Recent advances in cancer metabolism: a technological perspective

Abstract

Cancer cells are highly dependent on metabolic pathways to sustain both their proliferation and adaption to harsh microenvironments. Thus, understanding the metabolic reprogramming that occurs in tumors can provide critical insights for the development of therapies targeting metabolism. In this review, we will discuss recent advancements in metabolomics and other multidisciplinary techniques that have led to the discovery of novel metabolic pathways and mechanisms in diverse cancer types.

Fig. 1. Chromatography coupled to mass spectrometry-based metabolite profiling.

(Chromatography) The hydrophilic metabolites are separated by chromatography based on their chemical properties. For LC, the types of stationary phase and mobile phase, and the gradient of and composition of the mobile phase must be optimized to facilitate the optimal separation of metabolites of interest. In contrast, prior to GC-based separation, the metabolites must be derivatized to be volatile. Because the volatile-derivatized metabolites only interact with the stationary phase of the column and not the chemically inert gas mobile phase, the column oven temperature gradient needs to be optimized to achieve optimal metabolite separation. (Mass spectrometry) The separated metabolites are ionized and discriminated via mass spectrometry based on their molecular mass to charge ratio (m/z). The identity of metabolites is further determined by comparing with an authentic standard. (Metabolite ID) LC–MS analysis: after ionization by ESI, the ions with the same retention time can be discriminated via their MS/MS fragmentation (in triple Q) or via high-resolution mass accuracy (in Q-TOF or Orbitrap). In contrast, for GC–MS analysis, the metabolite ion specificity is based on their dissociation fragment pattern following EI-based hard ionization. TCA tricarboxylic acid, SGOC serine, glycine, and one carbon, LC liquid chromatography, GC gas chromatography, ESI electrospray ionization, EI electron (impact) ionization, Q-TOF quadrupole-time of flight, ACN acetonitrile; MeOH methanol, m/z mass to charge ratio, M(B)STFA N-methyl-N (or N,O-bis)-(trimethylsilyl) trifluoroacetamide, MTBSTFA N-(tert-butyldimethylsilyl)-N-methyltrifluoroacetamide, RT retention time, CE collision energy

Fig. 2. The use of stable isotopes for absolute quantification and tracing.

(Absolute quantification) Isotope-labeled metabolite standards can be used as internal standards (IS) to quantify the unknown amount of metabolites in biological samples. Since the IS has the same chemical properties as the target metabolite, variation during sample preparation and instrumental analysis can be corrected. Furthermore, the metabolite and IS can be distinguished during MS analysis due to the difference in their molecular weights. Finally, the quantity of the metabolite can be calculated by determining the peak area ratio of metabolite to IS, and multiplying by the known amount of the IS. (Metabolite tracing) The fate of isotope-labeled metabolite substrates can be traced into downstream products by chromatography coupled to MS. By calculating the isotope enrichment of a target metabolite, the relative contribution of different pathways to that metabolite pool can be evaluated. Furthermore, time-course analysis facilitates the determination of flux through metabolic pathways. MS mass spectrometry, m/z mass to charge ratio, LC liquid chromatography, GC gas chromatography, RT retention time

Fig. 3. Genetically encoded fluorescent biosensors facilitate dynamic monitoring of metabolism.

a (FRET) Many genetically encoded fluorescent biosensors are based on the FRET between donor (F_D) and acceptor (F_A) fluorophores flanking a bacterial or yeast metabolite-binding domain. In the absence of metabolite binding, excitation of the F_D results in the emission of its associated fluorescence. However, when the metabolite binds, the biosensor undergoes a conformational change that permits FRET between fluorophores and emission of F_A fluorescence. The F_A:F_D fluorescence ratio is thus indicative of the intracellular metabolite concentration. b (Ratiometric) Circularly permutated fluorophores associated with a metabolite-binding domain that recognizes two distinct but structurally similar metabolites can be employed for the ratiometric detection of these metabolite pairs (A & B). These fluorophores have two distinct excitation wavelengths (Ex₁ & Ex₂) that stimulate fluorescence of the same emission wavelength (Em). Occupancy of metabolite A increases the fluorescent signal associated with Ex₁ and vice versa, so that the fluorescence ratio of Ex₁:Ex₂ is representative of the ratio of metabolite A: metabolite B

Fig. 4. CRISPR/Cas9-based screening for the study of metabolism.

(CRISPR–Cas9) The CRISPR/Cas9 system induces DNA double strand breaks at target genes of interest, thereby inducing gene knockouts. Cas9 specificity is derived from the complementarity of the single-guide RNA (sgRNA) sequence to the target gene of interest, as well as the requirement of the PAM sequence DNA binding and cleavage. (CRISPR/Cas9-based loss-of-function screening) The CRISPR/Cas9 system can be applied to whole genome loss-of-function screening to identify metabolic genes that are essential under defined metabolic states. First, a library of sgRNAs targeting the whole human genome are cloned into a to lentiviral vector encoding Cas9, and followed by packaging into lentiviral particles. Following infection, cells are treated with vehicle or a metabolic inhibitor, serially passaged, and sequenced. By comparing the abundance of sgRNAs in the final cell populations (both vehicle or inhibitor) to their initial representation, the selective depletion of sgRNAs in the inhibitor treated population can be determined. CRISPR clustered regularly interspaced short palindromic repeats, PAM protospacer adjacent motif, sgRNA single-guide RNA