks of cancer’ [1], echoes from the last (at least) 60 years of research already suggested a crucial correlation between chronic and uncontrolled cell proliferation and deregulated metabolism [2]. The ‘Warburg effect’, named after Otto Warburg, the first researcher to document an exception to the Pasteur effect (inhibition of glycolysis in presence of oxygen) in highly proliferating cancer cells, is based upon the appreciation of an increased glycolytic rate, at the expenses of mitochondrial metabolism (preferentially exploited by normal differentiated cells for energy production purposes), even in the presence of oxygen [2]. This phenomenon, often referred to as ‘aerobic glycolysis’, was at first deemed to be counterintuitive, since rapidly proliferating cells are supposed to have higher energy requirements, while a strictly glycolytic metabolism is less efficient than one relying


upon mitochondrial oxidative phosphorylation in terms of ATP production (~18-fold lower efficiency) [3]. However, since generalization of a Warburg-like metabolism seems to be also applicable to many rapidly dividing embryonic tissues, a tentative explanatory and evidence-based theory posits that aerobic glycolysis might have evolved to meet the elevated anabolic demand (for biosynthetic purposes) and favor the uptake and incorporation of nutrients into biomass by rapidly dividing cells [3]. Conversely, over generalizations should be avoided as well, since tumor cells do not always display a Warburg-like metabolism. Indeed, some tumors are characterized by two subpopulations of cancer cells, one consisting of glucose-dependent cells that secrete lactate (Warburg-wise), while a second subpopulation almost symbiotically relies upon the secreted lactate to sustain their energy production via the tricarboxylic acid cycle (TCA cycle, also known as Krebs cycle) [1]. During the last decade, molecular evidences have underpinned a role for genetic reprogramming in the metabolic regulation observed in cancer cells, a phenomenon that is often accompanied by the preferential expression of cancer-specific isoforms of certain metabolic enzymes, or rather by peculiar and recurrent cancer-associated mutations, especially in genes coding for TCA cycle enzymes [4]. In the light

 2013 Informa UK Ltd

ISSN 1478-9450









D’Alessandro & Zolla

of these observations, Warburg’s hypothesis was recently revived 55 and expanded, as it became to be considered one key target for therapeutic treatments [5]. The implementation of novel mass spectrometry (MS)-based metabolomics and proteomics approaches has boosted this area of research, delivering promising results and suggesting new avenues for further research 60 development in the field of cancer biology, as we will attempt to review in this paper. Mass spectrometry-based proteomics & metabolomics





Cancer proteomics still represents a mainstay in cancer research since the dawn of the post-genomic era [6]. The underlying assumption is that proteins can be regarded as the ‘effectors’ of cellular functions and thus, in biological terms, protein profiling might be more informative than mRNA profiling [7]. Nevertheless, despite almost two decades of efforts, the ambitious agenda pursuing the complete annotation of the physiological role of all known genes still remains unfulfilled [8]. From a technical standpoint, recent technical advancements have opened new scenarios in the field of proteomics. While a decade ago separative and quantitative proteomics approaches mainly relied upon 2DE-based analyses and the implementation of HPLC-MS-based workflows was only auspicated, current proteomics analyses actually take advantage of quantitative analyses via chromatography-MS approaches. Novel instruments have been indeed implemented (both at the HPLC and MS level–the interested reader is referred to reference [7] for further details), as well as bioinformatics suites and tools, which simplified quantitative analysis by allowing peak alignment, detection, protein identification and attribution of posttranslational modifications (PTMs). Quantitative proteomics

85 Among quantitative proteomics approaches that have been extensively applied in the field of cancer research, three main strategies have gained momentum: i) in vivo labeling with stable isotopes; ii) in vitro labeling and iii) label-free approaches [7]. The basic concept behind labeling strategies is that a stable90 isotope labeled peptide shares identical chemical features with its native (unlabeled) counterpart, which results in identical behavior during chromatographic separation and mass spectrometric analysis, though it still allows them to be differentiated owing to their mass difference. Relative abundances can be 95 then grasped by measuring the ratio of signal intensities for the labeled and unlabeled peptide pairs under different biological conditions [7]. Stable isotope labeling of amino acid in culture (SILAC), is probably the most extensively adopted in vivo labeling 100 approach in cancer research [9]. The SILAC protocol envisages cell culturing in media containing either normal amino acids or amino acids labeled with heavy isotopes. The labeled amino acids are often lysine and arginine (with different combinations of 13C, 15N, and 3H). The choice to label these basic amino 105 acids stems from the broadly diffused adoption of trypsin as the protease of choice upstream of HPLC-MS proteomics 2

analyses. In SILAC, trypsin cleavage thus exposes C-terminally labeled arginine or lysine, which allows relative quantitation of each digestion-generated peptide, except for the C-terminus peptide of the protein [7,9]. While SILAC was at first optimized for unicellular model organisms [7,9], its experimental design makes it suitable for cell culture experiments and thus amenable for in vivo/ex vivo cancer research investigations. A direct evolution of SILAC is super-SILAC [10], a method that combines a mixture of multiple SILAC-labeled cell lines. Indeed, SILAC is a particularly accurate quantitative method, although until recently it was limited to cell lines or animals that could be metabolically labeled with heavy amino acids. This limitation of SILAC in studying patient tumor samples has been overcome through the use of a mix of multiple SILAC-labeled cell lines as an internal standard, a technique called superSILAC [10]. This mix achieved superior quantification accuracy compared with a single SILAC-labeled cell line, owing to the generation of hundreds of thousands of isotopically labeled peptides in appropriate amounts to serve as internal standards for MS [11]. Isotope-coded affinity tags (ICAT) are one of the most rapidly expanding in vitro labeling techniques for protein quantitation. ICAT is based upon specific tagging of cysteine residues with a reagent containing either eight or no deuterium atoms, along with a biotin group that enables affinity purification strategies to recover and enrich labeled peptides prior to MS analysis [7]. One major limitation of this technique is that it can be only applied to those proteins that contain cysteine residues. Isobaric tags for relative and absolute quantification (iTRAQ) is another important in vitro labeling strategy. In iTRAQ, tagging requires a reporter group, a balance group and a peptide reactive group. The reactive group binds the Nterminus and side-chain amines of peptides, while the reporter (up to eight different labeling patterns are possible) and the balance group are designed as to achieve isobaric balancing in MS mode, and discriminating fragments in collision induced dissociation (CID) mode for relative quantitation on the basis of the relative abundance of different reporter groups [7]. Another labeling strategy for quantitative proteomics implies the use of isotopomer labels, referred to as ‘tandem mass tags’ (TMTs) [12]. TMTs are designed to ensure that identical peptides labeled with different isotopomers exactly comigrate in all chromatographic separations. On the other hand, peptides from different samples can be identified and relatively quantified using CID-based analysis method. Relative abundance measurements made in the MS/MS mode using the new tags are accurate and sensitive [12]. Another strategy aims at quantifying protein abundances by targeting a so-called proteotypic peptide (defined as ‘an experimentally observable peptide that uniquely identifies a specific protein or protein isoform’ [13]) through selected reaction monitoring (SRM)-MS, which enables isolation and quantitative assaying of the expected mass to charge ratio (against standards or in silico predicted values) [14]. Proteotypic standard peptides could be used as external references to determine calibration Expert Rev. Proteomics 10(5), (2013)












Proteomics, metabolomics & drug development






curves, as resulting from the correlation of MS readings (either peak intensities or peak areas) in response to a variation in peptide concentrations. However, SRM is not necessarily a synonym for label free quantitation. Early evolutions of the SRM approach imply the use of isotopically labeled synthetic peptides (SPIKETIDES [15,16] or AQUA peptides) to enable absolute quantification of specific proteins in a targeted fashion. The clear advantage over non-labeled peptides is that isotopically labeled proteotypic peptides can be directly spiked in the sample to be used as an internal reference, which also helps coping with any untoward technical bias at the nano HPLC or MS level. While labeling-based quantitative strategies are rather expensive and often time consuming (especially in vivo, which might require four to six replication cycles of cultured cells to achieve full labeling), label free approaches are increasingly attracting a great deal of interest, owing to the reduced costs and ease of implementation to classic HPLC-MS proteomics workflows. Recently introduced post hoc algorithms allow calculation of the so-called ‘exponentially-modified protein abundance index’ (emPAI), where the number of identified peptides, normalized against the number of all the possibly identifiable peptides for a given protein, is used as an indicator of the absolute protein abundance on a logarithmic basis. Other indicators of absolute protein abundances include spectral counting and peak intensities [7]. Redox proteomics







Deregulation of metabolism in cancer cells is also interconnected with increased susceptibility to, and exacerbation of, oxidative stress (as it will be extensively described in the following paragraphs). The key approaches to redox proteomics have been recently reviewed [14,17]. Redox proteomics is a recent branch of proteomics that is devoted to the determination and, possibly, quantification of oxidative modifications to proteins (including protein carbonylation, oxidation/nitrosylation of thiol groups and nitrosylation of tyrosines). Among all redox modifications, oxidation of thiol groups might affect the functional activity of several key enzymes (such as metabolic and redox-homeostasis-related enzymes, including glyceraldehyde 3-phosphate dehydrogenase and peroxiredoxin 2) [14,17]. Owing to the labile nature of thiol groups-targeting oxidative modifications, an experimental strategy envisages the temporary quenching of free thiols (by means of trichloroacetic acid-based acidification or through the use of cell permeable reagents for thiol-groups, such as the alkylating agents iodoacetamide or N-ethylmaleimide) and subsequent specific reduction (also with dithiothreitol, sodium arsenite or dimedone) [14,17]. On the other hand, commercially available antibodies can be now exploited to determine the extent of S-nitrosylations, via enabling antibody-based enrichment strategies (immunoprecipitation) or direct immunoassay detection (ELISA, western blot). Other analytical approaches also include biotin labeling for the enrichment of S-glutathionylated peptides or isotope labeling (especially ICAT, see above) to enrich, determine and quantify S-oxidative modifications [14,17].


Further developments in the field of cancer redox proteomics are awaited in the next few years. Phosphoproteomics

Despite the significant body of accumulating knowledge about genes involved in the development of human cancer (at least 300 have been discovered so far), only a limited number of cancer genes encode for proteins that are suitable targets for effective drugs. In this view, protein kinases (such as Abl tyrosine kinase) are among the best eligible targets for small molecule inhibitors [18,19]. Indeed, sustaining proliferative signaling is a key hallmark of cancer, which is often achieved through kinase-triggered phosphorylation cascades [1]. It is thus pivotal to further our understanding of the biological role of protein phosphorylations through the introduction of novel analytical strategies to enhance their detection. Within this context, big strides have been recently made in the field of phosphoproteomics. Phosphorylation (mostly of S/T, and to a lesser extent to Y amino acid residues) is a reversible PTM that plays important regulatory functions in cellular signaling pathways, which can influence cell growth, differentiation, invasion, metastasis and apoptosis [18,19]. Since protein phosphorylations are often sub-stoichiometric, enrichment strategies are often necessary to enable determination of differential phosphorylation events. Enrichment strategies are either based upon affinity chromatography (immobilized metal ion affinity chromatography), titanium dioxide, zirconium dioxide, calcium phosphonate precipitation or strong cation exchange or immunoprecipitation strategies. Detection via non-MS methods mainly involves antibody-based approaches, while MS-based approaches rather rely on isotopic labeling (ICAT, iTRAQ or 32P/33P) and/or alternative (to collision induced dissociation–CID) fragmentation strategies, such as electron transfer dissociation (ETD), which favors generation of c and z ions upon peptide fragmentation (instead of b and y ions, which are predominant in CID) [19].









Evading growth suppression and activating invasion and metastasis (two key hallmarks of cancer [1]) is mainly achieved by cancer cells through the modulation of membrane proteins, which allow cancer cells to bypass contact-triggered growthinhibitory signaling. Glycosylation is one of the most common 255 PTMs, estimated to be found in over 50% of human proteins [20] and more than 80% of membrane proteins [21]. Most membrane biomarkers of cancer cells, which are amenable to antibody-based therapies, are glycosylated proteins. Other than extracellular membrane proteins, secreted proteins (which can 260 be thus searched for in the patient’s body fluids) are often N-glycosylated in the endoplasmic reticulum or Golgi apparatus. It is thus small a wonder that protein glycosylation is increasingly attracting a great deal of interest in the field of cancer research. Glycosylations can be further distinguished 265 into: i) N-linked glycosylation, ii) O-linked glycosylation and iii) C-glycosylations. In like fashion to phosphoproteomics, glycoproteomics approaches often rely upon preliminary 3


D’Alessandro & Zolla

enrichment strategies (lectin affinity or boronic acid chromatog270 raphy or immunoprecipitation) to selectively enrich those proteins bearing N-, O- or C-glycosylations. Glycosylated peptides can then be screened via the use of MS, either MALDI or ESIMS, the latter relying both on CID and ETD fragmentation modes [22]. Enriched glycosylated peptides can be thus released via enzy275 matic digestion (with PNGase F for N-glycosylations) and by chemical methods (for O-glycan release) [20]. Alternative enzymatic digestion combinations can help further define the arrangement of side-chain branches, the most challenging task 280 in glycoproteomics analyses to date [23]. Imaging mass spectrometry









One of the greatest advances over the last decade in the field of MS-based cancer research has been the introduction of MALDI imaging approaches. Imaging MS is a molecular analytical technology that enables the simultaneous measurement of multiple analytes directly from intact tissue sections [24]. Histological features of the sample, as gleaned through classic immunohistochemistry staining, can indeed be correlated with molecular species (proteins, peptides, lipids and metabolites) without the need for target-speci?c reagents such as antibodies. Imaging MS is based upon matrix spraying on suitablytreated cryostat sections mounted on conductive indium titainum oxide-coated slides. The preliminary treatment depends on the molecular species under investigation (e.g., organic solvents might interfere with on tissue lipid analyses). MALDI imaging can indeed be applied to determine molecular signatures that are specific of a tumor tissue, while theoretically easing the individuation of tumor biomarkers and their discrimination from tumor border biomarkers [25]. The possibility to combine it with to routine immunohistochemical approaches offers the opportunity to validate and complement the information attainable from a biopsy while obtaining additional complimentary proteomics/lipidomics/metabolomics-relevant results. Indeed, imaging MS can be applied to obtain tissue profiles of proteins [26], peptides [27], lipids and metabolites [28] (including drug metabolites, thus helping monitoring the efficiency of a therapeutical treatment). Additionally, protocols have been developed also to allow imaging analyses of formalin-fixed paraffin-embedded tissue slices [29]. One main limitation of imaging techniques is related to reproducibility (mainly affected by matrix depositing issues), although recently introduced automatic sprayers dramatically abated technical variability. Other limitations include the difficulty of monitoring high molecular weight compounds (especially proteins above the 50–60 kDa threshold) and the constraints related to the relative abundances of molecular species (e.g., most abundant compounds are often easily visible, while low abundance ones are hardly detectable through this approach). However, the flexibility of the method makes it suitable for targeted quantitative approaches (such as SRM to low molecular weight compounds, such as drugs [30]) directly on tissue. 4

From proteomics to metabolomics: MALDI imaging & MALDI-based metabolomics

Metabolomics is the global quantitative assessment of metabolites (low molecular weight compounds below the 1.5 kDa threshold, including sugars, phosphate compounds, organic acids, nucleosides, lipids and fatty acids or exogenous compounds) in a biological system [31]. While proteomics investigates the effectors influencing the phenotype, metabolites are the phenotype itself. Indeed, the historical precursor to metabolomics can be traced back to early clinical biochemistry approaches [32], while technical advancements in the field NMR and, subsequently, of MS [32,33] have boosted the refinement of this old/new omics discipline. NMR was at first favored by machine accessibility, established data handling and the conservative nature of the analysis, which allows further testing downstream of NMR analyses on the same samples. Conversely, MS has gradually complemented and often replaced NMR owing to its higher sensitivity, which results in MS being less demanding in terms of minimum detectable concentrations of the analyte [33]. Also, MS-based metabolomics holds the potential to better discriminate metabolites, thus improving coverage of the metabolome space, especially when performing upstream compound-class-specific chromatographic separations [32,33]. In likewise fashion to quantitative proteomics, quantitative MS-based metabolomics can also rely upon SRM or multiple reaction monitoring (MRM) approaches [34–36], which allow detection and absolute quantification of a compound and its fragments against a pure standard. However, most advanced metabolomics studies today rely upon post hoc alignment, peak detection and metabolite discrimination without any a priori restriction: this approach also goes by the name of untargeted metabolomics [37] and is already providing decisive insights into cancer biology. In particular, a rather recent application of MS-based untargeted metabolomics is based upon carbon flux during catabolism and anabolism, via supplying 13C-labeled metabolic substrates (mainly glucose and glutamine) [38]. This approach allows the kinetics of energy fluxes to be monitored during molecular biology experiments on cell lines (e.g., upon induction or silencing of an oncogene or tumor suppressor protein), thus helping further refine the understanding of metabolic networks in normal and cancer cells. Biomedical application of MALDI MS is technically suited to monitoring metabolic variations directly on tissue sections from biopsies, but it also allows the screening of whole body tissue sections from model organisms (e.g., mice) while looking for the metabolites derived from catabolism of a specific drug under testing. This is relevant in the context of cancer research since one of the key steps in drug discovery is the determination of the areas targeted by a therapeutic (body accumulation) [39]. MALDI-imaging has recently been applied in cancer metabolomics research, especially in lipidomics analyses, since lipids are highly preponderant and easily detectable through imaging Expert Rev. Proteomics 10(5), (2013)













Proteomics, metabolomics & drug development





approaches. One clear advantage with MALDI imaging approaches is that a molecule can be directly detected on tissue, with a lateral resolution down to 10–15 mm with commonly available instruments (while in secondary ion imaging MS it can be improved up to >50 nm [40]). These advancements have paved the way for a broader application of MALDI-based proteomics and metabolomics strategies [41]. MALDI-based metabolomics, on cell lysates or tissue homogenates instead of tissue sections, would have several advantages over routinely used MSbased metabolomics platforms (such as LC-MS), in that it would be amenable to automatization and multiplexing, especially in combination with robotic auto samplers. Anabolism & the Warburg effect














Glucose HXK

































































G6P GlcNAc-6P














Figure 1. An overview of the main catabolic pathways in normal and proliferat-

The recent introduction of specific ing cells. Glycolysis, Krebs cycle, PPP, serine synthesis, hexosamine synthesis and glutathione homeostasis are included. Enzymes are indicated in light grey, according to their metabolomics analytical platforms helped relative UniProt names. elucidating metabolic fluxes in highly PPP: Pentose phosphate pathway. proliferating cells [42,43]. In line with Kilburn’s observations [44], dating back to four decades ago, highly proliferating cells do not have extremely species (ROS) in the form of superoxide anions and hydroxyl higher energy demands (in terms of glucose metabolization) in radicals, which ends up ‘fertilizing’ the tumor microenvironcomparison to resting cells. This at least in part justifies the met- ment [46] and promotes the accumulation of further mutaabolic choice of oxidizing glucose via glycolysis while depressing tions to oncogenes and tumor suppressors [47]. Of note, the the TCA cycle in cancer cells. However, an elevated replication impairment in ROS modulation/production in mitochondria rate is based upon the accumulation of building blocks to build is often accompanied by mutations to electron transport up mass and cell constituents before replication. In this scenario, chain components [48]. In parallel, most cancers share distinct features such as metabolomics analyses contributed significant insights by demondefects in certain mitochondrial enzymes, including isocistrating how alternative metabolic pathways are indeed activated trate dehydrogenase (IDH), fumarate dehydrogenase (FD) along with glycolysis, which promote anabolism by constantly and succinate dehydrogenase (SD) [49,50]. Alterations to IDH providing key reducing coenzymes such as NADPH (that is pivotal, e.g., in lipid synthesis) (FIGURE 1). In parallel to aerobic glycol- or isoform switching (IDH1 vs IDH2) promote the utilizaysis, glucose utilization fuels the main NADPH-generating tion of a-ketoglutarate from glutamine metabolism via pathway, the (oxidative phase of the) pentose phosphate pathway reductive carboxylation to isocitrate, to fuel acetyl-CoA pro(PPP). Over-activation of the PPP at the non-oxidative phase duction for fatty acid biosynthetic purposes [51–53] or amino fuels the generation of ribose phosphate substrates for nucleoside acid synthesis via oxaloacetate intermediates (FIGURE 2). Of biosynthesis, another central step toward DNA replication in note, isocitrate to a-ketoglutarate conversion by cytosolic proliferating cells. It is perhaps worthwhile to recall that IDH is associated with the production of NADPH, analoNADPH also plays a fundamental role in the recycling of oxi- gous to the conversion of malate (another TCA cycle interdized glutathione (GSSG) back to the reduced form (GSH), mediate) to pyruvate by malic enzyme. Anomalies to IDH1 enzyme (R132H) result in 2-hydroxyglutarate prothus contributing substantially to the redox poise (FIGURE 1). In this complex metabolic scenario, mitochondria are not duction from a-ketoglutarate, which negatively affects ajust innocent and inactive bystanders in cancer cells [45]. ketoglutarate-dependent dioxygenase enzyme activity and Mitochondrial metabolism is indeed fueled by glutamine, promotes malignant progression of brain tumors and, in which is one major nitrogen source for biosynthesis reactions particular, gliomas [54]. Anomalies to FD and SD results in the accumulation of and carbon source (via glutamate-a-ketoglutarate intermediate conversion) for the TCA cycle (FIGURE 1). At the same time, fumarate and succinate, respectively. These metabolites (along mitochondrial activation fuels production of reactive oxygen with the aforementioned 2-hydroxyglutarate) have been recently








D’Alessandro & Zolla

view, fumarate and succinate have been recently referred to as oncometabolites. Also, HIF accumulation diverts IDHdependent reductive carboxylation fluxes toward fatty acid 470 anabolism [56], which further stresses the intertwinement of metabolic deregulation with proliferative capacity (FIGURE 3) [57].

Fatty acid Synthesis Reductive carboxylation ISOCIT






Cell proliferation, tumor suppressors, oncogenes & metabolism






Amino acid Synthesis

Figure 2. Cancer is often associated with mutations to TCA cycle enzymes, or expression of specific isoforms. In the case of IDH, this results in the promotion of reductive carboxylation from glutamine-derived a-ketoglutarate, instead of regular TCA cycle fluxing towards succinyl Co-A. This utterly results in acetyl Co-A accumulation, which promotes fatty acid synthesis and, thus, cell growth and proliferation. IDH: Isocitrate dehydrogenase; TCA: Tricarboxylic acid cycle.

460 found to play a key role as direct inhibitor of dioxygenase and AQ2 prolyl hydrolase, enzymes that indirectly catalyze the degradation of the hypoxia-inducible factor (HIF) [51–53]. This inhibitory phenomenon is relevant since accumulation of HIF, which normally occurs under hypoxia, mediates the activation of pyr465 uvate dehydrogenase kinase (PDK1), which in turn inhibits pyruvate dehydrogenase and thus hinders conversion of pyruvate to acetyl-CoA and shunts pyruvate to lactate [55]. In this Glucose


CITR Pyruvate


Acetyl-CoA PDH












Figure 3. Cancer is often associated to mutations to Krebs cycle enzymes, including SD, FD and MDH. These mutations end up blocking TCA cycle catabolic fluxes, and promote the accumulation of the respective substrates of these enzymes, succinate, fumarate ad malate. These oncometabolites inhibit prolyl hydrolase activity, thereby indirectly resulting in HIF stabilization. In turn, this promotes PDK1 activity, an inhibitory enzyme of pyruvate dehydrogenase. This results in increased lactic acid fermentation and decreased fluxes from glycolysis to the TCA cycle. FD: Fumarate dehydrogenase; HIF: Hypoxia-inducible factor; MDH: Malate dehydrogenase; PDK1: Pyruvate dehydrogenase kinase; SD: Succinate dehydrogenase; TCA: Tricarboxylic acid cycle.



Metabolic deregulation in cancer cells is partly the cause and mostly the effect of genetic deregulation, at the oncogene and tumor suppressor level [58]. As described in the previous paragraph, these metabolic adjustments serve to build up anabolic products to pursue cell proliferation [59]. At the same time, they promote deregulation of oxidative phosphorylation and mitochondrial events, thus favoring ROS accumulation and alterations to the tumor microenvironment. Exacerbation of oxidative stress promotes senescence-like phenomena in cancer cells, while decreasing glucose uptake, deregulating matrix attachment [59]. At the same time, cancer cells cope with the excess of ROS by activating pro-survival pathways, through the deregulation of specific oncogenes [60], often complementary to inactivating or gain of function mutations to tumor suppressor proteins, such as proteins from the p53 family (p53, p63 and p73), which normally act as the guardians of the genome stability. One paradigmatic example is indeed represented by mutations to p53 (e. g., R175H and R273H [61]), resulting in cell survival, increased proliferation and promotion of invasiveness.





The double role of p53 family members

While early cancer investigation studies indicated that p53 and retinoblastoma protein (RB) mutations (comprehensively detected in the great majority of tumors) mainly resulted in the loss of function of their tumor-suppressor activity [62], recent study indicate how these proteins (especially p53) might play a more complex role in modulating the balance of pro-survival and pro-apoptotic signaling, a function that escapes regulation upon the acquisition of specific mutations to these proteins [63]. Tp53, for example, is now known to take part in metabolic modulation at several levels [64]. Analogous roles are increasingly emerging for all the members of the p53 family [65]. For example, p53 can inhibit the expression of the glucose transporters GLUT1 and GLUT4 [64], and can increase the expression of Tp53-inducible glycolysis and apoptosis regulator (TIGAR), a fructose-bisphosphatase that inhibits glycolysis by reducing cellular levels of fructose-2,6-bisphosphate and thus promoting a shift backward to the PPP, a process that promotes NADPH accumulation and plays a role in anti-apoptotic signaling via ROS-damage protection and promotes anabolic pathways, as summarized in the previous paragraphs [66–68] (FIGURE 4). Besides, p53-responsive elements are present in the promoters of PGM and hexokinase II (HK2), which is suggestive of the fact that p53 can promote at least some steps in glycolysis. Of note, under hypoxic conditions, mitochondrial localization of TIGAR stimulates HK2 (often bound to mitochondrial membrane in tumors) [66]. Expert Rev. Proteomics 10(5), (2013)







Proteomics, metabolomics & drug development










Other than p53, a transactivation-proficient isoform of p73 has been recently shown to play a role in the promotion of a metabolic shift toward the PPP, via the upregulation of enzymes involved in the oxidative phase of the pentose phosphate shunt such as 6-phosphogluconolactonase [69]. Tp53 also regulates the expression of sestrin proteins, which activate AMPK to regulate growth and autophagy but also function as antioxidants, protecting cells from hydrogen peroxide-induced damage [65]. Furthermore, also in terms of antioxidant defenses, p53 triggers the activation of glutathione peroxidase, aldehyde dehydrogenase and tumor protein p53-inducible nuclear protein 1 (TP53INP1), all of them displaying antioxidant functions. However, it should not be forgotten that many p53-inducible proteins participate in apoptotic responses via the promotion of ROS production, including p53-induced gene 3 (PIG3), proline oxidase, BAX, PUMA and p66SHC [65], as well as of cytochrome C oxidase 2 [70]. Tp53 further influences mitochondrial metabolism by promoting the expression of glutaminase 2, glutaminases being a family of enzymes responsible for glutamine to glutamate conversion [71]. Analogous effects on glutaminase expression have been recently reported also for p63 [72]. In turn, glutamate is a pivotal constituent of the tripeptide GSH, or rather it can be further metabolized to a-ketoglutarate, that can either undergo reductive carboxylation to isocitrate or further oxidation to succinyl-CoA via the TCA cycle. Energy or nutrient deprivation, extreme environmental conditions or Ca2+ release from the lumen of the endoplasmic reticulum (ER) results in the disruption of proper proteinfolding activity in this organelle, a condition that promotes the so-called ER-stress. ER-stress has been observed to influence p53 stability by modulating its differential phosphorylation to serine 315 and serine 376, which promotes p53 localization in the cytoplasm and its degradation [73,74]. Conversely, other p53 family members, p63 and p73, have been shown to promote ER stress and scotin (protein shisa-5) [75,76], suggesting an intricate cross-talk among p53-family members, other than direct competition for p53 responsive elements or oligomerization through direct binding. In this view, it is interesting to note that almost pleiotropic metabolic effects are also expected for other p53 family members, since, for example, TAp73 deletion reduces cellular ATP levels, oxygen consumption and mitochondrial complex IV activity, with increased ROS production and oxidative stress sensitivity [77]. This phenomenon involves the mitochondrial complex IV subunit cytochrome C oxidase subunit 4 (Cox4i1), which is a direct TAp73 [77].

Metabolic starvation experiments 570 On the basis of the evidences for modest energy demands, albeit extreme substrate uptake, especially of glucose and glutamine (we hereby purposely neglect uptake of lipids from the medium, which would deserve an entirely dedicated review) for anabolic and redox homeostasis purposes, new trends in the


field of cancer drug research are aimed at evaluating the effects of starvation on cancer cells. One simple approach to promote cancer cell starvation in vivo would be to prevent angiogenesis, which would reduce the blood flux and thus oxygen and nutrient delivery [78]. Anti-angiogenic therapy mainly relies upon the administration of anti-VEGF-targeting monoclonal antibodies (e.g., bevacizumab). Within this framework, in vitro metabolomics experiments have recently provided a clearer understanding of the mechanisms underlying the effects of starvation on cancer cells. Since the main metabolic substrates for carbon and nitrogen build up in cancer cells are both glucose and glutamine, starvation experiments have been so far performed through the supplementation of cell media that are depleted of these two compounds [79–83]. A complex scenario emerged whereby glucose might represent the main energy source for certain cell lines (such as head and neck squamous carcinoma cells [79]), while only glutamine depletion actually triggered apoptosis in other cell lines [83]. Other than glucose and glutamine, cancer cells necessitate serine to support anabolism by providing precursors for biosynthesis of proteins, nucleotides, creatine, porphyrins, phospholipids and glutathione. Also, up-regulation of the serine synthesis pathway occurs in some breast cancers [84,85]. It has been recently demonstrated that p53-mediated cell responses to serine starvation involve over-activation of the serine synthesis pathway. Besides, it promotes inhibition of glycolysis, since a rate-limiting enzyme, in particular, the specific and less efficient cancer isoform, pyruvate kinase M2 (PKM2) [86], is allosterically activated by serine and thus, serine starvation, ends up inhibiting glycolysis (FIGURE 4) [87]. This prompts two main effects: increase in TCA cycle fluxes to cope with the decreased ATP production, and an increase in PPP fluxes, to generate NADPH and thus cope with oxidative stress arising from the elevation in TCA cycle-dependent energy production [85]. Of note, PKM2 expression in cancer cells results in relatively decreased glycolytic rates, which promotes accumulation of early glycolysis intermediates, including 3-phosphoglycerate, a precursor to serine de novo synthesis. Serine metabolism is also at the crossroads between p53 and starvation-induced autophagic responses [88]. Autophagy is a catabolic mechanism that promotes cell degradation of unnecessary or dysfunctional cellular components through the lysosomal machinery to cope with the decrease in energy and anabolic resources, and it is often regarded as: i) an alternative option cells might choose to commit suicide, other than apoptosis, ii) a cell’s defensive strategy upon cell damage or iii) a cell’s major adaptive (survival) strategy to cope with metabolic stress, such as nutrient deprivation, or starvation in general. Starvation, the most extensively investigated inducer of autophagic responses, triggers the activation of AMPK [89,90], a kinase that is activated by increased AMP/ATP ratios, and is regulated to some extent by mammalian target of rapamycin (mTOR), a key molecular sensor for nutrient availability and a regulator of cell growth and proliferation [91–95]. In particular, 7


















D’Alessandro & Zolla

phosphatidylinositol (4,5)-biphosphate (PIP2). In this way, PTEN modulates GLUT1, PIP3/PIP2 ratios and, indirectly, cell GLUT4 Anabolism and Antioxidant responses survival [96]. Glucose The list of oncogenes involved in metaNADPH NADPH HXK bolic regulation is not only limited to HIF G6P Oxidative phase PPP and mTOR, but includes many other players. One of those is MYC, which encodes F6P Nucleoside a transcription factor c-Myc that links TIGAR Non-oxidative phase PPP biosynthesis p53 FBP altered cellular metabolism to tumorigenesis [97]. Indeed, c-Myc regulates genes GLTM GLS2 involved in the biogenesis of ribosomes Bax GLUT and mitochondria, affects glucose and gluPUMA PEP 3PG Sestrins tamine metabolism, nucleotide metabolism p66SHC PKM2 Glutathione peroxidase Cytochrome C oxidase 2 Pyruvate Lactate and, along with E2F1, DNA replication Aldehyde dehydrogenase SER and miRNA expression [97]. Ectopic c-Myc TP53INP1 cooperates with HIF to promote the induction of a transcriptional program for ROS hypoxic adaptation, involving upMitochondria regulation of glycolytic genes including lactate dehydrogenase A, or the repression of Figure 4. Metabolic regulation by p53 at a glance. As a tumor suppressor, p53 is microRNAs (miRNAs) miR-23a/b to long known to promote apoptosis via enhancing pro-apoptotic factors and ROSincrease glutaminase protein expression generating mitochondrial metabolism. However, it recently emerged as a role for p53, as a pro-survival mediator under mild stress conditions. This phenomenon appears to involve and glutamine metabolism. a TIGAR, a fructose-2,6-bisphosphatase that promotes accumulation of early glycolytic Recent flux-balance analyses have indiprecursors thus boosting a diversion towards the PPP. The production of NADPH at the cated a key role in glycolytic modulation non-oxidative phase of the PPP sustains anabolic and antioxidant processes. Evidences (toward increase) and glutamine metaboalso indicate a role for p53 in the induction of anti-oxidant defenses (sestrins, glutathione lism (towards reductive carboxylation) for peroxidase, aldehyde dehydrogenase, TP53INP1). At the same time, non-oxidative phase products fuel de novo nucleoside synthesis. Pro-survival effects mediated by p53 under another oncogene, KRAS [98,99]. In particmild stress conditions appear to involve the homeostasis of serine metabolism (SER). ular, KRAS appears to be involved in the PPP: Pentose phosphate pathway; ROS: Reactive oxygen species; TIGAR: Tp53-induced increase in glycolytic metabolism and glycolysis and apoptosis regulator; TP53INP1: Tumor protein p53-inducible nuclear channeling of glycolytic intermediates to protein 1. non-oxidative phase PPP reactions and hexosamine biosynthesis (in turn promotautophagy initiation is associated with downregulation of ing protein glycosylation) [98]. Also, KRAS seems to promote an mTOR complex 1 (mTORC1) activity. mTOR is a well- alternative pathway for glutamine metabolism to glutamate and conserved serine/threonine kinase that belongs to the phosphoi- alternative downstream pathway. Such alternative route involving nositide 3-kinase (PI3K)-related kinase family, and it plays an glutamate metabolism appears to be dependent on transamiimportant role in the signaling network that controls growth nases, especially aspartate transaminases (GOT), which promote and metabolism in response to environmental cues. The activa- glutamate and oxalacetate production from aspartate and ation of mTORC1 (one of the two distinct multi-protein com- ketoglutarate (FIGURE 5). This results in oxaloacetate cytosolic accuplexes involving mTOR) requires glutamine and essential mulation (certain GOTs operate only in the cytosol), which amino acids such as leucine, while it has been recently demon- prompts its conversion to malate (via malate dehydrogenase) strated to also depend on availability of serine [87]. At the same and, through malic enzyme, to pyruvate, a reaction that concomtime, Akt and AMPK communicate directly with mTORC1, itantly fuels NADPH production [99]. by phosphorylating raptor (an mTORC1 component), leading Analogously, the tumor suppressor promyelocytic leukemia to 14-3-3 binding and the allosteric inhibition of mTORC1 [94]. (PML) gene acted as both a negative regulator of PPARg coacActivation of mTORC1 promotes protein synthesis by phos- tivator 1A (PGC1A) acetylation and a potent activator of phorylating the kinase S6K and the translation regulator 4E- PPAR signaling and fatty acid oxidation in breast cancer BP1 and lipid biogenesis, via activation of SREBP and PPARg cells [100]. Finally, it is at least worth mentioning the long time transcription factors [94,95]. The phosphatase and tensin homo- established role in metabolic regulation of the insulin-like log (PTEN) interferes with the whole phosphoinositide 3-kin- growth factors (IGF-I and IGF-II) system [101]. ase (PI3K)/Akt/mTORC1 axis, by dephosphorylating MicroRNAs (miRNAs) are small RNA molecules that reguphosphatidylinositol (3,4,5)-trisphosphate (PIP3) at the 3‘ posi- late gene expression post-transcriptionally [97]. As anticipated in tion of the phosphate of the inositol ring, thus producing the previous paragraphs, miRNA expression can be more or 8

Expert Rev. Proteomics 10(5), (2013)












Proteomics, metabolomics & drug development




Induction of apoptosis via gene 730 therapy & metabolic reprogramming






Cytosol CO2 NADPH











ctive Redu tion yla x carbo

α-KET Acetyl-CoA

TCA cycle OAA


Figure 5. Glutamine metabolism can follow different fates in normal and cancer cells. Glutamine can be converted to glutamate (via glutaminase enzymes–GLS), and thus be metabolized to ketoglutarate, whereby it enters TCA cycle or rather promotes reductive carboxylation. In parallel, glutamate can represent a building block of the tripeptide glutathione (GSH). An alternative route involves the malate-aspartate shuttle, a pathway that involves glutamine-derived aspartate towards the accumulation of oxaloacetate and malate intermediates in the cytosol. This pathway is relevant in that NADPH is produced to sustain anabolic and anti-oxidant pathways. GSH: Glutathione; TCA: Tricarboxylic acid cycle.

In the previous paragraphs we have summarized how cancer cells often suffer from mutations that provide a competitive edge over normal cells in terms of biomass accumulation and proliferative capacity. Starting from Warburg’s initial hypothesis, latest research has revealed that metabolic reprogramming occurs as a consequence of mutations in cancer genes and alterations in cellular signaling. Thus, we described how these mutations end up promoting certain pathways (glycolysis, PPP, nucleoside and fatty acid biosynthesis, NADPH-generating reactions) at the expenses of others (above all, oxidative phosphorylation). Appreciation of these phenomena was a step forward from the canonic conception of the Warburg effect and opened new avenues for future developments in the field of cancer treatment [106]. Therefore, ‘untuning the metabolic machine’ [99] might represent the new trend in cancer therapies, which might surprisingly be boosted by decades of research in the field of metabolism-related diseases, including diabetes and drugs for diabetes treatment (such as metformin) [107]. Other metabolic reprogramming strategies envisage the promotion of oxidative stress via mitochondrial uncoupling [108]. Another approach is related to caloric restriction, ketogenic diets and modulation of circulating nutrient levels through enzymes such as asparaginase, these strategies imply that tumor cells are more demanding in terms of nutrients in comparison to normal cells [106–110]. In this section, we will briefly describe the main metabolic reprogramming strategies [105–109] that are currently either under





less directly involved in metabolic modulation [102–105]. Small RNAs may have an intrinsic function in tumor suppression, since their levels are globally decreased in human cancers cells [88]. In line with this, the transcription of some miRNA genes (such as miR-34) is regulated by p53 [102]. Of note, miR-34a appears to be a key regulator of hepatic lipid homeostasis. Together with miR-34, other miRNAs play a key role in metabolic modulation, including miR-33a and miR-33b, which have a crucial role in controlling cholesterol and lipid metabolism in concert with their host genes, the sterol-regulatory element-binding protein (SREBP) transcription factors [104]. Other metabolic miRNAs, such as miR-103 and miR-107, regulate insulin and glucose homeostasis [104], while miR-143 regulates hexokinase-2 expression in cancer cells [105]. AQ4


laboratory testing or already under clinical evaluation (since 760 they are based on already commercialized drugs). Targeting glycolysis to modulate the Warburg effect

Substantiation of the Warburg effect derives from the increased glucose uptake by cancer cells and increased lactate production via glycolysis, even in presence of oxygen. On this ground, these characteristics have fostered the concept of new classes of drugs that can be used either to monitor cancer cell metabolism via state of the art diagnostic tools or to make it amenable to therapeutic interventions [111–114]. For example, since tumors consume higher levels of glucose, clinicians have long been able to monitor tumor uptake of a fluorine radioisotope of glucose, 18F-deoxyglucose, by FDG-PET. This technique has proven its usefulness in determining the cancer stage, to identify metastatic sites and monitor treatment effectiveness. In parallel, a correlation has been observed between the initial degree of FDG-PET positivity and the overall patient outcome across cancer types and subtypes [111]. Therapeutic intervention based upon drugs that target glycolytic enzyme activities are currently under evaluation. Among the possible targets, hexokinase (HXK), phosphofructokinase (PFK), glyceraldehyde 3-phosphate dehydrogenase (GADPH) and lactate dehydrogenase (LDH) represent ideal therapeutic 9






D’Alessandro & Zolla

targets, since chemical inhibitors are already known (though 785 they are often not extremely specific). Known inhibitors of hexokinase include 2-deoxyglucose, 3-bromopyruvate, 5-thioglucose and mannoheptulose. In particular, 3-bromopyruvate is a strong alkylating agent toward the free SH groups of cysteine residues in proteins, which 790 might thus also affect those enzymes with thiol groups in their active sites (such as GAPDH). Lonidamine is known to inhibit only the mitochondriabound hexokinase, which is a distinctive feature of cancer cells (please, refer to the previous paragraphs) [111–114]. Use of a PFK inhibition strategy implies the suppression of 795 the PFKB3 isozyme, which controls the cellular level of fructose-2,6-bisphosphate and thus affects the glycolytic flow by allosterically-modulating PFK activity. Known inhibitors of GAPDH include a-chlorohydrin, 800 ornidazole and iodoacetate, as well as the pentovalent arsenate [111]. LDH-A can be knocked down in tumor cells by shRNAs, thereby stimulating energy fluxes from pyruvate to mitochondria, which in turn promotes mitochondrial uncoupling 805 (ROS production, accumulation of pro-oxidant intermediates) in those tumors bearing mutation of mitochondrial enzymes. Targeting LDH might represent a cancer-cell suppressing preferential strategy, while it could be less toxic to normal cells [111]. Other therapeutical interventions might involve the use of 810 oxythiamine (a thiamine antagonist that inhibits transketolase and pyruvate dehydrogenase) or glufosfamide. Glufosfamide is a conjugate of glucose (highly consumed by cancer cells) and ifosfamide, an alkylating agent with cytotoxic effects. Of note, 815 glufosfamide is uptaken via the SAAT1 glucose transporter, which is overexpressed in cancer cells [111–114]. Owing to its pro-glycolytic potential, HIF-1-targeting drugs should be included, to a certain extent, in the same category of glycolysis inhibitors [115]. Several classes of drugs have been 820 designed and tested over the years, targeting HIF transcription, synthesis, stability, heterodimerization, DNA-binding activity or targets downstream to HIF-controlled signaling [115]. Targeting NAD-metabolism has recently emerged as a potential therapeutic approach to tackle cancer cell prolifera825 tion, since NAD undergoes crucial changes in cancer cells, whereby its use in transcription, DNA repair, cell cycle progression, apoptosis and metabolism processes is deregulated in comparison to normal cells [116]. Targeting autophagy

830 Although autophagy can result in the suppression of tumor development, it may also mirror an extreme and desperate attempt of the cancer cells to survive. Owing to the complexity of autophagic responses in mediating cancer survival/suppression, several strategies have been proposed to tackle autophagy 835 over the last few years [117–119]. Treatments with rapamycin (targeting mTOR), for example, induce glucose starvation-like effects in cancer cell [119]. Another 10

recent strategy to promote autophagic responses relies upon the administration of cannabinoid receptor agonists [120]. Starvation (via nutrient deprivation) of cancer cells has been 840 proposed as a viable strategy to tackle cancer cell proliferation [121,122], especially when used in combination with chemotherapy [123,124]. Old drugs, new benefits

Drug discovery is an extremely complicated, expensive and, most of the time a challenging (if not discouraging) area of research. Patenting, testing and commercializing new effective drugs is a prohibitive task even for multi-national companies, which need to invest resources (in terms of funds and trained personnel) for more than two decades, before (when lucky) receiving the final approval by US FDA. A record of currently available drugs, the latest version of DrugBank (release 2.0), includes a list of approximately 4900 drug entries, in which are enlisted both FDA-approved small molecule and biotech drugs [125]. Since metabolic diseases have long been investigated, especially those involving deregulation of glucose homeostasis, such as diabetes, a long list of currently commercialized drugs already exists that might be amenable for cancer treatment, in the light of the revisited role of the Warburg effect and metabolic reprogramming in cancer progression. Biguanides (such as metformin) belong to this category of old drugs with potential new benefits. Biguanides were first isolated in 1920 from the French lilac Galega officinalis, which was known to contain an agent that reduced the frequent urination associated with diabetes. Biguanides have step up to the spotlight for their ability to suppress liver gluconeogenesis, which is believed to occur through activation of hepatic AMP-activated protein kinase (AMPK) signaling [107]. It was but in recent years that the antitumor effect of metformin could be observed, an effect that is mediated by the activation of AMPK and thereby modulating the AMPK/mTOR pathways. At higher doses (than physiologically achievable in vivo), metformin appears to directly affect mitochondrial oxidative phosphorylation in cancer cells. Targeting fatty acid synthesis & metabolism








Owing to their highly proliferating nature, tumor cells need to build up new membrane to favor replication into daughter cells. In this view, fatty acid synthesis and uptake from the medium (tumor microenvironment) are two key pathways that have attracted a great deal of interest at least during the last 880 10 years [126]. Targeting fatty acid synthesis involves promoting lipid lowering PPAR-pathways (via fenofibrate, also decreasing local angiogenesis) [127] or rather by addressing fatty acid oxidation (FAO) [126,128]. While most cancer researchers focused on glycolysis, glutaminolysis and fatty acid synthesis, the role of 885 fatty acid oxidation in cancer cell metabolic transformation has not been hitherto carefully examined [129]. FAO is inhibited by oxidative stress and, though indirectly, it might contribute to counteracting ROS accumulation in cancer cells [129]. Indeed, FAO generates one molecule of acetyl 890 Expert Rev. Proteomics 10(5), (2013)

Proteomics, metabolomics & drug development






CoA in each oxidation cycle and two in the last cycle. Acetyl CoA enters the Krebs cycle, and combines with oxaloacetate to give rise to citrate. As described in the previous paragraphs, IDH-mediated cytosolic conversion of isocitrate to aketoglutarate also produces cytosolic NADPH (for anabolic and antioxidant purposes). This is the same pathway (though in the opposite direction), according to which fatty acid synthesis under hypoxic conditions (or when mitochondrial respiration is limited) might rely upon glutamine-glutamate-aketoglutarate generating reactions, thereby driving reductive carboxylation toward the accumulation of acetyl CoA, as a building block for fatty acid synthesis and elongation [130]. It is also worthwhile recalling that glutamine-derived glutamate generates NADPH when converted to a-ketoglutarate by glutamate dehydrogenase (GLUD1). Also, since glutaminederived a-ketoglutarate might fuel acetyl CoA accumulation and thus fatty acid synthesis, glutamine might represent a key therapeutic target (e.g., at the glutaminase level, through DON and azaserine) also when attempting to tackle lipid synthesis [131]. Potential pharmacological targets to inhibit FAO are represented by carnitine palmitoyl transferase (CPT1), the ratelimiting enzyme in FAO, and 3-ketoacylthiolase (3-KAT), which catalyzes the final step in FAO [129,132] and ATP-citrate lyase (ACLY), a cytosolic enzyme that catalyzes the generation of acetyl CoA from citrate [133]. Targeting protein markers of cancer






In the present paper we mainly focused on metabolites or protein targets mainly related to cancer metabolism. However, there is a long list of emerging molecular markers of cancer that is continuously expanding [134]. One of the most promising classes of protein biomarkers that are currently undergoing clinical testing is represented by molecular chaperones of the heat shock protein (HSP) family. A wide range of human cancers is accompanied by overexpression of HSPs, which are implicated in tumor cell proliferation, differentiation, invasion, metastasis, death and recognition by the immune system [135]. HSPs are useful biomarkers for carcinogenesis in some tissues and might be used as a valid indicators of the degree of differentiation and the aggressiveness of some cancers. Serum levels of HSP and HSP-specific antibodies in cancer patients might help tumor diagnosis and be related to prognosis of specific cancers. Overexpression of HSP27, for example, is associated with poor prognosis in gastric, liver and prostate carcinoma and osteosarcomas [135]. On the other hand, HSP70 [136] is correlated with poor prognosis in breast, endometrial, uterine cervical and bladder carcinomas. HSPs might interfere with therapetuic treatments and/or induce resistance to chemotherapy in breast cancer, leukemia patients and osteosarcomas. Two main strategies have been proposed to target HSPs, depending on whether they are related to an improved or worsened prognosis of a specific cancer. These strategies include: i) pharmacological modification of HSP expression or molecular chaperone activity and ii) use of


HSPs in anticancer vaccines, exploiting their ability to act as 945 immunological adjuvants [135]. Expert commentary

Big strides have been made over the last decade in the biological understanding of the phenomena underpinning cancer metabolism deregulation. This accumulating body of laboratory science has paved the way for designing new therapeutical strategies and re-discovering of older drugs, with metabolismregulatory aptitude. Mass spectrometry-based proteomics and metabolomics have been at the core of these basic science advancements, which will undoubtedly translate into actual pharmacological applications within the next decades. Without the broader distribution of highly sensitive and accurate MS instruments and new quantitative (isotope labeling-based) strategies, cancer research would have hardly had any chance to set even one single step into the deep forest of metabolic intricacies that we described in the previous sections. Many molecular biologists have revised their positions, firmly standing on a reductionistic ground, while starting to complement classic molecular biology experimental approaches (to put it cursorily, knock out one gene, induce another gene, knock down another) with emerging omics disciplines, theoretically encompassing the whole proteome and metabolome (within the current capabilities of the applied technique). Metabolism-targeting compounds already include a broad list of patented drugs and food derived nutrient/pharmaceutical-like molecules, also known as nutraceuticals [137]. Revisiting the Warburg effect with proteomics and metabolomics tools has revealed that we might already have an old answer (commercialized drugs) for a new, compelling question (tackling cancer proliferation via metabolic reprogramming).







Five-year view

Further advancements will soon be achieved through the implementation of in silico prediction tools, based upon machine learning algorithms, a wind of change that embraces the broader concept of systems biology [138,139]. Based upon bioinformatic models, computer predictions will also help predicting untoward effects (scarce efficiency, scarce specificity) of in vitro designed drugs [140]. Improvements in drug specificity and selectivity at the design phase will make it amenable to target cancer specific enzyme isoforms or mutations (PKM2, mitochondrial HXK2, monocarboxylate transporters MCT4 for lactate secretion [141]), by exploiting the concept of ‘synthetic lethality’. As summarized by Kaelin [142], ‘two genes are synthetic lethal if mutation of either alone is compatible with viability but mutation of both leads to death.’ Therefore, it could be possible to target a gene that is synthetic lethal to a cancerrelevant mutation, as to kill only cancer cells while sparing normal cells [142]. Finally, as we hope it emerged from this paper, cancer cells display an intricate network of intertwined and mutuallycompensating metabolic pathways. Altering one node of the 11






D’Alessandro & Zolla

network might result in the compensation effect promoted by the so-called plasticity of the metabolic network itself. There1000 fore, omics/systems biology-wise efforts should be pursued to unveil as many pathways as possible, and to gain an improved understanding of the role of already known, albeit underinvestigated ones, such as the folate and mevalonate pathways (both producing NADPH at some steps of the cycles). The final 1005 goal will be to determine, case by case, cancer by cancer, the most suitable targets for multi-targeted pharmaceutical interventions [143].


Financial & competing interests disclosure

A D’ Alessandro and L Zolla are supported by funds from the Italian National Blood Centre (Centro Nazionale Sangue–CNS–Istituto Superiore 1010 Sanita‘–Rome, Italy). The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pend- 1015 ing or royalties. No writing assistance was utilized in the production of this manuscript.

Key issues • The introduction of label-based and label-free quantitative proteomics and metabolomics analyses revived the field of cancer metabolism. • Eighty years after Warburg’s early observations, cancer metabolism is now deemed to be one key hallmark of cancer. • Most of the cancer-specific frequent mutations to tumor suppressors and oncogenes have the potential to affect cancer


cells metabolism. • These changes often promote glycolysis at the expenses of oxidative phosphorylation in terms of energy production, other than anabolic and NADPH-generating pathways. However, mitochondrial metabolism still plays a key role in producing oncometabolites with regulatory effects on key oncogenes, other than providing substrates for anabolic reactions (especially fatty acid de novo synthesis) and contributing to the redox poise (via NADPH generation).


• Starvation promotes autophagy, which could provide a therapeutical intervention opportunity for some cancers. Induction of autophagy might represent a viable alternative to induction of apoptosis. • Small molecule inhibitors of most of the key enzymes involved in the pathways described above are already available and deserve further testing to understand their specificity and effectiveness. Cancer isoform-targeting (‘synthetic lethality’) specific drugs must also be designed as well.


• Among the thousands of commercially available drugs, those having a validated effect in the treatment of metabolic diseases may also be suitable for the treatment of some cancers.

References • of interest •• of considerable interest 1

Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 144(5), 646–674 (2011).


An instant classic on cancer biology, whereby hallmarks of cancer are summarized and described into details.


Warburg O. On the origin of cancer cells. Science 123, 309–314(1956).


Warburg’s decade long observations are summarized in this paper.


Vander Heiden MG, Cantley LC, Thompson CB. Understanding the Warburg effect: the metabolic requirements of cell proliferation. Science 324, 1029–1033 (2009)



A thorough review encompassing metabolic reprogramming events in cancer cells from a Warburg-wise perspective.

cell culture (SILAC) as a tool for prostate cancer research. Mol. Cell. Proteomics 3(7), 729–735 (2004).


Frezza C, Pollard PJ, Gottlieb E. Inborn and acquired metabolic defects in cancer. J. Mol. Med. (Berl). 89(3), 213–220 (2011).


D’Alessandro A, Zolla L. Metabolomics and cancer drug discovery: let the cells do the talking. Drug Discov. Today 17(1–2), 3–9 (2012).


Geiger T, Cox J, Ostasiewicz P, Wisniewski JR, Mann M. Super-SILAC mix for quantitative proteomics of human tumor tissue. Nat. Methods 7(5), 383–385 (2010).


Bichsel VE, Liotta LA, Petricoin III EF. Cancer proteomics: from biomarker discovery to signal pathway profiling. Cancer J. 7, 69–78 (2001).



Bachi A, Bonaldi T. Quantitative proteomics as a new piece of the system biology puzzle. J. Proteomics 71(3), 357–367 (2008).

Geiger T, Wisniewski JR, Cox J et al. Use of stable isotope labeling by amino acids in cell culture as a spike-in standard in quantitative proteomics. Nat. Protoc. 6, 147–157 (2011).


Thompson A, Scha¨fer J, Kuhn K et al. Tandem mass tags: a novel quantification strategy for comparative analysis of complex protein mixtures by MS/MS. Anal. Chem. 75(8), 1895–1904 (2003).

Papers of special note have been highlighted as:

Quantitative proteomics approaches are reviewed and described into detail in this paper.


Legrain P, Aebersold R, Archakov A et al. The human proteome project: current state and future direction. Mol. Cell. Proteomics 10(7), M111.009993 (2011).


Kuster B, Schirle M, Mallick P, Aebersold R. Scoring proteomes with proteotypic peptide probes. Nat. Rev. Mol. Cell. Biol. 6(7), 577–583 (2005).


Everley PA, Krijgsveld J, Zetter BR, Gygi SP. Quantitative cancer proteomics: stable isotope labeling with amino acids in


D’Alessandro A, Rinalducci S, Zolla L. Redox proteomics and drug development. J. Proteomics 74(12), 2575–2595 (2011).

Expert Rev. Proteomics 10(5), (2013)

Proteomics, metabolomics & drug development




Ang CS, Rothacker J, Patsiouras H, Gibbs P, Burgess AW, Nice EC. Use of multiple reaction monitoring for multiplex analysis of colorectal cancer-associated proteins in human feces. Electrophoresis 32(15), 1926–1938 (2011).


Bachi A, Dalle-Donne I, Scaloni A. Redox proteomics: chemical principles, methodological approaches and biological/ biomedical promises. Chem. Rev. 113(1), 596–698 (2013).


The most thorough review on redox proteomics to date.


Ashman K, Villar EL. Phosphoproteomics and cancer research. Clin. Transl. Oncol. 11(6), 356–362 (2009).


Engholm-Keller K, Larsen MR. Technologies and challenges in large-scale phosphoproteomics. Proteomics 13(6), 910–931 (2013).


Kim EH, Misek DE. Glycoproteomics-based identification of cancer biomarkers. Int. J. Proteomics 2011, 601937 (2011).


Gupta R, Brunak S. Prediction of glycosylation across the human proteome and the correlation to protein function. Pac. Symp. Biocomput. 7, 310–322 (2002).


Pan S, Chen R, Aebersold R, Brentnall TA. Mass spectrometry based glycoproteomics– from a proteomics perspective. Mol. Cell. Proteomics 10(1), R110.003251 (2011).


Chen R, Jiang X, Sun D et al. Glycoproteomics analysis of human liver tissue by combination of multiple enzyme digestion and hydrazide chemistry. J. Proteome Res. 8(2), 651–661 (2009).


Schwamborn K, Caprioli RM. MALDI imaging mass spectrometry–painting molecular pictures. Mol. Oncol. 4(6), 529–538 (2010).




Schnatbaum K, Zerweck J, Nehmer J, Wenschuh H, Schutkowski M, Reimer U. SpikeTides—proteotypic peptides for large-scale MS-based proteomics. Nature Methods 8 (Application note) (2011).



imaging protocol for in situ characterization of tryptic peptide identity and distribution in formalin-fixed tissue. Rapid Commun. Mass Spectrom. 27(6), 655–670 (2013).

Scho¨ne C, Ho¨fler H, Walch A. MALDI imaging mass spectrometry in cancer research: Combining proteomic profiling and histological evaluation. Clin. Biochem. 46(6), 539–545 (2013). Franck J, Longuespe´e R, Wisztorski M et al. MALDI mass spectrometry imaging of proteins exceeding 30,000 daltons. Med. Sci. Monit. 16(9), BR293–299 (2010). Gustafsson OJ, Eddes JS, Meding S, McColl SR, Oehler MK, Hoffmann P. Matrix-assisted laser desorption/ionization

Monteiro MS, Carvalho M, Bastos ML, de Pinho PG. Metabolomics analysis for biomarker discovery: advances and challenges. Curr. Med. Chem. 20(2), 257–271 (2013). Casadonte R, Caprioli RM. Proteomic analysis of formalin-fixed paraffin-embedded tissue by MALDI imaging mass spectrometry. Nat. Protoc. 6(11), 1695–1709 (2011). Picotti P, Aebersold R. Selected reaction monitoring-based proteomics: workflows, potential, pitfalls and future directions. Nat. Methods 9(6):555–566 (2012).



van Kampen JJ, Burgers PC, de Groot R, Gruters RA, Luider TM. Biomedical application of MALDI mass spectrometry for small-molecule analysis. Mass Spectrom. Rev. 30(1), 101–120 (2011).


Klitzing HA, Weber PK, Kraft ML. Secondary ion mass spectrometry imaging of biological membranes at high spatial resolution. Methods Mol. Biol. 950, 483–501 (2013).


Issaq HJ, Fox SD, Chan KC, Veenstra TD. Global proteomics and metabolomics in cancer biomarker discovery. J. Sep. Sci. 34(24), 3484–3492 (2011).


Dang CV. Links between metabolism and cancer. Genes Dev. 26(9), 877–890 (2012).


Putri SP, Yamamoto S, Tsugawa H, Fukusaki E. Current metabolomics: technological advances. J. Biosci. Bioeng. 116(1), 9–16 (2013).


Lo´pez-La´zaro M. The warburg effect: why and how do cancer cells activate glycolysis in the presence of oxygen? Anticancer Agents Med. Chem. 8(3), 305–312 (2008).


D’Alessandro A, Giardina B, Gevi F, Timperio AM, Zolla L. Clinical metabolomics: the next stage of clinical biochemistry. Blood Transfus. 10(2), s19–24 (2012).


Kilburn DG, Lilly MD, Webb FC. The energetics of mammalian cell growth. J. Cell Sci. 4(3), 645–654 (1969).


Frezza C, Gottlieb E. Mitochondria in cancer: not just innocent bystanders. Semin. Cancer Biol. 19, 4–11 (2009).


Martinez-Outschoorn UE, Lin Z, Trimmer C et al. Cancer cells metabolically "fertilize" the tumor microenvironment with hydrogen peroxide, driving the Warburg effect: implications for PET imaging of human tumors. Cell Cycle 10(15), 2504–2520 (2011).


Cardaci S, Ciriolo MR. TCA Cycle Defects and cancer: when metabolism tunes redox state. Int. J. Cell Biol. 2012, 161837 (2012).


Guzy RD, Hoyos B, Robin E et al. Mitochondrial complex III is required for hypoxia-induced ROS production and cellular oxygen sensing. Cell Metab. 1, 401–408 (2005).


Frezza C, Tennant DA, Gottlieb E. IDH1 mutations in gliomas: when an enzyme loses its grip. Cancer Cell 17, 7–9 (2010).


Pollard PJ, Briere JJ, Alam NA et al. Accumulation of Krebs cycle intermediates and over-expression of HIF1alpha in tumours which result from germline FH and SDH mutations. Hum. Mol. Genet. 14, 2231–2239 (2005).


Yang M, Soga T, Pollard PJ, Adam J. The emerging role of fumarate as an oncometabolite. Front. Oncol. 2, 85 (2012).


Selak MA, Armour SM, MacKenzie ED et al. Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha

Summarizes the history of metabolomics and clinical biochemistry.


Milne SB, Mathews TP, Myers DS, Ivanova PT, Brown HA. Sum of the parts: mass spectrometry-based metabolomics. Biochemistry 52(22), 3829–3840 (2013).


D’Alessandro A, Gevi F, Zolla L. A robust high resolution reversed-phase HPLC strategy to investigate various metabolic species in different biological models. Mol. Biosyst. 7(4), 1024–1032 (2011).


Guo B, Chen B, Liu A, Zhu W, Yao S. Liquid chromatography-mass spectrometric multiple reaction monitoring-based strategies for expanding targeted profiling towards quantitative metabolomics. Curr. Drug Metab. 13(9), 1226–1243 (2012).


D’Alessandro A, Gevi F, Zolla L. Targeted mass spectrometry-based metabolomic profiling through multiple reaction monitoring of liver and other biological matrices. Methods Mol. Biol. 909, 279–294 (2012).



Tautenhahn R, Cho K, Uritboonthai W, Zhu Z, Patti GJ, Siuzdak G. An accelerated workflow for untargeted metabolomics using the METLIN database. Nat. Biotechnol. 30(9), 826–828 (2012). Crown SB, Ahn WS, Antoniewicz MR. Rational design of 13C-labeling experiments for metabolic flux analysis in mammalian cells. BMC Syst. Biol. 6, 43–57 (2012).



D’Alessandro & Zolla

prolyl hydroxylase. Cancer Cell 7, 77–85 (2005). 53


Hewitson KS, Lienard BM, McDonough MA et al. Structural and mechanistic studies on the inhibition of the hypoxiainducible transcription factor hydroxylases by tricarboxylic acid cycle intermediates. J. Biol. Chem. 282, 3293–3301 (2007). Navis A, Niclou S, Fack F et al. Increased mitochondrial activity in a novel IDH1-R132H mutant human oligodendroglioma xenograft model: in situ detection of 2-HG and a-KG. Acta Neuropathol. Commun. 1(1), 18 (2013).


Maxwell PH, Pugh CW, Ratcliffe PJ. Activation of the HIF pathway in cancer. Curr. Opin. Genet. Dev. 11, 293–299 (2001).


Wise DR, Ward PS, Shay JE et al. Hypoxia promotes isocitrate dehydrogenasedependent carboxylation of alpha-ketoglutarate to citrate to support cell growth and viability. Proc. Natl Acad. Sci. USA 108, 19611–19616 (2011).


Vance DE, Vance JE. Physiological consequences of disruption of mammalian phospholipid biosynthetic genes. J. Lipid Res. 50, S132 (2009).


Levine AJ, Puzio-Kuter AM. The control of the metabolic switch in cancers by oncogenes and tumor suppressor genes. Science 330, 1340–1344 (2010).



Berkers CR, Maddocks OD, Cheung EC, Mor I, Vousden KH. Metabolic Regulation by p53 Family Members. Cell. Metab. doi:10.1016/j.cmet.2013.06.019 (2013) (Epub ahead of print).


Cheung EC, Ludwig RL, Vousden KH. Mitochondrial localization of TIGAR under hypoxia stimulates HK2 and lowers ROS and cell death. Proc. Natl Acad. Sci. USA 109(50), 20491–20496 (2012).


Bensaad K, Cheung EC, Vousden KH. Modulation of intracellular ROS levels by TIGAR controls autophagy. EMBO J. 28(19), 3015–3026 (2009).


Wanka C, Steinbach JP, Rieger J. Tp53-induced glycolysis and apoptosis regulator (TIGAR) protects glioma cells from starvation-induced cell death by up-regulating respiration and improving cellular redox homeostasis. J. Biol. Chem. 287(40), 33436–33446 (2012).




Schafer ZT, Grassian AR, Song L et al. Antioxidant and oncogene rescue of metabolic defects caused by loss of matrix attachment. Nature 461, 109–113 (2009).

D’Alessandro A, Marrocco C, Rinalducci S et al. Analysis of TAp73-dependent signaling via Omics technologies. J. Proteome Res. 12(9), 4207–4220 (2013) Wanka C, Brucker DP, Ba¨hr O et al. Synthesis of cytochrome C oxidase 2: a p53-dependent metabolic regulator that promotes respiratory function and protects glioma and colon cancer cells from hypoxia-induced cell death. Oncogene 31(33), 3764–3776 (2012). Hu W, Zhang C, Wu R, Sun Y, Levine A, Feng Z. Glutaminase 2, a novel p53 target gene regulating energy metabolism and antioxidant function. Proc. Natl Acad. Sci. USA 107(16), 7455–7460 (2010).


Vander Heiden MG, Lunt SY, Dayton TL et al. Metabolic pathway alterations that support cell proliferation. Cold Spring Harb. Symp. Quant. Biol. 76, 325–334 (2011).


Giacobbe A, Bongiorno-Borbone L, Bernassola F et al. p63 regulates glutaminase 2 expression. Cell Cycle 12(9), 1395–1405 (2013).

Describes the importance of metabolic reprogramming in cancer cells and its role in sustaining anabolic reactions.


Stavridi ES, Halazonetis TD. p53 and stress in the ER. Genes Dev. 18(3), 241–244 (2004).


Muller PA, Vousden KH, Norman JC. p53 and its mutants in tumor cell migration and invasion. J. Cell Biol. 192(2), 209–218 (2011).



Sherr CJ, McCormick F. The RB and p53 pathways in cancer. Cancer Cell 2(2), 103–112 (2002).

Pluquet O, Qu LK, Baltzis D, Koromilas AE. Endoplasmic reticulum stress accelerates p53 degradation by the cooperative actions of Hdm2 and glycogen synthase kinase 3beta. Mol. Cell. Biol. 25(21), 9392–9405 (2005).


Zocchi L, Bourdon JC, Codispoti A et al. Scotin: A new p63 target gene expressed during epidermal differentiation. Biochem. Biophys. Res. Commun. 367(2), 271–276 (2008).


Terrinoni A, Ranalli M, Cadot B et al. p73 alpha is capable of inducing scotin and ER stress. Oncogene 23(20), 3721–3725 (2004).


Yeung SJ, Pan J, Lee MH. Roles of p53, MYC and HIF-1 in regulating glycolysis the seventh hallmark of cancer. Cell. Mol. Life Sci. 65, 3981–3999 (2008).


Vousden KH, Ryan KM. p53 and metabolism. Nat. Rev. Cancer 9, 691–700 (2009).



Rufini A, Niklison-Chirou MV, Inoue S et al. TAp73 depletion accelerates aging through metabolic dysregulation. Genes Dev. 26(18), 2009–2014 (2012).


Zulato E, Curtarello M, Nardo G, Indraccolo S. Metabolic effects of anti-angiogenic therapy in tumors. Biochimie 94(4), 925–931 (2012).


Wyld L, Tomlinson M, Reed MW, Brown NJ. Aminolaevulinic acid-induced photodynamic therapy: cellular responses to glucose starvation. Br. J. Cancer 86(8), 1343–1347 (2002).


Sandulache VC, Ow TJ, Pickering CR et al. Glucose, not glutamine, is the dominant energy source required for proliferation and survival of head and neck squamous carcinoma cells. Cancer 117(13), 2926–2938 (2011).


Ogiso Y, Tomida A, Kim HD, Tsuruo T. Glucose starvation and hypoxia induce nuclear accumulation of proteasome in cancer cells. Biochem. Biophys. Res. Commun. 258(2), 448–452 (1999).


Berardinis RJ, Mancuso A, Daikhin E et al. Beyond aerobic glycolysis: transformed cells can engage in glutamine metabolism that exceeds the requirement for protein and nucleotide synthesis. Proc. Natl Acad. Sci. USA 104, 19345–19350 (2007).


Yuneva M, Zamboni N, Oefner P, Sachidanandam R, Lazebnik Y. Deficiency in glutamine but not glucose induces MYC-dependent apoptosis in human cells. J. Cell. Biol. 178(1), 93–105 (2007).


McCarthy N. Metabolism: Sensitivity to serine starvation. Nat. Rev. Cancer 13(2), 77 (2013).


Maddocks OD, Berkers CR, Mason SM et al. Serine starvation induces stress and p53-dependent metabolic remodelling in cancer cells. Nature 493(7433), 542–546 (2013).

This elegant paper describes the linkage between serine starvation responses, metabolic reprogramming (increased PPP fluxes and maintenance of the redox poise, activation of the serine synthesis pathway and Krebs cycle) and the most important tumor suppressor, Tp53.


Christofk HR, Vander Heiden MG, Harris MH et al. The M2 splice isoform of pyruvate kinase is important for cancer metabolism and tumour growth. Nature 452, 230–233 (2008).


Chaneton B, Hillmann P, Zheng L et al. Serine is a natural ligand and allosteric activator of pyruvate kinase M2. Nature 491(7424), 458–462 (2012).

Expert Rev. Proteomics 10(5), (2013)

Proteomics, metabolomics & drug development




Ye J, Mancuso A, Tong X et al. Pyruvate kinase M2 promotes de novo serine synthesis to sustain mTORC1 activity and cell proliferation. Proc. Natl Acad. Sci. USA 109(18), 6904–6909 (2012). Dandapani M, Hardie DG. AMPK: opposing the metabolic changes in both tumour cells and inflammatory cells? Biochem. Soc. Trans. 41(2), 687–693 (2013). Li L, Chen Y, Gibson SB. Starvation-induced autophagy is regulated by mitochondrial reactive oxygen species leading to AMPK activation. Cell. Signal. 25(1), 50–65 (2013).


Kupiec M, Weisman R. TOR links starvation responses to telomere length maintenance. Cell Cycle 11(12), 2268–2271 (2012).


Lamming DW, Ye L, Sabatini DM, Baur JA. Rapalogs and mTOR inhibitors as anti-aging therapeutics. J. Clin. Invest. 123(3), 980–989 (2013).


Efeyan A, Zoncu R, Sabatini DM. Amino acids and mTORC1: from lysosomes to disease. Trends Mol. Med. 18(9), 524–533 (2012).


Laplante M, Sabatini DM. mTOR signaling in growth control and disease. Cell 149(2), 274–293 (2012).


Laplante M, Sabatini DM. mTOR Signaling. Cold Spring Harb. Perspect. Biol. 4(2) a011593 (2012).


Sun H, Lesche R, Li DM et al. PTEN modulates cell cycle progression and cell survival by regulating phosphatidylinositol 3,4,5,-trisphosphate and Akt/protein kinase B signaling pathway. Proc. Natl Acad. Sci. USA 96(11), 6199–6204 (1999).


Dang CV, Le A, Gao P. MYC-induced cancer cell energy metabolism and therapeutic opportunities. Clin. Cancer Res. 15(21), 6479–6483 (2009).


Ying H, Kimmelman AC, Lyssiotis CA et al. Oncogenic Kras maintains pancreatic tumors through regulation of anabolic glucose metabolism. Cell 149(3), 656–670 (2012).


Son J, Lyssiotis CA, Ying H et al. Glutamine supports pancreatic cancer growth through a KRAS-regulated metabolic pathway. Nature 496(7443), 101–105 (2013).


Carracedo A, Weiss D, Leliaert AK et al. A metabolic prosurvival role for PML in breast cancer. J. Clin. Invest. 122, 3088–3100 (2012).


Binoux M. The IGF system in metabolism regulation. Diabete. Metab. 21(5), 330–337 (1995).


Chiarugi A, Do¨lle C, Felici R, Ziegler M. The NAD metabolome–a key determinant of cancer cell biology. Nat. Rev. Cancer 12(11), 741–752 (2012).


Amaravadi RK, Lippincott-Schwartz J, Yin XM et al. Principles and current strategies for targeting autophagy for cancer treatment. Clin. Cancer Res. 17, 654–666 (2011).


Bursch W, Karwan A, Mayer M et al. Cell death and autophagy: cytokines, drugs, and nutritional factors. Toxicology 254(3), 147–157 (2008).


Peng T, Golub TR, Sabatini DM. The immunosuppressant rapamycin mimics a starvation-like signal distinct from amino acid and glucose deprivation. Mol. Cell Biol. 22(15), 5575–5584 (2002).


Dando I, Donadelli M, Costanzo C et al. Cannabinoids inhibit energetic metabolism and induce AMPK-dependent autophagy in pancreatic cancer cells. Cell Death Dis. 4, e664 (2013).


Lee C, Raffaghello L, Longo VD. Starvation, detoxification, and multidrug resistance in cancer therapy. Drug Resist. Updat. 15(1–2), 114–122 (2012).


DiPaola RS, Dvorzhinski D, Thalasila A et al. Therapeutic starvation and autophagy in prostate cancer: a new paradigm for targeting metabolism in cancer therapy. Prostate 68(16), 1743–1752 (2008).


Raffaghello L, Lee C, Safdie FM et al. Starvation-dependent differential stress resistance protects normal but not cancer cells against high-dose chemotherapy. Proc. Natl Acad. Sci. USA 105(24), 8215–8220 (2008).


Awale S, Lu J, Kalauni SK et al. Identification of arctigenin as an antitumor agent having the ability to eliminate the tolerance of cancer cells to nutrient starvation. Cancer Res. 66(3), 1751–1757 (2006).


Granchi C, Minutolo F. Anticancer agents that counteract tumor glycolysis. ChemMedChem 7(8), 1318–1350 (2012).

Wishart DS, Knox C, Guo AC et al. DrugBank: a knowledgebase for drugs, drug actions and drug targets. Nucleic Acids Res. 36, D901–D906 (2008).


Madhok BM, Yeluri S, Perry SL, Hughes TA, Jayne DG. Targeting glucose metabolism: an emerging concept for anticancer therapy. Am. J. Clin. Oncol. 34(6), 628–635 (2011).

Kuhajda FP, Jenner K, Wood FD et al. Fatty acid synthesis: a potential selective target for antineoplastic therapy. Proc. Natl Acad. Sci. USA 91, 6379–6383 (1994).


Panigrahy D, Kaipainen A, Huang S et al. PPARalpha agonist fenofibrate suppresses tumor growth through direct and indirect angiogenesis inhibition. Proc. Natl Acad. Sci. U.S.A. 105(3), 985–990 (2008).


Samudio I, Harmancey R, Fiegl M et al. Pharmacologic inhibition of fatty acid


Sassen S, Miska EA, Caldas C. MicroRNA: implications for cancer. Virchows. Arch. 452(1), 1–10 (2008).


Jansson MD, Lund AH. MicroRNA and cancer. Mol. Oncol. 6(6), 590–610 (2012).


Rottiers V, Na¨a¨r AM. MicroRNAs in metabolism and metabolic disorders. Nat. Rev. Mol. Cell Biol. 13(4):239–250 (2012).


Peschiaroli A, Giacobbe A, Formosa A et al. miR-143 regulates hexokinase 2 expression in cancer cells. Oncogene 32(6):797–802 (2013).



Soga T. Cancer metabolism: key players in metabolic reprogramming. Cancer Sci. 104(3), 275–281 (2013). Birsoy K, Sabatini DM, Possemato R. Untuning the tumor metabolic machine: targeting cancer metabolism: a bedside lesson. Nat. Med. 18(7), 1022–1023 (2012).


This paper reviews the main concepts behind cancer metabolism reprogramming.


Samudio I, Fiegl M, Andreeff M. Mitochondrial uncoupling and the Warburg effect: molecular basis for the reprogramming of cancer cell metabolism. Cancer Res. 69(6), 2163–2166 (2009).


Tennant DA, Duran RV, Gottlieb E. Targeting metabolic transformation for cancer therapy. Nat. Rev. Cancer 10, 267–277 (2010).


DeBerardinis RJ, Lum JJ, Hatzivassiliou G, Thompson CB. The biology of cancer: metabolic reprogramming fuels cell growth and proliferation. Cell Metab. 7, 11–20 (2008).


Chen Z, Lu W, Garcia-Prieto C, Huang P. The Warburg effect and its cancer therapeutic implications. J. Bioenerg. Biomembr. 39(3), 267–274 (2007).


The paper summarizes the main strategies available to tackle cancer metabolism via drug-mediated glycolysis inhibition.


Feron O. Pyruvate into lactate and back: from the Warburg effect to symbiotic energy fuel exchange in cancer cells. Radiother. Oncol. 92(3), 329–333 (2009).





Semenza GL. Hypoxia-inducible factors: mediators of cancer progression and targets for cancer therapy. Trends Pharmacol. Sci. 33, 207–214 (2012).



D’Alessandro & Zolla

oxidation sensitizes human leukemia cells to apoptosis induction. J. Clin. Invest. 120, 142–156 (2010). 129






Carracedo A, Cantley LC, Pandolfi PP. Cancer metabolism: fatty acid oxidation in the limelight. Nat. Rev. Cancer 13(4), 227–232 (2013). Anastasiou D, Cantley LC. Breathless cancer cells get fat on glutamine. Cell Res. 22(3), 443–446 (2012). Wise DR, Thompson CB. Glutamine addiction: a new therapeutic target in cancer. Trends Biochem. Sci. 35(8), 427–433 (2010). Middleton B. The mitochondrial long-chain trifunctional enzyme: 2-enoyl-CoA hydratase, 3-hydroxyacyl-CoA dehydrogenase and 3-oxoacyl-CoA thiolase. Biochem. Soc. Trans. 22(2), 427–431 (2002). Zaidi N, Swinnen JV, Smans K. ATP-citrate lyase: a key player in cancer

metabolism. Cancer Res. 72(15), 3709–3714 (2012). 134

Sidransky D. Emerging molecular markers of cancer. Nat. Rev. Cancer 2(3), 210–219 (2002).


Ciocca DR, Calderwood SK. Heat shock proteins in cancer: diagnostic, prognostic, predictive, and treatment implications. Cell Stress. Chaperones. 10(2), 86–103 (2005).


Shu CW, Huang CM. HSP70s: from tumor transformation to cancer therapy. Clin. Med. Oncol. 2, 335–345 (2008).


D’Alessandro A, Zolla L. We are what we eat: food safety and proteomics. J. Proteome Res. 11(1), 26–36 (2012).


Folger O, Jerby L, Frezza C, Gottlieb E, Ruppin E, Shlomi T. Predicting selective drug targets in cancer through metabolic networks. Mol. Syst. Biol. 7, 501 (2011).


Facchetti G, Zampieri M, Altafini C. Predicting and characterizing selective

multiple drug treatments for metabolic diseases and cancer. BMC Syst. Biol. 6(1), 115 (2012). 140

Chang R, Xie L, Bourne P, Palsson B, Dunbrack R. Drug off-target effects predicted using structural analysis in the context of a metabolic network model. PLoS Comput. Biol. 6, e1000938 (2010).


Draoui N, Feron O. Lactate shuttles at a glance: from physiological paradigms to anti-cancer treatments. Dis. Model. Mech. 4(6), 727–732 (2011).


Kaelin WG Jr. The concept of synthetic lethality in the context of anticancer therapy. Nat. Rev. Cancer 5, 689–698 (2005).


Leha´r J, Krueger AS, Avery W et al. Synergistic drug combinations tend to improve therapeutically relevant selectivity. Nat. Biotechnol. 27, 659–666 (2009).

Expert Rev. Proteomics 10(5), (2013)

Suggest Documents