Among K-ras mutations, codon 12 mutations have been identified as those conferring a more aggressive phenotype. This aggressiveness is primarily associated with slow proliferation but greatly increased resistance to apoptosis. Using transfected NIH3T3 fibroblasts with a mutated K-ras minigene either at codon 12 (K12) or at codon 13 (K13), and taking advantage of [1,2-13C2]glucose tracer labeling, we show that codon 12 mutant K-ras (K12)-transformed cells exhibit greatly increased glycolysis with only a slight increase in activity along pathways that produce nucleic acid and lipid synthesis precursors in the oxidative branch of the pentose phosphate pathway and via pyruvate dehydrogenase flux. K13 mutants display a modest increase in anaerobic glycolysis associated with a large increase in oxidative pentose phosphate pathway activity and pyruvate dehydrogenase flux. The distinctive differences in metabolic profiles of K12 and K13 codon mutated cells indicate that a strong correlation exists between the flow of glucose carbons towards either increased anaerobic glycolysis, and resistance to apoptosis (K12), or increased macromolecule synthesis, rapid proliferation, and increased sensitivity to apoptosis.
The position of the point mutation in the c-K-ras gene seems to be associated with different degrees of aggressiveness in human tumors. Several lines of evidence suggest that the malignant potential of tumor cells may be influenced not only by the presence or absence of ras mutations, but also by the molecular configuration and context of any mutation present (1–3). For example, codon 13 mutations have less transforming capacity compared with codon 12 mutations in both in vitro and in vivo experimental systems (4–6); and in human colorectal carcinomas, codon 12 mutations are associated with increased resistance to apoptosis, and are much more frequent than codon 13 mutations, yet this is not so in adenomas. Additionally, among cells in vitro, K-ras codon 12 mutations confer resistance to apoptosis—and are associated with increased AKT/protein kinase B activation, whereas codon 13 mutations do not have these characteristics (7). In vivo studies have also confirmed that K12-induced soft tissue sarcomas in nude mice show shorter latency times before appearance, as well as lower apoptotic and mitotic rates (8). Stable isotope-based dynamic metabolic profiling is the simultaneous assessment of substrate flux within and among major metabolic pathways of macromolecule synthesis and energy production under various physiologic conditions (9, 10). Tumor cells have altered metabolism, revealed in metabolic profiles that show accelerated rates of glucose uptake and protein and nucleic acid synthesis (11, 12). These metabolic changes may confer a common advantage to many different types of cancers, increasing cells' ability to survive, proliferate, and invade (13). In this regard, it has been recently reported that ras-transformed cells exhibit increased rates of anaerobic glycolysis (14). The use of [1,2-13C2]glucose tracer in combination with mass spectrometry allows the evaluation of metabolic flux through the main pathways facilitating energy production and biosynthetic metabolism of the cell. In search of an explanation for the increased aggressiveness of cells with K-ras codon 12 mutations, we have used this powerful methodology to gain insight into the metabolic consequences of those mutations.
Materials and Methods
Cell lines and culture. The NIH3T3 human fibroblast transfectants contained a K-ras minigene with a G:C→A:T mutation at the first position of codon 12 (K12 cells) and a G:C→A:T mutation at the second position of codon 13 (K13). All lines were maintained in DMEM in the presence of 10% fetal bovine serum, at 37°C in 95% air-5% CO2. Geneticin Selective Antibiotic (Life Technologies, Gaithersburg, MD) was used as a selective antibiotic in K12 and K13 cells. Cell cultures were started in T75 culture flasks with the same cell number (13,333 cells/cm2) in DMEM with 10% fetal bovine serum and 10 mmol/L of [1,2-13C2]glucose (50% isotope enrichment), for 72 hours. At the end of the incubations, cells were centrifuged (1,500 rpm for 5 minutes) and incubation medium and cell pellets were obtained. Glucose and lactate incubation medium concentrations were determined as previously described (15, 16) using a Cobas Mira Plus chemistry analyzer (ABX). Lactate from the cell culture media was extracted by ethyl acetate after acidification with HCl. Lactate was derivatized to its propylamide-heptafluorobutyric form and the m/z 328 (carbons 1-3 of lactate, chemical ionization) was monitored for the detection of m1 (lactate with a 13C in one position) and m2 (double-labeled lactate) for the estimation of pentose cycle activity versus anaerobic glycolysis (17). RNA ribose was isolated by acid hydrolysis of cellular RNA after Trizol-purification of cell extracts. Ribose isolated from RNA was derivatized to its aldonitrile acetate form using hydroxylamine in pyridine and acetic anhydride. We monitored the ion cluster around the m/z 256 (carbons 1-5 of ribose, chemical ionization), in order to find the molar enrichment of 13C labels in ribose (17). Glutamate was separated from the medium using ion-exchange chromatography (18). Glutamate was converted to its n-trifluoroacetyl-n-butyl derivative and the ion clusters m/z 198 (carbons 2-5 of glutamate, electron impact ionization) and m/z 152 (carbons 2-4 of glutamate, electron impact ionization) were monitored. Isotopomeric analysis of C2-C5 and C2-C4 fragments of medium glutamate was done in order to estimate the relative contributions of pyruvate carboxylase and pyruvate dehydrogenase to the Krebs cycle (9, 19).
Gas chromatography/mass spectrometry. Mass spectral data was obtained on the HP5973 mass selective detector connected to a HP6890 gas chromatograph. The settings were as follows: GC inlet 230°C, transfer line 280°C, mass spectrometry source 230°C, MS Quad 150°C. An HP-5 capillary column (30 m length, 250 μm diameter, 0.25 μm film thickness) was used for ribose, glutamate, and lactate analyses.
Data analysis and statistical methods. In vitro experiments were carried out using three cultures each time for each treatment regimen and then repeated twice. Mass spectral analyses were carried out by three independent automatic injections by the sampler and accepted only if the standard sample deviation was <1% of the normalized peak intensity. Statistical analyses were done using the parametric unpaired, two-tailed independent sample t test with 99% confidence intervals and P < 0.01 was considered to indicate significant differences in glucose carbon metabolism between NIH-3T3 and their derived transformant cells (K12 and K13).
Rates of anaerobic glycolysis and direct glucose oxidation in the pentose cycle. Lactate secreted into cell culture medium can be used to measure label incorporation into the three-carbon metabolite pool. Direct [1,2-13C2]glucose catabolism through anaerobic glycolysis results in m2 lactate, whereas passage through the oxidative and nonoxidative branches of the pentose phosphate pathway of glucose results in m1 lactate. The total label incorporation in lactate is measured as Σmn = m1 + 2*m2 and reveals the contribution of glucose to lactate, providing an estimate of the contribution of other unlabeled carbon sources via degradation of the amino acids glutamine (glutaminolysis) and serine (serinolysis). Because initial labeled glucose enrichment was 50% (maximum potential m2 lactate would be 25%), calculations of direct flux from glucose to lactate were done taking into account glucose consumption (3T3, 3.83 ± 0.11 mmol/L; K12, 3.47 ± 0.06 mmol/L; K13, 3.75 ± 0.15 mmol/L) and lactate production (3T3, 4.7 ± 0.07 mmol/L; K12, 5.06 ± 0.17 mmol/L; K13, 5.35 ± 0.04 mmol/L). K12 cells showed only a slight increase in oxidizing glucose directly in the pentose cycle (Fig. 1A, gray column), whereas K13 cells had doubled pentose flux compared with the other cell lines (Fig. 1A, black column). Therefore, although lactate production was similar in both transfected lines, the 13C isotopomer distribution showed a significant increase in direct glucose oxidation and pentose recycling only in K13 cells.
Traffic of glucose carbons to nucleic acid synthesis is not altered in K12 cells. Stable isotope glucose labeling analysis allows us to measure the relative enrichment of 13C in a given metabolite pool. Σmn in ribose, calculated as m1*1 + m2*2 + m3*3 + m4*4, is understood as 13C content in a mole of RNA ribose molecules. Because this relative value cannot be compared with total ribose content, we corrected proportionally by cell number (data not shown) in order to assess carbon flux through the pentose cycle from labeled glucose to nucleic acid ribose. Normalized data showed a significant increase of ∼40% in the use of glucose for ribose synthesis in K13-transfected cells. K13 mutants preferentially used the nonoxidative branch of the pentose cycle for nucleic acid ribose synthesis as shown in Fig. 1B (black column), whereas K12 mutants primarily used oxidative ribose synthesis from glucose (Fig. 1C, gray column).
K12 transfectants present low Krebs cycle activity and null pyruvate carboxylase flux. The relative fluxes through pyruvate dehydrogenase and pyruvate carboxylase pathways were estimated from levels of m2 isotopomers of C2 to C4 and C2 to C5 fragments and 13C label enrichment in C2 to C5 fragment. We observed that in K13 cells Σmn was increased 2.3-fold with respect to control 3T3 cells, indicating a more active Krebs cycle in K13 mutants, whereas in K12 transfectants Krebs cycle activity was slightly decreased with respect to control cells. Pyruvate carboxylase flux, which is indicative of the use of Krebs cycle for amino acid synthesis, is zero in K12-transfected cells, indicating that the Krebs cycle is not used in anabolic processes in this mutant. Results also show a significant increase of pyruvate dehydrogenase flux in both transfectants (Fig. 1D), and a more intense pyruvate dehydrogenase flux in K13 transfectants.
Different phenotypes with varying tumor aggressiveness and apoptosis resistance characteristics as well as phenotype-unique alterations in signaling pathways have been reported for the two most important mutations in K-ras gene (codon 12 and 13; refs. 7, 8). We have used stable isotope-labeled glucose tracer technology and two in vitro K-ras-transfected NIH-3T3 cell models, K12 cells expressing increased levels of codon 12 cysteine point mutation, and K13 cells expressing an aspartic acid mutation at codon 13, to show that the specific molecular configuration of different single point mutations in the same oncogene can result in different, codon-distinctive metabolic profiles indicative of significantly different patterns of metabolic activities. These significantly different metabolic profiles and the very different patterns of substrate utilization they represent are the likely explanations for the differences in aggressiveness and resistance to apoptosis reported in these two K-ras-induced malignant phenotypes.
Our results show that K12 cells mainly route glucose to anaerobic glycolysis. We have estimated that 54.6 ± 1.01% of glucose is directly oxidized to lactate in K12 cells, whereas control and K13 cells showed rates of direct oxidation of glucose to lactate of 42.9 ± 0.36% and 50.36 ± 0.89%, respectively. This increased direct oxidation of glucose through anaerobic glycolysis occurs to the detriment of rerouting of glucose for anabolic purposes. This K-12 cell metabolic phenomenon seems to confer the increased resistance to apoptosis observed both in vitro and in vivo (7, 8). Conversely, K12 transfectants do not need high activity of glucose anabolic pathways because they do not show highly accelerated growth rates (8). Our results are also in line with the proposed integration of glycolysis and apoptosis by Akt/protein kinase B (20). In this regard, it is of note that K12 transfectants show higher AKT/protein kinase B activation.
On the other hand, K13 cells display a completely different metabolic profile. K13 transfectants show an increased use of anabolic pathways to satisfy the metabolic requirements associated with an increased growth rate observed both in vitro and in vivo. The increased use of anabolic pathways occurs to the detriment of anaerobic glycolysis and is likely to confer vulnerability to apoptosis. Thus, we observed a much more significant increase in the use of the nonoxidative pentose phosphate pathway and Krebs cycle in K13 cells than in K12 transfectants. Increased use of the oxidative branch of the pentose cycle enables cells not only to synthesize more ribose for nucleic acid requirements, but to recruit reducing power in the form of NADPH. This fact, coupled with the increased availability of acetyl-CoA associated with an increased use of the pyruvate dehydrogenase pathway, may be indicative of increased new membrane lipid synthesis.
We conclude that the specificity of phenotypic transformations in K12 and K13 mutants are related not only to altered signaling pathways but also to distinct alterations in metabolic profiles which, as manifestations of gene expression, provide metabolic “fingerprints” that are correlated with different tumor phenotypes. Thus, K12 mutants avoid cell death by increasing glycolysis and oxidative ribose synthesis, which confers resistance to apoptosis. On the other hand, the altered metabolism of the K13 mutants enhances cell proliferation by channeling glucose carbons to synthetic processes rather than via oxidative ribose synthesis, at the cost of losing resistance to apoptosis (Fig. 2). The present study not only shows that metabolomics can be a powerful tool in understanding mechanisms underlying different tumor phenotypes, but also shows that metabolic profiles may be used to identify aggressive tumors, in a manner complementary to the older, predictive tools of proteomic, and genomic profiles. However, further work is needed before directly linking the observed metabolic changes to transformation functions of the two mutant forms.
Grant support: SAF20002-02785 from the Ministerio de Ciencia y Tecnología, Spain and the European Union FEDER funds; FIS 01-1264 from the Fondo de Investigaciones Sanitarias, Spain; PHS M01-RR00425 of the General Clinical Research Unit; P01-CA42710 of the UCLA Clinical Nutrition Research Unit Stable Isotope Core; and “La Caixa” Oncology Grant Program (Fundació “La Caixa”, ON03/070-00).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.