Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Aug 9.
Published in final edited form as: Cancer Res. 2020 Jun 24;80(16):3251–3264. doi: 10.1158/0008-5472.CAN-19-3281

p38γ MAPK is essential for aerobic glycolysis and pancreatic tumorigenesis

Fang Wang 1,2, Xiao-Mei Qi 1, Ryan Wertz 1, Matthew Mortensen 1, Catherine Hagen 3, John Evans 3, Yuri Sheinin 3, Michael James 4, Pengyuan Liu 5, Susan Tsai 4, James Thomas 6, Alexander Mackinnon 3, Michael Dwinell 7, Charles R Myers 1, Ramon Bartrons 8, Liwu Fu 2, Guan Chen 1,9
PMCID: PMC9358694  NIHMSID: NIHMS1607187  PMID: 32580961

Abstract

KRAS is mutated in most pancreatic ductal adenocarcinomas (PDAC) and yet remains undruggable. Here we report that p38gamma MAPK, which promotes PDAC tumorigenesis by linking KRAS signaling and aerobic glycolysis (also called the Warburg effect), is a novel therapeutic target. p38gamma interacted with a glycolytic activator PFKFB3 that was dependent on mutated KRAS. KRAS transformation and overexpression of p38gamma increased expression of PFKFB3 and glucose transporter GLUT2; conversely, silencing mutant KRAS and p38gamma decreased PFKFB3 and GLUT2 expression. p38gamma phosphorylated PFKFB3 at S467, stabilized PFKFB3, and promoted their interaction with GLUT2. Pancreatic knockout (KO) of p38gamma decreased p-PFKFB3/PFKFB3/GLUT2 protein levels, reduced aerobic glycolysis, and inhibited PDAC tumorigenesis in KPC mice. PFKFB3 and GLUT2 depended on p38gamma to stimulate glycolysis and PDAC growth and p38gamma required PFKFB3/S467 to promote these activities. A p38gamma inhibitor cooperated with a PFKFB3 inhibitor to blunt aerobic glycolysis and PDAC growth, which was dependent on p38gamma. Moreover, overexpression of p38gamma, p-PFKFB3, PFKFB3, and GLUT2 in PDAC predicted poor clinical prognosis. These results indicate that p38gamma links KRAS oncogene signaling and aerobic glycolysis to promote pancreatic tumorigenesis through PFKFB3 and GLUT2, and that p38gamma and PFKFB3 may be targeted for therapeutic intervention in PDAC.

Keywords: p38γ MAPK, KRAS, PFKFB3, GLUT2, Aerobic glycolysis, Pancreatic cancer

Introduction

Pancreatic ductal adenocarcinoma (Pdac) is among the deadliest malignancies with more than 90% of the cases containing KRAS mutations (1). KRAS signals through multiple signal transduction pathways including PI3K and ERKs (2,3). However, KRAS is still undruggable and there is an urgent need to identify its key signaling pathways for drug discovery to improve the clinical outcome of Pdac (4,5).

Metabolic reprogramming is a hallmark of cancer (6). Normal cells rely primarily on mitochondrial oxidative phosphorylation to generate ATP, whereas cancer cells generate energy mostly by aerobic glycolysis (also called “the Warburg effect”) (7), even in the presence of sufficient oxygen. Although the Warburg effect has been observed for almost a century, the molecular mechanisms driving this metabolic switch are largely unknown (8). RAS oncogenesis is a process of selective stimulation of aerobic glycolysis through transcriptional activation of multiple genes including glucose transporter 1 (GLUT1) (9) and phosphofructokinase-2/fructose-2,6-bisphosphatase 3 (PFKFB3) (10), these activities have not been amenable to specific and simultaneous blockade (11,12). It is therefore essential to identify druggable links between the KRAS oncogene and the Warburg effect for therapeutic intervention (Fig. 1A).

Figure 1. p38γ interacts with PFKFB3 and mediates KRAS signaling to stimulate PFKFB3 and GLUT2 expression and to increase Pdac growth.

Figure 1.

Open in a new tab

A, RAS mutation is known to stimulate the Warburg effect via transcriptional activation of multiple genes including GLUT1, PFKFB3 and others; however, a druggable link (indicated by “?”) had not been identified. B, the indicated cells were depleted of p38γ by lentiviral infection with shRNAs and effects on soft-agar growth were assessed (mean ± SD, n = 3, * P < 0.05, ** P < 0.01); WB as inserts. C, D, Cells were cultured in the presence of PFD (100 μg/ml) or SB203580 (50 μM) for 10–14 days for soft-agar growth (C) or treated with PFD for 72 h for effects on cell proliferation (D) (mean ± SD, n = 3, * p < 0.05, ** p < 0.01, vs DMSO). E, Flag-p38γ was isolated from HPNE/KRAS cells, precipitates were separated via SDS-PAGE, and gels containing proteins between 45 and 65 KD were subjected to proteomic analysis, with the identified glycolytic proteins shown at right and their associated peptide sequences shown in Supplementary Fig. 1C. Flag-p38γ did not bind these proteins in HPNE/Vector cells (not shown). F, HPNE cells stably transfected with p38γ or mutant KRAS were analyzed for protein expression. G, Total RNAs were prepared for assessing of RNA levels by quantitative RT-PCR and relative changes over HPNE cells were shown (mean ± SD, n = 3, * p < 0.05, ** p < 0.01 vs HPNE). H, KRAS mutated MIA-PaCa-2 cells were depleted of mutant KRAS by siRNAs oligos and analyzed for protein expression. I, p38γ was silenced by lentiviral mediated shRNA and effects on protein expression and/or phosphorylation were analyzed.

p38 mitogen-activated protein kinases (MAPKs) (p38α, β, γ, and δ) are highly conserved inflammatory proteins and play an isoform-specific role in RAS oncogenesis (13–15). p38γ (the gene name: MAPK12) is expressed in several tissues (15–18) and plays a role in glucose transport, KRAS transformation and inflammation/carcinogen-induced colon and liver cancer (15,17–22). This report tests the hypothesis that p38γ may be required for pancreatic tumorigenesis by linking KRAS oncogene signaling and the Warburg effect. Our results show that p38γ is essential for Pdac tumorigenesis by linking KRAS signaling and the Warburg effect via PFKFB3 and GLUT2 and that targeting p38γ alone and together with PFKFB3 may have great therapeutic potential for treating Pdac.

Materials and Methods

Mouse Strains

Conditional LSL-Trp53R172H/+ (23), LSL-KrasG12D/+, and Pdx-1-Cre (24) strains were interbred to obtain LSL-KrasG12D/+; LSL-Trp53R172H/+; Pdx-1-Cre triple mutant animals (KPC) on a mixed 129/SvJae/C57BL/6 background. p38γflox/flox mice (provided by Boehringer Ingelheim Pharmaceuticals, Inc., Columbus, OH, USA) were bred with Pdx-1-Cre mice to generate pancreas-specific p38γ knockout (KO) mice. Experimental LSL-KrasG12D/+; LSL-Trp53R172H; p38γ KO; Pdx-1-Cre mice (KPC/p38γ KO) were generated by crossing p38γflox/flox -Pdx-1-Cre mice to Trp53R172H/+-KrasG12D/+ mice. All animal experiments were conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee-approved protocols of the Medical College of Wisconsin (IACUC number: AUA000042). Male and female mice were used indiscriminately. Mice were examined for pancreatic cancer development as previously described (25). If mice displayed symptoms of impending death such as cachexia, abdominal distension, rapid weight loss, or labored breathing, they were euthanized.

Genotyping

The following primers were used to genotype wild-type (WT), heterozygous mutant, and knockout (KO) mice: Kras WT, F 5’- GTCGACAAG CTCATGCGGG-3’, Kras Mutant, 5’- CCATGGCTTGAGTAAGTCTGC-3’, Kras Common, 5’- CGCAGACTGTAGAGCAGCG-3’; p53R172H WT, F 5’- TTACACATCCAGCCTCTGTGG-3’, p53R172H Mutant, 5’- AGCTAGCCACCATGGCTTGAGTAAGTCTGCA-3’, p53R172H Common, 5’- CTTGGAGACATAGCCACACTG-3’; Pdx-1-Cre, F 5’- GCGGTCTGGCAGTAAAAACTATC-3’, R 5’- GTGAAACAGCATTGCTGTCACTT-3’; p38γ WT, F 5’-TGGGCTGCGAAGGTAGAGGTG-3’, p38γ flox, 5’-GTGTCACGTGCTCAGGGCCTG-3’, p38γ common, 5’-CCAGGAGGTGACCAAAACGGC-3’. Tail DNA was used as the PCR template with the following conditions: for Kras and p53R172H, denaturing (95°C, 30s), annealing (66°C, 1min) and extension (72°C, 1 min), 35 cycles; for Cre, denaturing (95°C, 30s), annealing (52°C, 30s) and extension (72°C, 30s), 35 cycles; for loxP-flanked p38γ allele, denaturing (95°C, 30s), annealing (62°C, 30s) and extension (72°C, 30s), 35 cycles. KC and KC/p38γ KO mice were generated by cross-breeding Pdex-1-Cre, LSL-KrasG12D/+, with p38γfl/fl mice. Three month-old mice were euthanized for PanIN analyses (24).

Treatment with inhibitors

LSL-KrasG12D/+; LSL-Trp53R172H/+; Pdx-1-Cre triple mutant mice (KPC) and the p38γ KO counterparts (KPC/p38γ KO) were treated with 500 mg/kg pirfenidone (PFD) alone daily or with 300 mg/kg in combination with PFK15 by oral gavage. PFD used for treating mice with primary KPC tumors was provided by Genentech, while the drug for cell culture was purchased from Sigma. The mice were treated starting at the age of 4 weeks and continued for a total of 12 weeks. The entire tumor tissues were harvested, weighed and immediately frozen in dry ice and stored at −80°C for subsequent studies or fixed in 10% phosphate buffered formalin for histologic analysis.

Cell lines and Authentication

Human pancreatic cancer cells Capan-1, Panc-1 and MIA PaCa-2 were obtained from the American Type Cell Culture (ATCC, Manassas, VA, USA). All cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin. The hTERT-immortalized human pancreatic ductal progenitor (hTERT-HPNE) cells modified to express E6/E7 alone or in conjunction with oncogenic Kras (referred to as HPNE-Ras cells) were generous gifts from Dr. Richard Klemke (26). The culture medium for HPNE and HPNE-Ras cells was: 75% DMEM, 25% medium M3 base, 5% FBS, 10 ng/ml human recombinant EGF, and 750 ng/ml puromycin. Mouse pancreatic cancer cells (KPC and KPC/p38γ KO) from primary tissues of individual mice were generated as described (27). Briefly, tissue was minced and digested with collagenase V for 15 min at 37°C to obtain single cell suspensions. All experiments reported in this study were conducted in early passage cells of KPC or KPC/p38γ KO lines (< p10) grown in complete medium (IMDM +10% FBS + 1% penicillin/streptomycin).

Cell Proliferation, Migration assays, Matrigel invasion assays, Cell cycle analyses

Cells were plated in 96-well plates at a density of 500 cells/well. The MTT assay was operated at days 1, 2, 3, 4 and 5. For cell migration, 1×106 serum starved cells were plated on six-well plate and grown to confluence. A straight wound track was introduced by scraping the cell monolayer with a p200 pipette tip. Cells were maintained in serum free conditions to minimize proliferation and cultured for 24 h. Images were obtained at the indicated time points using a 10X objective on an Olympus IX73 inverted microscope. For Matrigel invasion assays, cells (5×104) were resuspended in 100 μl of serum free growth medium and plated in triplicate into Transwell precoated with 8 μM Matrigel membrane filter inserts (Corning) of 24-well plates. Filters were precoated with 10 μl of Matrigel (BD) diluted 1:4 in ice-cold serum-free medium and allowed to solidify for 1 h at 37°C before use. The lower chamber contained 500 μL of growth medium with 10% FBS as chemoattractant. After incubation for 24 h, non-invading cells were removed from the upper chamber and the transwell membranes were fixed with 100% methanol and stained with Crystal Violet. Cell number was analyzed using Image J software. To determine cell cycle distribution, cells were seeded and cultured in 6-well plates (5×105 cells/well) for 24 h, and then harvested, washed with PBS and fixed in 70% ethanol at 4°C overnight. Subsequently, cells were incubated with 1% RNase A at 37°C for 30 min and stained with propidium iodide (50 μg/mL) at 4°C for 30 minutes. The DNA content of stained cells was measured using flow cytometry (Beckman Coulter, CytoFLEX) and data were analyzed by ModFit LT V4.1 Software.

Statistics

All results shown are the average of triplicates (at least) and presented as mean ± SD. Differences were analyzed by Student’s t-test or one-way ANOVA test using SPSS statistical software (Version 16.0) unless specified, with * and ** indicating p < 0.05 and p < 0.01, respectively. Survival was measured using the Kaplan-Meier method and statistical significance was determined by log-rank using GraphPad Prism software (Version 5). Linear relationships were tested by Pearson’s correlation to assess the relationship between p38γ and p-PFKFB3 or between p-PFKFB3 and GLUT2 levels in Pdac specimens.

Detailed Materials and Methods are given in Supplemental Information. Research reagents will be provided upon a written request.

Results

p38γ mediates KRAS signals to stimulate PFKFB3 and/or GLUT2 expression and/or phosphorylation and is required for Pdac cell growth

Previous studies have demonstrated a required role for p38γ in colon and liver cancer development (18,22). To determine if p38γ is overexpressed in human Pdac tumors, we generated two tissue microarrays containing 49 pathologically verified and clinically annotated Pdac samples together with 16 matched normal controls. IHC analysis shows that p38γ is significantly elevated in human Pdac tissues over normal controls (arbitrary units: 1.62 ± 0.096 in tumor vs 0.86 ± 0.08 in normal, p < 0.001, Supplementary Fig.S1A), indicating that p38γ may play a promoting role in Pdac development and progression. To test if p38γ is required for Pdac growth, p38γ was silenced in two Pdac cell lines and effects on colony formation were assessed. Fig. 1B shows that p38γ depletion inhibits Pdac growth, indicating p38γ’ proliferative activity in Pdac. The p38γ pharmacological inhibitor pirfenidone (PFD) also inhibits Pdac growth, whereas the p38α/β inhibitor SB203580 has no effect (Fig.1C). PFD dose-dependently inhibits the growth in Pdac or KRAS transformed cells, but only suppresses proliferation of normal HPNE/Vect cells at the higher concentration (400 μg/ml) (Fig.1D; Supplementary Fig. S1B). These results together demonstrate that p38γ is overexpressed in human Pdac and is a druggable target to inhibit Pdac growth.

p38γ increases glucose uptake, stimulates glucose transporter 1 (GLUT1) expression in muscle cells (19), and promotes metabolic adaption (17) and glucose tolerance (28). To search for p38γ-interacting proteins that may be involved in aerobic glycolysis in KRAS-mutated Pdac cells, Flag-p38γ was expressed in KRASG12D transformed human pancreas nesting-expressing cells (HPNE/Kras) (26) and Flag precipitates were subjected to mass spectrometry analyses. These analyses (Fig. 1E; Supplementary Fig. S1C) show that p38γ co-precipitates PFKFB3 and several other glycolytic proteins. Pathway analyses further show that transformation with mutant KRASG12D increases the protein amounts of p38γ and PFKFB3 as well as GLUT2, a β cell-specific glucose transporter (29), whereas has no major effects on GLUT1 (Fig. 1F). Because GLUT1 is frequently involved in Kras oncogenesis (8), we further measured RNA levels of GLUT1 vs GLUT2 by qRT-PCR. Fig. 1G shows that Kras transformation in HPNE cells decreases GLUT1 and increases GLUT2 RNA levels as observed in WB (Fig. 1F). Kras-induced GLUT1 downregulation is similar to GLUT1 upregulation by silencing mutant Kras in MIAPaPa-2 cells (30), whereas p38γ RNA induction by Kras is consistent with our previous observation in Kras-transformed IEC-6 cells (15). Similar increases in PFKFB3 and GLUT2 protein were also seen in p38γ overexpressed HPNE cells (Fig. 1F), indicating that KRAS may signal to PFKFB3 and GLUT2 via p38γ. Moreover, silencing either endogenous KRASG12C (31) or p38γ decreases protein levels of p-PFKFB3, PFKFB3 and GLUT2 (Fig. 1H/1I). PFKFB3 silencing, however, has no major effects on p38γ or GLUT2 protein levels (Supplementary Fig. S1D). These results together indicate that p38γ may mediate KRAS signaling to increase glycolytic PFKFB3 and GLUT2 protein expression, which may play a role in linking KRAS oncogenesis and aerobic glycolysis.

p38γ phosphorylates PFKFB3 at S467, stabilizes PFKFB3, and promotes interaction with GLUT2

PFKFB3 and its phosphorylated forms are upregulated in human cancers, but mechanisms for its activation are unknown (32–34). Because p38γ is phosphorylated and overexpressed in HPNE/Kras cells and silencing of Kras and p38γ decreases PFKFB3 expression and phosphorylation (Fig. 1F–1H), we tested if p38γ phosphorylates PFKFB3 by mass spectrometry screening (35). Fig. 2A/B show that His-p38γ phosphorylates Flag-isolated PFKFB3 protein at S467 in vitro, which was further confirmed by using a specific phospho-PFKFB3/S467 antibody (p-PFKFB3). Analyses of Flag precipitates further show that PFKFB3/S467A has a decreased activity in binding p38γ and GLUT2 as compared with wild-type (WT) PFKFB3 (Fig. 2B). The significant difference between PFKFB3 and its S467A mutant in interactions with p38γ and GLUT2 cannot be explained by a slight increase in total (in input, middle) or precipitated PFKFK3 proteins (Flag IP, right), and suggests that phosphorylated PFKFB3 may have a higher affinity in binding p38γ and GLUT2. p38γ also forms a complex and/or colocalizes with p-PFKFB3, PFKFB3, and GLUT2 in Pdac, KPC or KRAS-transformed cells but not in normal HPNE cells as compared with negative control KPC/p38γ KO cells (Supplementary Fig. S1E–1G). These results together indicate that p38γ phosphorylates PFKFB3/S467, which promotes PFKFB3 interactions with p38γ and GLUT2 as a ternary complex in Pdac cells.

Figure 2. p38γ phosphorylates PFKFB3 at S467 leading to its stabilization and PFKFB3 requires S467 to bind p38γ and GLUT2.

Figure 2.

Open in a new tab

A, Flag-PFKFB3 precipitates were separated on SDS-PAGE after incubation with His-p38γ in vitro (left); gel corresponding to the PFKFB3 band was digested and subjected to Mass Spectrometric analysis (right). B, Flag-isolated PFKFB3 and PFKFB3/S467A expressed in 293T cells were incubated with His-p38γ (p-p38 band, at bottom) for in vitro phosphorylation, which was detected with a p-PFKFB3/S467 specific antibody (p-PFKFB3). The indicated constructs were also expressed in 293T cells and Flag precipitates were analyzed for PFKFB3 phosphorylation and interaction with p38γ and Glut2 proteins (right) with a portion of the lysates analyzed by direct WB as input (middle). C, The indicated constructs were co-expressed in 293T cells, which were then treated with the protein synthesis inhibitor cycloheximide (CHX, 100 μg/ml) for different times and analyzed by WB (top); the means of 3 experiments are plotted at bottom (± SD, * p < 0.05; ** p < 0.01). D, PFKFB3/S467D is more stable than PFKFB3 and PFKFB3/S467A, and p38γ, but not p38γ/AGF, binds PFKFB3/p-PFKFB3 in 293T cells. The indicated constructs were transfected, and Flag precipitates were analyzed for interaction with PFKFB3 and p-PFKFB3 whereas total lysates were analyzed for protein stability as in C (curves, mean ± SD, * p < 0.05, ** p < 0.01). E. Mouse pancreatic tumor cells (KPC) and p38γ knockout counterparts (KPC/p38γ KO) were treated with CHX and analyzed by WB (left) with the summarized results at right (mean ± SD, n = 3, * p < 0.05; ** p < 0.01).

Because KRAS and p38γ are both required for PFKFB3 phosphorylation and expression (Fig. 1F–1I), we examined if p38γ phosphorylates and stabilizes PFKFB3. Figs. 2C shows that p38γ co-transfection increases the stability of PFKFB3, but not PFKFB3/S467A, protein, indicating that p38γ stabilizes PFKFB3 via S467. Moreover, the phospho-mimetic S467D mutant is more stable than PFKFB3 and PFKFB3/S467A, the stability of PFKFB3/S467D is further increased by p38γ co-transfection, and a dominant negative p38γ/AGF fails to bind to PFKFB3 (Fig. 2D), indicating that p38γ may further stabilize PFKFB3 via phosphorylation at additional residues. Endogenous PFKFB3 is also degraded more rapidly in KPC/p38γ KO tumor cells as compared to KPC cells (Fig. 2E). These results together indicate that p38γ stabilizes PFKFB3 by stimulating S467 phosphorylation and binds p-PFKFB3/PFKFB3 by a mechanism depending on its phosphorylation.

p38γ is required for G1-to-S transition in cell-cycle progression, for cell proliferation and motility, for PFKFB3/GLUT2 expression in pancreatic tissues, and for cerulein-induced PFKFB3 expression and phosphorylation, cytokine expression, and pancreatitis

Recent studies showed that p38γ promotes cell cycle progression (22). Because p38γ is required for Pdac growth (Fig. 1), we tested if p38γ stimulates cell cycle progression that is involved in cell proliferation and motility. Supplementary Fig. S2A–S2C show that there are decreased levels of p-RB, RB, and CDK6 proteins in association with increased G1 but decreased S and G2/M populations in KPC/p38γ KO cells as compared to KPC line. Moreover, there is a decrease in cell proliferation, migration and invasion in KPC/p38γ KO as compared to KPC cells (Supplementary Fig. S2D–2G). These results are consistent with the growth-promoting role of p38γ in Pdac cells (Fig. 1B–D) and indicate that p38γ may be required for G1-to-S transition of cell cycle progression of Pdac cells. Thus, p38γ may be fundamentally important for proliferative, invasive, and oncogenic activities of pancreatic cancer through promoting cell cycle progression.

We next examined systemic effects of conditional p38γ KO in the pancreas. To specifically knock out p38γ in pancreatic epithelial cells, p38γfl/fl mice (17) were crossbred with Pdx-1-Cre transgenic animals (25) to generate pancreas-specific p38γ knockout mice (Pdx-1-Cre-p38γfl/fl, or p38γ KO) (Fig. 3A). Fig. 3B/C show that p38γ was specifically deleted from the pancreas in p38γ KO mice, causing decreased PFKFB3 and GLUT2 protein expression without consistent effect on GLUT1. Inflammation plays an important role in pancreatic tumorigenesis (36). Because both p38γ (18) and PFKFB3 (37) are involved in inflammation, we tested if p38γ KO impacts cerulein-induced pancreatitis through PFKFB3 pathways (38). Cerulein treatment increases relative pancreas weight in both groups but the pancreas in control Pdx-1-Cre mice displayed more severely disorganized parenchymal structure and more inflammatory cell infiltration (Fig. 3D/E) (39). Cerulein-induced histological changes and increases in pro-inflammatory cytokines in pancreas, which were significantly attenuated in p38γ KO mice in which protein expression and phosphorylation of PFKFB3 and another p38γ substrate PTPH1 (35) were also decreased (Fig. 3F–3I). These results together indicate that pancreatic p38γ is required for PFKFB3 and GLUT2 expression, for cerulein-induced PFKFB3/PTPH1 expression and phosphorylation, and for cerulein-induced cytokine expression and pancreatitis.

Figure 3. p38γ is required for PFKFB3/GLUT2 expression in pancreatic tissues, for inflammation induced PFKFB3 and PTPH1 expression/phosphorylation, for pro-inflammatory cytokine expression and pancreatitis.

Figure 3.

Open in a new tab

A, PCR shows genotypes of pancreatic-specific p38γ KO. B, the indicated tissues were prepared from two-month-old mice and analyzed for p38γ protein expression. C, Pancreatic p38γ KO decreases protein levels of PFKFB3 and GLUT2. D, Mice were treated with cerulein or saline and effects on pancreas weight relative to body weight were determined (mean ± SD, n = 3 mice, * p < 0.05). E, Pancreas tissues from the indicated mice were processed for H&E staining after the treatment with cerulein or saline. F, RNA was prepared from pancreas tissues and analyzed by qRT-PCR (mean ± SD, n = 3 mice, * p < 0.05). G, the pancreatic tissues were analyzed for histological changes after treatment with cerulein in mice (mean ± SD, n = 3, * < 0.05). H, I, the indicated pancreatic tissues were analyzed by WB (H, 3 mice per group) and IHC (I) for protein expression and/or phosphorylation.

Studies have shown that pancreatic expression of KRAS (G12D) transgene induces pancreatic intraepithelial neoplasia (PanIN), a putative precursor to invasive pancreatic cancer (24). Because p38γ is required for KRAS-induced transformation in intestinal epithelial cells (15), we tested if pancreatic p38γ KO impacts PanIN in three-month old KC mice (LSL-KRASG12D/+; Pdx-1-Cre). Supplementary Fig. S2H–S2J show that 100% of KC mice (6/6) developed PanIN lesions whereas only 71% (5/7) of KC/p38γ KO mice had PanIN, albeit this difference is not statistically significant. IHC analyses showed that there are also no substantial differences in the levels of p-ERK and p-AKT proteins in PanIN lesions between KC and KC/p38γ KO mice (Supplementary Fig.S2J). These results indicate that conditional p38γ KO only moderately attenuates KRAS-induced PanIN. Additional studies with a large group of mice at several time points may be needed to demonstrate a definite role of p38γ in early stages of Pdac tumorigenesis.

p38γ is required for pancreatic tumorigenesis in KPC mice, for aerobic glycolysis and for PFKFB3/GLUT2 expression and/or phosphorylation in KPC tumors and/or cells

Mice with conditional expression of mutant K-Ras and mutant p53 alleles in pancreas ductal epithelial cells (KPC) faithfully reproduce the histological lesions and clinical profiles of human Pdac (25). We next investigated if conditional p38γ KO impacts tumorigenesis in KPC mice and decreases p-PFKFB3/PFKFB3/GLUT2 protein levels. To test if pancreatic p38γ KO impacts tumorigenesis, p38γfl/fl mice were crossbred with LSL-KrasG12D/+, LSL-Trp53R172H/+, Pdx-1-Cre transgenic animals (KPC) (25) to generate p38γ homozygous KO counterparts (KPC/p38γ KO) (Fig. 4A). KPC mice developed pancreatic cancer with a median overall survival (OS) of 160 days; in KPC/p38γ KO animals, survival increased to at least 400 days (Fig. 4B). While the average tumor weight was similar in both groups, the metastasis rate was significantly decreased in KPC/p38γ KO mice (Supplementary Fig. S3A/B). WB and/or IHC analyses further show that p38γ KO results in decreased levels of p-PFKFB3, PFKFB3 and GLUT2 proteins and other mitogenic pathways and pro-inflammatory cytokines in primary tumors and/or lung metastases (Fig. 4C/D and Supplementary Fig. S3C/D). The role of p38γ in pancreatic cancer, metabolic and inflammation pathways was further demonstrated by RNA seq analyses of KPC and KPC/p38γ KO cells (Supplementary Fig. S3E). These results together demonstrate that pancreatic p38γ is required for pancreatic tumorigenesis and for glycolytic/inflammatory pathway activities.

Figure 4. Conditional p38γ KO suppresses pancreatic tumorigenesis in KPC mice and decreases protein amounts of p-PFKFB3, PFKFB3 and GLUT2.

Figure 4.

Open in a new tab

A, PCR shows genotypes of KPC and pancreatic-specific p38γ knockout KPC mice (KPC/p38γ KO). B, Kaplan survival curves of KPC and KPC/p38γ KO mice are shown with the number of mice and medium OS given in brackets and representative tumor images at the indicated times are shown as insets. C, Tumor tissues were prepared from the indicated mice and analyzed by WB (3 mice per group). D, IHC analyses show protein expression or phosphorylation in tumor tissues with CK19 as an epithelial marker and α-SMA as a marker of stromal fibroblast activation.

To determine if p38γ KO affects aerobic glycolysis, KPC and KPC/p38γ KO tumor cells were analyzed. Fig. 5A–C show that KPC/p38γ KO cells have decreased glucose uptake, reduced lactate secretion, and suppressed PFK activity as compared to KPC cells. Seahorse analyses further reveal that there is a decreased extracellular acidification rate (ECAR) (an indicator for glycolysis) but not a decreased oxygen consumption rate (OCR, an indicator of respiration) in KPC/p38γ KO cells (Fig. 5D/E), indicating a required role of p38γ for aerobic glycolysis. Consistent with the WB/IHC data (Fig. 4C/D), GLUT2 RNA is also significantly decreased in KPC/p38γ KO cells (Fig. 5F). We further assessed if p38γ KO affected the levels of glycolytic intermediates specifically from [U-13C6] glucose. Fig. 5G shows that the relative enrichment of 13C labeled lactate was reduced in KPC/p38γ KO cells as compared to KPC cells at 2 min time point in which another glycolytic intermediate pyruvate appears to decrease to a great extent. These results reinforce the ability of p38γ to promote glycolysis. Consistent with this premise, relative lactate and pyruvate abundance were also significantly decreased in KPC/p38γ KO as compared to KPC cells in this experiment (Fig. 5H). Similar patterns were also observed at 24 h time point (Fig. 5G/H, bottom). Moreover, PET scans reveal that there is significant decrease in 18F radio-labeled fluorodeoxyglucose (18FDG) uptake in KPC/p38γ KO allografts in nude mice (Fig. 5I–5K). These results, together with the suppressive effect of p38γ KO in KPC tumorigenesis, indicate that p38γ may be a functional link between the KRAS oncogene and pancreatic tumorigenesis by stimulating of the Warburg effect.

Figure 5. p38γ is required for aerobic glycolysis in KPC cells/allografts. A-C, KPC and KPC/p38γ KO cells were analyzed for glucose uptake.

Figure 5.

Open in a new tab

(A), lactate secretion (B), and PFK activity (C) (results from 3 separate cell lines derived from 3 separate mice are shown with each from 3 experiments (mean ± SD, ** p < 0.01). D, and E, cells were subjected to Seahorse analysis (D, ECAR; E, OCR, average of 3 separate cell lines with each measured in triplicate and normalized by protein content, ** p < 0.01). F, Total RNAs were prepared to measure GLUT1, GLUT2, MAPK12 (p38γ) and Kras levels by qRT-PCR (mean ± SD, n = 3, ** p < 0.01). G, H, cells were exposed to media containing 25 mM D-Glucose (U13C6) for 2 min (top) and 24 h (bottom) and sample extraction and analysis by GC/MS were performed as previously described using liquid chromatography. The mole percentage of 13C-labeled lactate/pyruvate in the mixture is presented in G, whereas the relative intracellular abundance of 13C-labeled lactate and pyruvate is shown in H (left and right panels, respectively) (mean ± SD, n = 3, * p < 0.05; ** p < 0.01). I-K, cells were s.c. injected into nude mice and FDG uptake was measured before the tumor was removed and photographed (I, bar graph, mean ± SD, n = 3, * p < 0.05; tumor image, bottom). For Positron Emission Tomography (PET), 18F-FDG was injected into tumor-bearing nude mice and 45 minutes later images were taken (J and K, top, images with arrows indicating the tumor locations; bottom, mice with tumor grown at right side).

PFKFB3 and GLUT2 depend on p38γ to promote aerobic glycolysis and Pdac growth, and PFKFB3/S467 is required for p38γ glycolytic and oncogenic activity

To investigate if PFKFB3 and GLUT2 depend on p38γ to promote aerobic glycolysis and Pdac growth, both were depleted in KPC and KPC/p38γ KO cells and effects on glycolysis and colony formation were determined. Fig. 6A–6E and Supplementary Fig. S4A/B show that PFKFB3 and GLUT2 depletion decreases glucose uptake, lactate secretion, PFK activity, ECAR, and/or colony formation in KPC cells, but not in KPC/p38γ KO cells, indicating their p38γ-dependent glycolytic and oncogenic activities. PFKFB3 silencing further decreases FDG uptake in KPC allografts and inhibitory effects of PFKFB3 and/or GLUT2 knockdown in glycolysis and/or growth were further demonstrated in human Pdac cells (Fig. 6F/6G; Supplementary Fig. S4C–4H). These results together demonstrate that silencing of PFKFB3 and GLUT2 only inhibits glycolysis and Pdac growth when p38γ is present.

Figure 6. PFKFB3 and GLUT2 depend on p38γ to stimulate glycolysis and/or Pdac growth, and p38γ promotes aerobic glycolysis and Pdac growth through PFKFB3/S467.

Figure 6.

Open in a new tab

A-E, PFKFB3 and GLUT2 were stably depleted by lentiviral mediated shRNA with negative target control (NTC) as control and cells were analyzed by WB (A), for soft-agar growth (B), PFK activity (C), glucose uptake (D), and lactate secretion (E) (mean ± SD, n = 3, * < 0.05; ** p < 0.01). F, G, shPFKFB3 and NTC were introduced into KPC cells by lentivirus and cells were then s.c. inoculated into nude mice. Allograft growth was assessed by 18F-FDG uptake and tumors are shown in F (top), whereas PET images are shown in G (top), with mice bearing tumors being shown in G (bottom) and the summarized FDG uptake is given in F (bottom, mean ± SD, n = 3, * p < 0.05). H-M, p38γ was re-expressed in KPC/p38γ KO cells that had been stably expressed with PFKFB3/S467A or pcDNA3.1 vector and these engineered cells were analyzed for protein expression by WB (H), lactate secretion (I), and colony formation (J, K) (for I and J, mean ± SD, ** p < 0.01), and allograft growth in nude mice (L, M) (mean ± SD, n = 6 mice, ** p < 0.01).

Next, we assessed the role of PFKFB3/S467 phosphorylation in glycolysis by analyzing KPC and KPC/p38γ KO cells that stably express PFKFB3 and its S467A or S467D mutants. There was increased PFK activity and elevated ECAR in KPC cells expressing the S467D mutant, whereas opposite effects were observed in cells transfected with the S467A variant (Supplementary Fig. S4I/M/O) as compared to cells expressing PFKFB3; these results are consistent with the constitutive active (S467D) and dominant negative (S467A) glycolytic activity of these PFKFB3 variants. Consistent with conclusion, rescue experiments further show that S467D expression in KPC/p38γ KO cells more significantly increases PFK activity, lactate secretion, glucose uptake and allograft growth in nude mice as compared with the expression of wild-type PFKFB3 (Supplementary Fig. S4J–4R). These results together indicate a necessary and sufficient role of PFKFB3/S467 in aerobic glycolysis and pancreatic tumor growth downstream of p38γ. To determine if PFKFB3/S467 is required for p38γ glycolytic and oncogenic activities, KPC/p38γ KO cells were stably transfected with PFKFB3/S467A or an empty vector; p38γ was then overexpressed in these cells. Fig. 6H–6M show that p38γ re-expression in KPC/p38γ KO cells increases lactate secretion, colony formation and allograft-growth over Vector cells and that these stimulatory effects were greatly diminished in cells with mutant PFKFB3/S467A. These results indicate that p38γ requires PFKFB3/S467 to stimulate the Warburg effect and to increase Pdac growth in vitro and in mice.

The pharmacological p38γ inhibitor PFD inhibits tumorigenesis and decreases p-PFKFB3 protein levels in KPC, but not KPC/p38γ KO, tumors

To determine if pharmacologically targeting p38γ has a therapeutic potential, PFD was orally administered in mice (18) and its effects on pancreatic tumorigenesis were determined. Fig. 7A/B shows that PFD increased the median OS of KPC mice by 35 days over water control, whereas the same treatment had no significant effects in KPC/p38γ KO mice, indicating that PFD suppresses KPC tumorigenesis by targeting p38γ. Consistent with this conclusion, PFD decreases PFKFB3 (and PTPH1) phosphorylation, Ki-67 expression, and inhibits growth of KPC cells/allografts, but not of KPC/p38γ KO counterparts (Fig. 7A/B; Supplementary Fig. S5A–C), indicating that PFD depends on p38γ to inhibit the phosphorylation of p38γ substrates and to suppress Pdac growth. PFD also inhibits growth of human Pdac xenografts in nude mice and decreases p-PFKFB3 and Ki-67 levels in tumor tissues (Supplementary Fig. S6A). These results together indicate that PFD suppresses Pdac development and growth and inhibits PFKFB3 phosphorylation by targeting p38γ, indicating its potential as a novel targeted therapy for Pdac.

Figure 7. PFD depends on p38γ to inhibit Pdac tumorigenesis and to cooperate with PFK15 to suppress aerobic glycolysis and Pdac growth, and p38γ/p-PFKFB3/PFKFB3/GLUT2 overexpression in clinical specimens predicts a poor prognosis.

Figure 7.

Open in a new tab

A, B, the indicated mice were treated daily with PFD (500 mg/kg, P.O., from 4 weeks after birth for 12 weeks) with water as a control and their survival time was compared. Kaplan survival curves were generated with the number of mice and median OS given as insets; tumor lysates were analyzed by WB (4 and 3 mice per group for A and B, respectively, bottom). C, KPC and KPC/p38γ KO cells were incubated with the indicated inhibitors for 24 h and analyzed by WB. D, E, KPC and KPC/p38γ KO cells (2 × 104) were s.c. injected in 6-week-old nude mice and therapy with PFD (300 mg/kg, P.O daily) and/or PFK15 (10 mg/kg daily) or water was initiated 7 days later for the next 3 weeks. Tumor volume was measured every 4 days; tumor images at the end of the experiments are shown at right and growth curves shown at left (mean ± SD, n = 6 mice, ** p < 0.01 vs either alone). Mouse body weight was also measured during the therapy and the therapy-induced body weight change is presented in the middle. F, Representative IHC images for p38γ, p-PFKFB3, PFKFB3 and GLUT2 in Pdac specimens are shown. G, Kaplan survival curves for Pdac patients whose tumors express high or low levels of the indicated proteins; the number of patients in each group is given in brackets; the log rank test was used to assess the relationship between the protein expression level and patient overall survival time (OS).

p38γ and PFKFB3 inhibitors depend on p38γ to collaboratively inhibit the Warburg effect and Pdac growth, and p38γ, p-PFKFB3, PFKFB3, and Glut2 are upregulated in Pdac specimens, which predicts a poor clinical prognosis

In addition to PFKFB3, p38γ may phosphorylate and activate other oncogenic proteins, whereas PFKFB3 may have glycolytic and oncogenic activities that are independent of p38γ. We therefore tested if p38γ and PFKFB3 inhibitors collaborate to suppress the Warburg effect and to inhibit Pdac growth. We chose a lower concentration of PFD that alone has little effects on growth by itself and analyzed its effect when combined with PFK15, a second generation of PFKFB3 inhibitor (40). Consistent with the effect in KPC tumors in mice, incubation with PFD decreases p-PFKFB3 levels only in KPC, but not in KPC/p38γ KO, cells (Fig. 7C). Significantly, PFD and PFK15 collaborate to deplete p-PFKFB3 and to inhibit glucose uptake, lactate secretion and colony formation in KPC, but not in KPC/p38γ KO, cells (Fig. 7C; Supplementary Fig. S6B–D). Most importantly, PFD and PFK15 collaborate to inhibit the allograft growth of KPC, but not KPC/p38γ KO cells, without significant impacts on mouse body weight (Fig. 7D/E). Similar effects of the PFD plus PFK15 combination on glycolysis and malignant growth were further demonstrated in human Pdac cells/xenografts (Supplementary Fig. S7A–C). Together with the p38γ-dependent oncogenic and glycolytic activity of PFKFB3, these results indicate that dual-targeting of p38γ and PFKFB3 may have a great potential for Pdac therapy by more efficiently disrupting the Warburg effect (Supplementary Fig. S7F).

To demonstrate the clinical significance of p38γ activation of p-PFKFB3, PFKFB3 and GLUT2 pathways, a separate set of clinical Pdac specimens was analyzed for their protein levels by IHC and the results were further compared with patient survival to explore the prognostic value. Results (Fig. 7F) confirmed the overexpressed p38γ as observed in TMA analyses (Supplementary Fig. S1A) and further revealed upregulation of p-PFKFB3, PFKFB3 and GLUT2 in the same cohorts of specimens. Importantly, increased expression of each component of the p38γ/p-PFKFB3/PFKFB3/GLUT2 pathway is associated with decreased patient survival (Fig. 7G). Additional analyses show that the IHC intensity of p-PFKFB3/S467 is positively correlated with p38γ and GLUT2 (Supplementary Fig. S7D/E), suggesting their cooperative role as a glycolytic and oncogenic pathway and/or complex in Pdac. These results together indicate that overexpressed p38γ may signal downstream of KRAS to activate p-PFKFB3/PFKFB3/Glut2 and cooperate with them to promote Pdac tumorigenesis and progression (Supplementary Fig. S7F).

Discussion

The coupling of KRAS oncogenesis with aerobic glycolysis has suggested that blocking this metabolic pathway may be an effective therapeutic strategy for treatment of KRAS-dependent cancers such as Pdac. However, until now, a druggable linker had not been identified (5,11). Our results have provided several pieces of evidence that together indicate that p38γ may be this missing link between KRAS and aerobic glycolysis by activating PFKFB3 and GLUT2 (Supplementary Fig. S7F). First, p38γ expression and phosphorylation are induced by KRAS; second, both KRAS and p38γ increase PFKFB3 and GLUT2 protein expression and/or phosphorylation in cells and/or tissues; and third, p38γ is required for KRAS-induced aerobic glycolysis and Pdac tumorigenesis (Fig. 1–5). Moreover, both PFKFB3 and GLUT2 depend on p38γ to stimulate glycolysis and/or Pdac growth, which may be coordinated and/or even further enhanced through p38γ-induced PFKFB3 phosphorylation and stabilization, induction of GLUT2 expression and their complex formation (Fig. 2/3/6; Supplementary Fig. S1E–1G/S4). Furthermore, the p38γ inhibitor PFD, alone and in combination with PFKFB3 inhibitor, suppresses aerobic glycolysis and inhibits Pdac growth that is dependent on p38γ (Fig. 7A–E and Supplementary Fig. S5A–5C/S6A–6D and S7A–7C). These results together indicate that p38γ promotes Pdac tumorigenesis by activating PFKFB3/GLUT2-dependent aerobic glycolysis and that p38γ and PFKFB3 may be targeted for therapeutic intervention (Supplementary Fig. S7F).

Consistent with the recent finding (22), we found that there is a blockade of cell cycle progression from G1 to S phase in association with decreased CDK6 (cyclin dependent kinase 6) and p-Rb protein levels in KPC/p38γ KO cells as compared to KPC (Supplementary Fig. S2A–2C). The role of p38γ in G1/S transition in cell cycle progression may be essential for its stimulation of cell proliferation and cell motility, activation of glycolysis, promotion of inflammatory responses, and Pdac development and growth (41). p38γ is known to phosphorylate Rb to promote liver proliferation in compensation for the loss of CDK1 or CDK2 (22). In KPC cells, however, we found that while p38γ KO decreases p-Rb, p-PFKFB3 and CDK6 levels, it does not affect the abundance of CDK2 or 4 (Supplementary Fig.S2A; Fig.7C). Because p38γ and CDK6 both phosphorylate PFKFB3 at S467 (42) (Fig.2), it would be interesting in the future to investigate if p38γ promotes cell cycle progression and stimulates aerobic glycolysis in pancreatic cancer by directly phosphorylating PFKFB3/S467 and by indirectly stimulating this phosphorylation via activating CDK6.

The important role of the p38γ pathway in aerobic glycolysis is supported by the fact that silencing of p38γ or PFKFB3 or GLUT2 reduces glucose uptake, suppresses PFK activity, reduces EACR and/or decreases lactate secretion in KRAS-dependent Pdac cells (Fig. 5/6). A decrease in Pdac growth, pancreatic tumorigenesis, PFKFB3 and GLUT2 expression and/phosphorylation, and aerobic glycolysis by knockout of p38γ (Fig. 1/3–5) further indicates that p38γ is a master activator of the Warburg effect downstream of KRAS to promote Pdac tumorigenesis through PFKFB3 and GLUT2. This conclusion is further supported by the fact that depletion of PFKFB3 and/or GLUT2 only inhibits aerobic glycolysis and Pdac growth in cells/allografts expressing p38γ (Fig. 6A–G; Supplementary Fig. S4A–H). Targeting the p38γ pathway may therefore be a specific strategy to disrupt the Warburg effect and to inhibit Pdac tumorigenesis by suppressing PFKFB3/GLUT2-dependent glycolytic pathways.

p38γ may activate PFKFB3 predominantly by stimulating its phosphorylation at S467, as p38γ phosphorylates PFKFB3 at S467 in vitro and in vivo, and p38γ stabilizes PFKFB3 via S467 (Fig. 1/2). Moreover, p38γ depletion or KO decreases p-PFKFB3 levels and the dominant-negative PFKFB3/S467A mutant blocks p38γ-induced glycolytic/oncogenic activity (Fig. 1–3/4/6H–6M). Furthermore, expression of the phospho-memetic PFKFB3/S467D mutant in KPC/p38γ KO cells more effectively rescues aerobic glycolysis and allograft growth in mice than in cells that express wild-type PFKFB3 (Supplementary Fig. 4I–4R). These results together indicate a necessary and sufficient role of PFKFB3/S467 phosphorylation to stimulate aerobic glycolysis and Pdac growth. Although PFKFB3 can be phosphorylated by several kinases such as AMPK (43) and IKKβ (44), p38γ may be the first oncogenic kinase that stimulates PFKFB3/S467 phosphorylation causing its stabilization, which may be the mechanism for increased p-PFKPB3 and PFKFB3 levels in pancreatic cancer specimens (34) (Fig. 7F/G). Both KRAS and p38γ also positively regulate GLUT2 gene expression in pancreatic cells and tissues (Fig. 1F/1G/5F). In addition to transcriptional regulation, p38γ may stimulate GLUT2 phosphorylation and/or regulate GLUT2 stability or its cytoplasmic membrane localization as observed with a constitutively active MAPK (45). These possibilities remain to be determined. However, PFKFB3 depends on S467 to bind p38γ and GLUT2, p38γ requires phosphorylation to interact with PFKFB3 (Fig. 2B/2D), and p38γ binds p-PFKFB3/PFKFB3/GLUT2 in several Pdac cell lines, but in not normal pancreatic HPNE cells (Supplementary Fig. S1E–G). These results together indicate that the p38γ-activated, PFKFB3/S467-excuted and GLUT2-containing ternary complex may more efficiently promote aerobic glycolysis and Pdac tumorigenesis through their locally elevated concentrations.

The p38γ inhibitor PFD has a stronger inhibitory activity in vitro on p38γ MAPK (also called SAPK3) than on other p38 family members (46). While PFD may also have off-target effects, we found that PFD depends on p38γ to decrease PFKFB3 phosphorylation, to suppress glucose uptake, to reduce lactate secretion, and to inhibit Pdac growth (Fig. 7A/B; Supplementary Fig. S5A–C). PFD has an established anti-fibrotic activity in the clinic (47), which may be also important for its therapeutic effects in KPC mice. For example, PFD reduces α-SMA expression in KPC, but not KPC/p38γ KO, allografts (Supplementary Fig. S5C) and there is also deceased α-SMA expression in primary KPC/p38γ KO tumors (Fig. 4D). These results suggest that epithelial p38γ may regulate Pdac tumorigenesis through signaling crosstalk with fibroblastic cell compartments, which remains to be tested in a fibroblast-specific p38γ KO model. Of great interest, PFD depends on p38γ to collaborate with the PFKFB3 inhibitor PFK15 to further suppress the Warburg effect and to inhibit Pdac growth (Fig. 7C–E; Supplementary Fig. S6B–D), thus revealing a new therapeutic strategy through dual-targeting of p38γ and PFKFB3. Because PFD is currently used clinically (47) and PFK15 (48) is in clinical trials for treating solid tumors (48), their combination may have an immediate potential to improve clinical outcomes of pancreatic cancer. This notion is further supported by the positive correlation of up-regulated p38γ with p-PFKFB3 in clinical Pdac specimens (Fig.7F/G; Supplementary Fig. S7D–S7F).

Supplementary Material

1
2
3
4
5
6
7
8

Significance:

Findings show that p38gamma links KRAS oncogene signaling and the Warburg effect through PFKBF3 and Glut2 to promote pancreatic tumorigenesis, which can be disrupted via inhibition of p38gamma and PFKFB3.

Acknowledgements

We would like to thank Drs. Ning Yin and Donna McAllister for the help with breeding and maintaining mice in the early stages of the study, and Dr. Xiuxu Chen for the help in TMA analyses. We also thank Drs. David Tuveson and Tyler Jacks for providing KPC mice and Richard Klemke for HPNE/Kras cells. We thank Genentech for providing the PFD drug and Dr. Zhen Yan and Boehringer Ingelheim Pharmaceuticals for providing p38γflox/flox mice. The Redox and Bioenergetics Shared Resource of the MCW Cancer Center conducted parts of the Seahorse analyses for ECAR and OCR. This work was supported by grants to G.C. (Department of Veterans Affairs, Merit Review 1I01BX002883; Department of Defense, BC141898; NIH R01 CA91576 and the Cancer Center of the Medical College of Wisconsin); by funding from the Natural Scientific Foundation of China (No. 81402503), Science and Technology Program of Guangzhou, China (No. 201802020002), The Basic Scientific Fund of Sun Yat-Sen University (19ypy196), and NIH to M. D. (NIH R01 CA178960 and CA226279).

Footnotes

Conflict of Interest: MD is the co-founder of a biotech startup, Protein Foundry, LLC, that produces recombinant cytokines for biomedical research purposes and all other authors declared no potential conflict of interest.

References

  • 1.Hidalgo M Pancreatic cancer. N Engl J Med 2010;362:1605–17. [DOI] [PubMed] [Google Scholar]
  • 2.Conway JRW, Hermann D, Evans TRJ, Morton JP, Timpson P. Combating pancreatic cancer with PI3K pathway inhibitors in the era of personalised medicine Gut 2019;68:742–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Hayes TK, Neel NF, Hu C, Gautam P, Chenard M, Long B. et al. Long-term ERK inhibition in KRAS-mutant pancreatic cancer is associated with MYC degradation and senescence-like growth suppression. Cell 2016;29:75–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Halbrook CJ, Lyssiotis CA. Employing metabolism to improve the diagnosis and treatment of pancreatic cancer. Cancer Cell 2017;31:5–19. [DOI] [PubMed] [Google Scholar]
  • 5.Ying H, Dey P, Yao W, Kimmelman AC, Draetta GF, Maitra A. et al. Genetics and biology of pancreatic ductal adenocarcinoma. Genes & Develop 2016;30:355–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell 2011;144:646–74. [DOI] [PubMed] [Google Scholar]
  • 7.Warburg O, Wind F, Negelein E. The metabolism of tumors in the body. J Gen Physiol 1927;8:519–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Potter M, Newport E, Morten KJ. The Warburg effect: 80 years on. Biochem Soc Trans 2016;44:1499–505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Phay JE, Oladapupo S, Koch S, Short C, Granner D, Printz RI et al. Oncogenic H-Ras increases glucose metabolism with an induction of glucose transporter GLUT1 and hexokinase 2 expression. J Amer College Surg 2007;205:S89. [Google Scholar]
  • 10.Telang S, Yalcin A, Clem AL, Bucala R, Lane AN, Eaton JW, et al. Ras transformation requires metabolic control by 6-phosphofructo-2-kinase. Oncogene 2006;25:7225–34. [DOI] [PubMed] [Google Scholar]
  • 11.Ying H, Kimmelman AC, Lyssiotis CA, Hua S, Chu GC, Fletcher-Sananikone F. et al. Oncogenic Kras maintains pancreatic tumors through regulation of anabolic glucose metabolism. Cell 2012;149:656–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Vander Heiden MG, DeBerardinis RJ. Unerstanding the intersections between metabolism and cancer biology. Cell 2017;168:657–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Loesch M, Chen G. The p38 MAPK stress pathway as a tumor suppressor or more? Front Biosci 2008;13:3581–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Chen G, Hitomi M, Han J, Stacey DW. The p38 pathway provides negative feedback to Ras proliferative signaling. J Biol Chem 2000;275:38973–80. [DOI] [PubMed] [Google Scholar]
  • 15.Tang J, Qi X, Mercola D, Han J, Chen G. Essential role of p38γ in K-Ras transformation independent of phosphorylation. J Biol Chem 2005;280:23910–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Li Z, Jiang Y, Ulevitch RJ, Han J. The primary structure of p38γ: a new member of p38 group of MAP kinases. Biochem Biophys Res Commun 1996;228:334–40. [DOI] [PubMed] [Google Scholar]
  • 17.Pogozelski A, Geng T, Li P, Lira V, Zhang M, Chi JT, et al. p38γ mitogen-activated protein kinase is a key regulator in skeletal muscle metabolic adaptation in mice. Plos One 2009;4:e7934. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Yin N, Qi X, Tsai S, Lu Y, Basir Z, Oshima K. et al. p38γ MAPK is required for inflammation-associated colon tumorigenesis. Oncogene 2016;35:1039–48. [DOI] [PubMed] [Google Scholar]
  • 19.Ho RC, Alcazar O, Fujii N, Hirshman MF, Goodyear LJ. p38γ MAPK regulation of glucose transporter expression and glucose uptake in Ly myotubes and mouse skeletal muscle. Am J Physio Regul Integr Comp Physiol 2003;286:R342–R9. [DOI] [PubMed] [Google Scholar]
  • 20.Hou SW, Zhi H, Pohl N, Loesch M, Qi X, Li R, et al. PTPH1 dephosphorylates and cooperates with p38γ MAPK to increases Ras oncogenesis through PDZ-mediated interaction. Cancer Res 2010;70:2901–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Del Reino P, Alsina-Beauchamp D, Escos A, Cerezo-Guisado MI, Risco A, Aparicio N. et al. Pro-oncogenic role of alternative p38 mitogen-activated protein kinases p38γ and p38δ, linking inflammation and cancer in colitis-associated colon cancer. Cancer Res 2014;74:6150–60. [DOI] [PubMed] [Google Scholar]
  • 22.Tomas-Loba A, Manieri E, Gonzalez-Teran B, Mora A, Leiva-Vega L, Santamans AM. et al. p38γ is essential for cell cycle progression and liver tumorigenesis. Nature 2019;568:557–60. [DOI] [PubMed] [Google Scholar]
  • 23.Olive KP, Tuveson DA, Ruhe ZC, Yin B, Willis NA, Bronson RT. et al. Mutant p53 gain of function in two mouse models of Li-Fraumeni syndrome. Cell 2004;119:847–60. [DOI] [PubMed] [Google Scholar]
  • 24.Hingorani SR, Petricoin III EF, Maitra A, Rajapakse V, King C, Jacobetz MA. et al. Preinvasive and invasive ductal pancreatic cancer and its early detection in the mouse. Cancer Cell 2003;4:437–50. [DOI] [PubMed] [Google Scholar]
  • 25.Hingorani SR, Wang L, Multani AS, Combs C, Deramaudt TB, Hruban RH, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell 2005;7:469–83. [DOI] [PubMed] [Google Scholar]
  • 26.Kelber JA, Reno T, Kaushall S, Metildi C, Wright T. et al. KRAS induces a src/PEAK1/ErbB 2 kinase amplification loop that drives metastatic growth and therapy resistance in pancreatic cancer Cancer Res 2012;72:2554–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Collins MA, Bednar F, Zhang Y, Brisset J, Galban S, Galban CJ, et al. Oncogenic Kras is required for both the initiation and maintenance of pancreatic cancer in mice. J Clin Invest 2012;122:639–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Koh A, Molinaro A, Stahlman M, Khan MT, Schmidt C, Manneras-Holm L. et al. Microbially produced imidazole propionate impairs insulin signaling through mTORC1. Cell 2018;175:947–61. [DOI] [PubMed] [Google Scholar]
  • 29.Thorens B, Wu Y, Leahy JL, Weir GC. The loss of GLUT2 expression by glucose-unresponsive b cells of db/db mice is reversible and is induced by the diabetic environment. J Clin Invest 1992;90:77–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Fleming JB, Shen G, Holloway SE, Davis M, Brekken RA. Molecular consequences of silencing mutant K-ras in pancreatic cancer cells: Justification for kras-directed therapy. Mol Cancer Res 2005;3:413–23. [DOI] [PubMed] [Google Scholar]
  • 31.Young A, Lou D, McCormick F. Oncogenic and wild-type Ras play divergent roles in the regulation of mitogen-activated protein kinase signaling. Cancer Discov 2013;3:112–23. [DOI] [PubMed] [Google Scholar]
  • 32.Bando H, Atsumi T, Nishio T, Niwa H, Mishima S, Shimizu C. et al. Phosphorylation of the 6-phosphofructo-2-kinase/fructose 2,6-bisphosphatase/PFKFB3 family of glycolytic regulators in human cancer. Clin Cancer Res 2005;11:5784–92. [DOI] [PubMed] [Google Scholar]
  • 33.Atsumi T, Chesney J, Metz C, Leng L, Donnelly S, Makita Z. et al. High expression of inducible 6-phosphofructo-2-kinase/fructose-2,6-bisphophatase (iPFK-2; PFKFB3) in human cancers. Cancer Res 2002;62:5881–7. [PubMed] [Google Scholar]
  • 34.Bartrons R, Simon-Molas H, Rodriguez-Garcia A, Castario E, Navarro-Sabate A, Manzano A. et al. Fructose 2,6-bisphosphate in cancer cell metabolism. Front in Oncol 2018;8:331. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Hou S, Padmanaban S, Qi X, Leep A, Mirza S, Chen G. p38g MAPK signals through phosphorylating its phosphatase PTPH1 in regulating Ras oncogenesis and stress response. J Biol Chem 2012;287:27895–905. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Guerra C, Schuhmacher AJ, Canamero M, Grippo PJ, Verdaguer L, Perez-Gallego L, et al. Chronic pancreatitis is essential for induction of pancreatic ductal adenocarcinoma by K-Ras oncogenes in adult mice. Cancer Cell 2007;11:291–302. [DOI] [PubMed] [Google Scholar]
  • 37.Schoors S, Bock KB, Cantelmo AR, Georgiadou M, Ghesquiere B, Cauwenberghs S. et al. Partial and transient reduction of glycolysis by PFKFB3 blockade reduces pathological angiogenesis. Cell Metab 2014;17:39–48. [DOI] [PubMed] [Google Scholar]
  • 38.Stumpf F, Algul H, Thoeringer CK, Schmid RM, Wolf E, Schneider MR, et al. Metamizol relieves pain without interfering with cerulein-induced acute pancreatitis in mice. Pancreas 2016;45:572–8. [DOI] [PubMed] [Google Scholar]
  • 39.Lugea A, Man L, French SW, Bezerra JA, Gukovskara AS, Pandol SJ. Pancrease recovery following cerulein-induced pancreatitis is impaired in plasminogen-deficient mice. Gastroenterol 2006;131:885–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Clem BF, O’Neal J, Tapolsky G, Clem AL, Imbert-Femandez Y, Kerr II D, et al. Targeting 6-phosphofrcuto-2-kinase (PFKFB3) as a therapeutic strategy against cancer. Mol Cancer Ther 2013;16:1461–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Icard P, Fournel WZ, Wu Z, Alifano M, Lincet H. Interconnection between metabolism and cell cycle in cancer. Treands in Biochemical Sciences 2019;44:490–501. [DOI] [PubMed] [Google Scholar]
  • 42.Xing Z, Zhang Y, Liang K, Yan L, Xiang X, Li C, et al. Expression of long noncoding RNA YIYA promotes glycolysis in breast cancer. Cancer Res 2018;78:4524–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Marsin AS, Betrand L, Rider MH, Deprez J, Beauloye C, Vincent MF, et al. Phosphorylation and activation of heart PFK2 by AMPK has a role in the stimulation of glycolysis during ischaemia. Curr Biol 2000;10:1247–55. [DOI] [PubMed] [Google Scholar]
  • 44.Reid MA, Lowman XH, Pan M, Tran TQ, Warmoes MO, Gabra MBI. IKKβ promotes metabolic adaption to glutamine deprivation via phosphorylation and inhibition of PFKFB3. Genes & Develop 2016;30:1837–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Yamamoto Y, Yoshimasa Y, Koh M, Suga J, Masuzaki H, Ogawa Y, et al. Constitutively Active Mitogen-Activated Protein Kinase Kinase Increases GLUT1 Expression and Recruits Both GLUT1 and GLUT4 at the Cell Surface in 3T3-L1 Adipocytes. Diabetes 2000;49:332–9. [DOI] [PubMed] [Google Scholar]
  • 46.Ozes O, Blatt LM, Seiwert SD. Use of pirfenidone in therapeutic regimens. United States Patent-US 7,407,973 B2 2008;Aug. 5th:1–46.
  • 47.King TE, Bradford WZ, Castro-Bernardini S, Fagan E,A,, Glaspole I, Glassberg MK, et al. A phase 3 trial of pirfenidone in patients with idiopathic pulmonary fibrosis. N Engl J Med 2014;370:2083–92. [DOI] [PubMed] [Google Scholar]
  • 48.Clem BF, O’Neal J, Klarer AC, Telang S, Chesney J. Clinical development of cancer therapeutics that target metabolism. Inter J Med 2016;109:367–72. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

1
NIHMS1607187-supplement-1.docx (77.5KB, docx)
2
NIHMS1607187-supplement-2.png (3MB, png)
3
NIHMS1607187-supplement-3.png (5MB, png)
4
NIHMS1607187-supplement-4.png (3.9MB, png)
5
NIHMS1607187-supplement-5.png (692.6KB, png)
6
NIHMS1607187-supplement-6.png (5.5MB, png)
7
NIHMS1607187-supplement-7.png (1.8MB, png)
8
NIHMS1607187-supplement-8.png (294.3KB, png)

RESOURCES