|Home | About | Journals | Submit | Contact Us | Français|
A common metabolic change in cancer is the acquisition of glycolytic phenotypes. Increased expression of glycolytic enzymes is considered as one contributing factor. The role of mitochondrial defects in acquisition of glycolytic phenotypes has been postulated but remains controversial. Here we show that functional defects in mitochondrial respiration could be induced by oncogenic H-RasQ61L transformation, even though the mitochondrial contents or mass was not reduced in the transformed cells. First, mitochondrial respiration, as measured by mitochondrial oxygen consumption, was suppressed in NIH-3T3 cells transformed with H-RasQ61L. Second, oligomycin or rotenone did not reduce the cellular ATP levels in the H-RasQ61L transformed cells, suggesting a diminished role of mitochondrial respiration in the cellular energy metabolism. Third, inhibition of glycolysis with iodoacetic acid reduced ATP levels at a much faster rate in H-RasQ61L transformed cells than in the vector control cells. The reduction of cellular ATP levels was reversed by exogenously added pyruvate in the vector control cells but not in H-RasQ61L transformed cells. Finally when compared to the HRasQ61L transformed cells, the vector control cells had increased resistance toward glucose deprivation. The increased resistance was dependent on mitochondrial oxidative phosphorylation since rotenone or oligomycin abolished the increased survival of the vector control cells under glucose deprivation. The results also suggest an inability of the H-RasQ61L transformed cells to reactivate mitochondrial respiration under glucose deprivation. Taken together, the data suggest that mitochondrial respiration can be impaired during transformation of NIH-3T3 cells by oncogeneic H-RasQ61L.
Pioneering studies by Warburg over seven decades ago demonstrated that a vast majority of human and animal tumors display a high rate of glycolysis under aerobic conditions, a phenomenon known as the Warburg effect.1,2 The preferential utilization of aerobic glycolysis for cellular energy needs by cancer cells has been confirmed in a wide spectrum of cancers and successfully developed for tumor imaging with [18F] fluorodeoxyglucose positron emission tomography.3,4 Contributing factors toward acquisition of glycolytic phenotypes by cancer cells include, but not limited to, an increased expression of glucose transporters, glycolytic enzymes, lactate dehydrogenases, and dysfunctions in oxidative phosphorylation in mitochondria.1,2 Although originally posited by Warburg, the contribution of mitochondrial defects toward the glycolytic phenotypes of cancer cells remains controversial, partly because mitochondria as a cellular organelle can be readily observed in a majority of cancer cells and mutations or depletions of mitochondrial DNA cannot entirely account for the aerobic glycolysis observed in cancer cells.1,2
As integrators of cellular energy metabolism, generation of reactive oxygen species (ROS), aging, and the initiation of apop-tosis, mitochondria can contribute approximately 90% of cellular ATP through the citric acid cycle and the electron transport chain coupled with oxidative phosphorylation (OXPHOS). Generation of ATP through OXPHOS in the mitochondria is an efficient and preferred metabolic process, which produces far more ATP molecules from a given amount of glucose compared with glycolysis. The major substrates for mitochondria OXPHOS are NADH as the primary electron donor, oxygen as the terminal electron acceptor, and pyruvate, the end product of glycolysis, as the primary carbon source. It has been shown that hypoxiainducible factor 1 (HIF-1) can actively downregulate mitochondrial respiration during cellular adaptation to hypoxia.5–7
Emerging evidence suggests that oncogenes and/or tumor suppressors can regulate mitochondrial respiration and by extension, the cellular energy metabolism. It has been recently demonstrated that p53, a putative tumor suppressor frequently mutated in cancers, regulates mitochondrial respiration.8 Among various oncogenes, ras, myc and akt mutations are known to increase glycolysis. For example, in yeast, the Ras-cAMP pathway can activate the glycolytic enzyme 6-phosphofructo-1-kinase via cAMP-dependent protein kinase A.9 While the effects of oncogenes on the expression of glycolytic enzymes have been examined in a number of studies, it remains unknown whether mitochondrial respiration can be impacted by oncogenes during carcinogenesis.
The present study was undertaken to test the hypothesis whether oncogenes such as ras can affect mitochondrial respiration. Ras was chosen in our study because activating mutations in the K-, H- or N-Ras proto-oncogene family occur in a high percentage of human epithelial cancers. It is estimated that mutated Ras proteins are involved in 20% of all human cancers.10,11 Deregulated Ras activation in tumors is accounted for by mutations in codons 12, 13 and 61 that cause Ras to remain in the constitutive GTP-bound, activated state. Using mouse fibroblast NIH-3T3 cells as a model of oncogenic transformation, we herein demonstrate that oncogenic H-RasQ61L caused severe impairment in mitochondrial respiration and shifted the energy metabolism toward independence of mitochondria in NIH-3T3 cells. Our studies suggest that mitochondrial respiration is actively suppressed in cells transformed by oncogene H-RasQ61L.
To study the changes in cellular energy metabolism as a result of oncogenic transformation, NIH-3T3 cells were transfected with the activated form of H-RasQ61L or its backbone vector as the control. Ras proteins are converted from inactive (GDP-bound) to active (GTP-bound) states for cellular signaling pathways downstream of receptor tyrosine kinases, such as the epidermal growth factor receptor family.11,12 The mutation at position 61 of H-RasQ61L can result in a five-fold increase in exchange of GTP for GDP in addition to decreased GTP hydrolysis,13 leading to a high transformation rate in NIH-3T3 cells.14 As shown in Figure 1A, there was a large increase in GTP-bound Ras proteins in H-RasQ61L transfected cells (designated as H-Ras), compared to the vector control (designated as pUSE).
To study the effect of oncogenic transformation on mitochondria, we examined mitochondrial localization, content and mass in H-RasQ61L transformed cells. As shown in Figure 1B, mitochondria were dispersed in the live vector control cells as revealed by Mitotracker Green FM staining. In contrast, the mitochondria in H-RasQ61L transformed cells tended to form aggregates in the perinuclear regions (Fig. 1B), consistent with the observations made in cells transfected with K-Ras oncogene.15 Flow cytometrical analysis of cells stained with Mitotracker GreenFM revealed that H-RasQ61L transformed cells had slightly, but significantly, higher mitochondrial mass than pUSE control cells (Fig. 1C). Interestingly, there was a distinct sub-population of H-RasQ61L transformed cells with smaller mitochondrial mass (Fig. 1C and arrowed), suggesting a greater heterogeneity in mitochondrial mass in the H-RasQ61L transformed cells than those in the pUSE cells. To further study whether mitochondria content can be altered as part of oncogenic transformation, we examined mitochondria protein and DNA content in H-RasQ61L transformed cells in comparison with pUSE cells.
As shown in Figure 1D, there was no significant difference in the amounts of mitochondrial proteins isolated from NIH-3T3, pUSE or H-RasQ61L cells. Neither did we find significant differences in the amount of mitochondrial DNA in the mitochondrial preparations (Fig. 1E). The results suggest that H-RasQ61L transformation slightly increased mitochondrial mass and most significantly, altered cellular distribution of mitochondria.
To further examine the effects of H-RasQ61L transformation on mitochondria, we evaluated the levels of mitochondrial gene expression by RT-PCR. As shown in the Figure 2A, there was no appreciable difference in the RNA levels of various mitochondrion-encoded genes. Western blot analysis of select mitochondrial proteins revealed a general increase in Complex II-Ip subunit, Complex III core 2 protein, and Complex V F1α subunit in the mitochondrial preparations isolated from the H-RasQ61L transformed cells than those from the vector control cells (Fig. 2B, left). In contrast, the Complex IV COX II was reduced in mitochondria isolated from H-RasQ61L transformed cells when compared to those from the vector control (Fig. 2B, left). To determine whether the changes in the levels of proteins in the mitochondrial observed are due to the changes of gene expression, we analyzed the protein levels in total cell lys-taes. Analysis of total cell lysates, however, did not confirm the increases in the levels of Complex II-Ip subunit or Complex III core 2 protein as observed in mitochondrial preparations (Fig. 2B, right). The reduction of Complex IV COX II subunit was confirmed in the analysis of total cell lystates. The discrepancy in the relative levels of select mitochondrial proteins observed in mitochondrial preparations versus total lysates may be due to an increased mitochondrial mass or biogenesis in the H-RasQ61L transformed cells.
Immunocytochemical staining further confirmed a reduction in Complex IV COXII subunit and an increase in Complex V F1α subunit in the H-RasQ61L transformed cells when compared to the pUSE vector control cells (Fig. 2C). In addition to Complex V F1α subunit, other subunits of Complex V were also found increased in the H-RasQ61L transformed cells (Fig. 2C and bottom two panels). We next determine whether the reduced levels of Complex IV COXII subunit can impact on the activities of Complex IV. As shown in the Figure 2D, there was a clear reduction in the complex IV activities in the mitochondrial preparations from the H-RasQ61L transformed cells (Red line) when compared to those from the pUSE vector control cells (Green line). However, after normalization with the protein levels of COX, there was no difference in the specific activities of Complex IV oxidase (Fig. 2E), suggesting that the reductions in the Complex IV activities observed in the mitochondrial preparations from H-RasQ61L transformed cells are probably due to the reduced levels of COX proteins, rather than the potential mutations or posttranslational modifications of Complex IV oxidases.
The activities of Complex V were found slightly higher in the mitochondrial preparations from the H-RasQ61L cells than those from pUSE vector control cells (Data not shown), in consonance with the increased levels of Complex V proteins in the H-RasQ61L transformed cells (Fig. 2C).
As originally postulated by Otto Warburg, impairment in mitochondrial respiration is a major contributing factor to the increased glycolysis observed in culture tumor slices.1 A number of studies have examined the effects of Ras on oxygen consumption at the cellular level with different results. In yeast the Ras-cAMP pathway was shown to increase the levels of mitochondrial oxidative phosphorylation complexes.16 H-Ras transformation was shown to increase cellular oxygen consumption17 and increase mitochondrial metabolism in human bronchial epithelial cells.18
However, in rat embryo fibroblast cells, H-Ras stimulated glycolysis and inhibited oxygen consumption, but when combined with c-myc, stimulated oxygen consumption.19 Since other cellular elements, in addition to mitochondria, can also consume oxygen, it is important to measure oxygen consumption resulting from the respiration from mitochondria.
To this end, we utilized inhibitors of the mitochondrial electron transport chain to discern the effects of H-RasQ61L transformation on mitochondrial oxygen consumption. As shown in the Figure 3A, under the unperturbed condition, the rate of oxygen consumption was lower in H-RasQ61L transformed cells than that of the vector control cells (pUSE) or parental NIH-3T3 cells (not shown). Perturbation with antimycin A, an inhibitor of the complex III, reduced oxygen consumption in the pUSE and NIH-3T3 cells to the levels comparable to those in the H-RasQ61L cells (Fig. 3A and B). Similar results were obtained with rotenone, an inhibitor of the ‘ Complex I of mitochondrial electron transport chain (Fig. 3A). The results suggest an essential absence of mitochondrial oxygen consumption in the H-RasQ61L transformed cells. The results also suggest that the oxygen consumption in H-RasQ61L transformed cells was not due to mitochondrial respiration. Taken together, the data suggest that mitochondrial respiration, as measured by oxygen consumption, is impaired in H-RasQ61L transformed cells.
The impairment of mitochondrial respiration as result of H-RasQ61L transformation led us to examine the resultant changes in cellular energy metabolism. First we examined the changes in cellular ATP levels in the presence of rotenone, antimycin A, or FCCP. As shown in Figure 4A, rotenone reduced cellular ATP levels in a time dependent manner in pUSE cells but not in H-RasQ61L transformed cells. In pUSE cells, rotenone treatment significantly reduced the cellular ATP levels within 5 to 15 min (Fig. 4A). The cellular ATP levels rebounded back to the normal level after 60 min (p > 0.05, Fig. 4A). Similar results were obtained with antimycin A, another inhibitor of mitochondrial electron transport chain (Data not shown), or with FCCP, an uncoupler of oxidative phosphorylation (Fig. 4B).
Next we examined the cellular ATP levels after treatment with graded levels of rotenone or antimycin A for 10 min. As shown in Figure 4C, the ATP levels in the pUSE control cells were reduced by treatment with rotenone or antimycin A at doses as low as 1 µM. In contrast, neither rotenone nor antimycin A had significant effects on the ATP levels in the H-RasQ61L transformed cells, even at highest dose tested (10 µM). The results further suggest that unlike the pUSE control cells, the energy metabolism in the H-RasQ61L transformed cells is independent on the mitochondrial electron transport chain.
We further examined the effects of oligomycin, an inhibitor of F0-F1 ATPase, on the ATP levels in H-RasQ61L transformed cells. As shown in Figure 4D, oligomycin reduced the ATP levels in the pUSE control cells. In contrast, under same treatment conditions, the ATP levels in H-RasQ61L transformed cells were only slightly reduced (Fig. 4D). The results further suggest that energy metabolism in H-RasQ61L transformed cells were less dependent upon mitochondrial oxidative phosphorylation for the energy generation than the pUSE control cells.
The temporal patterns of cellular ATP levels, in the presence of inhibitors of the oxidative phosphorylation chain in mitochondria, suggest that the reductions in cellular ATP levels observed within 5~15 min of treatment were due to inhibition of mitochondrial respiration. The rebound of cellular ATP levels after prolonged treatment (60 min) suggests that the cellular ATP levels are tightly controlled and other compensatory systems, such as reduced ATP consumption or increased ATP generation via mitochondrion independent pathways, can be activated to keep the steady state level of cellular ATP. The rebounds in cellular ATP levels after prolonged treatment also suggest that the reductions in cellular ATP levels within 5~10 min treatment were not due to a non-specific reduction in the number of viable cells. The difference in the temporal patterns observed between pUSE and HRasQ61L cells (Fig. 4A and B) suggest a diminished role for mitochondrial oxidative phosphorylation in cellular energy metabolism in H-RasQ61L transformed cells.
To study whether the reduced contribution of mitochondria to cellular energy metabolism in HRasQ61L transformed cells is compensated by an increase in glycolysis, we measured glucose consumption. As shown in Figure 5A, the H-RasQ61L-transformed cells consumed glucose at higher rate than the vector control or parental NIH-3T3 cells. To determine the contribution of glycolysis toward cellular energy metabolism, we measured the changes in cellular ATP levels after treatment with IAA, an inhibitor of glyceraldehyde 3-phosphate dehydrogenase in the glycolytic pathway.20 As shown in Figure 5B, IAA reduced ATP levels in both H-RasQ61L transformed cells and their vector control cells. However, the kinetics in the changes of ATP levels revealed that H-RasQ61L transformed cells were more susceptible to IAA treatment when compared to the vector control cells, especially at earlier time points. The data suggest an increased dependence of the H-RasQ61L transformed cells on glycolysis to synthesize cellular ATP.
Pyruvate is the product of glycolytic pathway that enters the TCA cycle in mitochondria. We examined whether pyruvate can restore ATP production when glycolysis is inhibited. H-RasQ61L transformed cells or pUSE control cells were treated with IAA, in the absence or presence of pyruvate, and then cellular ATP levels were measured. As shown in Figure 5C, pyruvate significantly prevented the depletion of cellular ATP induced by IAA in the pUSE cells. In contrast, pyruvate did not significantly attenuate the IAA-induced ATP depletion in the H-RasQ61L transformed cells (Fig. 5C), suggesting that mitochondria in the H-RasQ61L transformed cells could not utilize the exogenously added pyruvate for ATP generation. The results further suggest a diminished role of mitochondria in the energy metabolism in the H-RasQ61L transformed cells.
The switch in energy metabolism from mitochondrial oxidative phosphorylation toward glycolysis in the H-RasQ61L transformed cells led us to examine whether the altered energy metabolism can be targeted to kill the transformed cells selectively. As shown in Figure 6A, at low concentrations (5 and 10 µM), IAA killed H-RasQ61L transformed cells at much higher rates than the vector control cells, suggesting an enhanced sensitivity of H-RasQ61L transformed cells to inhibition of glycolysis. However, at higher concentration (25 µM), IAA killed both H-RasQ61L and pUSE cells at similar rates, indicating an essential role of glycolysis in energy metabolism in normal cells as well.
Next we examined the effects of glucose deprivation on the survival of H-RasQ61L transformed cells. As shown in Figure 6B, when cultured in glucose-free media, significantly higher death rates were found in H-RasQ61L transformed cells when compared with those in pUSE cells, suggesting an increased dependence of H-RasQ61L transformed cells on glucose for survival.
To examine the importance of the mitochondrial respiratory chain in the survival of pUSE cells in glucose-free media, the cells were treated with rotenone or oligomycin in combination with glucose deprivation. As shown in Figure 6C and D, rotenone or oligomycin treatment increased cell death only slightly in the presence of glucose. However, when cultured in glucose free media, rotenone or oligomycin induced massive cell death, suggesting an essential role for mitochondrial respiration chain in the survival of cells under glucose deprivation.
Taken together, the data suggest that the impairment of mitochondrial respiration, either due to H-RasQ61L transformation or to treatment with inhibitors of mitochondrial respiration, causes an enhanced dependence on glucose, or glycolysis for their survival. On the other hand, the data also confirm the inability of the H-RasQ61L transformed cells to reactivate or utilize oxidative phosphorylation in mitochondria to cope with glucose deprivation, as pUSE vector control cells did.
Otto Warburg postulated as early as the 1930s that mitochondrial respiration is impaired in cancer cells, leading to the observed elevation of glycolysis.1 Although the increased glycolysis later has been well documented in a spectrum of cancers and further utilized for imaging of tumors in situ via [18F] fluorodeoxyglucose positron emission tomography, the presumed impairment of mitochondrial respiration in cancer cells has never been convincingly demonstrated. This report provides direct evidence suggesting that although mitochondrial content and mass are not reduced by H-RasQ61L transformation, mitochondrial respiration is impaired in the transformed cells. First, oxygen consumption due to mitochondrial respiration was reduced to minimal levels in the H-RasQ61L transformed cells. Second, mitochondrial oxidative phosphorylation did not contribute to cellular ATP homeostasis in the H-RasQ61L transformed cells. Third, the H-RasQ61L transformed cells had reduced ability to utilize pyruvate to synthesize ATP, when compared to the vector control cells. Fourth, mitochondrial respiration was required for the vector control cells to resist glucose deprivation. The increased sensitivity of the H-RasQ61L transformed cells toward glucose deprivation, or inhibition of glycolysis, suggests that the transformed cells could not activate or utilize oxidative phosphorylation for their survival. Our data suggest that defects in mitochondrial respiration can be acquired as part of cellular transformation induced by oncogene H-RasQ61L.
In our study, we found that the mitochondria tended to form aggregates in the perinuclear region of H-RasQ61L transformed cells, while they were evenly distributed in the vector control cells. Similar observations regarding the altered cellular localization of mitochondria were also made in the K-Ras transformed cells.15 It should be noted that the formation of mitochondria aggregates was not due to the apoptotic process since we did not observe any significant reduction in viability in HRasQ61L transformed cells. It remains to be determined whether the altered cellular distribution of mitochondria is due to possible defects in the cytoskeleton in transporting mitochondria or altered fission and fusion of mitochondria in the H-RasQ61L transformed cells.
While mitochondrial respiration was significantly compromised in the H-RasQ61L transformed cells, we did not detect striking changes in mitochondrial DNA content or mitochondrial gene expression. In fact, we detected a slight increase in mitochondrial mass and increased incorporation of mitochondrial proteins that encoded by nuclear genes into mitochondria, as evidenced by the relative contents of the proteins in total cell lystaes and in mitochondrial preparations. In yeast, activation of Ras signaling pathway has been shown to increase the content of mitochondrial respiration complexes.16 In rat embryo cells, H-Ras transformation was found to reduce oxygen consumption, but when combined with c-myc, stimulated oxygen consumption.19 Oncogenic H-RasV12 was found to increase mitochondrial metabolism in human bronchial epithelial cells18 and may stimulate oxygen consumption.17 However, since the oxygen consumption can be ascribed to mitochondria (80~90% in normal cells) or to elements not related to mitochondria,22 we used a perturbation approach to differentiate the mitochondrial oxygen consumption from those due to other sources. Using this approach, we were able to determine that the respiration of mitochondria was significantly compromised in H-RasQ61L transformed cells. It was found that the rate of the oxygen consumption in the H-RasQ61L transformed cells was not significantly reduced by rotenone or antimycin A, suggesting that most of oxygen consumption in the transformed cells is due to cellular elements other than the mitochondrial respiratory chain. In contrast, 40~50% of oxygen consumption was inhibited by rotenone or antimycin A in the vector control cells. The data suggest that mitochondrial respiration, as measured by oxygen consumption, is suppressed in the H-RasQ61L transformed cells.
In consonance with the suppression of mitochondrial respiration, we observed that the energy metabolism of H-RasQ61L transformed cells was essentially independent of OXPHOS in mitochondria.
The cellular ATP levels in H-RasQ61L transformed cells were essentially recalcitrant to treatment with rotenone or antimycin A. Neither did treatment with oligomycin A, an inhibitor of F0-F1 ATP synthase (Complex V), have any significant effects on cellular ATP levels. In contrast, the cellular ATP levels in the pUSE control cells can be rapidly, albeit transiently, reduced by treatment with rotenone, antimycin A or oligomycin. Taken together, the findings further support a diminished role for mitochondrial respiratory chain in the energy metabolism in H-RasQ61L transformed cells.
Despite the diminished role of mitochondria in energy metabolism, the steady state levels of ATP in unperturbed H-RasQ61L transformed cells were found remarkably similar to those in the vector control cells. H-RasQ61L transformed cells presented a glycolytic phenotype as evidenced by increased consumption of glucose and increased acidosis. Perturbation experiments with a glycolysis inhibitor revealed that H-RasQ61L transformed cells were dependent on glycolysis for their energy needs to a greater extent than the pUSE control cells. Interestingly we found that exogenously added pyruvate could restore cellular ATP depleted by IAA in pUSE cells, but not in the H-RasQ61L transformed cells. As the end product of glycolysis, pyruvate enters mitochondria to generate ATP via the TCA cycle and oxidative phosphorylation. The inability of pyruvate to attenuate the IAA-induced ATP depletion in the H-RasQ61L transformed cells further suggest a diminished role for mitochondria in the energy metabolism in the transformed cells. Taken together, the data strongly suggest a switch from mitochondrial respiration to glycolysis in cellular energy metabolism as result of H-RasQ61L transformation. Recently it is shown that mitochondrial localization of STAT3 was required for oncogenic Ras to transform cells.21 Further studies are required to determine whether STAT3 is a downstream effector for Ras to exert its suppressive functions on mitochondrial respiration.
On one hand, the increased dependence on glycolysis for their energy needs makes the transformed cells vulnerable to glucose deprivation. When cultured in glucose-free but glutamine containing media, the pUSE control cells survived reasonably well. In contrast, the H-RasQ61L transformed cells were found highly susceptible to glucose deprivation. Functional mitochondrial respiration was required for the vector control cells to survive under glucose deprivation. Both rotenone and oligomycin abolished the resistance of the vector control cells toward glucose deprivation. The susceptibility of the H-RasQ61L transformed cells toward glucose deprivation, on the other hand, further suggests impairment in mitochondria respiration as result of transformation.
In conclusion, the experimental data presented herein demonstrate that mitochondria respiration can be functionally impaired during cellular transformation, even though mitochondria as organelles are abundantly present. In the cells transformed by the H-RasQ61L oncogene, mitochondrial respiration was suppressed as evidenced by the lack of mitochondrial oxygen consumption, the lack of mitochondrial contribution to cellular ATP, and the inability to utilize pyruvate or glutamate when deprived of glucose. Our studies demonstrate a profound change in cellular metabolism as result of oncogenic transformation, which can be potentially targeted to develop new approaches to treat cancer.
All cell culture, immunochemistry, and molecular biology reagents were purchased from Invitrogen unless otherwise indicated. The GenePORTER® liposome transfection reagent was from Genlantis (San Diego, CA). ATP assay kit was ordered from Promega. All other chemicals and reagents were ordered from Sigma unless indicated otherwise.
NIH-3T3 cells were purchased from ATCC and were cultured in DMEM medium containing 10% (v/v) bovine serum (BS) with 100 µg/ml each of penicillin/streptomycin at 37°C under an atmosphere of 95% air and 5% CO2. NIH-3T3 cells (~80% confluence) were transfected with a plasmid harboring activated Q61L mutant H-Ras13 or backbone vector pUSE (Millipore, Billerica, MA) formulated in GenePORTER transfection liposomes. An equal volume of complete medium without any antibiotics was added after 6 hours. After 24 h, the medium was changed to complete medium with 10% bovine serum, supplemented with 500 µg/ml of G418 to select stable transfectants. To rule out possible artifacts introduced during selection of single clones, stable transfectants were pooled and used for studies. Cells transfected with construct expressing constitutively active H-RasQ61L were designated H-RasQ61L or H-Ras or those with the backbone vector as pUSE.
The levels of GTP-bound Ras (active form) in pUSE or H-RasQ61L cells were measured using EZ-Detect™ Ras Activation Kit (Pierce), according to the manufacturer’s instruction. Briefly, the cells were grown to 90–100% confluence in one 100 mm culture dish and harvested in 500 µl lysis/binding/washing buffers. The lysates (500 µg) were incubated with GST-Raf1-RBD (to pull down active Ras) in the presence of SwellGel Immobilized Glutathione at 4°C for 1 hour in a spin column. After incubation, the mixture was centrifuged at 8,000 xg to remove the unbound proteins. The resins were washed three times with lysis/binding/wash buffer and the sample was eluted by adding 50 µl of 2X SDS sample buffer and boiled at 95°C for 5 minutes. Half (25 µl) of the sample volumes were used for western blot analysis.
For analysis of Ras proteins, the samples above were resolved by SDS-PAGE (14% Polyacrylamide mini-gel) and transferred to a PVDF membrane. The level of Ras was detected by western blotting using a specific antibody. The membrane was washed (3x) with TBS-Tween (0.1%) and probed with goat anti-rabbit fluorescently labeled secondary antibody (1:5,000) for 1 h at room temperature and washed (3x) with TBS-T for a total of 15 min. The immunoblots were visualized by an Infrared imaging system (Odyssey, Lincoln, Nebraska).
For analysis of OXPHOS proteins, extracts of mitochondrial proteins or total cellular proteins were resolved by SDS-PAGE, blotted onto PVDF membrane and probed with the following antibodies: Complex I subunit NDUFB8 (MS105), Complex II subunit 30 kDa (MS203), Complex III subunit Core 2 (MS304), Complex IV subunit II (MS405), and ATP synthase subunit alpha (MS507) as an optimized premixed cocktail (1:200 dilutions) (MitoSciences, Eugene, Oregon). For staining, goat anti-mouse fluorescently-labeled secondary antibody (diluted 1:5,000 in blocking buffer) was used. The immunoblots were visualized and quantified by Infrared imaging system.
Cells were cultured in DMEM-10% BS for up to seven days without changing media. Culture supernatants were collected at 24 h intervals, and glucose levels were measured using a Glucose (HK) Assay kit, according to manufacture’s instruction (Sigma-Aldrich, St. Louis, MO). Basically, glucose was phosphorylated by hexokinase in the presence of ATP. Glucose-6-phosphate was then oxidized to 6-phospho-gluconate in the presence of oxidized nicotinamide adenine dinucleotide (NAD) in a reaction catalyzed by glucose-6-phosphate dehydrogenase. During this oxidation, an equimolar amount of NAD was reduced to NADH, which was measured by the absorbance at 340 nm using a spectrophotometer.
MitoTracker GreenFM dye (Molecular Probes, Invitrogen), which becomes fluorescent once it accumulates in the membrane lipids of mitochondria independent of membrane potential, was used to evaluate the distribution of mitochondria. Cells were cultured on coverglasses overnight, and then stained with MitoTracker Green dye at 37°C in the CO2 incubator for 30 minutes. The labeling solution was aspirated away and the cells were washed with pre-warmed media three times. The coverglasses were then removed and mounted on a slide for observation of the distribution of mitochondria under 40X or 60X objectives of an epifluorescence microscope (Olympus). Images were captured with a CCD camera linked with a computer.
The cells were first seeded on the 6-well plates with coverglasses. After reaching to 50% to 70% confluency, the cells were fixed with 2% paraformaldhyde at room temperature for 20 minutes. After washing with 1X PBS for three times, the fixed cells were treated the 0.1% Triton X-100 for 1 minute at room temperature for increasing the permeability. The blocking step with 1% BSA for 30 minutes was followed. The cells were incubated with the primary antibodies indicated at room temperature for 2 hours and then after washing with 1X PBS for 3 times, incubated with Alexa Fluor® 488 goat anti-mouse IgG(H + L) secondary antibody for 1 h (1:100 dilution). After rinsing with IX PBS for 4 times, the slides were mounted with Prolong® Gold antifade reagent (Invitorgen) and observed under microscope.
Mitochondrial mass was analyzed with a MitoTracker Green FM dye, which stains mitochondria independent of membrane potential. Live cells were stained with 10 µM MitoTracker Green FM for 15 min and subjected to flow cytometry analysis.
Intact mitochondria were isolated by Cell Mitochondria Isolation Kit (Sigma). pUSE and HRasQ61L cells were grown to ~90% confluence. Cells were harvested by trypsin digestion, washed, resuspended in ice cold PBS, and counted. Equal numbers of cells were pelleted by centrifugation for 5 minutes at 600 xg at 4°C, and the cell pellets were re-suspended in 0.65–2 ml of Lysis Buffer per 3 × 107 cells. After incubation on ice for 5 min, the cells were pipeted up and down at 1 min intervals three times. Then 2 volumes of 1x extraction buffer were added. The homogenate was centrifuged at 600 xg for 10 min at 4°C. The supernatants were carefully transferred to a fresh tube and centrifuged at 3,500 xg for 10 min at 4°C. After careful removal of the supernatant, the pellet was resuspended in CelLytic M buffer for analysis including protein content using BCA kit (Pierce).
Mitochondrial DNA was extracted from mitochondria extracted from exponentially growing cells. Mitochondria preparations were resuspended in 0.5 ml 10% SDS extract buffer (containing 5% Proteinase K and 1% RNase), incubated at 37°C for 10 hours, and then incubated at 50°C water bath for 5 hours for thorough degradation of RNA and proteins. The mitochondrial DNA was extracted using 0.2 ml cholorform/isoamyl alcohol buffer, followed by isopropanol precipitation, and resuspended in TE buffer. The mitochondrial DNA contents were analyzed by the levels of M7 amplicons using GGA GCA GTG TTT GCT ATC ATA GC and TGA TGG CTA CAA CGA TTG GGA ATC as forward and reverse primer, respectively. For reference of cell number, genomic DNA content was extracted from identical sets of cells and analyzed by the levels of amplicons of 18 s RNA gene using ACC AAC TGG GAC GAT ATG GAG AAG A and TAC GAC CAG AGG CAT ACA GGG ACA A, as forward and reverse primer, respectively. The amplification was conducted at the following settings: 94°C × 3′ for one cycle; 94°C × 30″, 60–65°C × 1′, 72°C × 1′ for 30 cycles; 72°C × 5′ for one cycle. The PCR products were analyzed by 1.2% agarose gel and the density of bands was analyzed by Alpha Innotech software.
Mitochondria RNA were extracted from the mitochondria preparations using TRI Reagent. To eliminate the possible contamination of mitochondrial DNA, the isolated mitochondrial RNA was treated with RNAase-free DNAase I at 37°C for 30 minutes. The RNA was precipitated by 60% volume of isopropyl alcohol, washed by 75% ethanol, and dissolved in 20 µl water (treated by DEPC). The RNA was reverse transcribed with random primers using SuperScript II reverse transcriptase according to manufacturer’s instruction (Invitrogen). The 1st strand cDNA was used for PCR under the following conditions: 94°C × 3′ for one cycle; 94°C × 30″, 60–65°C × 1′, 72°C × 1′ for 30 cycles; 72°C × 5′ for one cycle of extension using the primer sets listed in Table 1. The PCR products were analyzed using agarose gel. The primers used to assess the expression of mitochondrial genes at RNA levels were listed in Table 1.
The activities of Complex IV and V in mitochondrial preparations were measured using kits from Mitosciences (MS443 and MS543), according to manufacture’s instructions.
Measurement of cellular respiration was performed in a high performance Oxygraph (OROBOROS®, www.oroboros.at). The supplier’s DatLab software was used for data acquisition and analysis. pUSE and H-RasQ61L cells were suspended at a density of 1 million cells per ml in 1 ml stirred chambers in CO2 equilibrated DMEM medium with 10% bovine serum. Oxygen consumption was measured at 37°C polaragraphically over a period of 5 min. To differentiate oxygen consumption due to mitochondrial respiration or to non-mitochondrial sources, the following inhibitors of mitochondria respiration were added: Rotenone (Inhibitor of complex I, final concentration, 1 µM), and/or antimycin A (Inhibitor of complex III, 1~10 µM).
Cells (pUSEandH-RasQ61L) were harvested by Trypsin-EDTA, counted, and adjusted to 1 × 105 cells/ml in DMEM with 10% bovine serum. Cells (1 × 105) were added to each well in an opaque 96-well plate and then treated with oligomycin (final concentration 1 µg/ml), rotenone (1 µM), antimycin A (1 µM), or iodoacetic acid (IAA, 1 µM, or otherwise indicated) at different time intervals and incubate at 37°C. At the end of incubation, 100 µl CellTiter-Glo® reagent was added to each well. After mixing and incubation for 10 min at room temperature, the luminescence in each well was measured using Lumionoskan Ascent plate reader (Thermo Electro Corp., Waltham, MA).
Cells were cultured in glucose-free DMEM or glucose-containing media in the absence or presence of IAA, rotenone, antimycin A or oligomycin. At different time intervals, cells were harvested with trypsin-EDTA, combined with non-adherent cells in the culture supernatants, and the viability determined using trypan blue exclusion assay with Beckman Coulter ViaCell Analyzer. For each sample, at least 50 images were collected and analyzed.
Student’s t-test (two-tails) was used to analyze the difference between two groups. For comparisons among three or more groups, ANOVA was used. The statistical tests were performed using GraphPad Prism 4.00 for Windows (San Diego, CA). A p value less than 0.05 was considered statistically significant.
This work was partially supported by National Institute of Health Grant R01CA131445 (D.N.), R01AG013435 (G.B.), R01CA121904 (K.K.S.), and start-up funds from Southern Illinois University School of Medicine and SimmonsCooper Cancer Institute (D.N.).