Regulation of Pyruvate Kinase M Switch towards PKM1 by LKB1-AMPK Axis Tolerates Hypoglycemic Stress in Cancer Cells

Prakasam Gopinath1, Rajnish Kumar Singh1, 2, Mohammad Askandar Iqbal1, 3, Ashu Bhan Tiku4, Rameshwar N.K. Bamezai1,*

Author’s affiliations

1. National Center for Applied Human Genetics, School of Life Sciences, Jawaharlal Nehru University,New Delhi 110067, India. 2. Department of Microbiology and Tumor Virology Program of the Abramson Cancer Center, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania 19104. 3. Department of Biotechnology, Faculty of Natural Sciences, Jamia Millia Islamia, New Delhi 110025, India. 4. Radiation and Cancer Therapeutics Laboratory, School of Life Science, Jawaharlal Nehru University, New Delhi 110067, India.

* Correspondence: School of Life Sciences, Jawaharlal Nehru University, New Delhi-110067, India.Fax: 0091-11-26742211. E-mail: bamezai@hotmail.com.

Running Title: LKB1-AMPK-PKM1 axis and Hypoglycemic Tolerance in Cancer Cells

Key Words: Cancer Biology, Pyruvate Kinase, Warburg Effect, AMP‐Activated Kinase (AMPK), Hypoglycemia, Energy Metabolism, Liver Kinase B1 (LKB1), heterogeneous nuclear ribonucleoproteins (hnRNP).

Abstract

Cancer cells alter expression of core glycolytic enzymes to rewire metabolism for limitless proliferative competence. Unchecked division and poor vascularization of cancer cells creates a hypoglycemic tumor-microenvironment; where the outcome of glucose-depletion on aerobic glycolysis and associated key glycolytic enzymes remains elusive. We show here that glucose depleted cancer cells activate LKB1-AMPK pathway to regulate the switch of pyruvate kinase M isoform expression from PKM2 to PKM1, and increase its activity through enhanced subunit association. Remarkably, AMPK promotes the alternative splicing of PKM mRNA in favour of PKM1 by antagonizing MTOR and down regulating hnRNPs expression. AMPK driven switch towards PKM1 was critical for glucose depleted cancer cells to augment ATP levels through aerobic glycolysis, validated by AMPK and PKM1 silencing experiments. Accordingly, knockdown of PKM1 attenuated proliferation and enhanced induction of apoptosis under depleted-glucose condition. Collectively, our results demonstrate that AMPK-PKM1 axis favours cancer cell survival by preserving the energy homeostasis under hypoglycemic stress, a

conclusion supported in preliminary Immunohistochemical studies of tumor tissues.

Introduction

Apart from six cardinal alterations in cell physiology that collectively dictate transformation (1), the additional hallmark feature of cancer cells is its unique abnormal bioenergetics with a shift in glucose metabolism from oxidative phosphorylation to aerobic glycolysis (2). The tumor cells consume ample amount of glucose and largely break it down to lactate under well oxygenated conditions, a phenomenon known as Warburg effect (3). Although, our understanding of Warburg effect (aerobic Glycolysis) is still inconclusive, accumulating evidence suggest the role of oncogenic mutations in reprogramming metabolism (4). Such a rewiring of metabolism confers multitude of advantages to cancer cells including, macromolecular synthesis, rapid ATP generation, and in maintaining redox balance (2),

(5). Glycolytic enzymes like HK2 (Isoform 2 of Hexokinase), PKM2 (Isoform M2 of Pyruvate kinase) and LDHA (Lactate dehydrogenase A subunit), have been proposed as critical regulators of Warburg effect (6-8). The expression and oligomer status of these enzymes is suggested to determine the flux of glycolysis and the fate of

1

http://www.jbc.org/cgi/doi/10.1074/jbc.M116.729079The latest version is at JBC Papers in Press. Published on April 26, 2016 as Manuscript M116.729079

Copyright 2016 by The American Society for Biochemistry and Molecular Biology, Inc.

This article has been withdrawn by the authors. The PKM2 immunoblot in Fig. 2E was reused as part of the caspase-3 immunoblot in Fig. 9C. The PKM2 immunoblot from 5 mM Glu, fractions 1–10 was reused as the PKM2 immunoblot from 1 mM Glu, fractions 1–10. The actin immunoblot from A549 cells from Fig. 5A was reused as the actin blot from Fig. 7C.

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

glucose. Oncogenic signals are known to regulate these enzymes at transcription, post-transcription, translation and post-translation levels to provide an adaptive sustenance, metabolically and non-metabolically, to proliferating cancer cells that ultimately develop into a tumor (9-13).

Glucose addicted and rapidly proliferating cancer cells, in comparison to their normal counterparts, are more vulnerable to metabolic stress in vitro and in vivo, possibly because of the lack of tumor suppressive networks, which sense and uncouple proliferation in relation to limited availability of nutrient (14). During nutrient deprivation and hypoxia, AMP-activated protein kinase (AMPK) - a serine/threonine protein kinase, is known to get activated by sensing bioenergetic stress of decreasing intracellular ATP and increasing intracellular AMP and ADP (15). AMPK plays a pivotal role in conserving the cellular energetic homeostasis by remodeling the metabolic phenotype to resist nutritional stress. Upon activation, AMPK increases catabolic ATP generating processes, such as fatty-acid oxidation and inhibits ATP consuming biosynthetic processes, such as protein, cholesterol and fatty-acid synthesis (15,16). In recent years, extensive studies have appreciated the importance of LKB1-AMPK (Liver Kinase B1, an upstream kinase of AMPK) pathway in tumorigenesis, where the axis of LKB1-AMPK provides tolerance to nutrient deprivation in cancer cell (17,18). Similarly lack of LKB1 expression has been suggested to result in inhibition of tumorigenesis (19); suggesting LKB1-AMPK has tumor supportive properties along with already known tumor suppressive functions (20). However, how AMPK contributes to metabolic phenotype of cancer cells is largely unknown. Evidently, it is pertinent to investigate the role of LKB1-AMPK axis in regulation of glycolytic enzymes, like Hexokinase (HK), Pyruvate kinase (PK) and Lactate Dehydrogenase (LDH).

Due to high proliferation rate and poor vasculature within deep-seated core cells and certain pockets, tumor cells face the challenge of nutrient starvation. This situation raises important question of how nutrient insufficiency affects the metabolism of these cells. Also, what remains unanswered is whether these cells rely on

glycolysis, like their richly vascularized counterparts and how do the critical glycolytic enzymes respond to nutrient deprivation? Here, we attempt to address these questions and unravel a mechanism that comes into play to ensure survival of nutritionally (glucose) starved cancer cells. We demonstrate a mechanism which allows rapidly proliferating cancer cells to adapt glucose insufficiency by switching the expression of pyruvate kinase to PKM1 isoform in an LKB1-AMPK dependent manner with concomitant down-regulation of the expression of PKM2. We reveal the significance of AMPK regulated PKM1 expression in maintaining energy level in cancer cells through enhanced aerobic glycolysis and show how knock-down of PKM1 under glucose depleted conditions results in enhanced apoptosis.

Experimental Procedures

Reagents and Antibodies

AICAR, Compound C, Metformin, Phenformin and Oligomycin were procured from Sigma-Aldrich AICAR and Compound C was dissolved in DMSO, and aliquots of stock were stored at -20°C. Rapamycin was procured from Cell Signaling Technology (Beverly, MA). Primary antibodies used in the study were as follows. PKM1 (#SAB4200094), PKM2 (#SAB4200095), Myc-tag (#C 3956), β-Actin (#A1978) and HA Tag (# H 6908) were procured from Sigma-Aldrich. AMPKa (#5831), p-AMPKa(T172) (#2535), ACC (#3676), p-ACC (Ser79) (#11818), p-PKM2 (Y105) (#3827), P70 S6 Kinase (#2708), p-P70 S6 Kinase(T389) (#9234), p-S6 Ribosomal Protein (Ser 235/236) (#4858), p-S6 Ribosomal Protein (Ser 240/244) (#5364), anti-Caspase-3 (#9662) and anti- Cleaved Caspase-3 (#9664) (Cell Signaling Technology) c-Myc (# sc-47694) and anti PARP (#sc-7150) from (Santa Cruz Biotechnology, Inc.). Secondary antibodies, anti-Rabbit HRP (# 7076) and anti-Mouse HRP (# 7074) were from Cell signaling Technology.

Cell Culture and Tumor tissue Samples

Cell lines, MCF-7 (Human Breast Adenocarcinoma), MDA-MB-231 (Human Breast Adenocarcinoma), PC-3 (Human Prostate Adenocarcinoma), H1299 (Non-Small Cell Lung Carcinoma), A549 (Lung Carcinoma), HeLa

2

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

(Human Cervical Adenocarcinoma), HEK293T (Human Embryonic Kidney Cell line) and L6 (Rat Skeletal Muscle), procured from American Type Culture Collection (ATCC) or National Centre for Cell Science (NCCS), Pune, India, were grown in DMEM medium (Sigma-Aldrich), supplemented with 1X from 100X stock solution of Penicillin/Streptomycin (Sigma-Aldrich), 10 % Fetal Bovine Serum (FBS) (Gibco - Life technologies) and maintained in the incubator (Thermo Scientific Heraeus® - UK) at 37°C and 5% CO2. For experimental purposes cells were cultured in a glucose free DMEM medium (Sigma-Aldrich), and required glucose was supplemented exogenously to attain glucose concentration independently. Tumor tissues from sporadic breast cancer patients (provided by Dr. Gaurav Agarwal from Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow), were collected as detailed earlier (21,22) with the prior approval from Jawaharlal Nehru University ethical committee.

Cloning and Site Directed Mutagenesis

The Coding Sequence of AMPKa2, c-Myc, PKM1 and PKM2 isoforms were PCR amplified from cDNA of H1299 cell line. The PCR amplified ORFs were sequenced and cross checked for background mutations and were cloned in pcDNA™3.1/myc-His (-) A or pcDNA™3.1/HA vectors. Expression vector pcDNA™3.1/myc-His (-) A, Lentiviral transfer vector (pLKO.1), and packaging vectors (psPAX and pMD2.G), was a kind gift from Prof. Shyamal K Goswami and Dr. Goutam K Tanti (SLS, JNU, New Delhi). Full-length constructs of ‘pcDNA-LKB1-Myc tag’ was a generous gift from Dr. Shaida Andrabi (University of Kashmir, Srinagar, J&K). Lentiviral transducing vector, pLKO.1 encoding shPKM1 and shPKM2 was a generous gift from Dr. Marta Cortes Cros (Novartis, Basel). Lentiviral shRNA vector targeting the expression of c-Myc (pLKO.1 shc-Myc) and AMPK a1 and 2 (pLKO.1 shAMPKa1/2) was constructed in-house using synthetic oligonucleotides ((shc-Myc 5’-ccggcctgagacagatcagcaacaactcgagttgttgctgatctgtctcaggtttttg-3’) and (shAMPKA1/2 5’-ccggatgatgtcagatggtgaatttctcgagaaattcaccatctgacatcattttttg-3’)). Site Directed Mutagenesis (SDM) was performed using Quick Change Site directed

mutagenesis Kit (Agilent Technologies), according to the manufacturer’s protocol. In brief, Wild Type (WT) construct of AMPK a2 was truncated and AMPK a2 constitutively active T172D mutant generated using the site directed mutagenesis.

Generation of Stable Gene Expressing, And Knockdown Cell Lines

To establish stable gene expression, constructs (pcDNA - PKM1-Myc, PKM2-Myc, c-Myc-HA and LKB1-Myc) were transfected using Lipofectamine® LTX reagent (Life Technologies), as per the manufacturer’s instructions. Briefly, 48 hours of the post transfection, cells were selected in G418 (1 mg/mL) containing selection medium for 2 weeks to generate stable cell lines. For stable gene knockdowns, the lentiviral particles were generated as described previously (23). In brief, HEK293T cells were transfected with transfer vector (LKO.1) harboring shRNAs, along with packaging vectors, psPAX and pMD2.G, using Lipofectamine® LTX. After 48 hours of post transfection viral particles were harvested and used to infect the target cells. Infected cells were selected in puromycin (2µg/mL) containing DMEM medium for the course of 14 days for generating cell lines with stable gene knockdown.

Immunoblotting and Immunohistochemistry

Immunoblotting analysis were performed by lysing the cells in modified RIPA buffer (50 mM Tris-HCl pH 7.2, 150 mM NaCl, .5% Sodium deoxycholate, .1% SDS, 1% Triton X 100) supplemented with 1mM PMSF, Protease inhibitor cocktail and phosphatase Inhibitor Cocktail II and III (Sigma-Aldrich). Protein lysates were cleared by centrifugation and concentrations quantified using BCA kit (Thermo scientific). Equal amount of protein lysate was resolved in SDS-PAGE, transferred to nitrocellulose membrane and probed with primary antibodies of interest. Protein bands were detected using Luminata forte (Millipore). Immunohistochemistry (IHC) was performed as described previously (24). In brief clinically and histologically defined sporadic breast cancer tissue sample was fixed for 24 hours in phosphate buffer saline with 10% formalin and embedded in paraffin, followed by sectioning (4µm serial section), and staining with anti-pACC (Ser79),

3

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

anti-PKM1, anti-PKM2 and (H & E) staining respectively (with the assistance of Dr. Chitra Sarkar, Department of Pathology, AIIMS, New Delhi).

RT-PCR

Total RNA isolated from cells with TRI Reagent (Sigma-Aldrich) was reverse transcribed into cDNA, using Superscript® III Reverse Transcriptase kit (Life Technologies). qRT-PCR analysis was carried out in BIORAD CFX96 TouchTM Real-time PCR Detection system,

using SYBR Green PCR master mixture (Applied Biosystems). The relative gene expression was calculated using the comparative CT method (2−ΔΔCT). Results were analyzed and presented as fold change (Log10 relative quantification) after normalizing with the control group (25 mM Glucose). The reactions were run in triplicate using ACTIN as an endogenous control. Primers used for the RT-PCR analysis are as follows: HK1 ((Forward 5’-tacttcacggagctgaaggatg-3’) (Reverse 5’-agccatcaggaatggacctt-3’)), HK2 ((Forward 5’-ccaaccttaggcttgccatt-3’) (Reverse 5’- cttggacatgggatggggtg-3’)), PKM1 (Exon9) ((Forward 5’-aggcagccatgttccac-3’) (Reverse 5’-tgccagactccgtcagaact-3’)), PKM2 (Exon10) ((Forward 5’-tgcaattatttgaggaactcc-3’) (Reverse 5’-cactgcagcacttgaaggag-3’)), LDHA ((Forward) 5’-gacctacgtggcttggaaga-3’ (Reverse) 5’-tccatacaggcacactggaa-3’), LDHB ((Forward 5’- ccaacccagtggacattctt-3’) (Reverse 5’-aaacacctgccacattcaca-3’)), c-Myc ((Forward 5’-gcttttttgccctgcgtgac-3’) (Reverse 5’-cgcacaagagttccgtagc-3’)), PTBP1 ((Forward 5’-acggaccgtttatcatgagc-3’) (Reverse 5’-catcaggaggttggtgacct-3’)), hnRNPA1 ((Forward 5’-ttgtgaactcagccaagcac-3’) (Reverse 5’-cagcgtcacgatcagactgt-3’)), hnRNPA2/B1 ((Forward 5’-ggctacggaggtggttatga-3’) (Reverse 5’-cccatggcaaataggaagaa-3’)) and ACTIN ((Forward 5’-actcttccagccttccttc-3’) (Reverse 5’- atctccttctgcatcctgtc-3’)). For semi-quantitative RT-PCR, Images of RT-PCR products resolved in agarose gel were obtained by SYNGENE gel documentation system and densitometric analysis performed using ImageJ software (http://imagej.nih.gov/ij/) to calculate the changes in the ratio of gene expression, which were plotted

after normalizing with the expression of ACTINB (ACTB).

Metabolic Assays

Glucose consumption and lactate release levels were measured in cell culture medium, using Glucose (Hexokinase) assay kit (Sigma-Aldrich) and Lactate Colorimetric/Fluorometric assay Kit (BioVision) following the manufacturer’s specifications. Pyruvate was extracted using Trichloroacetic acid (TCA) precipitation of cells and pyruvate concentration was assessed spectrophotometrically with LDH coupled assay. The concentration of the ATP was measured using the ATP bioluminescence assay kit (BioVision) as per the manufactures instruction.

Glycolytic Enzyme Assays and Glycerol Gradient Centrifugation

Hexokinase (HK) activity was measured through an enzyme glucose-6-phosphate dehydrogenase (G6PD) coupled assay, following the reduction of NADP. 5 µg of whole cell protein lysate was added with the mixture of 100 mM Tris-HCl (pH 8), 5 mM MgCl2, 100 mM glucose, 0.8 mM ATP, 1 mM NADP and 2 units of G6PD and reaction was measured spectrophotometrically at 340nm for 5 min (25). Lactate Dehydrogenase (LDH) catalytic activity was measured spectrophotometrically at 340 nm, through the conversion of pyruvate into lactate, following the oxidation of NADH. To the mixture of 200mM Tris-HCl (pH 7.3), 6.6 mM NADH and 30 mM Sodium pyruvate in the final volume of 1mL, 5 µg of protein lysate was added to measure the LDH activity. Pyruvate kinase (PK) activity was measure as described previously (26). Specific activity of enzymes (HK, LDH and PK) per mg of cell lysate was calculated as:

Glycerol gradient ultracentrifugation experiment was performed by loading 750µg of Whole cell protein lysates on top of 15-33% step gradient and

OD340/min Units mg =

6.22× mg lysate / mL reaction mixture

4

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

centrifuged at 50,000 rpm for 18 hours at 4°C in SW55Ti rotor (Beckman Coulter) and fractions were collected and examined for PK activity as previously mentioned (27) and subjected to Immunoblot analysis.

Cell proliferation Assay

To each well of 24 well plates, 50,000 cells were seeded and after 24 hours, adherent cells were washed and replaced with fresh medium with differential concentrations of glucose (25, 5, 1 mM). Cell proliferation was measured every 24 hours, starting from 0 to 78 hours, as described (23). In brief cells were fixed with glutaraldehyde, stained with methylene blue and quantified at 650 nm using microplate reader (Molecular Devices).

Immuno Precipitation

Co-Immunoprecipitation (Co-IP) was performed using the kit, procured from Thermo Scientific. In brief, the antibodies were cross-linked with the amine activated Agarose A beads, and incubated with cell lysates at 4°C overnight. Protein complex bound with the beads were washed thrice with lysis buffer and then eluted by adding elution buffer. The resultant IP products and inputs were subjected to LC/MS analysis.

LC-MS Studies

We Immunoprecipitated Myc-tagged, PKM1 or PKM2 from the lysates of stable H1299 cells grown in glucose deplete condition and were subjected to in-solution Trypsin digestion. In brief, the eluted proteins were mixed with surfactant RapiGest (Waters) and were reduced with DTT and alkylated, using iodoacetamide. Samples were digested with 2µg of sequencing grade trypsin gold (Promega) at 37°C overnight and subjected to LC/MS-MS analysis (LC MS/MS Waters SYNAPT G2 with 2D nano ACQUITY System). Mass spectrum obtained from the LC/MS was analysed using Protein Lynx Global SERVER (PGLS) and the interactome for PKM1 and PKM2 were generated.

Statistical Analysis

Data were represented as mean ± SEM. Level of significance was tested using Student’s t-test or one-way analysis of variance with Dunnett’s

multiple comparisons test or two-way analysis of variance with Tukey’s multiple comparisons test (GraphPad Prism). P<.05 was considered to be statistically significant. Tested significance is displayed in the Fig.’s as * P<0.05; ** P<0.01; *** P<0.001.

Results

Glucose inadequacy influences preferential expression of PKM1 isoform.

H1299 and MCF-7 cells were cultured in glucose free medium and supplemented with varying concentrations (25, 5 & 1mM) of glucose to mimic the standard cell culture, physiological and near glucose deprived conditions, to comprehend the effect of nutritional stress on crucial glycolytic enzymes that govern aerobic glycolysis. Cell lysates assayed for the activity of critical enzymes (HK, PK and LDH) associated with aerobic glycolysis showed a significant increase in the PK activity with decreasing glucose concentrations; whereas the activity of HK and LDH remained unaltered (Fig. 1A). Immunoblots showed a consistent decrease in PKM2 protein expression with unaltered PKM2-Tyr105 phosphorylation and a shift in expression from PKM2 to PKM1 (Fig. 1B), suggesting that the enhanced PK activity apparently resulted due to an increased expression of alternatively spliced PKM1. The results of qRT-PCR in H1299 and MCF-7 cells also depicted the switch towards PKM1, without altering significantly the expression of isoforms of hexokinase (HK1 & HK2) and lactate dehydrogenase (LDHA and LDHB) at RNA level (Fig. 1C).

Appearance of PKM1 isoform at mRNA and protein level in experimental lung (H1299) and breast cancer (MCF-7) cell lines concomitantly with PKM2 contradicted the exclusiveness of the latter in cancer cells as reported earlier (7,28). With this reason, we characterized six cancer cell lines (MCF-7, MDA-MB-231, PC-3, H1299, A549 and HeLa), of four different tissue (Breast, Lung, Prostate and Cervical) origins and compared with two other non-cancerous control (L6 -rat skeletal muscle and HEK293 - Human Embryonic Kidney) cell lines, establishing the presence of PKM1 protein besides PKM2 in most of the studied cancer cell lines (Fig. 2A). This was also

5

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

true for a representative set of type II and III stage sporadic breast tumor samples, depicting their co-expression (Fig. 2B). Recombinant GST tagged PKM1 and PKM2 proteins were used to rule out any cross-reactivity between the two (PKM1 & PKM2) antibodies and validating their specificity (Fig. 2C). Further confirmation of the switching of the expression from PKM2 to PKM1, affecting the PK activity, was shown to be due to the depletion of glucose alone and not glutamine or pyruvate (Fig. 2D and 2E).

The observation of co-existence of both PKM1 and PKM2 in cell lines and tumor samples necessitated correlating the observed changes to the oligomeric (dimer: tetramer, using glycerol gradient density ultracentrifugation) status of the two isoforms, which showed a significant shift in response to the hypoglycemic culture conditions (Fig. 3A). H1299 cells cultured in 25mM of glucose exhibited predominantly dimeric form of PKM, and showed a shift towards tetramer peak under lower glucose concentrations (5mM – 1mM). Further analysis, of the glycerol gradient fractions in the immunoblots with PKM1 and PKM2 specific antibodies, revealed that under insufficient glucose presence (5-1mM glucose) there was a drop in dimer and tetramer PKM2 and an enrichment of PKM1 in tetrameric fractions (Fig. 3B), confirming that the shift in tetramer: dimer ratio was solely due to a switch favoring PKM1 expression.

Glucose insufficiency triggers LKB1 dependent AMPK activation to regulate the isoform switch from PKM2 to PKM1.

The involvement of AMP-activated protein kinase, known for preserving cellular energetic homeostasis upon sensing nutritional stress (16), in regulating PKM isoform switch under experimental conditions of glucose depletion, was observed by the activation of AMP-Kinase with simultaneous expression switch toward PKM1 (Fig. 4A). This was shown in cells cultured under declining concentrations of glucose followed by immunoblotting to detect PKM1, PKM2, AMPK, p-AMPK (T172) and p-ACC (Acetyl CoA carboxylase) (Ser79). The p-ACC (Ser79) profile served as a marker to show the degree of AMPK activation. To validate if AMPK played a role in

PKM isoform switch, we adopted two approaches- i) pharmacological activation or inhibition of AMPK and (ii) AMPKa1/2 knock down in cells grown under glucose insufficient conditions. A concentration gradient of AICAR (5-Aminoimidazole-4-carboxamide ribonucleotide), an agonist of AMPK, and other AMPK inducers Metformin and Phenformin under enriched glucose condition, replicated the observation of the PKM isoform switch (Fig. 4B and 4C), as witnessed under insufficient-glucose experimental conditions. Pretreatment with Compound C, an inhibitor of AMPK kinase, followed by glucose depletion (1mM) or AICAR treatment, impaired the function of AMPK, as assessed by p-ACC (ser79) and abrogated PKM2 to PKM1 isoform switch (Fig. 4D). Finally, silencing of AMPKa1/2 (catalytic alpha 1 and 2 subunits) in H1299 cells using shRNA prevented the switch towards PKM1 under low-glucose condition (Fig. 4E), validating the role of AMPK in the process. Replicated experiments in MCF-7 cells also showed similar results (Data Not Shown).

To study the significance of functional LKB1, upstream to AMPK, to regulate PKM isoform switch; the cell lines, A549 (29) and MDA-MB-231 (30), lacking LKB1 were transfected with vector or Myc-tagged-LKB1 (LKB1-Myc) constructs and subjected to AMPK activation by glucose depletion (1mM). Both cell lines harboring LKB1-Myc showed switch towards PKM1, not observed in vector transfected controls (Fig. 5A). Further, we explored if the protein kinase activity of AMPK (i.e., Threonine 172 phosphorylation, pT172, mediated by LKB1) was critical to activate PKM isoform switch. A549 cells cultured with enriched glucose (25mM) and transiently transfected with T172D AMPKa2 constitutively active (CA) mutant (AMPKa2 T172D-Myc), showed PKM2 to PKM1 switch with increased Ser79 ACC phosphorylation. However, A549 cells transfected with vector or wild type AMPKa2 (AMPKa2-Myc) failed to show the switch (Fig. 5B), suggesting the involvement of LKB1-AMPK axis in regulation of PKM isoform switch.

We also examined if the activated AMPK interacted and phosphorylated PKM isoforms to influence its glycolytic activity. We predicted

6

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

computationally two sites (Threonine 45 and 93), both in M1 and M2 isoforms of PK; and one additional site (Serine 403) exclusive to PKM1, as consensus phosphorylation sites of AMPK with moderate stringency (Fig. 6A). For subsequent validation, we generated H1299 stable cells, over expressing either Myc-His-tagged- PKM1 (PKM1-Myc-His) or PKM2 (PKM2-Myc-His) (Fig. 6B). These isoforms when independently immunoprecipitated (IP) from whole cell lysates of H1299 stable cells cultured under depleted glucose (1mM) condition and subjected to LC-MS/MS analysis, revealed neither a physical interaction between PKM isoforms and AMPK, nor phosphorylation on the predicted sites in either of the PKM isoforms, in accordance with the PKM1 and PKM2 interactome (Data not shown) and the coverage map (Fig. 6C).

AMPK regulates alternative splicing of PKM isoform towards PKM1.

Given the observation of LKB1-AMPK pathway driven preferential expression of PKM1 isoform, we investigated the link between LKB1-AMPK pathway and the PKM1- expression switch. Since c-Myc, a downstream effector of MTOR (Mammalian Target of Rapamycin) which controls transcription activation of hnRNPs (Heterogeneous nuclear ribonucleoproteins) and results in PKM1 repression and preferential expression of PKM2 (9,31,32), it was pertinent first to investigate if the activated LKB1-AMPK pathway antagonized MTOR-c-Myc-hnRNPs axis to regulate alternative splice switch towards PKM1 isoform under insufficient glucose condition. We observed that glucose deprived (1mM) cells resulted in the switch in expression towards PKM1, repressing MTOR signaling pathway and the expression of its downstream effector, c-Myc, an observation which was similar to the cells cultured under enriched glucose (25mM) condition in presence of the MTOR inhibitor, Rapamycin (Fig. 7A). Likewise, after examining the transcripts of c-Myc, hnRNPs (hnRNPA1, hnRNPA2 and PTBP1) and PKM isoforms, it was observed that both glucose deprivation (1mM) and the treatment of Rapamycin down-regulated c-Myc and its transcriptional activation of hnRNPs, favoring PKM2 to PKM1 switch (Fig. 7B).

Further, to establish the involvement of AMP-Kinase in MTOR-c-Myc-hnRNPs axis controlling the PKM isoform expression status, stably transduced H1299 cells with Vector (pLKO.1) alone or shAMPKa1/2 was studied. Glucose-depletion-induced AMPK stimulation in vector transfected cells exhibited a significant downregulation of c-Myc at protein level, conceivably in MTOR dependent manner, as indicated by a decrease in phosphorylation of p70 ribosomal protein S6 kinase (RPS6K) and ribosomal protein S6 (RPS6) in addition to PKM1 switch (Fig. 7C). Whereas, under similar experimental conditions knockdown of AMPKa1/2, reversed the phenomenon and prevented the inhibition of MTOR pathway, c-Myc expression and PKM1 switch. Accordingly, examining the transcripts of c-Myc, hnRNPs (hnRNPA1, hnRNPA2 and PTBP1) and PKM isoforms in glucose depleted (1mM) H1299 cells revealed that AMPK activation down-regulates c-Myc and hnRNPs, thus favoring PKM2 to PKM1 switch. This switch in expression was prevented by AMPKα1/2 silencing (Fig. 7D). To examine whether the ectopic expression of c-Myc was sufficient to prevent AMPK induced PKM switch, we over-expressed c-Myc and detected that c-Myc expression did not inhibit AMPK mediated switch towards PKM1, under glucose insufficient (1mM) culture conditions (Fig. 7E). Remarkably, neither c-Myc over expression nor knockdown altered the basal level expression of PKM isoforms in H1299 cells grown in glucose replete (25mM) medium (Fig. 7F). Together, these results provided an evidence of the involvement of AMPK in regulating alternative splicing towards PKM1 by antagonizing the MTOR signaling and down regulating the expression of hnRNPA1, hnRNPA2 and PTBP1, in a c-Myc independent manner.

AMPK driven PKM1 switch influences glycolysis and ATP production.

Since AMPK is known to be involved in maintaining energy homeostasis (15,16), we hypothesized that the AMPK driven PKM switch (AMPK-PKM1 axis) would influence cellular glucose metabolism and the energy status of cancer cells in culture; and generated stable knockdowns of PKM1 and PKM2 in H1299 cells (Fig. 8A) (in addition to stable knock downs of

7

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

AMPKa1/2 - as discussed in our previous results). The features of aerobic glycolysis and energy status when measured, there was a significant increase in glucose uptake, intracellular level of pyruvate, lactate release and most interestingly ATP levels, in cells transduced with vector alone (pLKO.1) and cultured under glucose insufficient (1mM) condition when compared to those cultured with enriched (25mM) glucose (Fig. 8B-E). However, these features were markedly reduced upon either AMPKa1/2 or PKM1 knock down in cells cultured under insufficient glucose condition (Fig. 8B-E), unraveling the crucial role of AMPK-PKM1 axis in enhancing aerobic glycolysis to augment ATP status in glucose deprived cancer cells. Expectedly, knock down of PKM2 reduced the glucose uptake, intracellular pyruvate and lactate release. However, ATP production remained unaltered in cells cultured with enriched (25mM) glucose (Fig. 8B-E). Whereas, under glucose depletion, PKM2 knock-down did not affect metabolism, probably because of the replacement of PKM1 isoform upon AMPK induction.

AMPK-PKM1 axis augmenting ATP production in glucose depleted cancer cells (Fig. 8E), prompted us to examine the glycolytic and OXPHOS contribution to the ATP production in nutritionally deprived cancer cells in culture. We examined glycolytic ATP levels in stable H1299 cells transduced with vector (shPKM1) or shPKM2, and cultured under enriched (25mM) and insufficient (1mM G) glucose conditions, followed by treatment with Oligomycin (a specific inhibitor of mitochondrial ATP synthase). It was clear in Oligomycin treated experiments under insufficient glucose conditions with knock-down expression of PKM2 that PKM1 contributed to a substantial increase in ATP production through glycolysis (Fig. 8F). The large drop in ATP upon PKM1 knockdown, compared to vector and shPKM2, substantiated this further (Fig. 8F). However, results also suggest that basal OXPHOS always operates in the cells regardless of glucose and PKM status (Fig. 8F). Together, our data suggests that under glucose insufficient conditions, AMPK driven PKM2 to PKM1 switch in cancer cells promotes aerobic glycolysis to enhance the ATP production.

PKM1 is essential for cancer cell survival under glucose depletion.

We next assessed the preferential requirement of PKM isoforms (M1 or M2) under enriched (25mM) and insufficient (1mM) glucose conditions in H1299 cells, stably transduced with vector (LKO.1), shPKM1 and shPKM2. As is known, knockdown of PKM2 retarded the cell proliferation rate (P< 0.01; 48 hrs, P< 0.01; vector vs. shPKM2 at 72 hrs), but PKM1 knockdown did not affect the proliferation of cells, under 25 mM glucose (Fig. 9A). However, glucose depleted cells showed remarkable sensitivity to PKM1 silencing (P< 0.001; vector vs. shPKM1 and P<0.01; shPKM1 vs. shPKM2 at 72 hrs) (Fig. 9B). Though, the viability of the cells differed markedly in prolonged glucose depletion (1mM), resulting in cell death, which did not differ between PKM2 knockdown cells and the vector (Fig. 9B). Further, immunobloting from the lysates of stable H1299 cells transduced with vector, shPKM1 and shPKM2 and cultured under limited glucose (1mM), with apoptosis markers revealed the cleavage of Caspase-3 and PARP only in PKM1 knockdown cells. Cleavage began at 24 hours and stabilized rapidly at 48 hours (Fig. 9C). To further corroborate our in vitro observations, we performed Immunohistochemistry (IHC) on sporadic breast tumor tissue serial sections (4 µm) to examine the spatial distribution of PKM1 and PKM2. PKM1 was found to be confined to a subset of cells which also showed the presence of pACC (Ser79) (a marker we used to locate hypoglycemic tumor microenvironment). PKM2 distribution, however, was spatially different from pACC (Ser79) positive regions, though with an appreciative co-expression of both PKM1 and PKM2 (Fig. 10A). Together, these results demonstrate that the expression, an AMPK dependent PKM switch towards PKM1 is necessary for cancer cells to survive in glucose insufficient conditions (Fig. 10B).

Discussion

A key challenge in understanding tumor development and its progression is to comprehend how rapidly dividing cancer cells in solid tumors overcome the checkpoints that maintain tissue homeostasis, including that of nutritional stress.

8

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

Unraveling the signaling pathways and their downstream adaptive metabolic phenotype that benefit cancer cells to survive the fluctuations in available nutrient conditions is of interest. Our study elucidates a novel signaling axis of LKB1-AMPK-PKM1 with metabolic advantage that could nurse cancer cells to overcome the unfavorable condition of nutritional stress.

We demonstrate an integral role of LKB1-AMPK signaling pathway in regulating the switch from M2 to M1 isoform of the glycolytic enzyme, pyruvate kinase, in response to glucose depletion in cancer cells (Fig. 1, 3 and 4), which did not alter the expression of isoforms of hexokinase (HK1 and HK2) and lactate dehydrogenase (LDHA and LDHB). The role played by AMPK in PKM isoform switch towards PKM1 under hypoglycemic condition was verified by the use of AMPK activator (AICAR), inhibitor Compound C, and finally by AMPKa1/2 knockdown. The cell lines lacking LKB1 and the experiments conducted with a constitutively active mutant AMPK (T172D) further demonstrated the importance of LKB1 upstream to AMPK in regulating PKM1 switch (Fig. 4 and 5). This work reveals that AMPK regulates preferential expression of PKM1 by antagonizing MTOR signaling and down-regulating splicing repressor-hnRNPs (hnRNPA1, hnRNPA2 and PTBP1) to execute alternative splicing in favour of PKM1 (Fig. 7) in a c-Myc independent manner. The latter was confirmed by ectopic expression of c-Myc in glucose deprived cancer cells (H1299 and MCF-7) which failed to inhibit AMPK induced PKM1 switch. Further, the knockdown as well as over-expression of c-Myc in glucose enriched cell culture condition had no effect on the expression of both PKM isoforms, endorsing that hnRNPs regulate PKM alternative splicing without the involvement of c-Myc, an observation contrary to what has been reported earlier (9,32). Besides unraveling a novel LKB1-AMPK regulation upstream of MTOR signaling pathway and their effector hnRNPs in influencing the alternative splicing of PKM, a novel contribution of this work, we also largely support and revalidate some of the previous works of David et al.; Clower et al and Sun et al. (9,31,32). However, our work evidently disagrees with the involvement of c-Myc in regulation of the alternative splicing of PKM

isoforms as documented earlier (9,32). Our results also highlight the missing link that connects MTOR signaling pathway with the expression regulation of hnRNPs and widens the scope for building future perspective.

A preferential expression of PKM2 over other tissue-specific PK isoforms has been proposed as one of the metabolic hallmarks of cancer (2), where PKM2 expression serves a pivotal role in cancer growth (7,11) by governing aerobic glycolysis and performing non-metabolic role of; co-transcriptional activation (33,34), protein kinase function (35,36) and chromosomal segregation (37). In addition, the expression of PKM2 in tumors has been considered to be a strategic step, since PKM2 retains an intrinsic feature of interchangeable oligomeric states of catalytically active tetramer and an inactive dimer in response to numerous extracellular and intracellular factors (38) The dynamic oscillation between PKM2 dimer : tetramer is believed to replenish the biosynthetic and bioenergetic needs of cancer cells by redirecting the flux of glycolysis towards anabolism or catabolism. However, its role in dealing with cell survival under varying oxygen and nutrient conditions in tumor microenvironment is not clear. We report here the co-expression of M1 and M2 isoforms of pyruvate kinase, in a representative set of sporadic breast cancer tissues and several cancer cell lines (Fig. 2) and elucidate their role by finding how nutrient status of insufficient glucose under culture conditions results in a preferential switch of expression towards the PKM1 isoform. The preferential expression of PKM1 and its increased activity in glucose depleted cancer cells results in enhanced aerobic glycolysis, a conclusion drawn from our observation of increased glucose uptake, pyruvate level and lactate release rate, to maintain energy (ATP) homeostasis (Fig. 8). We conjecture that this switch to PKM1 and increased aerobic glycolysis helps cancer cells survive the nutrient-deprived conditions. In other words, this mechanism may help cancer cells to generate glycolytic ATPs for survival and remain dormant. Abolishment of either AMPK or PKM1 expression by knock-down resulted in impaired aerobic glycolysis, disturbed energy homeostasis (Fig. 8). Further, knock-down of PKM1 reduced survival of glucose depleted cancer cells and induced

9

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

apoptosis (Fig. 9). Observations of impaired aerobic glycolysis and reduced cell proliferation were also made in PKM2 knock-down cells under glucose-enriched (25mM) conditions (Fig. 8 and 9), consistent with the role known (7,23,39) and observed by us (26,40) of PKM2 in literature.

We suggest that the switch from PKM2 to PKM1 expression apparently shifts the flux of glycolysis from biosynthesis to bioenergetics, essential to cell survival under hypoglycemic conditions. Notably, despite the association of AMPK pathway with mitochondrial biogenesis (41), we find that the LKB1-AMPK-PKM1 axis in glucose depleted cancer cells, shows reliance on aerobic glycolysis for ATP generation, as evidenced from our observation of knock-down expression of PKM2 under insufficient glucose conditions and presence of Oligomycin (a selective mitochondrial ATP synthase inhibitor), where PKM1 contributed to a substantial increase in ATP production through glycolysis (Fig. 8F). The large drop in ATP upon PKM1 knockdown, compared to vector and shPKM2, substantiated this further (Fig. 8F). It may appear enigmatic, but contemporary studies support our observation of accelerated aerobic glycolysis associated with nutrient depleted cancer cells (41,42). Consistent with our results, Wu et al. (42) also showed that AMPK inhibits the entry of pyruvate in mitochondria in nutrient depleted cancer cells. Dependency of cancer cells on aerobic glycolysis during hypoglycemic conditions appears pertinent as some of the pockets or deep-seated (core) tumor cells perceive recurrent hypoglycemia along with hypoxia, due to poor vasculature, which makes such cells prefer aerobic glycolysis over mitochondrial oxidative phosphorylation.

Our IHC results support this notion, where immunostaining of the sporadic breast tumor tissue serial sections showed the presence of pACC (pACC-Ser79) (a marker we used to locate hypoglycemic tumor microenvironment) and PKM1 in an overlapping subset of cells within a region positive for both. Areas with low to no PKM1 positive cells showed relatively higher expression of PKM2; confined to distinct region, apparently representing nutritionally rich and probably rich vascularized area of the tumour (Fig. 10A). These results largely support the notion of

recent work of Israelsen et al. (43) where cancer cells in tumor possess differential requirement in expressing PKM isoforms. However, contrary to their conclusion of PKM1 takeover after complete abolition of the expression of PKM2 (PKM2 knockouts), In our study, we demonstrate that cancer cells of tumor retain an integral bioenergetic sensing mechanism (LKB1-AMPK), which monitors poorly vascularized, nutritionally deprived, regions and express the M1 isoform of pyruvate kinase to augment rapid ATP synthesis through aerobic glycolysis for their survival.

PKM2 has been validated as a potential candidate of glycolysis for therapeutic interventions to target cancer metabolism (38,44). However, in the background of our observations of co-existence of the two isoforms of PK and the preferential expression of PKM1 isoform under nutrient deficient conditions, a model (Fig. 10B) is proposed of the two PKM isoforms in tumors. We suggest where cells nourished with nutrients in the periphery of tumor and in proximity to blood vessels express PKM2, predominantly in a dimeric state, with concomitant low expression of PKM1 tetramer, these attain a proliferation advantage. Whereas, the cells of poorly vascularized tumor regions (hypoglycemic/hypoxic centers), starved of nutrients, switch to express PKM1 isoform in an AMPK dependent manner to provide the cancer cells a survival advantage (Fig. 10B). Thus, the two processes, one driven by PKM2 dimer/tetramer state of attending to the biosynthetic/bioenergetic needs and the other determined by PKM1 tetramer to attend to the bioenergetic requirement alone, contribute to the metabolic heterogeneity of real-life tumor cells. We therefore propose on the basis of our observations that considering both PKM2 and PKM1 as simultaneous targets would be essential to contain tumor promotion and progression.

Acknowledgements This work was partially supported by University Grant Commission (UGC), Government of India, the grant (F. No 17-4/2001 (NS/PE)) provided to National Center for Applied Human Genetics (NCAHG) was upto March 2012. P.G. acknowledges UGC for providing meritorious science research Fellowship.

10

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

Conflict of interest

The authors declare that they have no conflicts of interest with the contents of this article.

Author Contributions

P.G. and R.N.K.B designed the work, P.G. performed research, R.K.S and M.A.I contributed new reagents/analytical tools; P.G., A.B.T, M.A.I and R.N.K.B analysed the data and wrote the paper.

References

1. Hanahan, D., and Weinberg, R. A. (2000) The hallmarks of cancer. Cell 100, 57-702. Cairns, R. A., Harris, I. S., and Mak, T. W. (2011) Regulation of cancer cell metabolism. Nature

reviews. Cancer 11, 85-953. Warburg, O. (1956) On the origin of cancer cells. Science 123, 309-3144. Vander Heiden, M. G., Cantley, L. C., and Thompson, C. B. (2009) Understanding the Warburg

effect: the metabolic requirements of cell proliferation. Science 324, 1029-10335. DeBerardinis, R. J., Lum, J. J., Hatzivassiliou, G., and Thompson, C. B. (2008) The biology of

cancer: metabolic reprogramming fuels cell growth and proliferation. Cell Metab 7, 11-206. Patra, K. C., Wang, Q., Bhaskar, P. T., Miller, L., Wang, Z., Wheaton, W., Chandel, N., Laakso,

M., Muller, W. J., Allen, E. L., Jha, A. K., Smolen, G. A., Clasquin, M. F., Robey, R. B., andHay, N. (2013) Hexokinase 2 is required for tumor initiation and maintenance and its systemicdeletion is therapeutic in mouse models of cancer. Cancer cell 24, 213-228

7. Christofk, H. R., Vander Heiden, M. G., Harris, M. H., Ramanathan, A., Gerszten, R. E., Wei, R.,Fleming, M. D., Schreiber, S. L., and Cantley, L. C. (2008) The M2 splice isoform of pyruvatekinase is important for cancer metabolism and tumour growth. Nature 452, 230-233

8. Fantin, V. R., St-Pierre, J., and Leder, P. (2006) Attenuation of LDH-A expression uncovers alink between glycolysis, mitochondrial physiology, and tumor maintenance. Cancer cell 9, 425-434

9. David, C. J., Chen, M., Assanah, M., Canoll, P., and Manley, J. L. (2010) HnRNP proteinscontrolled by c-Myc deregulate pyruvate kinase mRNA splicing in cancer. Nature 463, 364-368

10. Roberts, D. J., Tan-Sah, V. P., Smith, J. M., and Miyamoto, S. (2013) Akt phosphorylates HK-IIat Thr-473 and increases mitochondrial HK-II association to protect cardiomyocytes. The Journalof biological chemistry 288, 23798-23806

11. Yang, W., Zheng, Y., Xia, Y., Ji, H., Chen, X., Guo, F., Lyssiotis, C. A., Aldape, K., Cantley, L.C., and Lu, Z. (2012) ERK1/2-dependent phosphorylation and nuclear translocation of PKM2promotes the Warburg effect. Nature cell biology 14, 1295-1304

12. Hitosugi, T., Kang, S., Vander Heiden, M. G., Chung, T. W., Elf, S., Lythgoe, K., Dong, S.,Lonial, S., Wang, X., Chen, G. Z., Xie, J., Gu, T. L., Polakiewicz, R. D., Roesel, J. L., Boggon,T. J., Khuri, F. R., Gilliland, D. G., Cantley, L. C., Kaufman, J., and Chen, J. (2009) Tyrosinephosphorylation inhibits PKM2 to promote the Warburg effect and tumor growth. Sciencesignaling 2, ra73

13. Fan, J., Hitosugi, T., Chung, T. W., Xie, J., Ge, Q., Gu, T. L., Polakiewicz, R. D., Chen, G. Z.,Boggon, T. J., Lonial, S., Khuri, F. R., Kang, S., and Chen, J. (2011) Tyrosine phosphorylation oflactate dehydrogenase A is important for NADH/NAD(+) redox homeostasis in cancer cells.Molecular and cellular biology 31, 4938-4950

14. Jones, R. G., and Thompson, C. B. (2009) Tumor suppressors and cell metabolism: a recipe forcancer growth. Genes & development 23, 537-548

15. Hardie, D. G., Ross, F. A., and Hawley, S. A. (2012) AMPK: a nutrient and energy sensor thatmaintains energy homeostasis. Nature reviews. Molecular cell biology 13, 251-262

11

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

16. Shackelford, D. B., and Shaw, R. J. (2009) The LKB1-AMPK pathway: metabolism and growthcontrol in tumour suppression. Nature reviews. Cancer 9, 563-575

17. Kato, K., Ogura, T., Kishimoto, A., Minegishi, Y., Nakajima, N., Miyazaki, M., and Esumi, H.(2002) Critical roles of AMP-activated protein kinase in constitutive tolerance of cancer cells tonutrient deprivation and tumor formation. Oncogene 21, 6082-6090

18. Jeon, S. M., Chandel, N. S., and Hay, N. (2012) AMPK regulates NADPH homeostasis topromote tumour cell survival during energy stress. Nature 485, 661-665

19. Bardeesy, N., Sinha, M., Hezel, A. F., Signoretti, S., Hathaway, N. A., Sharpless, N. E., Loda,M., Carrasco, D. R., and DePinho, R. A. (2002) Loss of the Lkb1 tumour suppressor provokesintestinal polyposis but resistance to transformation. Nature 419, 162-167

20. Liang, J., and Mills, G. B. (2013) AMPK: a contextual oncogene or tumor suppressor? Cancerresearch 73, 2929-2935

21. Pal, R., Gochhait, S., Chattopadhyay, S., Gupta, P., Prakash, N., Agarwal, G., Chaturvedi, A.,Husain, N., Husain, S. A., and Bamezai, R. N. (2011) Functional implication of TRAIL -716 C/Tpromoter polymorphism on its in vitro and in vivo expression and the susceptibility to sporadicbreast tumor. Breast cancer research and treatment 126, 333-343

22. Pal, R., Srivastava, N., Chopra, R., Gochhait, S., Gupta, P., Prakash, N., Agarwal, G., andBamezai, R. N. (2010) Investigation of DNA damage response and apoptotic gene methylationpattern in sporadic breast tumors using high throughput quantitative DNA methylation analysistechnology. Molecular cancer 9, 303

23. Cortes-Cros, M., Hemmerlin, C., Ferretti, S., Zhang, J., Gounarides, J. S., Yin, H., Muller, A.,Haberkorn, A., Chene, P., Sellers, W. R., and Hofmann, F. (2013) M2 isoform of pyruvate kinaseis dispensable for tumor maintenance and growth. Proc Natl Acad Sci U S A 110, 489-494

24. Mohseni, M., Sun, J., Lau, A., Curtis, S., Goldsmith, J., Fox, V. L., Wei, C., Frazier, M., Samson,O., Wong, K. K., Kim, C., and Camargo, F. D. (2014) A genetic screen identifies an LKB1-MARK signalling axis controlling the Hippo-YAP pathway. Nature cell biology 16, 108-117

25. Marini, C., Salani, B., Massollo, M., Amaro, A., Esposito, A. I., Orengo, A. M., Capitanio, S.,Emionite, L., Riondato, M., Bottoni, G., Massara, C., Boccardo, S., Fabbi, M., Campi, C.,Ravera, S., Angelini, G., Morbelli, S., Cilli, M., Cordera, R., Truini, M., Maggi, D., Pfeffer, U.,and Sambuceti, G. (2013) Direct inhibition of hexokinase activity by metformin at least partiallyimpairs glucose metabolism and tumor growth in experimental breast cancer. Cell cycle 12, 3490-3499

26. Iqbal, M. A., Siddiqui, F. A., Gupta, V., Chattopadhyay, S., Gopinath, P., Kumar, B., Manvati, S.,Chaman, N., and Bamezai, R. N. (2013) Insulin enhances metabolic capacities of cancer cells bydual regulation of glycolytic enzyme pyruvate kinase M2. Molecular cancer 12, 72

27. Gupta, V., Kalaiarasan, P., Faheem, M., Singh, N., Iqbal, M. A., and Bamezai, R. N. (2010)Dominant negative mutations affect oligomerization of human pyruvate kinase M2 isozyme andpromote cellular growth and polyploidy. The Journal of biological chemistry 285, 16864-16873

28. Mazurek, S., Boschek, C. B., Hugo, F., and Eigenbrodt, E. (2005) Pyruvate kinase type M2 andits role in tumor growth and spreading. Seminars in cancer biology 15, 300-308

29. Ji, H., Ramsey, M. R., Hayes, D. N., Fan, C., McNamara, K., Kozlowski, P., Torrice, C., Wu, M.C., Shimamura, T., Perera, S. A., Liang, M. C., Cai, D., Naumov, G. N., Bao, L., Contreras, C.M., Li, D., Chen, L., Krishnamurthy, J., Koivunen, J., Chirieac, L. R., Padera, R. F., Bronson, R.T., Lindeman, N. I., Christiani, D. C., Lin, X., Shapiro, G. I., Janne, P. A., Johnson, B. E.,Meyerson, M., Kwiatkowski, D. J., Castrillon, D. H., Bardeesy, N., Sharpless, N. E., and Wong,K. K. (2007) LKB1 modulates lung cancer differentiation and metastasis. Nature 448, 807-810

30. Shen, Z., Wen, X. F., Lan, F., Shen, Z. Z., and Shao, Z. M. (2002) The tumor suppressor geneLKB1 is associated with prognosis in human breast carcinoma. Clinical cancer research : anofficial journal of the American Association for Cancer Research 8, 2085-2090

12

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

31. Clower, C. V., Chatterjee, D., Wang, Z., Cantley, L. C., Vander Heiden, M. G., and Krainer, A.R. (2010) The alternative splicing repressors hnRNP A1/A2 and PTB influence pyruvate kinaseisoform expression and cell metabolism. Proc Natl Acad Sci U S A 107, 1894-1899

32. Sun, Q., Chen, X., Ma, J., Peng, H., Wang, F., Zha, X., Wang, Y., Jing, Y., Yang, H., Chen, R.,Chang, L., Zhang, Y., Goto, J., Onda, H., Chen, T., Wang, M. R., Lu, Y., You, H., Kwiatkowski,D., and Zhang, H. (2011) Mammalian target of rapamycin up-regulation of pyruvate kinaseisoenzyme type M2 is critical for aerobic glycolysis and tumor growth. Proc Natl Acad Sci U S A108, 4129-4134

33. Luo, W., Hu, H., Chang, R., Zhong, J., Knabel, M., O'Meally, R., Cole, R. N., Pandey, A., andSemenza, G. L. (2011) Pyruvate kinase M2 is a PHD3-stimulated coactivator for hypoxia-inducible factor 1. Cell 145, 732-744

34. Yang, W., Xia, Y., Ji, H., Zheng, Y., Liang, J., Huang, W., Gao, X., Aldape, K., and Lu, Z.(2011) Nuclear PKM2 regulates beta-catenin transactivation upon EGFR activation. Nature 480,118-122

35. Gao, X., Wang, H., Yang, J. J., Liu, X., and Liu, Z. R. (2012) Pyruvate kinase M2 regulates genetranscription by acting as a protein kinase. Molecular cell 45, 598-609

36. Yang, W., Xia, Y., Hawke, D., Li, X., Liang, J., Xing, D., Aldape, K., Hunter, T., Alfred Yung,W. K., and Lu, Z. (2012) PKM2 phosphorylates histone H3 and promotes gene transcription andtumorigenesis. Cell 150, 685-696

37. Jiang, Y., Li, X., Yang, W., Hawke, D. H., Zheng, Y., Xia, Y., Aldape, K., Wei, C., Guo, F.,Chen, Y., and Lu, Z. (2014) PKM2 regulates chromosome segregation and mitosis progression oftumor cells. Molecular cell 53, 75-87

38. Iqbal, M. A., Gupta, V., Gopinath, P., Mazurek, S., and Bamezai, R. N. (2014) Pyruvate kinaseM2 and cancer: an updated assessment. FEBS letters 588, 2685-2692

39. Goldberg, M. S., and Sharp, P. A. (2012) Pyruvate kinase M2-specific siRNA induces apoptosisand tumor regression. The Journal of experimental medicine 209, 217-224

40. Chaman, N., Iqbal, M. A., Siddiqui, F. A., Gopinath, P., and Bamezai, R. N. (2015) ERK2-Pyruvate Kinase Axis Permits Phorbol 12-Myristate 13-Acetate-induced MegakaryocyteDifferentiation in K562 Cells. The Journal of biological chemistry 290, 23803-23815

41. Reznick, R. M., and Shulman, G. I. (2006) The role of AMP-activated protein kinase inmitochondrial biogenesis. The Journal of physiology 574, 33-39

42. Wu, C. A., Chao, Y., Shiah, S. G., and Lin, W. W. (2013) Nutrient deprivation induces theWarburg effect through ROS/AMPK-dependent activation of pyruvate dehydrogenase kinase.Biochimica et biophysica acta 1833, 1147-1156

43. Israelsen, W. J., Dayton, T. L., Davidson, S. M., Fiske, B. P., Hosios, A. M., Bellinger, G., Li, J.,Yu, Y., Sasaki, M., Horner, J. W., Burga, L. N., Xie, J., Jurczak, M. J., DePinho, R. A., Clish, C.B., Jacks, T., Kibbey, R. G., Wulf, G. M., Di Vizio, D., Mills, G. B., Cantley, L. C., and VanderHeiden, M. G. (2013) PKM2 isoform-specific deletion reveals a differential requirement forpyruvate kinase in tumor cells. Cell 155, 397-409

44. Gupta, V., Wellen, K. E., Mazurek, S., and Bamezai, R. N. (2014) Pyruvate kinase M2:regulatory circuits and potential for therapeutic intervention. Current pharmaceutical design 20,2595-2606

Figure Legends

Figure Legends

Figure 1. Effect of glucose depletion on glycolytic pathway enzymes and the pyruvate kinase isoform switch. (A) Relative enzyme activity of glycolytic enzymes: Hexokinase, HK; Pyruvate Kinase,

13

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

PK; and Lactate Dehydrogenase, LDH; in the protein lysates of H1299 and MCF-7 cells cultured in glucose free medium, supplemented with (25mM, 5mM, 1mM) glucose for a period of 8 hours; with statistical analysis (where n=3; mean ± SEM), performed using one-way ANOVA with Dunnett’s multiple comparisons test (GraphPad Prism); **P < 0.01, ***P < 0.001. (B) Immunoblot of PKM2, p-PKM2 (Try105) and PKM1 from protein lysates of cells as in ‘A’ with relative expression levels (signals) quantified using ImageJ and normalized to β-Actin. (C) qRT-PCR analysis to show the relative expression of genes involved in glucose metabolism (HK1, HK2, PKM1, PKM2, LDH A, LDHB) from H1299 (Left) and MCF-7 (Right) cells, cultured in glucose free medium, supplemented with (25 mM, 5mM, 1mM) glucose for a period of 8 hours. The bars represent the fold-change (Log10 relative quantification) after normalizing with the control group (25 mM Glucose), with statistical analysis (n=3; mean ± SEM), performed using student’s t-test (GraphPad Prism). *P < 0.05, **P < 0.01, ***P < 0.001.

Figure 2. Co-expression of PKM1 and PKM2 in cancer cell lines and tumor cells of Stage II and III. (A) Expression status of PKM1 and PKM2 in six cancer cell lines of four different tissue origins and two non-cancerous control cell lines. (B) Differential co-expression of PKM1 and PKM2 isoforms of pyruvate kinase in stage II and stage III sporadic breast tumors in comparison to their normal tissue pair (N – Normal tissue, T – Tumor tissue). (C) Immunoblotting with anti-PKM1 and anti-PKM2 to show the specificity for purified recombinant PKM1 (rGST-PKM1) and PKM2 (rGST-PKM1), C.B.B. (Coomassie Brilliant Blue) stained. (D) Pyruvate kinase enzyme activity in H1299 cell lysates, grown under 1 mM or 25 mM glucose, in presence and absence of 4mM glutamine and / or 1mM pyruvate for a period of 8 hours; with statistical analysis (where n=3; mean ± SEM), performed using one-way ANOVA with Dunnett’s multiple comparisons test (GraphPad Prism). (E) Immunoblot of PKM1 and PKM2 to show their expression in protein lysates of H1299 cells as in ‘D’.

Figure 3. Glucose inadequacy shifts the tetramer: dimer ratio of PKM. (A) Dimeric and Tetrameric peaks of PK, resolved by spectrometric PK enzyme assay of the fractions collected from glycerol density gradient ultracentrifugation, loaded with protein lysates of H1299 cells cultured in glucose free medium, supplemented with (25mM, 5mM, 1mM) glucose for a period of 8 hours. (B) Immunoblot of PKM1 and PKM2 to qualitatively and quantitatively measure the distribution of PKM1 and PKM2 in dimer and tetramer peaks, using glycerol density gradient fractions as in ‘A’.

Figure 4. PKM1 expression in hypoglycemic conditions is regulated by AMPK pathway. (A-E) Immunoblots of: (A) p-AMPKa (T172), AMPKa, p-ACC (S79), ACC, PKM1 and PKM2 from the protein lysates of H1299 cells, grown in glucose free medium, supplemented with relative lowering concentration of glucose (25, 10, 5, 1, .1, 0 mM) for a period of 8 hours to show AMPK activation and PKM expression status. (B) the protein lysates of H1299 cells cultured in enriched glucose (25mM) medium and treated with a gradient of AICAR (0, .5, .1, .2, .5, 1 mM) for 8 hours, using the indicated antibodies, to show AMPK activation and PKM isoform expression status. (C) the protein lysates of H1299 cells, cultured in enriched glucose (25mM) medium and treated with Metformin (2mM) or Phenformin (2mM) for 8 hours, using the indicated antibodies to show the AMPK activation and PKM isoform expression status. (D) p-ACC (S79), PKM1 and PKM2 to show AMPK activation and PKM expression status in protein lysates of H1299 cells, pretreated with or without 10 µM Compound C (inhibitor of AMPK) for 30 minutes and cultured in the presence and absence of glucose or 500µM of AICAR as indicated, for the period of 8 hours. (E) the protein lysates of H1299 cells stably transduced with vector (pLKO.1) or AMPKa1/2 targeting shRNA (shAMPKa1/2), grown under enriched (25mM) or insufficient (1mM) glucose medium for 8 hours, using the indicated antibodies, to show PKM isoform expression in the background of AMPK activation and knock down. The relative expression levels (signals) of PKM1 and PKM2 from the above immunoblots were quantified using ImageJ and normalized to β-Actin.

Figure 5. AMPK requires LKB1 kinase to regulate the pyruvate kinase M2 to M1 isoform switch in cancer cells (A and B) Immunoblots of: (A) the protein lysates of A549 and MDA-MB 231 cells stably

14

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

transfected with empty vector (pcDNA3.1) or Myc-tagged LKB1 (LKB1-Myc), cultured in glucose enriched (25mM) and deplete (1mM) medium, for the period of 8 hours, probed with Myc-tag, PKM1 and PKM2 antibodies, to show overexpression of LKB1-Myc and PKM expression status. (B) the protein lysates of A549 cells transfected with empty vector (pcDNA3.1), Myc-tagged wild type AMPKa2 (AMPKa2-Myc) or AMPKa2 constitutively active T172D mutant (AMPKa2T172D-Myc) and cultured under enriched (25mM) glucose conditions and probed with Myc-tag and pACC (ser79) antibodies to show the expression and functional validation of AMPKa2 Wild type and Mutant; also, showing the results of probing with PKM1 and PKM2 antibodies to show their expression status. The relative expression levels (signals) of PKM1 and PKM2 were quantified using ImageJ and normalized to β-Actin.

Figure 6. Identification and characterization of AMPK phosphorylation sites of M1 and M2 isoforms of pyruvate kinase. (A) The AMPK consensus phosphorylation motif shown in top panel were acquired from Gwinn et al. (2008) where the level of stringency represents the optimal match with motif. The potential AMPK consensus phosphorylation site in both PKM1 & PKM2 (Threonine 45 & 93) and PKM1 alone (Serine 403) were predicted using Scansite3. ACC1, TSC-2 and PFK2 are reference targets, previously recognized phosphorylation site of AMPK. (B) Immunoblots of the protein lysates of H1299 stable cell lines, expressing empty vector, Myc-His-tagged PKM1 (Left) or Myc-His-tagged PKM2 (Right) with Myc-tag, PKM1 and PKM2 antibodies to show over-expression of PKM1 or PKM2. (C) LC/MS, coverage map of M1 and M2 isoforms of pyruvate kinase, amino acid sequence highlighted in yellow color shows the sequence covered by M/S analysis; Green color represents amino acid predicted to be phosphorylated as in A., red color shows the M1 and M2 specific sequences obtained from independent M/S analysis of PKM1 and PKM2.

Figure 7. AMPK regulates the alternative splicing of PKM isoforms. (A) Immunoblots of the protein lysates of H1299 cells, grown under enriched (25mM) or insufficient (1mM) glucose medium or enriched (25mM) glucose medium with 10 nM Rapamycin for 8 hours, using the indicated antibodies to show an antagonistic relation between AMPK and MTOR signaling pathways. The relative expression levels (signals) of PKM1 and PKM2 were quantified using ImageJ and normalized to β-Actin. (B) RT-PCR of c-Myc, hnRNPA1, hnRNPA2, PTBP1, PKM1 and PKM2 using H1299 cells grown under enriched (25mM) or insufficient (1mM) glucose medium or enriched (25mM) glucose medium with 10 nM Rapamycin for 8 hours; RT-PCR bands were subjected for densitometric analysis using Image J software and the ratio was plotted using the loading control ACTINB (ACTB), (C) Immunoblots of the protein lysates of H1299 cells stably transduced with vector (pLKO.1) or shAMPKa1/2, grown under enriched (25mM) or insufficient (1mM) glucose medium for 8 hours, using the indicated antibodies to show an antagonistic relation between AMPK and MTOR signaling pathways. The relative expression levels (signals) of PKM1 and PKM2 were quantified using ImageJ and normalized to β-Actin. (D) RT-PCR for c-Myc, hnRNPA1, hnRNPA2, PTBP1, PKM1 and PKM2 using stable H1299 cells transduced with vector (pLKO.1) or shAMPKa1/2 and grown under enriched (25mM) or insufficient (1mM) glucose medium for 8 hours; RT-PCR bands were subjected for densitometric analysis using Image J software and the ratio was plotted using the loading control ACTINB (ACTB), (E-F) Immunoblots of: (E) the protein lysates of H1299 cells, transfected with vector (pcDNA3.1) or HA tagged c-Myc (c-Myc-HA), grown under replete (25mM) or depleted (1mM) glucose conditions and probed with HA-tag, PKM1 and PKM2 antibodies, to show the overexpression of HA-tagged c-Myc; and PKM1 and PKM2 expression status. (F) the protein lysates of H1299 cells, transfected with empty vector (pcDNA3.1) or HA tagged c-Myc (C-Myc-HA) and shRNA Vector (pLKO.1) or shc-Myc, grown in enriched glucose medium; and probed with HA-tag, c-Myc, PKM1 and PKM2 antibodies, to show c-Myc over expression or knockdown along with the expression status of PKM isoforms in the respective experimental background.

Figure 8. Loss of AMPK or PKM1 negatively affects the aerobic glycolysis and ATP in glucose depleted cancer cells. (A) Immunoblots to validate stable knockdown of PKM1 (Left) and PKM2

15

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

expression (right) in H1299 cell lines transduced with lentiviruses containing empty vector (pLKO.1) or shPKM1 or shPKM2; (B, C, D and E) Glucose uptake (B); Intracellular pyruvate (C); Lactate release (D) and Intracellular ATP (E) levels in H1299 cells stably expressing with vector (pLKO.1), shAMPKa1/2, PKM1 or PKM2 targeting shRNAs, cultured in enriched (25mM) or insufficient (1mM) glucose medium for 24 hours. (F) ATP levels in Vector, shPKM1 and shPKM2 transduced H1299 stable cells is shown under enriched (25mM) or insufficient (1mM) glucose culture condition, with or without 100nM Oligomycin (Oligo) treatment for 24 hours. For all the experiments above (n=3; mean ± SEM), statistical analyses was performed using two-way ANOVA with Tukey’s multiple comparison test (GraphPad Prism),*P < 0.05, **P < 0.01, ***P < 0.001.

Figure 9. Loss of PKM1 expression inhibits survival, enhances apoptosis of glucose deprived cancer cells. (A and B) Bars represent the proliferation rate of stable H1299 cells, transduced with lentiviruses expressing empty vector (pLKO.1), shPKM1 or shPKM2, and cultured under glucose enriched (25mM) or insufficient (1mM) conditions for the period of 72 hours; where cell proliferation rate was assayed every 24 hours. For all the experiments above (n=4; mean ± SEM), statistical analyses were performed using two-way ANOVA with Tukey’s multiple comparison test (GraphPad Prism),*P < 0.05, **P < 0.01, ***P < 0.001. (C) Immunoblot for PARP, Caspase 3 and Cleaved Caspase 3 to assess the apoptosis and PKM1 and PKM2 to measure their expression status in lysates of H1299 cells stably transduced with vector (pLKO.1), shPKM1 or shPKM2 and grown in glucose free medium supplemented with 1mM of glucose for the indicated time periods.

Figure 10. Immuno Histochemistry reveals differential distribution of PKM1 and PKM2 in tumor tissues (A) Representative, hematoxylin and eosin (H & E), anti-pACC, anti-PKM1 and anti-PKM2 staining of tumor serial sections from sporadic breast cancer tissues (representative images of 8-Stage III ductal invasive carcinoma) is shown. Scale bars shown are of 200µm. (B) Schematic illustration of the dynamic expression regulation of PKM1 and PKM2 by antagonistic signaling pathways of LKB1-AMPK and MTOR-hnRNPs in response to the nutritional status of tumor microenvironment which provides proliferative (PKM2) and survival (PKM1) advantage. The latter is proposed to be under nutrition deprivation (poor vascularized regions of a tumor) state; where ATP is generated through aerobic glycolysis predominantly. OXPHOS apparently operates in the cells regardless of glucose and PKM status.

16

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

5 mm Glucose 1 mm Glucose-3

-2

-1

0

1

2HK1

HK2

PKM1

PKM2

LDHA

LDHB

**

*

**

***

5 mm Glucose 1 mm Glucose-3

-2

-1

0

1

2HK1

HK2

PKM1

PKM2

LDHA

LDHB

**

**

***

**L

acat

e d

ehyd

roge

nas

e A

ctiv

ity

(U/m

g) PKM2

p-PKM2 (Tyr105)

PKM1

β-ACTIN

25 5 1 (Glucose mM)

0.95 0.81 0.55

0.47 0.65 0.90

PKM2/β-ACTIN

PKM1/β-ACTIN

Pyr

uva

te K

inas

e A

ctiv

ity

(U/m

g)

A

Figure 1

25 5 1

PKM2

PKM1

p-PKM2 (Tyr105)

.93 0.76 0.61

0.67 0.83 .95

(Glucose mM)

β-ACTIN

PKM2/β-ACTIN

PKM1/β-ACTIN

25 mM G 5 mM G 1 mM G0.0

0.2

0.4

0.6

0.8

1.0

Hex

okin

ase

Act

ivit

y (U

/mg)

25 mM G 5 mM G 1 mM G0.0

0.2

0.4

0.6

Lac

ate

deh

ydro

gen

ase

Act

ivit

y (U

/mg)

B

C

H1299

MCF-7

H1299 MCF-7

H1299

MCF-7

H1299 H1299

MCF-7 MCF-7

17

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

PKM2

PKM1

β - ACTIN

PKM2

PKM1

β - ACTIN

N T N T N T N T N T N T

Stage 2 Stage 3

Anti-PKM2

C.B.B. stained

Glucose (mM) 25 1 25 25

Glutamine (4mM) + + - +

Pyruvate (1mM) + + + -

PKM 2

β - ACTIN

Glucose (mM) 25 1 25 25

Glutamine (4mM) + + - +

Pyruvate (1mM) + + + -

0.85 0.57 0.89 0.90 PKM2/β-ACTIN

Anti-PKM1 PKM 1

0.65 0.85 0.62 0.64 PKM1/β-ACTIN

C

A

D

B

E

Figure 2

18

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

-0.05

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0 5 10 15 20 25 30 35

Py

ruv

ate

Kin

ase

Act

ivit

y (U

/ml)

Fraction No.

25 mM Glucose

5 mM Glucose

1 mM Glucose

Dimer

Tetramer

1 10 20 30

PKM1

PKM2 25 mM Glucose

PKM1

PKM2

PKM1

PKM2

5 mM Glucose

1 mM Glucose

Fractions

Figure 3

A

B

19

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

(Glucose mM) 25 10 5 1 .1 0

AMPKa

p- AMPKa (Thr172)

p-ACC (Ser79)

ACC

PKM1

PKM2

β -ACTIN

0.24 0.32 0.42 0.59 0.65 0.87

PKM2/β-ACTIN 0.90 0.81 0.72 0.52 0.44 0.32

PKM1/β-ACTIN

(AICAR mM)

PKM1

PKM2

β -ACTIN

AMPKa

p-ACC (Ser79)

ACC

p-AMPKa (Thr172)

(Glucose mM) 25 25 25 25 25 25

0 .05 .1 .2 .5 1

0.56 0.85 0.91 1.02 1.14 1.23 PKM1/β-ACTIN

1.12 1.07 0.84 0.78 0.55 0.34 PKM2/β-ACTIN

Glucose 25 mM + + - - + +

AICAR 500 µM - - - - + +

Compound C (10 µM) - + - + - +

β - ACTIN

p-ACC (Ser79)

PKM2

PKM1

0.41 0.51 0.74 0.20 0.76 0.29 PKM1/β-ACTIN

1.16 1.06 0.43 1.16 0.42 1.26 PKM2/β-ACTIN

PKM1

PKM2

β-ACTIN

AMPKa

p-ACC (Ser79)

ACC

p-AMPKa (Thr172)

Glucose (mM) 25 25 25

Metformin (2mM) - + -

Phenformin (2 mM) - - +

0.44 0.88 0.92

0.66 0.42 0.22 PKM2/β-ACTIN

PKM1/β-ACTIN PKM2

β -ACTIN

p-ACC (Ser79)

ACC

p-AMPKa(Thr172)

(Glucose mM) 25 1

Vector shAMPKa1/2

25 1

PKM2/β-ACTIN0.96 0.63 0.88 0.84

AMPKa

0.59 0.82 0.46 0.30

PKM1

PKM1/β-ACTIN

Figure 4

A B

C D E

20

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

55 KDa

35 KDa

40 KDa

PKM1

β-ACTIN

Myc tag

p-ACC (Ser79)

PKM2

A549

PKM2/β-ACTIN

PKM1/β-ACTIN

25 25 25 (Glucose mM)

0.66 0.66 0.86

0.83 0.80 0.65

β - ACTIN

Myc-tag

PKM1

PKM2

25 1

Vector LKB1-myc

25 1

0.54 0.47 0.61 0.84

0.71 0.66 0.66 0.34

(Glucose mM)

PKM2/β-ACTIN

PKM1/β-ACTIN

25 1

Vector LKB1-myc

25 1

0.31 0.35 0.38 0.65

0.55 0.61 0.66 0.34

Figure 5

A549 MDA-MB-231

A B

21

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Protien Targets Site -5 -4 -3 -2 -1 0 1 2 3 4

ACC S79 I R S S M S G L H L

TSC2 S1385 L S K S S S S P E L

PFK2 S466 R M R R N S F T P L

PKM 1/2 T45 I T A R N T G I I C

PKM 1/2 T93 I K N V R T A T E S

PKM 1 S403 L V R A S S H S T D

Scansite -5 -4 -3 -2 -1 0 1 2 3 4

High Stringency L R R V X S/T X X N L

Moderate Stringency M K K S X S/T X X D V

Low Stringency I/R P H/S R/N X S/T X X X I/F

PKM1

Myc - tag

PKM 2

β Actin

PKM2

Myc - tag

Endogenous

Exogenous

PKM 1

β Actin

A

C

B

Figure 6

22

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

p-ACC (Ser79)

ACC

p-AMPKa (Thr172)

AMPKa1/2

p-p70 S6 Kinase

p70 S6 Kinase

P-S6 (Ser 240/244)

P-S6 (Ser 235/236)

c-Myc

(Glucose mM) 25 1

Vector shAMPKa1/2

25 1

β - ACTIN

PKM2/β-ACTIN

0.68 0.94 0.65 0.60 PKM1/β-ACTIN

0.74 0.21 0.72 0.76

PKM1

PKM2

(Glucose mM) 25 1

Vector shAMPKa1/2

25 1

c-Myc

hnRNPA1

hnRNPA1/ACTB

c-Myc/ACTB 0.55 0.25 0.85 0.85

0.50 0.27 0.77 0.69

hnRNPA2

0.97 0.58 1.15 1.11 hnRNPA2/ACTB

PTBP1

PKM1

PKM2

ACTB

PTBP1/ACTB

PKM1/ACTB

PKM2/ACTB

0.84 0.46 0.92 1.15

0.50 0.96 0.47 0.41

0.81 0.28 0.79 0.82

RT-PCR

Glucose (25 mM) + - +

Glucose (1 mM) - + -

Rapamycin (10 nM) - - +

p-S6 (Ser 240/244)

P70 S6 Kinase

p-P70 S6 Kinase

β-ACTIN

c-Myc

PKM1

PKM2

0.5 0.92 0.89 PKM1/β-ACTIN

0.8 0.25 0.50 PKM2/ β -ACTIN

p-S6 (Ser 235/236)

Glucose (25 mM) + - +

Glucose (1 mM) - + -

Rapamycin (10 nM) - - +

ACTB

0.77 0.20 0.43

PKM2

PKM2/ACTB

0.30 0.67 0.62 PKM1/ACTB

PTBP1

0.61 0.32 0.38

PKM1

0.94 0.70 0.48

hnRNPA2

0.67 0.33 0.43

hnRNPA1

0.84 0.45 0.51

c-Myc

PTBP1/ACTB

hnRNPA2/ACTB

hnRNPA1/ACTB

c-Myc/ACTB

25 1

HA-tag

β-ACTIN

PKM1

PKM2

(Glucose mM) 25 1

0.45 0.92 0.49 0.88

0.80 0.34 0.79 0.45

HA-tag

β-ACTIN

PKM 1

PKM 2

25 25

c- Myc

β-ACTIN

PKM 1

PKM 2

25 25 (Glucose mM)

A B

E

Figure 7

C

D F

RT-PCR

23

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

Figure 8

A B

C D

E F

PKM2

β-ACTIN

PKM2

PKM1

β-ACTIN

PKM1

24

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

A B

C

PKM1

PKM2

β - Actin

PARP

Cleaved PARP

Vector shPKM 1 shPKM 2

Caspase -3

Cleaved Caspase -3

0 24 48 0 24 48 0 24 48 Hours

Figure 9

25

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

H & E pACC PKM1 PKM2 A

B

Figure 10

26

at Cadm

us Professional Com

munications on M

ay 13, 2016http://w

ww

.jbc.org/D

ownloaded from

by guest on February 4, 2020

http://ww

w.jbc.org/

Dow

nloaded from

and Rameshwar N. K. BamezaiPrakasam Gopinath, Rajnish Kumar Singh, Mohammad Askandar Iqbal, Ashu Bhan Tiku

Tolerates Hypoglycemic Stress in Cancer CellsRegulation of Pyruvate Kinase M Switch towards PKM1 by LKB1-AMPK Axis

published online April 26, 2016J. Biol. Chem. 

10.1074/jbc.M116.729079Access the most updated version of this article at doi:

 Alerts:

When a correction for this article is posted•When this article is cited•

to choose from all of JBC's e-mail alertsClick here

http://www.jbc.org/content/early/2016/04/26/jbc.M116.729079.full.html#ref-list-1This article cites 0 references, 0 of which can be accessed free at

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from

and Rameshwar N. K. BamezaiPrakasam Gopinath, Rajnish Kumar Singh, Mohammad Askandar Iqbal, Ashu Bhan Tiku

LKB1-AMPK Axis Tolerates Hypoglycemic Stress in Cancer CellsWITHDRAWN: Regulation of Pyruvate Kinase M Switch towards PKM1 by

published online April 26, 2016 originally published online April 26, 2016J. Biol. Chem. 

  10.1074/jbc.M116.729079Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

by guest on February 4, 2020http://w

ww

.jbc.org/D

ownloaded from