Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 16.
Published in final edited form as: NMR Biomed. 2024 Apr 8;37(9):e5157. doi: 10.1002/nbm.5157

Metabolic fingerprinting by nuclear magnetic resonance of hepatocellular carcinoma cells during p53 reactivation-induced senescence

Philipp Knopf 1,2, Jesus Pacheco-Torres 2, Laimdota Zizmare 1,3, Noriko Mori 2, Flonne Wildes 2, Benyuan Zhou 1, Balaji Krishnamachary 2, Yelena Mironchik 2, Manfred Kneilling 1,3,4, Christoph Trautwein 1,3, Bernd J Pichler 1,3,5, Zaver M Bhujwalla 2,6,7
PMCID: PMC12356598  NIHMSID: NIHMS2094870  PMID: 38589764

Abstract

Cellular senescence is characterized by stable cell cycle arrest. Senescent cells exhibit a senescence-associated secretory phenotype that can promote tumor progression. The aim of our study was to identify specific nuclear magnetic resonance (NMR) spectroscopy-based markers of cancer cell senescence. For metabolic studies, we employed murine liver carcinoma Harvey Rat Sarcoma Virus (H-Ras) cells, in which reactivation of p53 expression induces senescence. Senescent and nonsenescent cell extracts were subjected to high-resolution proton (1H)-NMR spectroscopy-based metabolomics, and dynamic metabolic changes during senescence were analyzed using a magnetic resonance spectroscopy (MRS)-compatible cell perfusion system. Additionally, the ability of intact senescent cells to degrade the extracellular matrix (ECM) was quantified in the cell perfusion system. Analysis of senescent H-Ras cell extracts revealed elevated sn-glycero-3-phosphocholine, myoinositol, taurine, and creatine levels, with decreases in glycine, o-phosphocholine, threonine, and valine. These metabolic findings were accompanied by a greater degradation index of the ECM in senescent H-Ras cells than in control H-Ras cells. MRS studies with the cell perfusion system revealed elevated creatine levels in senescent cells on Day 4, confirming the 1H-NMR results. These senescence-associated changes in metabolism and ECM degradation strongly impact growth and redox metabolism and reveal potential MRS signals for detecting senescent cancer cells in vivo.

Keywords: cellular homeostasis, metabolomics, MR spectroscopy, p53, senescence

1 |. INTRODUCTION

Cellular senescence, commonly defined as stable and permanent cell cycle arrest, was first described by Leonard Hayflick and Paul Moorhead in 1961.1,2 Senescent cells can proliferate in response to growth signals and proliferation-stimulating factors.3 Cellular senescence in tumors can either inhibit or promote tumor progression.4 Cytotoxic drugs such as bleomycin, camptothecin, cisplatin, doxorubicin, or etoposide, as well as radiation, induce senescence via DNA damage. This type of senescence is also termed therapy-induced senescence and inhibits tumor growth.4 By contrast, the senescent cell secretome, which includes small extracellular vesicles, promotes cancer cell proliferation and therefore has a protumorigenic function.5,6 For combining cytotoxic drugs with senolytic agents such as navitoclax7 in a timely coordinated manner, markers identifying senescent cells within tumors noninvasively in vivo need to be identified. Metabolites that can be detected noninvasively in vivo by magnetic resonance spectroscopy (MRS) represent promising biomarkers.

Multiple senescence types have been described, namely, replicative,8 stress-induced,9 and oncogene-induced.10 Here, a genetically engineered murine liver carcinoma Harvey Rat Sarcoma Virus (H-Ras) cell line was used to study cancer senescence. In this model, the tumor suppressor gene p53 plays a critical role, with senescence being induced upon p53 reactivation.11,12 In addition to lacking proliferation, senescent cells are characterized by a senescence-associated secretory phenotype (SASP), which can promote cancer cell proliferation. The SASP includes the proinflammatory cytokines interleukin (IL)-6 and IL-8,13 which synergistically promote cell migration.14 The secretion of these soluble signaling factors and proteases shapes the local tumor microenvironment and interactions with immune cells and promotes cancer progression.15 Therefore, senolytic drugs, including navitoclax, a Bcl-2 inhibitor that targets senescent but not nonsenescent cells, are used experimentally and have been studied in clinical trials.7,16 The identification of senescent tumors is crucial for administering senolytic drugs that target senescent cells in cancer patients. Furthermore, the effect of senolytic drugs needs to be monitored longitudinally and noninvasively to avoid overdosing.

Compared with normal cells, senescent cells exhibit an altered metabolism. Studies across different types of senescence, including replicative, DNA damage-induced, and oncogene-induced senescence in human lung fibroblasts, revealed a greater ratio of sn-glycero-3-phosphocholine (GPC) to O-phosphocholine (PC).17,18 These studies imply that the two metabolites GPC and PC could be used as markers for senescence across different senescence types. Although the metabolism of senescent cells has been studied in fibroblasts,17–21 the metabolism of senescent cancer cells is largely unexplored.

The current study focused on investigating how in vitro-based metabolite alterations in cell extracts could be translated to in vivo MRS markers to differentiate proliferating from senescent cancer cells using a cell perfusion system. The in vitro metabolism of senescent cells was studied using high-resolution (600 MHz) proton (1H)-NMR spectroscopy-based metabolomics of cell extracts. Intact control and senescent H-Ras cells were grown in a three-dimensional conformation on microcarrier beads and studied for longitudinal metabolic changes and their ability to degrade the extracellular matrix (ECM) using an MR-compatible cell perfusion system.22,23 Our experiments revealed increased levels of GPC, myoinositol, taurine, and creatine (Cr), and decreased levels of glycine, PC, threonine, and valine with p53 reactivation-induced senescence in H-Ras cancer cells. Furthermore, senescent H-Ras exhibited a greater degree of ECM degradation.

2 |. MATERIALS AND METHODS

2.1 |. Cell culture and senescence induction

For analysis of metabolism during tumor senescence, customized murine shp53;H-RasV12 liver carcinoma cells (hereafter named H-Ras cells) kindly provided by the laboratory of Lars Zender were used. The cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Thermo Fisher Scientific, Waltham, MA, USA) containing 10% fetal bovine serum (Sigma–Aldrich, St. Louis, MO, USA), 100 U/mL penicillin and streptomycin (Corning, Tewksbury, MA, USA), 1% nonessential amino acids (Sigma–Aldrich), 1 mM sodium pyruvate, and 5 μg/mL doxycycline hyclate (BD, Franklin Lakes, NJ, USA). The medium was changed every second day and was supplemented with 5 μg/mL doxycycline hyclate. Doxycycline hyclate withdrawal induced senescence via p53 reactivation within 3 days. Therefore, the media were removed from H-Ras cells grown in tissue culture (TC) dishes or on microcarrier beads, and the cells were thoroughly washed with phosphate-buffered saline (PBS) and subsequently maintained in culture media without doxycycline hyclate (Figure S1A). To mimic the greater cell density and therefore greater proximity of H-Ras cells growing on microcarrier beads, we seeded cells at low (control = 0.081*106 cells, senescence = 0.162*106 cells, 100-mm TC-dish) and high (control = 0.162*106 cells, senescence = 0.324*106 cells, 100-mm TC-dish) initial seeding cell density and confirmed senescence induction.

2.2 |. Fluorescence imaging

To confirm short hairpin Ribonucleic Acid (shRNA) downregulation upon doxycycline hyclate withdrawal, fluorescence microscopy of green fluorescent protein (GFP) was performed. Murine shp53;H-RasV12 liver carcinoma cells were plated either in a monolayer or on microcarrier beads (2 × 106 H-Ras cells added to 0.5 mL packed microcarrier bead volume) in a 100-mm plate. Before senescence was induced, bright field and fluorescence fields of view were imaged at 10X using a Nikon inverted microscope equipped with a Nikon Coolpix digital camera (Nikon Instruments, Melville, NY, USA). The exposure times used for bright field (70 ms) and fluorescence imaging (monolayer: 20 s, microcarrier beads: 4 s) on Day 3 and Day 5 after doxycycline hyclate withdrawal were kept constant for capturing images at later time points.

2.3 |. Immunoblot analysis

Reactivation of p53 upon doxycycline hyclate withdrawal in H-Ras cells was confirmed by immunoblotting. Control and senescent cells were trypsinized and washed with Dulbecco’s phosphate-buffered saline (DPBS), and cell lysis was performed with radioimmunoprecipitation assay (RIPA) lysis buffer supplemented with a protease inhibitor (PI), phenylmethylsulfonylfluoride (PMSF), dithiothreitol (DTT), sodium fluoride (NaF), and sodium orthovanadate (Na3VO4). Protein extracts were prepared after short centrifugation, and the protein concentration was determined using a Bradford Bio-Rad protein assay kit (Bio-Rad, Hercules, CA, USA). Total cellular protein was resolved via SDS–PAGE, and the expression levels of β-actin (A1978; Sigma–Aldrich), Ki67 (NB500–170; Novus Biologicals, Littleton, CO, USA), and p53 (2524; Cell Signaling Technology, Danvers, MA, USA) were determined by immunoblotting after blocking with 5% nonfat milk and visualization with horseradish peroxidase-conjugated secondary antibodies using the SuperSignal West Pico Chemiluminescent Substrate Kit (Thermo Scientific, Rockford, IL, USA).

2.4 |. High-resolution 1H-NMR spectroscopy-based metabolomics

For analysis of the metabolic differences between control and senescent H-Ras cells, high-resolution 1H-NMR spectroscopy-based metabolomics of cell extracts was performed. Approximately 0.5*106 cells (control) or 0.25*106 cells (senescent) were seeded in a T175 flask (Greiner Bio-One, Frickenhausen, Germany) in the presence of 5 μg/mL doxycycline hyclate (BD). After 24 h, the cells were washed with PBS to remove residual doxycycline hyclate for senescence induction. For the control cells, the medium was replaced with 5 μg/mL doxycycline hyclate-supplemented medium 2 days after seeding (Figure S1A). Cells were harvested 4 days after seeding, which is equivalent to 3 days post-senescence induction, with a cell scraper in ice-cold PBS, then centrifuged (Figure S1A). Polar metabolites were extracted from the cells using a dual-phase extraction method, as previously described.24 Briefly, 400 μL of ice-cold methanol (Thermo Fisher Scientific) was added, after which the cells were vigorously vortexed. The cells in the methanol of five T175 cell culture flasks were combined and stored at −80°C until further processing. Following the addition of 300 μL of methyl tert-butyl ether (MTBE; Sigma–Aldrich), the samples were sonicated using an UltraSonic Heatsystem W-385, and 300 μL of water (Lonza, Basel, Switzerland) was added. All procedures were performed on ice, and dual-phase separation was performed by centrifugation (8000 ×g at 4°C for 5 min). The aqueous phase containing water-soluble metabolites was collected and evaporated to dryness (SP Scientific, Warminster, PA, USA). Then the aqueous phase extracts were resuspended in deuterated phosphate buffer (0.2 M KH2PO4, 0.2 mM NaN3, and 100 μM 3-(trimethylsilyl) propionic 2,2,3,3-d4 acid sodium salt [TSP]) in deuterated water (D2O); the pH was adjusted to 7.4, the mixture was centrifuged, and the supernatant was used for NMR spectroscopy analysis. TSP dissolved in D2O was used as an internal standard. Given that initial experiments showed high variation within individual replicates, we performed a total of six independent experiments. Five experiments (n = 5 for control and senescence) were analyzed with 600 MHz, and one experiment (n = 4 for control and senescence) was analyzed with 750 MHz, providing us with a total of 29 observations for both the control and senescence groups.

High-resolution 1H-NMR spectra were recorded on Bruker Biospin Avance III 600 MHz (TXI probe) and 750 MHz (BBI probe) NMR spectrometers (Bruker Biospin, Billerica, MA, USA). For the quantitative analysis of metabolites, processed spectra were imported into the ChenomX NMR suite (version 8.2) and referenced to the TSP signal at 0 ppm. A total of 20 metabolites were quantified in all the samples.

2.5 |. MR data acquisition with an MR-compatible cell perfusion system

Longitudinal metabolic changes in control and senescent H-Ras cells, as well as degradation of the ECM, were monitored using an MR-compatible cell perfusion system. For MR control experiments, 2 × 106 H-Ras cells were seeded on 0.5 mL of packed Plastic Plus (Solohill Engineering, Inc., Ann Arbor, MI, USA) microcarrier beads and cultured in the presence of 5 μg/mL doxycycline hyclate (BD) in noncell culture Petri dishes (Labtec, Nunc, Denmark) for 4 days. For senescence experiments, 2 × 106 H-Ras cells were cultured on 0.5 mL of packed microcarrier beads starting on Day 1. Control cells were loaded into the MR-compatible cell perfusion system on Day 3 and were maintained in medium supplemented with 5 μg/mL doxycycline hyclate. Senescence was induced on Day 0, and cells were maintained in medium in the absence of doxycycline hyclate (Figure S1B). Microcarrier beads were loaded into the cell perfusion system for control and senescence experiments as previously described,22,23 and as illustrated in Figure S1C. A detailed overview of the perfusion system setup was previously described.25 Control or senescent H-Ras cells grown on microcarrier beads were loaded above and below a custom-made chamber consisting of a porous filter material (polyethylene; Small Parts, Inc., Miami Lakes, FL, USA) and a polycarbonate membrane (Millipore, Bedford, MA, USA) conjoined to a Delrin ring (McMater-Carr Supply Company, Chicago, IL, USA) filled with Matrigel (Sigma–Aldrich) at a concentration of 8.8 mg/mL to image and quantify degradation of the ECM by the H-Ras cell-microcarrier layer above. Two layers of perfluorocarbon alginate beads were placed between the cell-microcarrier layer above and underneath the Matrigel chamber to monitor and subsequently maintain oxygen tension at [O2] 20% or higher using 19F MR relaxometry over the 2-day course of the experiment (Figure S1C). Medium pH in the cell perfusion system was between 7.4 and 7.7 (Figure S2) and did not differ between the control and senescence experiments. Furthermore, the temperature in the cell perfusion system was maintained at 37°C for all MR experiments.

For MR data acquisition, a 9.4-T MR spectrometer (Bruker, Billerica, MA, USA) was used, and T1-weighted imaging and 1H and 31P spectroscopy were performed every 12 h over the course of 2 days. The acquisition, processing, and analysis of MR imaging and spectroscopy data have been described previously.25 ECM Matrigel degradation by control and senescent H-Ras cancer cells grown on microcarrier beads was determined based on 1H-MR images. Quantification of Matrigel degradation was determined by drawing regions of interest (ROIs) around the Matrigel in a T1-weighted image using NIH ImageJ software. The degradation indices were calculated based on the following formula: Degradation Index = −([ROIt = xh − ROIt = 0h]/ROIt = 0h)*10.

Lactate and lipid content were quantitatively determined using diffusion-weighted (DW) one-dimensional (1D) 1H-MR spectra acquired via lactate editing. Cell proliferation was determined by DW 1D 1H-MR spectra without water suppression, assuming a proportional relationship between the intracellular water content and the cell number. Intracellular levels of several metabolites were determined every 0.5 days.

2.6 |. Statistical analysis

The full metabolite raw concentration matrix of 29 senescent and 29 control samples (Figure S3) was normalized by the probabilistic quotient normalization (PQN) method to account for dilution effects and transformed using the square root method to account for heteroskedasticity. The statistical significance of the differences was evaluated using a combination of fold changes and p values (volcano plot analysis), the Mann–Whitney test, Fisher’s least significant difference (LSD) test, or an unpaired t-test. p values less than 0.05 were considered statistically significant. Volcano plots, heatmaps, scores plots, and variance importance in projection (VIP) plots were generated using MetaboAnalyst 5.0,26 and dot/box plots were produced with Prism Graph Pad version 9.3.1. The heatmap shows autoscaled (−0.6; 0.6) concentrations of eight statistically significant metabolites correlated with the Ward clustering method using the Euclidean distance measure. Pattern hunter analysis was based on the Pearson correlation distance measure.

3 |. RESULTS

3.1 |. Reactivation of p53 in H-Ras cells reduces cell proliferation, suggesting senescence induction

In the current study, we used a genetically engineered murine shp53;H-RasV12 liver carcinoma cell line (hereafter named H-Ras cells) as a model for cancer senescence. In the presence of doxycycline hyclate, p53 expression was repressed by shRNA, and the reporter GFP was expressed. When doxycycline hyclate was withdrawn, GFP and shRNA were downregulated, reactivating p53 expression, which induced senescence.

To confirm senescence induction under the different conditions and at different time points, we measured shRNA expression by fluorescence microscopy of GFP and p53 as well as Ki67 levels by Western Blot (WB) analysis. To mimic the proximity of cells in the cell perfusion system, we grew cells in monolayers at low and high densities, and the cells exhibited reduced GFP expression 3 days after doxycycline hyclate withdrawal (Figure 1A). When grown on microcarrier beads, the cells showed a heterogeneous reduction in GFP expression after 3 days of doxycycline withdrawal. This reduction in the GFP signal for the senescent H-Ras cells was even more pronounced after culturing for 2 additional days (equal to Day 5) in the absence of doxycycline hyclate in the cell perfusion system (Figure 1B). In good agreement with these results, p53 reactivation and proliferation inhibition, as measured by total Ki67 protein levels upon doxycycline hyclate withdrawal, was confirmed by WB analysis (Figure 1C,D). Cells grown in a monolayer showed a marked increase in p53 and a significant decrease in Ki67 protein expression upon doxycycline hyclate withdrawal (Figure 1C), indicating that p53 reactivation and proliferative inhibition occurred independently of cell density. Surprisingly, H-Ras cells grown on microcarrier beads had higher Ki67 levels 3 days after doxycycline hyclate withdrawal in the control group than in the senescence group, but had similar Ki67 levels on Day 5 (Figure 1D). This finding indicates heterogeneous senescence induction in cells grown on microcarrier beads in a Petri dish. Senescent H-Ras cells maintained in the cell perfusion system until Day 5 exhibited p53 reactivation and lower total Ki67 protein levels upon doxycycline withdrawal than control cells did (Figure 1D). These results suggest that cells cultured on microcarrier beads in a Petri dish exhibited delayed senescence induction, most likely because of insufficient washout of doxycycline hyclate. By contrast, the cell perfusion system corrected this issue through the constant flow of media through microcarrier beads that carry H-Ras cells. In summary, shRNA downregulation, p53 reactivation, and proliferative inhibition upon senescence induction were confirmed in H-Ras cells grown in monolayers, but these effects were delayed and heterogeneous in cells grown on microcarrier beads.

FIGURE 1.

FIGURE 1

Open in a new tab

Senescence is induced in H-Ras cells upon doxycycline hyclate withdrawal. (A and B) Bright field (70 ms) and GFP fluorescence images (20 s for cell monolayer, 4 s for cells coated on beads) of H-Ras cells at Day 3 and Day 5 postdoxycycline hyclate withdrawal, where GFP indicates the expression of a shRNA against p53 (n = 1 independent experiment; scale bar: 100 μm). (C) Western blot analysis of p53, the proliferation marker Ki67 and β-actin of control (c, red) and senescent (s, green) H-Ras cells grown as cell monolayer at low (control = 0.081*106 cells, senescence = 0.162*106 cells, 100-mm TC-dish) and high (control = 0.162*106 cells, senescence = 0.324*106 cells, 100-mm TC-dish) initial seeding cell density. (D) Western blot analysis of p53, the proliferation marker Ki67 and β-actin of control (c, red) and senescent (s, green) H-Ras cells grown and harvested on beads, according to the scheme illustrated in Figure S1, and cells that have been in the cell perfusion system (CP) for 48 h. Note that “3d” and “5d” indicate cells harvested at Day 3 and Day 5 postdoxycycline hyclate withdrawal, respectively. Different exposure times have been applied (n = 1 independent experiment). GFP, green fluorescent protein; TC, tissue culture.

3.2 |. Senescence induction results in metabolic changes quantified by 1H-NMR spectroscopy

Senescent cells differ from proliferating tumor cells in various ways, with specific metabolic characteristics being one factor for discriminating between two cellular states. High-resolution 1H-NMR spectroscopy is widely applied in metabolomics to identify and quantify metabolic alterations in preclinical or clinical studies.24,27–30

High-resolution 1H-NMR spectroscopy of control and senescent H-Ras cells, a model for p53 reactivation-induced senescence, revealed several altered hydrophilic metabolites (Figure 2, Figure S4, Figure S5). Senescent H-Ras cells presented elevated GPC and reduced PC levels compared with those of control cells, suggesting altered phospholipid metabolism. A switch from PC to GPC during senescence was previously described by Gey and Seeger.17 Furthermore, senescent H-Ras cells exhibited an altered cellular homeostasis profile, indicated by decreased glycine and threonine levels (needed for glutathione synthesis via one-carbon metabolism and the transsulfuration pathway) and increased taurine levels (Figure 2). In addition to phospholipid metabolism and cellular homeostasis, senescent cells exhibited alterations in Cr levels, which further indicated changes in energy metabolism. More interestingly, senescent H-Ras cells characterized by reactivation of p53 expression exhibited higher myoinositol levels than did proliferating control H-Ras cells, in which p53 expression was suppressed (Figure 2). Myoinositol is also involved in the cellular growth metabolism as a basic substrate for the generation of phosphatidylinositol, a major component of cell membranes. The VIP scores of orthogonal partial least squares discriminant analysis (o-PLSDA) and pattern hunter for glycine identified further connections between individual metabolites (Figure 3). A strong correlation between glycine and lactate, glutathione and branched-chain amino acids (isoleucine and valine) indicated a reduced Warburg effect phenotype under senescence and increased oxidative stress by consumption of glutathione.

FIGURE 2.

FIGURE 2

Open in a new tab

Senescent H-Ras cells exhibit distinct metabolomic phenotype. (A) A total of eight metabolites were significantly altered, as illustrated by the volcano plot analysis (raw p < 0.05, fold change [FC] > 1.2). sn-Glycero-3-phosphocholine, myo-inositol, taurine, and creatine (blue) concentrations were increased, while glycine, O-phosphocholine, threonine, and valine (red) concentrations were decreased in the senescence group compared with control. (B) Individual metabolite dot/box plots of control (red box, black dots) and senescence (blue box, black squares) groups show a high dispersion in individual replicated concentrations over all experiments (n = 29 pooled from six independent experiments); p values: **** < 0.0001; *** < 0.001; * < 0.05, based on an unpaired parametric t-test.

FIGURE 3.

FIGURE 3

Open in a new tab

Senescent H-Ras cell metabolite concentration changes correlate with alterations in redox and growth metabolism. (A) Variance importance in projection (VIP) scores from multivariate orthogonal partial least squares discriminant analysis (o-PLSDA) indicates glycine, snglycero-3-phosphocholine (GPC), O-phosphocholine, and myo-inositol as the most important metabolites leading to senescence and control group separation. (B) Pattern hunter investigation for glycine, the metabolite with the most significant concentration decrease in senescence group compared with control in volcano plot analysis, shows a positive correlation with lactate, valine, glutathione, and isoleucine, but a negative correlation with GPC, creatine phosphate, and taurine, based on Pearson’s r correlation distance measure. ADP, adenosine diphosphate; AMP, adenosine monophosphate; ATP, adenosine triphosphate.

Taken together, these findings indicate that, compared with proliferating H-Ras cancer cell extracts, growth, energy, and redox metabolism are altered in senescent cells.

3.3 |. Senescent H-Ras cells exhibit greater degradation of the ECM and metabolic alterations in an MR-compatible cell perfusion system

An MR-compatible cell perfusion system was used to investigate the interplay between metabolic changes in senescent H-Ras cells and ECM gel degradation. Briefly, H-Ras cells were coated on microcarrier beads, and senescence was induced 3 days prior to loading them into the MR-compatible cell perfusion system to account for the delay in senescence induction in H-Ras cells coated on beads. A chamber with ECM was placed in the cell perfusion system with layers of cells on top as well as below the chamber (Figure S1) to evaluate cell proliferation from the intracellular water profile. For sufficient media buffering and oxygen supply in the cell perfusion system, the medium pH was measured by 31P MRS, and oxygen tension was monitored daily by 19F MR relaxometry of perfluorcarbon beads. Representative 1H MR images and degradation indices are shown in Figure 4A. Senescent H-Ras cells exhibited elevated ECM degradation, suggesting that senescent H-Ras cancer cells might be more invasive in vivo.

FIGURE 4.

FIGURE 4

Open in a new tab

Senescent H-Ras cells exhibit an enhanced degradation phenotype. Dynamic quantification of ECM degradation and invasion of control and senescent H-Ras cells. (A) Representative T1-weighted 1H MR images of the ECM chamber. (B) Calculated degradation index indicating increased degradation of the ECM by senescent compared with control cells at 4.5 days (# p ≤ 0.143, unpaired t-test). (C) Degradation index of the ECM over time (outlier excluded; n = 3–5; values represent mean ± SEM, Wilcoxon test, * p < 0.05 vs. control cells). ECM, extracellular matrix.

In addition to ECM degradation, metabolic alterations in senescent cells in the MR-compatible cell perfusion system were monitored via longitudinal MRS between Day 3 and Day 5. The increase in the ECM degradative phenotype in senescent H-Ras cancer cells compared with control H-Ras cancer cells was accompanied by multiple metabolic changes (Figures 5, S6, and S7) that were only partly in line with the high-resolution 1H-NMR spectroscopy results outlined previously. In the MR-compatible cell perfusion system, GPC levels decreased with time and were indistinguishable between the two groups (Figure 5). By contrast, PC increased between Day 3 and Day 4 in the senescent group, as well as in the control group (Figure 5). Nevertheless, there was no significant difference in GPC or PC levels between the control group and the senescence group on Day 5 in the cell perfusion system, in contrast to the 1H-NMR spectroscopy results. This difference might be attributed to the delay and heterogeneity of senescence induction when cells are grown on microcarrier beads, as well as the investigation of relative changes in metabolites over time in the cell perfusion system. Furthermore, relative choline and triglyceride levels did not change over time and were comparable between senescent and control cells (Figure 5). With the MR-compatible cell perfusion system, a relative increase in the energy metabolite creatine phosphate (PCr) was detected over time in senescent H-Ras cells, which was significantly greater than that in control cells (Figure 6). An increase in PCr was accompanied by an increase in the relative Cr level between Day 3 and Day 5 in both groups (Figure 6). In line with the 1H-NMR spectroscopy data obtained from cells grown in a monolayer conformation, senescent cells grown on microcarrier beads in the MR-compatible cell perfusion system exhibited a significantly greater increase in relative Cr levels than control cells on Day 4 (Figure 6), implicating changes in energy metabolism.

FIGURE 5.

FIGURE 5

Open in a new tab

Senescent H-Ras cells grown in a 3D-like conformation exhibit temporal alterations of metabolites. Quantification of GPC, PC, choline, and triglycerides from 1H and 31P MR spectra of control and senescent H-Ras cells grown on microcarrier beads and maintained in a MR-compatible cell perfusion system between Day 3 and Day 5 postsenescence induction. Spectra were quantified, normalized to the (1) water phase and to the (2) initial time point. n = 3–5 biological replicates, values represent mean ± SEM, statistics: Tukey’s multiple comparison test and unpaired t-test; p values: *** < 0.0001; * < 0.05. GPC, sn-glycero-3-phosphocholine; PC, O-phosphocholine.

FIGURE 6.

FIGURE 6

Open in a new tab

Senescent H-Ras cells grown in a 3D-like conformation exhibit temporal alterations of energy metabolites. Quantification of PCr and Cr from 1H and 31P MR spectra of control and senescent H-Ras cells grown on microcarrier beads and maintained in a MR-compatible cell perfusion system between Day 3 and Day 5 postsenescence induction. Spectra were quantified, normalized to the (1) water phase and to the (2) first time point (n = 3–5, values represent mean ± SEM, statistics: Tukey’s multiple comparison test and unpaired t-test; p values: *** < 0.0001; * < 0.05). Cr, creatine; PCr, creatine phosphate.

In summary, senescent H-Ras cancer cells exhibited significant metabolic changes accompanied by elevated ECM degradation.

4 |. DISCUSSION

The aim of the current study was to characterize the metabolome of H-Ras cancer cells with p53 reactivation-induced senescence by 1H-NMR spectroscopy-based metabolomics and to identify a potential MRS biomarker for cancer cell senescence. Furthermore, we wanted to characterize the effect of senescence on cancer cell ECM degradation. Our data revealed several metabolic alterations upon senescence induction, which were accompanied by an increase in ECM degradation.

Senescent H-Ras cancer cells exhibited alterations in phospholipid metabolism, including increased GPC levels. It has been reported that malignant transformation is characterized by activated choline metabolism in various cancer entities, including increased PC levels.31–33 A higher PC/GPC ratio is a characteristic of tumor progression.31,34 By contrast, increased GPC levels are a common characteristic of senescence, as reported by Gey and Seeger for immortalized WI-38 cells.17 Increased GPC together with reduced phosphatidylcholine (PtdC) during senescence suggest increased phospholipase A1 and/or A2 (PLA1/2) and lysophospholipase activity.17,35

The analysis of energy metabolites revealed increased Cr levels in the senescence group, as measured by 1H-NMR spectroscopy. These findings were substantiated by the increase in the relative Cr and PCr on Day 4 in intact perfused senescent cells. Cr and PCr play important roles in energy storage and transmission. Therefore, an increase in PCr indicates a decreased energy demand in senescent H-Ras cancer cells. Similar results have been reported for proliferative exhaustion-induced cellular senescence, where increased Cr metabolism was reported.19 James et al. attributed these metabolic changes to altered energy storage and utilization in senescent cells.19 In contrast to our results, another study reported a decrease in the PCr during replicative and oncogene-induced senescence in human diploid fibroblasts.17

We also detected higher myoinositol levels in senescent H-Ras cancer cells than in control H-Ras cancer cells, suggesting changes in cellular growth metabolism. Koguchi et al. revealed, via a microarray screen, that p53 induced the expression of genes related to myoinositol metabolism, including the enzyme inositol 3-phosphate synthase (ISYNA1).36 The expression of this enzyme, which is critical in the synthesis of myoinositol,37–39 is regulated by activated p53.36 Because senescence is induced by p53 reactivation in the H-Ras cancer cell line, this may explain the increase in myoinositol in senescent cells.

Metabolic alterations in growth, redox, and energy metabolism were accompanied by increased ECM degradation, suggesting increased invasion by senescent cancer cells. The greater ECM degradation of senescent cells might be attributed to the SASP. As senescence induction in H-Ras cells grown on microcarrier beads was heterogeneous, it cannot be determined whether senescent cells educate nonsenescent cells to degrade the ECM, or whether senescent H-Ras cells themselves elicit this increase in the ECM degradative phenotype. The invasive characteristics of senescent cells have been described for papillary thyroid carcinoma and are linked to the SASP, namely, by the generation of a C-X-C-motif ligand 12 (CXCL12) chemokine gradient in this model.40 Another study showed that phospholipase D catalyzed the transformation of PtdC to choline to increase human breast cancer cell invasion.41 Similar connections between phosphatidylserine-specific phospholipase A1 expression and colorectal cancer invasion have been reported,42 indicating that a correlation between phospholipid metabolism and invasion might be enforced by senescence.

5 |. CONCLUSION

Taken together, the results of our study indicate changes in growth, redox, and energy metabolism during cancer senescence in a model of p53 reactivation-induced senescence. These metabolic changes were accompanied by elevated ECM degradation, underlining the importance of identifying and targeting senescent cancer cells in vivo.

Supplementary Material

Supplementary Information

Additional supporting information can be found online in the Supporting Information section at the end of this article.

ACKNOWLEDGMENTS

PK was supported by a grant from the German Academic Exchange Service (DAAD PPP USA 2018, Project-ID 57387312). All the studies were supported by NIH R35 CA209960. This work was further supported by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation, Germany’s Excellence Strategy-EXC2180–390900677) and the Werner Siemens Foundation.

Abbreviations:

1D

one-dimensional

1H

proton

Cr

creatine

CXCL12

C-X-C- motif ligand 12

DMEM

Dulbecco’s modified Eagle’s medium

DPBS

Dulbecco’s phosphate buffered saline

DW

diffusion-weighted

ECM

extracellular matrix

GFP

green fluorescent protein

GPC

sn-glycero-3-phosphocholine

H-Ras

Harvey Rat Sarcoma Virus

IL

interleukin

ISYNA1

inositol 3-phosphate synthase

LSD

least significant difference

MRS

magnetic resonance spectroscopy

MTBE

methyl tert-butyl ether

NMR

nuclear magnetic resonance

o-PLSDA

orthogonal partial least squares discriminant analysis

PBS

phosphate-buffered saline

PC

O-phosphocholine

PCr

creatine phosphate

PE

phosphoethanolamine

PLA1/2

cytosolic phospholipase A1/2

PQN

probabilistic quotient normalization

PtdC

phosphatidylcholine

RIPA

radioimmunoprecipitation assay

ROI

region of interest

SASP

senescence-associated secretory phenotype

shRNA

short hairpin Ribonucleic Acid

SM

sphingomyelin

TC

tissue culture

TSP

3-(trimethylsilyl) propionic 2,2,3,3-d4 acid sodium salt

VIP

variance importance in projection

WB

Western Blot

Footnotes

CONFLICT OF INTEREST STATEMENT

PK is now an employee of Boehringer Ingelheim Biopharmaceuticals GmbH. This work was part of his PhD thesis at the University of Tuebingen. CT reports a research grant from Bruker BioSpin GmbH.

REFERENCES

  • 1.Hayflick L, Moorhead PS. The serial cultivation of human diploid cell strains. Exp Cell Res. 1961;25:585–621. doi: 10.1016/0014-4827(61)90192-6 [DOI] [PubMed] [Google Scholar]
  • 2.Collado M, Serrano M. Senescence in tumours: evidence from mice and humans. Nat Rev Cancer. 2010;10:51–57. doi: 10.1038/nrc2772 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Sherwood SW, Rush D, Ellsworth JL, Schimke RT. Defining cellular senescence in IMR-90 cells: a flow cytometric analysis. Proc Natl Acad Sci U S A. 1988;85:9086–9090. doi: 10.1073/pnas.85.23.9086 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Schosserer M, Grillari J, Breitenbach M. The dual role of cellular senescence in developing tumors and their response to cancer therapy. Front Oncol. 2017;7:278. doi: 10.3389/fonc.2017.00278 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Takasugi M, Okada R, Takahashi A, Virya Chen D, Watanabe S, Hara E. Small extracellular vesicles secreted from senescent cells promote cancer cell proliferation through EphA2. Nat Commun. 2017;8:15729. doi: 10.1038/ncomms15728 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Lehmann BD, Paine MS, Brooks AM, et al. Senescence-associated exosome release from human prostate cancer cells. Cancer Res. 2008;68:7864–7871. doi: 10.1158/0008-5472.CAN-07-6538 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Zhu Y, Tchkonia T, Fuhrmann-Stroissnigg H, et al. Identification of a novel senolytic agent, navitoclax, targeting the Bcl-2 family of anti-apoptotic factors. Aging Cell. 2016;15:428–435. doi: 10.1111/acel.12445 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Campisi J, d’Adda di Fagagna F. Cellular senescence: when bad things happen to good cells. Nat Rev Mol Cell Biol. 2007;8:729–740. doi: 10.1038/nrm2233 [DOI] [PubMed] [Google Scholar]
  • 9.Passos JF, Simillion C, Hallinan J, Wipat A, von Zglinicki T. Cellular senescence: unravelling complexity. Age. 2009;31:353–363. doi: 10.1007/s11357-009-9108-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Serrano M, Lin AW, McCurrach ME, Beach D, Lowe SW. Oncogenic ras provokes premature cell senescence associated with accumulation of p53 and p16INK4a. Cell. 1997;88:593–602. doi: 10.1016/S0092-8674(00)81902-9 [DOI] [PubMed] [Google Scholar]
  • 11.Yasaei H, Gilham E, Pickles JC, Roberts TP, O’Donovan M, Newbold RF. Carcinogen-specific mutational and epigenetic alterations in INK4A, INK4B and p53 tumour-suppressor genes drive induced senescence bypass in normal diploid mammalian cells. Oncogene. 2013;32:171–179. doi: 10.1038/onc.2012.45 [DOI] [PubMed] [Google Scholar]
  • 12.Xue W, Zender L, Miething C, et al. Senescence and tumour clearance is triggered by p53 restoration in murine liver carcinomas. Nature. 2007;445: 656–660. doi: 10.1038/nature05529 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Kuilman T, Michaloglou C, Vredeveld LC, et al. Oncogene-induced senescence relayed by an interleukin-dependent inflammatory network. Cell. 2008; 133:1019–1031. doi: 10.1016/j.cell.2008.03.039 [DOI] [PubMed] [Google Scholar]
  • 14.Jayatilaka H, Tyle P, Chen JJ, et al. Synergistic IL-6 and IL-8 paracrine signalling pathway infers a strategy to inhibit tumour cell migration. Nat Commun. 2017;8:15584. doi: 10.1038/ncomms15584 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Faget DV, Ren Q, Stewart SA. Unmasking senescence: context-dependent effects of SASP in cancer. Nat Rev Cancer. 2019;19:439–453. doi: 10.1038/s41568-019-0156-2 [DOI] [PubMed] [Google Scholar]
  • 16.Shoemaker AR, Mitten MJ, Adickes J, et al. Activity of the Bcl-2 family inhibitor ABT-263 in a panel of small cell lung cancer xenograft models. Clin Cancer Res. 2008;14:3268–3277. doi: 10.1158/1078-0432.CCR-07-4622 [DOI] [PubMed] [Google Scholar]
  • 17.Gey C, Seeger K. Metabolic changes during cellular senescence investigated by proton NMR-spectroscopy. Mech Ageing Dev. 2013;134:130–138. doi: 10.1016/j.mad.2013.02.002 [DOI] [PubMed] [Google Scholar]
  • 18.Gey C, Seeger K. Metabolic changes investigated by proton NMR spectroscopy in cells undergoing oncogene-induced senescence. Methods Mol Biol. 2017;1534:155–163. doi: 10.1007/978-1-4939-6670-7_15 [DOI] [PubMed] [Google Scholar]
  • 19.James EL, Lane JA, Michalek RD, Karoly ED, Parkinson EK. Replicatively senescent human fibroblasts reveal a distinct intracellular metabolic profile with alterations in NAD+ and nicotinamide metabolism. Sci Rep. 2016;6:38489. doi: 10.1038/srep38489 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.James EL, Michalek RD, Pitiyage GN, et al. Senescent human fibroblasts show increased glycolysis and redox homeostasis with extracellular metabolomes that overlap with those of irreparable DNA damage, aging, and disease. J Proteome Res. 2015;14:1854–1871. doi: 10.1021/pr501221g [DOI] [PubMed] [Google Scholar]
  • 21.Zwerschke W, Mazurek S, Stockl P, Hutter E, Eigenbrodt E, Jansen-Durr P. Metabolic analysis of senescent human fibroblasts reveals a role for AMP in cellular senescence. Biochem J. 2003;376:403–411. doi: 10.1042/bj20030816 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Penet MF, Shah T, Wildes F, et al. MRI and MRS of intact perfused cancer cell metabolism, invasion, and stromal cell interactions. NMR Biomed. 2019; 32:e4053. doi: 10.1002/nbm.4053 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Pilatus U, Ackerstaff E, Artemov D, Mori N, Gillies RJ, Bhujwalla ZM. Imaging prostate cancer invasion with multi-nuclear magnetic resonance methods: the Metabolic Boyden Chamber. Neoplasia. 2000;2:273–279. doi: 10.1038/sj.neo.7900089 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Bus C, Zizmare L, Feldkaemper M, et al. Human dopaminergic neurons lacking PINK1 exhibit disrupted dopamine metabolism related to vitamin B6 co-factors. iScience. 2020;23:101797. doi: 10.1016/j.isci.2020.101797 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Ackerstaff E, Gimi B, Artemov D, Bhujwalla ZM. Anti-inflammatory agent indomethacin reduces invasion and alters metabolism in a human breast cancer cell line. Neoplasia. 2007;9:222–235. doi: 10.1593/neo.06673 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Chong J, Soufan O, Li C, et al. MetaboAnalyst 4.0: towards more transparent and integrative metabolomics analysis. Nucleic Acids Res. 2018;46: W486–W494. doi: 10.1093/nar/gky310 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Letertre MPM, Giraudeau P, de Tullio P. Nuclear magnetic resonance spectroscopy in clinical metabolomics and personalized medicine: current challenges and perspectives. Front Mol Biosci. 2021;8:698337. doi: 10.3389/fmolb.2021.698337 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Trautwein C, Zizmare L, Mäurer I, et al. Tissue metabolites in diffuse glioma and their modulations by IDH1 mutation, histology, and treatment. JCI Insight. 2022;7. doi: 10.1172/jci.insight.153526 [DOI] [PMC free article] [PubMed]
  • 29.Chen C, Gao J, Wang T-S, et al. NMR-based metabolomic techniques identify the toxicity of emodin in HepG2 cells. Sci Rep. 2018;8:9379. doi: 10.1038/s41598-018-27359-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Singh Y, Trautwein C, Dhariwal A, et al. DJ-1 (Park7) affects the gut microbiome, metabolites and the development of innate lymphoid cells (ILCs). Sci Rep. 2020;10:16131. doi: 10.1038/s41598-020-72903-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Aboagye EO, Bhujwalla ZM. Malignant transformation alters membrane choline phospholipid metabolism of human mammary epithelial cells. Cancer Res. 1999;59:80–84. [PubMed] [Google Scholar]
  • 32.Ackerstaff E, Pflug BR, Nelson JB, Bhujwalla ZM. Detection of increased choline compounds with proton nuclear magnetic resonance spectroscopy subsequent to malignant transformation of human prostatic epithelial cells. Cancer Res. 2001;61:3599–3603. [PubMed] [Google Scholar]
  • 33.Jacobs MA, Barker PB, Bottomley PA, Bhujwalla Z, Bluemke DA. Proton magnetic resonance spectroscopic imaging of human breast cancer: a preliminary study. J Magn Reson Imaging. 2004;19:68–75. doi: 10.1002/jmri.10427 [DOI] [PubMed] [Google Scholar]
  • 34.Iorio E, Mezzanzanica D, Alberti P, et al. Alterations of choline phospholipid metabolism in ovarian tumor progression. Cancer Res. 2005;65:9369–9376. doi: 10.1158/0008-5472.CAN-05-1146 [DOI] [PubMed] [Google Scholar]
  • 35.Kim HJ, Kim KS, Kim SH, et al. Induction of cellular senescence by secretory phospholipase A2 in human dermal fibroblasts through an ROS-mediated p53 pathway. J Gerontol A Biol Sci Med Sci. 2009;64:351–362. doi: 10.1093/gerona/gln055 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Koguchi T, Tanikawa C, Mori J, Kojima Y, Matsuda K. Regulation of myo-inositol biosynthesis by p53-ISYNA1 pathway. Int J Oncol. 2016;48:2415–2424. doi: 10.3892/ijo.2016.3456 [DOI] [PubMed] [Google Scholar]
  • 37.Chauvin TR, Griswold MD. Characterization of the expression and regulation of genes necessary for myo-inositol biosynthesis and transport in the seminiferous epithelium. Biol Reprod. 2004;70:744–751. doi: 10.1095/biolreprod.103.022731 [DOI] [PubMed] [Google Scholar]
  • 38.Guan G, Dai P, Shechter I. cDNA cloning and gene expression analysis of human myo-inositol 1-phosphate synthase. Arch Biochem Biophys. 2003;417: 251–259. doi: 10.1016/S0003-9861(03)00388-6 [DOI] [PubMed] [Google Scholar]
  • 39.Ju S, Shaltiel G, Shamir A, Agam G, Greenberg ML. Human 1-D-myo-inositol-3-phosphate synthase is functional in yeast. J Biol Chem. 2004;279: 21759–21765. doi: 10.1074/jbc.M312078200 [DOI] [PubMed] [Google Scholar]
  • 40.Kim YH, Choi YW, Lee J, Soh EY, Kim JH, Park TJ. Senescent tumor cells lead the collective invasion in thyroid cancer. Nat Commun. 2017;8:15208. doi: 10.1038/ncomms15208 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Henkels KM, Boivin GP, Dudley ES, Berberich SJ, Gomez-Cambronero J. Phospholipase D (PLD) drives cell invasion, tumor growth and metastasis in a human breast cancer xenograph model. Oncogene. 2013;32:5551–5562. doi: 10.1038/onc.2013.207 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Iida Y, Sunami E, Yamashita H, et al. Phosphatidylserine-specific phospholipase A1 (PS-PLA1) expression in colorectal cancer correlates with tumor invasion and hematogenous metastasis. Anticancer Res. 2015;35:1459–1464. [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information
NIHMS2094870-supplement-Supplementary_Information.docx (2.3MB, docx)

RESOURCES