by Jana Vree
A v a n s U n i v e r s i t y o f A p p l i e d S c i e n c e ,
S c h o o l o f L i f e S c i e n c e s a n d E n v i r o n m e n t a l T e c h n o l o g y
by Jana Vree
A v a n s U n i v e r s i t y o f A p p l i e d S c i e n c e ,
S c h o o l o f L i f e S c i e n c e s a n d E n v i r o n m e n t a l T e c h n o l o g y
Bachelor Thesis
The role of GSTP1 in KRAS induced lung tumorigenesis
Bachelor’s Thesis, September 1
st 2014– June 30
th2015
By Jana Vree
Supervised by
Jos van der Velden Ph.D and Prof. Yvonne Janssen- Heininger Ph.D Department of Pathology and Laboratory Medicine
University of Vermont, USA
Pascal Hommelberg Ph.D
School of Life Sciences and Environmental Technology Avans University of Applied Science, The Netherlands
Abstract
Lung cancer is known for its aggressive and metastatic behavior. Non-Small Cell Lung cancer represent 85%
of patients diagnosed with lung caner. Despite novel advances in medical treatment and research the survival rate did not change in the last three decades and is less than 5 years. Additionally it’s acknowledged that lung cancer exhibits a substantially resistance against chemotherapeutic drugs. Therefor it’s vital to develop new therapeutics that systematically work in combination with already existing chemotherapeutic drugs. 20-30% of patients diagnosed with NSCLC harbor a KRAS mutation. Furthermore, it is implied that KRAS-induced tumorigenesis impacts resistance to cancer treatment, cellular motility and invasiveness. Expression of Glutathione S-transferase P1 (GSTP1) is significantly higher in tumor tissue of patients with NSCLC. Therefore a better understanding of the contribution of GSTP1 in KRAS induced transformation, could lead to the development of more effective therapeutic strategies in lung cancer treatment.
To determine the role of GSTP1 in KRASG12D induced tumorigenesis, we used in vitro as well in vivo models.
In our in vivo experiments we exposed human lung cancer cells with the GSTP1 inhibitor TLK-117 and analyzed mRNA, protein and lactate production as well as total GST activity and protein s-glutathionylation (PSSG).
Pharmacological inhibition of GSTP1 by TLK-117 resulted in a significant decrease in lactate production and GST activity in A549 cells. Furthermore we discovered that GSTP1 expression varied in different lung cancer cell lines.
TLK-117 significantly reduced cell migration in A549 and H292 cells and this appears to correlate with GSTP1 expression levels. In A549 cells TLK-117 reduced protein s-glutathionylation of PKM2, an important key regulator of the Warburg Effect. We also developed a CCSP-rtTA/tetON-Cre/LSL-KRASG12D triple transgenic mouse.
Following initiation of the KRASG12D mutant using an adenovrius expressing Cre or doxycycline containing food tumor burden was evaluated. Further, gene expression was analyzed by RT-qPCR and protein analysis by Western blotting. In addition changes in lactate production and GST activity were determined. In KRASG12D expressing mice gene expression of PKM2, HKII, HKIII, Glut2 and PDHKI increased and LDHA decreased.
Additionally an increased lactate production in these mice could be detected. Mice expressing KRASG12D that were exposed to TLK-117 showed an slight decrease in GST activity and showed fewer tumor regions than untreated mice.
Together, these findings demonstrate that GSTP1 plays an important role in KRASG12D induced tumorigenesis, cancer cell migration, tumor progression and protein s-glutathionylation. It also implicates the importance of GSTP1 in KRAS induced tumorigenesis and the relevance of TLK-117 as new important drug for the treatment of NSCLC.
Table of Contents
Abstract ... 3
1. Introduction ... 6
2. Theoretical Background ... 7
2.1 Lung cancer ... 7
2.2 Classification of Lung cancer types ... 7
2.3 Kirsten-rous avian sarcoma (KRAS) ... 8
2.4 GSTP1 ... 8
2.5 Protein S-Glutathionylation (PPSG) ... 10
2.6 TLK-199 ... 11
2.7 PKM2, the Warburg effect and Redox balance ... 12
2.8 Impact of this project ... 13
3. Materials and methods ... 14
3.1 Culturing of cell lines ... 14
3.2 Cell treatment ... 14
3.3 Animals ... 14
3.4 Microsprayer Instillation of Adenovirus Cre ... 15
3.5 Oropharyngeal Instillation of TLK-117 ... 16
3.6 Lung harvest and bronchoalveolar lavage ... 16
3.7 GSH-Immunoprecipitation ... 17
3.8 Protein Harvest and analysis (whole cell lysate) ... 17
3.9 Lung tissue homogenization and Western blotting ... 18
3.10 Lactate assay ... 18
3.11 GST-activity assay ... 19
3.12 Migration/Invasion assays ... 19
3.13 Quantification of gene expression by real-time quantitative reverse transcriptase PCR ... 19
3.14 Assessment of cancer development and tumor growth ... 20
3.15 GSTP1 immunohistochemistry ... 20
3.16 Immunofluorescence staining for S-Glutathionylation ... 20
3.17 Statistical analysis
... 21
4. Results ... 22
4.1 GSTP1 expression in NSCLC cell lines ... 22
4.2 Pharmacological GSTP1 inhibition in NSCLC cell lines ... 24
4.3 Protein S-glutathionylation (PSSG) in a NSCLC cell line ... 24
4.4 KRASG12D induced tumorigenisis ... 25
4.5 Pharmacological GSTP1 inhibition in KRASG12D induced tumorigenesis ... 26
4.6 GSTP1 in NSCLC ... 27
5. Discussion and Conclusion ... 29
6. Future perspectives ... 32
References ... 33
Appendix ... 36
Appendix 1: Supplemental figures ... 36
Appendix 2.Mouse model ... 37
Appendix 3: Primer sets ... 38
1. Introduction
Lung cancer and particularly non- small cell lung cancer (NSCLC) is one of the leading lethal diseases worldwide, with an average survival rate of 5 years. Despite novel advances in medical treatment of lung cancer, the survival rate remained unchanged in the last 3 decades. Approximately 85% of lung cancer patients are diagnosed with NSCLC and in 20% to 30% of these cases a KRAS mutation is discovered. A mutation in the KRAS gene results in constitutive activation, which leads to an increased proliferation and decreased apoptosis, all known cancer hallmarks [1]. Glutathione S- transferase P1 (GSTP1) is one of the enzymes of the Glutathione S-transferase (GST) super family.
Although GSTP1 is overexpressed in a variety of cancers, little is known about GTSP1 and its involvement in lung cancer. Hence it’s importance to improve the understanding of the contribution of GSTP1 in NSCLC and it’s contribution in KRASG12D induced transformation.
An important hallmark of cancer is that tumor cells ferment glucose and produce a large amount of lactate even in the presence of oxygen. This is known as “the Warburg effect” and despite its widespread appreciation the exact mechanisms remains to be elucidated. Preliminary data obtained in our lab reveals an increase in glutathionylation of PKM2, together with an increase in lactate production following KRAS-induced transformation of primary epithelial cells. Furthermore activation of KRASG12D induced overall protein S-glutathionylation (PSSG). The finding that GSTP1 expression is increased in NSCLC together with a novel identified role for GSTP1 as a catalyst of protein S- glutathionylation warrants to study the molecular mechanisms of the role of GSTP1 in KRAS- induced lung tumorigenesis.
This study aimed to gain knowledge of the role of GSTP1 in KRAS induced transformation and tumor cell behavior in in vitro as in in vivo models. To evaluate GSTP1 expression and overall S- glutathionylation in non-small cell lung tumors and to determine whether GSTP1 inhibition attenuates KRAS-induced lung tumorigenesis, we hypothesize that GSTP1-catalyzed S- glutathionylation is enhanced in KRAS induced lung tumorigenesis. This could lead to subsequent activation of the pentose phosphate pathway (PPP), tumor cell growth and invasion.
2. Theoretical Background
2.1 Lung cancer
Lung cancer is one of the leading causes of dead in both male and female worldwide. Cigarette smoke is the main risk factors for lung cancer development, not only active smoking causes cancer but also passive exposure to tobacco can result in the development of lung cancer. Cigarette smoking accounts for about 85% of lung cancer cases worldwide. Cancer risk is dependent on age, smoking duration and intensity. About 15 to 20 % of lung cancer patients minimally or never smoked.
Prolonged exposure to cancer-promoting agents (e.g cigarette smoke, asbestos, radiation, radon, arsenic, chromates, nickel, chloromethyl ethers, polycyclic aromatic hydrocarbons, mustard gas, coke- oven emissions, primitive cooking, heating huts) and multiple genetic mutations are required to develop neoplastic respiratory epithelial cells, which leads to lung cancer development. [1]
2.2 Classification of Lung cancer types
Lung cancer can be divided into two major classifications. It is categorized as small cell lung cancer (15%) or non–small cell lung cancer (85%) [2]. Both types can be further differentiated into four major histological classes: adenocarcinoma, squamous cell carcinoma, small cell carcinoma, and large cell carcinoma [3]. Small cell carcinomas are aggressive and usually located centrally in the lung, with extensive mediastinal involvement. They are also associated with early stage extra thoracic metastases [4]. Small cell carcinomas are often advanced at the time of diagnosis, which leads to a poor prognosis for the patient, despite its responsiveness to chemotherapeutics.
The three main subtypes of non-small cell lung cancer are adenocarcinomas, Squamous cell carcinomas and large cell carcinomas. Adenocarcinomas are peripheral tumors that metastasize early and are diagnosed in patients with underlying lung disease [3]. Squamous cell carcinomas unlike adenocarcinomas metastasize in later stages of the disease and are mostly located endobronchial.
Large cell carcinomas are associated with cigarette smoking, they grow rapidly and metastasized early. As mentioned earlier, Non-small cell lung cancer (NSCLC) represents 85 % of it. Most patients with advanced NSCLC are treated with chemotherapy regimens that provide an overall survival of less than 1 year. The 5-year survival rate in lung cancer cases did not change in the last 20 years and is less than 20 % [5]. Due to molecular analysis, tumor cells may be distinguished based on various
markers for example, the absence or presence of e.g mutations or oncogenic fusion rearrangements.
One of those markers is the KRAS gene, which KRAS belongs to the RAS family. These genes encode proteins, which are central mediators of growth factor receptor signaling [6]. In about 20-30 % of NSCLC cases a KRAS mutation is existent [7].
2.3 Kirsten-rous avian sarcoma (KRAS)
Of the numerous oncogenes involved in human cancers RAS mutations are the most common ones and is therefore a favorite to target pharmacologically. There are 3 different RAS genes that encode four proteins: HRas, KRAS 4a/b and NRas. In normal cells RAS signaling is crucial to cellular proliferation, cell survival and differentiation. In lung cancer the most commonly mutated gene is the KRAS gene [8]. Mutated KRAS proteins have lost the ability to become inactive due to the fact that the KRAS protein is constitutively bound to guanosine triphosphate (GTP) such that its GTPase activating protein (GAP) cannot bind to the RAS protein. This results in decreased apoptosis, increased cell proliferation, increased angiogenesis and disrupted cellular metabolism [9]. Yun et al. described that cancer cells with a KRAS mutant phenotype, have a consistently up regulated gene expression of the glucose transporter-1 (GLUT1) [10]. Recent studies also described a link between KRAS induced tumorigenesis and its role in metabolic changes. Furthermore it was described that the mRNA expression of Glutathione S- transferase P1 (GSTP1), was elevated in several treatment resistant cancer types. Little is known about the potential role of GSTP1 in the tumorigenesis of NSCLC harboring a KRAS mutant. It is possible that GSTP1 activation has a major impact on cellular regulation in metabolic pathways [11].
2.4 GSTP1
Glutathione S-transferases (GST’s) belong to the family of phase II detoxification enzymes. They catalyze the conjugation of glutathione to endo- or exogenous electrophilic compounds. GST‘s can be divided into 2 super-family members: the membrane bound, microsomal and the cytosolic family.
Cytosolic human GST’s can be derived into six classes: α, µ, ω, π, θ and ζ. GSTπ (in this report referred as GSTP1) and GSTµ have a regulatory role in the mitogen-activated protein kinase (MAPK) pathway. This pathway participates in cellular proliferation and death signals via protein/protein interactions with apoptosis signal-regulating kinase (ASK1) and c-Jun N-terminal kinase (JNK1).
GSTP1 binds to and inhibits Jun-N-terminal kinase (JNK), which is key regulator of cellular
proliferation, differentiation, and apoptosis. Following stimulation by growth factors or cellular stress, JNK gets phosphorylated and activated. Active JNK phosphorylates the transcription factor c-Jun, which then promotes the transcription of genes involved in cell proliferation and differentiation [12].
Most of the GST’s show a high degree of polymorphism. Furthermore dependent on the type of GST, GST’s can have several subunits. Each subunit is approximately 199- 244 amino acids (22 to 29 kD) long. More than a dozen cytosolic GST subunits have been identified. Those subunits contain a catalytic active site, the GSH- binding site (G-site) in the amino- terminal domain, which binds a hydrophobic substrate (‘H-site’) in the carboxy-terminal domain. Expression varies and is dependent on the type of GST. The functional GST enzymes are dimeric, with the exception of GSTα and GSTµ, which can form heterodimers in addition to homodimers. Due to the different subunits isoenzymes can be generated form those subunits. Isoenzymes are named according to their class and composition of present subunits, which is designated with an Arabic number. In case of GSTP1-1 this refers to the heterodimer of the two subunits 1 and 1. GSTπ has in total 2 subunits referred to as subunit 1 and subunit 2 [13]. GSTP1 for example is found mainly in brain, lung, and heart; its expression in liver decreases during embryonic development, becoming very low in non-malignant adult tissue [14].
Recently studies have shown that GST’s act as modulators of signal transduction pathways that control cell proliferation. Their involvement in the development of resistance to anti-cancer drugs has made them attractive research targets. GST isoforms have recently been identified as negative regulators of signal transduction pathways, and thereby protecting cells against apoptosis in stress response, such as Reactive Oxygen Species (ROS). GSTs are well known for its role in phase II drug metabolism, as it catalyzes conjugation of GSH to electrophiles. A variety of human cancers, including of breast, colon, kidney, lung, and ovarian, usually express high levels of GSTP1 compared to normal surrounding tissues. However recently a new role has been described for GSTP, as a catalyst in PSSG reactions [15].
2.5 Protein S-Glutathionylation (PPSG)
Reactive Oxygen Species (ROS) are chemical reactive molecules containing oxygen. These molecules include H2O2, O2-, 1O2, HO-, NO2, RO-, and ROO-. They are formed as a natural byproduct of the normal metabolism of oxygen [16]. Elevated levels of ROS and down regulation of ROS scavengers and antioxidant enzymes are associated with various human diseases including cancer [17]. ROS are normally present in all aerobic cells and are in balance with biochemical antioxidants.
Oxidative stress occurs when the balance is disrupted, because of antioxidants depletion, ROS excess or both. This can result in significant damage to cell structure like e.g. DNA damage (mutagenesis), lipid peroxidation (lipid degradation), amino acid oxidation (protein) and oxidative inactivation of specific enzymes [18]. ROS play an important role in cell signaling and homeostasis.
H2O2 for example is able to control cellular homeostasis and plays a role in de induction of the posttranslational modification of proteins, by selectively oxidizing its cysteine residues, [19,20]. It specifically targets the cysteine’s with low pKa’s, which results in a sulfenic acid intermediate protein.
Protein S-Glutathionylation (PSSG) is the formation of a disulfide bond between the protein cysteine and the glutathione (GSH) (fig. 1). Conjugation of GSH by GST’s such as GSTP1, leads to PSSG of the sulfenic acid intermediate. Due to the conjugation of protein cysteines, GSH changes the size of
the protein by 3 amino acids. This confers a negative charge. This suggest that PSSG can alter the structure and function of the target protein which can than result in an activation or inhibition of this protein [21,22]. Kinase mediated signaling pathways; also play key roles in controlling cell survival, differentiation, apoptosis and stress response. Addressing the complex interactions between GSTP1 and regulatory kinases will help to get a better understanding of the role GSTP plays in tumor pathogenesis and its cancer resistance [23,24].
Figure 1 PKM2 S-glutathionylation: Prior to s-glutathionylation, PKM2 gets oxidized at its cysteine residue, which results in the formation a sulfenic acid (RSOH) derivative form of the protein. Glutathion S Transferase P1 (GSTP1), catalyzes the s-glutathionylation reaction by forming a disulfide bond, between the oxidized cysteine and the thiol residue of the glutathione (GSH).
2.6 TLK-199
Human GSTs are of significant interest because of their high expression levels in tumor tissue and, their possible role in Protein S-Glutathionylation of target proteins. Identifying new inhibitory compounds against human GSTs therefore could utilize new anti cancer drugs [25]. As previous mentioned GSTs exists in homo or heterodimers. Each subunit has its own active site. These active sites consist of a GSH binding site (G-site), which is specific to GSH and the H-site, which binds xenobiotics [26]. At the G-site the substrate enzyme complex (GSTP1-GSH) forms when the hyrdoxyl group of Tyr7, of the GSTP1 enzyme forms a hydrogen bond with the thiol group of the GSH molecule (fig. 2A) [27]. TLK-199 or Ezatiostat hydrochloride is a prodrug and was in phase 2 clinical trial for the treatment of myelodysplastic syndrome. This prodrug has been reported to be a potent inhibitor of GSTP1-1 [28]. The actual working compound is the tripeptide TLK-117 (g-glutamyl(benzyl)cysteinyl-
R(-)-phenylglycine). It’s binding affinity to the G-site of the GSTP1-1 enzyme, is greater than that of GSH and GSTP1-1 itself. It’s also highly selective to GSTP1-1. TLK-117 itself interacts and forms a bond with GSTP1. When TLK-199 passes the cell membrane it is hydrolyzed to the phenyl glycine monoethyl ester form TLK-117 [27, 29]. Inside the cell TLK-117 interacts with GSTP1 due to the formation of a hydrogen bond between the Tyr7, of GSTP1 and the thiol group of the TLK-117 molecule (fig. 2 B). This prohibits the binding of GSH to GSTP1, which imaginably inhibits s- glutathionylation of proteins. It also keeps GSTP1 from inhibiting JNK through direct protein-protein Figure 2 Schematic drawing of the residues that interact between the inhibitor/GSH and GSTP1. A) Interaction between the tyrosine of the GSTP1 enzyme and the thiol group of GSH. B) Di-esterfication of the prodrug TLK-199 to the active compound TLK-117 and schematic interaction between GSTP1 and TLK-117.
interactions [26,27,30]. TLK-117 will be used in this project to determine its effect on protein s- Glutathionylation and it’s role in KRASG12D induced tumorigenesis. TLK-117 will be tested in vitro and in vivo.
2.7 PKM2, the Warburg effect and Redox balance
In spite of the presence of an abundant amount of oxygen, cancer cells produce a large amount of lactate, also known as the Warburg effect or aerobic glycolysis. In a rich oxygen environment, normal differentiating cells metabolize glucose to pyruvate by glycolysis. The product is completely oxidized in the mitochondria to CO2, during oxidative phosphorylation. If the oxygen access is limited, cells redirect the produced pyruvate away from the mitochondria and lactate is generated. Cancer cells have the tendency to convert glucose to lactate [31]. It does not matter if oxygen is present or not. This results in minimal ATP production but a high production of waste products such as lactate. Pyruvate kinases are important regulators in the production of lactate [32]. High enzymatic activity results in the production of pyruvate, which enters the TCA cycle. Low enzyme activity of pyruvate kinase, promotes lactate production and therefore promotes the Warburg effect (fig 3). Pyruvate kinases 2 or PKM2 belongs to the pyruvate kinases and is a regulator of anti-oxidative metabolism. It plays an important role in cancer. When the activity of PKM2 is low the substrate phosphoenol pyruvate accumulates. This results in the inhibition of glycolytic enzyme triose phosphate isomerase and leads to activation of the pentose phosphate pathway (PPP). Due to increased activity of the PPP, the cancer cell is protected and promotes the adaptation against oxidative stress and ROS [33]. Anastasiou et.al showed that the activation of the PPP and the associated anti-oxidative activity are essential to the growth of cancer cells. They also established that the oxidation of PKM2 on cysteine (Cys 385), results in a lower enzyme activity of PKM2, in lung cancer cells [34]. This data suggest, that by inducing the Warburg Figure 3 Activation of the pentose
phosphate pathway and its anti-oxidative activity in cancer-cell growth. [20]
effect, cancer cell growth is promoted by activation of the PPP. This maintains the balance of the redox equivalents, providing NADPH an activation the antioxidant defense systems of cancer cells.
2.8 Impact of this project
The precise molecular based mechanisms, where changes in the redox environments of cells promote tumor growth are unknown. This study provides a complete new understanding of the role of GSTP1 as a key regulator of the Warburg effect, due to its ability to catalyze protein S- glutathionylation reactions. Significantly, activation of the PPP provides the cell with GSH [34]. This is required in the catalytic cycle of GSTP1. It provides a feed forward amplification mechanism whereby GSTP1 could induce S- glutathionylation of PKM2, leading to activation of the PPP (fig 4). This leads to a continual increases in GSH required to fuel GSTP1 mediated protein s-glutathionylation (fig. 4).
Thus, this application has the potential to discover a “new” role for an “old” enzyme. Of additional relevance is the potential direct clinical application of this study. TLK-117, a highly selective inhibitor for GSTP1 has been extensively used in patients. Furthermore, TLK-117 is non-toxic and highly suitable for therapy purposes. The information obtained in this study could provide a better understanding of the role of GSTP1 in KRAS induced tumorigenesis, which could lead to the re-purpose of TLK-117 as a new therapeutic for patients with NSCLC, which express high levels of GSTP1 and PSSG.
Figure 4 The role of PSSG in KRAS induced tumorigenesis: (Top) Glucose enters the cell and gets phosphorylated by Hexokinase (HK) to Glucose 6-Phosphate, which initiates glycolysis. Glycolysis converts glucose to pyruvate via a series of intermediate metabolites. The final step of glycolysis, phosphoenolpyruvate gets converted into pyruvate by Pyruvate Dehydrogenase M2 (PKM2). High PKM2 enzymatic activity results in the production of pyruvate which fuels the TCA cycle. Low PKM2 activity results in an accumulation of phosphoenolpyuvate and shunts G6P into the PPP. Consequently NADPH and GSH are produced, which feasibly fuels additional PSSG (feed forward mechanism). This process is possibly mediated by PSSG of PKM2 by GSTP1 (Bottom).
3. Materials and methods
3.1 Culturing of cell lines
A549, H460, H292 and H1437 human NSCLC cell lines were cultured to study KRAS-induced tumorigenesis and the role of GSTP1 in KRAS induced transformation. A549 (KRASG12D mutation, high KRAS activity) cells where cultured in DMEM/F12 ([+] L-Glutamine and [+] 15 mM Hepes) and H460 (KRASQ61H mutation, low KRAS activity), H292 and H1437 (no KRAS mutation) in RPMI 1640 ([+] L-Glutamine) (Life Technologies, Grand Island, NY) with 10% fetal bovine serum (Life Technologies), 50 U/ml Penicillin and 50 U/ml streptomycin (Life Technologies) in a humidified 37˚C, 5% CO2 incubator.
Human tumor tissue samples were obtained from adult donors undergoing biopsy surgery at the University of Vermont Medical Center. Tissues of these patients were treated with RNase and elastase to disrupt tumor tissue and obtain cells. The cell suspension was plated on collagen coated 35mm dishes and cultured in DMEM/F12 ([+] L-Glutamine and [+] 15 mM Hepes) ) in a humidified 37˚C, 5% CO2 incubator.
3.2 Cell treatment
For cell experiments, a cell density of either 1x104, 5x104 2x105 per 35 mm cell culture dish (Falcon) was used. For TLK treatment cells were cultured during 24 hours with a retreatment of TLK- 117 6 hours prior to the harvest. Dependent on the experiment, cells were incubated with the GSTP1 inhibitor TLK-117, which was dissolved in RPMI 1640 or DMEM/F12 medium with 0,02% DMSO. TLK- 117 is made in house (Department of Pathology and Laboratory Medicine, Janssen-Heiniger lab, UVM) following the protocol described by Anathy et al. [35]. Three different TLK-117 concentrations were used, 50µM, 100µM and 200µM.
3.3 Animals
For animal experiments eight-week-old C57B/6NJ mice were used. CCSP-rtTA/tetON-Cremice (University of Vermont, Burlington, VT) and LoxP-Stop-LoxP/KRASG12D mice (The Jackson Laboratory, Bar Harbor, ME) were bread in house. Oncogenic KRASG12D expression, specific in lung epithelial cells was achieved by breeding mice expressing CCSPrtTA/tetON-Cre with mice that express the LoxP-Stop-LoxP system and the KRASG12D oncogene (fig. 5). You can find more
information about the mechanism in the appendix 2. Eight-week-old triple and double transgenic mice expressing either the CCSPrtTA/tetON-Cre/LSL-KRASG12D or CCSPrtTA/tetON-Cre/KRASWT were fed food with 6g/kg Dox (Purina Diet Tech, St. Louis, MO, USA). Double transgenic mice expressing KRASWT were used as controls.
One group of the double and triple transgenic mice were treated (every three day’s) with TLK-117 the other group with the vehicle as control, starting 14 days after the first administration of Adenovirus expressing Cre recombinase (described below). The Institutional Animal Care and Use Committee (IACUC) at the University of Vermont approved each animal experiment for this study.
3.4 Microsprayer Instillation of Adenovirus Cre
We also induced KRAS mediated tumorigenesis in mice lacking either the CCSP rTetA or the Tet- OpCre transgenes. In these mice we introduce the Adenovirus expressing Cre recombinase. Like in the triple transgenic mice, these mice will express the KRASG12D gene due to the presents of the LoxP-Stop-LoxP sequence that gets terminated due to the presence of the Cre recombinase. We administred the Aenovirus Cre (AdCre) by micro sprayer instillation (Vector BioLabs, Malvern, PA), LSL-KRASG12D and KRASWT expressing mice were anesthetized as described above. Tongues were retracted, and an otoscope with a specula and light source was gently maneuvered into place to make the trachea clearly visible. Then A Microsprayer (Penn Century) was placed into the trachea to deliver 50 µL AdCre (1x108pfu/mouse) aerosolized directly into the trachea and subsequently the lungs. The instruments were then removed; the mice were then briefly rotated and allowed to take 3 deep breaths in the left and right lateral decubitus position. Mice recovered from anesthesia within 30 to 60 seconds, and were monitored periodically thereafter.
Figure 5 Triple transgenic CCSP-rtTA/tetON-Cre LSL- KRASG12D mouse.
3.5 Oropharyngeal Instillation of TLK-117
To anesthetize experimental mice, mice were placed in a Plexiglas induction chamber and anesthetized with 5% isoflurane. Once breathing slowed down to a rate less than one breath per second, mice were extracted from the chamber and positioned on a 60° incline board. Their tongues were retracted, and 50 µl of solution was delivered into the distal part of the oropharynx. The tongue was kept in a retracted position to prevent swallowing, until the fluid had been fully aspirated. Mice were then briefly rotated and allowed to take 3 deep breaths in the left and right lateral decubitus position. Mice recovered from anesthesia within 30 to 60 seconds, and were monitored periodically thereafter. Agents to be delivered were 25 mg/kg TLK-117 (in 1,5M Tris-HCl and 0,02% DMSO) and vehicle (in 1,5M Tris-HCl 0,02 % DMSO) as a control.
3.6 Lung harvest and bronchoalveolar lavage
Mice were anesthetized by an IP injection with 0.1 ml Pentobarbital (stock 50 µg/ml). Upon effect where they are immobile and the heart rate has stopped, a cervical dislocation was performed as a secondary method. A with ethanol dampened gauze was used to wipe the fur on and around future cut area. An incision was made in the skin above the trachea. Afterwards the trachea was exposed, and freed from surrounding fascia. The trachea was than raised with forceps and a 4cm long suture was placed underneath it. After this a tiny cut was made, at the top of trachea. The suture was tide around the cannula, to assure sure that the cannula stays in place. A syringe filled with PBS was inserted into the cannula. The lung of the mouse is slowly filled up with 1 ml sterile PBS (Life Technologies) and the chest was gentle massaged to ensure that a fair amount of cells were collected. Next the just inserted PBS was slowly pulled out back into the syringe with an total end volume of 0.7-0.8 ml. The syringe was carefully removed and the cannula without removing it from the trachea. The cell suspension is stored on ice or in the -80˚C. The rib cage is opened up, to harvest the lung lobes. For this an incision a long the sternum was made and the rib cage was opened just above the trachea. Ribs were grabbed with forceps and separate, to open up the rib cage. Cutting through the bone above the trachea exposed the rest of the trachea. A cut was made above the suture and the lung was gently lifted up by the trachea. Fine tip scissors were placed flat underneath the trachea and the lung dissection finished by cutting out the lobes and all attached tissue around the trachea. A new syringe was inserted in the cannula to inflate the lung, to ensure that the lung wasn’t punctured during the procedure. Lung lobes
where separated from each other by sutures. Holding the suture ends, each lobe was cut off carefully.
The Lobes were afterwords embedded in paraffin for histological analysis.
3.7 GSH-Immunoprecipitation
S- glutathionylated proteins were immuno-precipitated (IP) using an anti-glutathione (GSH) antibody (Virogen, Watertown, MA) followed by Western Blotting of PKM2 (Cell Signaling, Baverly, MA). For this procedure cells were harvested using the GSH IP lysis buffer. This buffer contains: 50 mM Tris-HCl (pH = 7.4), 150 mM NaCl, 1% NP-40, 0.2% SDS, 0.2% CHAPS, 20 mM N- ethylmaleimide (NEM), 100 uM DTPA, 100 uM Neocuproine, 1% Protease inhibitors, 1% Phosphatase inhibitors (Sigma- Aldrich Co.). Cells were plated at a density of 2x105 cells and harvested after a 24- hour treatment with 50µM TLK-117. Protein concentration was determined and a total of 200 µg protein was used for the IP procedure. As a control one protein sample (200µg of protein) was incubated with DTT (final concentration of 50mM) (Sigma- Aldrich Co.), for 1 hour at 4 ˚C. IP- samples were over night incubated with 2 µl GSH primary antibody at 4˚C. In the next step Protein G Agarose beads (50-60 µl beads for each sample), were prepared by washing and centrifugation at 20,000xg for 30s. GSH IP lysis buffer was subsequently added; this step was repeated for a total of 3 times. After the preparation, beads were added to each sample tube. Sample tubes than were incubated for 2 hours on a rotator at 4˚C and centrifuged at 14,000 xg for 30 sec. Supernatant was aspirated and 500 µl GSH IP lysis buffer was added, the samples were spun down and the supernatant was removed, this was repeated once. Samples were prepared for SDS-Page and western blotting. Therefore 20 µl of GSH IP lysis buffer and 7µl of 4x leamli buffer were added to the protein bound beads. Cell lysate samples were separated by SDS-Page for Western blotting. Therefore a Bio-Rad Mini-PROTEAN®
Electrophoresis System (Bio-Rad) was used. Proteins were transferred on a PVDF membrane for 1 hour at 100 V. To identify the impact of TLK-117 on S- glutathionylation of PKM2, the PKM2 specific primary antibody, was used. A secondary peroxidase labeled rabbit antibody (GE, Pittsburg, PA, USA) was used for detection. The antibody was visualized by enhanced chemiluminescence with LumiGLO chemiluminescent substrates (Thermo Scientific Inc.)
3.8 Protein Harvest and analysis (whole cell lysate)
For protein experiments cells were plated at a density of 2x105 cells per 35 mm cell culture dish (Falcon by Fisher Scientific). Cells were treated with TLK-117 or transfected with GSTP1 siRNA
dependent on the experiment. For the protein harvest a whole cell lysate extract lysis buffer was used.
This buffer contains: 20 mM tris pH 7.4, 150 mM NaCl, 1 % Nonidet (NP-40), ddH2O, 1mM DTT (Sigma- Aldrich Co, St. Louis, MO, USA), Phosphatase inhibitor cocktail and a Protease inhibitor cocktail. For each 35 mm dish 200µl of lysis buffer was added and cells were scraped with a cell scraper (Sarstedt, Newton, NC, USA). Lysates were transferred to eppendorf tupes (Eppendorf, Hauppauge, NY, USA). After a 30-minute incubation on ice, cells were centrifuged for 30 minutes by 14000 rpm at 4˚C. After the spin 150 µl of the supernatant was transferred to a new tube and 50 µl of 4x laemli buffer was added. To assess protein concentration a detergent compatible protein assay was used (Bio-Rad, Hercules, CA, USA), following the manufacturer’s instructions.
3.9 Lung tissue homogenization and Western blotting
Lung tissues for the protein analysis were prepared by mincing lung tissue in liquid nitrogen to powder. Lung powder was added into the cold RIPA buffer immediately followed by freeze- thawing samples at -80˚C. RIPA buffer contains: 20mM Tris-HCl (pH=7,4), 130mM NaCl, 1mM EDTA, 1% NP- 40, 0.05% Natriumdioxycholate and 0,1% SDS. After the freeze-thaw step, lung lysates were centrifuged for 30 min at 14,000 g. A portion of the supernatant was saved for protein determination, before the addition of Laemli sample buffer. Total protein, was assessed by the Bio-Rad DC Protein Assay kit (Bio-Rad). Total G6PD, PFKP, HKI, HKII, PKM2, PFKFB3, GSTP1 and β-actin protein abundance was evaluated by Western blotting.
3.10 Lactate assay
For the lactate assay cells were plated at a density of 2x105 cells and harvested after a 24-hour treatment with 50µM TLK-117. To determine intracellular and extracellular lactate in those cells, 500µl cell culture medium was collected and cells were washed wit PBS. After the washing step, 200µl RIPA buffer was added to the cell culture dishes and the scraped cells were transferred to eppendorf tubes.
In the next step 200 µl of cell culture medium and/or cell lysates were transferred to Amicon Ultra -0.5 ml centrifugal filters with a molecular weight cut-off of 10 kD (EMD Millipore, Billerica, MA, USA).
Samples were centrifuged at 10000xg for 30 min. Supernatant was collected to perform the lactate assay using the L-Lactate Assay Kit II (Eton Biosciences, Charlestown, MA, USA), following the manufacturer’s instructions.
3.11 GST-activity assay
To assess GST activity in A549 and H460 cell samples, cells were either treated with TLK-117 or transfected with GSTP1 siRNA. Cells were plated at a density of 2x105 cells and harvested after at the indicated time points. Cells were harvested by washing cells twice with PBS and scraping them by adding 200µl GSH lysis buffer consisting of 100mM Potassium Phosphate Monobasic, 100mM EDTA, 1 mM GSH (added fresh prior to experimental use), pH 6.5 (Sigma- Aldrich Co.). Samples were centrifuged at 10,000x g for 15 min at 4°C. The supernatant was collected and use for the assay.
Before starting the procedure protein concentration of the cell samples is determined. Samples were prepared in a final volume of 100uL with GST Assay buffer (for duplicate) including a blank sample containing GST Assay buffer alone. GSH was added to each sample at a final concentration of 5 mM.
To detect GST activity a CDNB substrate mix was used. This substrate mix, which consists of 1 mM CDNB in GST Assay buffer (+GSH), was added to each 50µl sample, to initiate the reaction of the GST enzyme with the CDNB substrate. The assay was carried out in a 96 well plate (Bio-Rad), reading the absorbance once every minute at 340 nm using a plate reader to obtain at least 10 time points.
3.12 Migration/Invasion assays
To examine cell migration and determine whether TLK-117 can affect growth and migration a Scratch assays was carried out. For the Scratch assay cells were plated at a density of 2x105 cells per dish. Confluent A549, H460, H292, H1437 and HLCB1 cell monolayers were scratched using a sterile 200µl pipette tip, and the closure of the scratched areas were monitored for at least 72 hours, dependent on the closure rate. Scratches were analyzed by microscopy. For a better visualization of the scratchs, cells were stained with the May Grunewald- Giemsa staining (Sigma- Aldrich Co
3.13 Quantification of gene expression by real-time quantitative reverse transcriptase PCR
To assess gene expression in CCSPrtTA/tetON-Cre/LSL-KRASG12D (KRASG12D) or CCSPrtTA/tetON-Cre/KRASWt (KRASWt) mice lung lobes were harvested and RNA was isolated with the RNeasy Mini Kit (Qiagen, Valencia, CA, USA), according to the manufacturer’s instructions. cDNA was synthesized from 1 ug of RNA using oligo dT-primers and M-MLV Reverse Transctiptase (Promega BioSciences, LLC, San Luis Obispo, CA, USA). Quantitative PCR using SYBR Green (Bio-
Rad, Hercules, CA) was carries out, to assess gene PKM2, H6PD, G6PDX, HKII, HKIII, Glut1, Glut2, Glut3, PFKF, PFKL, LDHA, LDHB PDHKI and PFKFB3 mRNA expression. The Quantitative PCR
experiments were performed on the C1000 Thermal Cycler (CFX96 real-time system) (Bio-Rad). The following PCR program was used: Pre-incubation for 3 minutes at 95 °C followed by 40 cycles of denaturation at 95 °C for 10 seconds, annealing at 50 °C for 10 seconds and amplification at 72 °C for 30 seconds. Followed by a melting curve at 95 °C for 10 seconds 65 °C for 5 seconds and 95˚C for 5 sec. Primer sets used in this study are listed in the appendix. Expression values were normalized to the housekeeping gene beta-actin and calculated with the ddCt method.
3.14 Assessment of cancer development and tumor growth
Lungs harvested form mice were inflated to 25 cm H2O and fixated with 4% paraformaldehyde in PBS, followed by paraffin imbedding. Tissue blocks were cut into 5-µm sections and mounted to slides. Tissue histopathology and GSTP1 levels were assessed by a Hematoxylin and Eosin staining and immunohistochemistry staining (IHC) for GSTP1 and protein s-glutathionylation (pssg-staining).
3.15 GSTP1 immunohistochemistry
GSTP1 staining was performed on harvested lung sections after antigen retrieval during an 20- minute incubation in 0.01 M sodium citrate (pH 6.0) at 95˚C. Tissue slides were blocked with normal goat serum (Vectastein ABC-AP Kit, Vector laboratories, Burlingame, CA, USA) for 30 minutes accompanied by over night incubation with GSTP1 monoclonal rabbit antibody against GSTP1 (1:1000 dilution) at 4˚C. Biotinylated universal antibody was added and slides were incubated for 30 min at room temperature. Additionally slides were incubated with avidin-biotin complex-alkaline phosphatase (Vector Laboratories) for another 30 min at room temperature. Following slides were rinsed in distilled water and the substrate Vector Red (Vector Laboratories) was added and incubated for 20 minutes. After a color change in tissue to an intense red, slides were counter stained with Mayer’s hematoxylin. Tissue slides were mounted and cover slipped. Images were acquired on a EVOS xl Core microscope (Life Technologies) and analyzed using a 10x magnification.
3.16 Immunofluorescence staining for S-Glutathionylation
To determine protein s-glutathionylation in human lung tumor and normal tissue, 106 TMA’s obtained from biopsies performed in the University of Vermont Medical Center, were stained with the
pssg-staining. Tissue blocks were cut into 5-µm sections and mounted to slides and prior to the staining deparafinized. To deglutathionylate and label S-glutathionylated proteins, free thiol groups were blocked with 40mM NEM in HEM (HEN (25mM Hepes, 0.1mM EDTA, 0.01mM Neocuproine) and 1 M NEM in DMSO), with 1% Triton and HRP Streptavidin (dilution 1:2000), for 1 hour on room
temperature. After blocking, slides were washed three times for 5 minutes with PBS. The tissues were deglutathionylated by adding the GRX reaction mix (137 mM Tris-HCl pH=7.5 (Sigma), 0.1875 mM EDTA pH= 8.0 (Sigma), 50 mM NaCl (Sigma) and 100mM stocks of GSH and NADPH reduced (Sigma), to prepare the reaction mix 10 µl/ml of these stock solution was added). Slides were washed three times for 5 minutes with PBS. The newly exposed thiol groups were labeled with 1mM MBP in HEN buffer for 1 hour at room temperature. Slides were washed again 3 times for 5 minutes is PBS.
Streptavidin conjugated Alexaflour-647 (diluted 1:400 with PBS) was added and slides were incubated at room temperature for 1 hour. After a 1-hour incubation slides were washed with PBS as described before. The tissues were counter stained with DAPi (diluted 1:4000 in PBS) for 10 minutes at room temperature. Tissue slides were washed, mounted and cover slipped. Images were acquired on a Zeiss CSM 510 META Laser scanning Imaging System microscope (Carl Zeiss Microscopy, LLC, Thornwood, NY, USA) and analyzed using a 10x magnification.
3.17 Statistical analysis
Statistical analyses were performed using Graphpad Prism software (GraphPad, San Diego, Ca, USA) using ONE-Way ANOVA and the student T-test for multiple comparisons. All experiments were conducted at least two times and data is presented as mean values plus the standard error of the mean.
4. Results
4.1 GSTP1 expression in NSCLC cell lines
Little is known about the involvement of GSTP1 in lung tumorigenesis. First we determined protein expression of GSTP1 in A549, H460, H292, H1437 and a patient cell line (HLCB1), obtained from a lung biopsy. As expected GSTP1 basal protein expression varied among cell lines. The highest GSTP1 expression could be observed in A549 and H292 cells, moderate expression could be observed in H460 cells and low expression in HLCB1 and H1437 cells (fig. 6A). Before testing TLK- 117 in vitro we tested whether TLK-117 effectively inhibited GSTP1. GSTP1 activity inhibition was measured by using a recombinant GSTP1 protein. 1 µg of recombinant GSTP1 was incubated with two different TLK-117 doses 50µM and 100µM. A dose dependent decrease was observed following TLK administration. Additionally we determined GST activity in cell lysates from in A549, H460, H292, H1437 and HLCB1 cells that were treated with 50µM TLK-117 for 24-hours and re-stimulated 6 hours prior to the harvest. TLK-117 had a significant inhibitory effect on GST activity only in A549 cells. In all other cell lines there was a trend towards a decrease in GST activity, which was not significant (fig.
6C). Several studies have shown that despite of the presence of an abundant amount of oxygen, a large amount of lactate is produced in cancer cells, also known as the Warburg effect or aerobic glycolysis. Lactate production is an important hallmark of cancer cell metabolism. Intracellular lactate expression significantly decreased in A549 cells and trended towards a decrease in H292 cells (fig.
6D). In H460, H1437 and HLCB1 cells intracellular lactate expression seemed not affected (fig. 6D),
Figure 6: GSTP1 expression and the effect of TLK-117 on NSCLC cell lines. A) Basel GSTP1 protein expression in untreated cells. B) Determination of the effect of TLK-117 on GSTP1 and enzyme activity. *** p<0.02, *p<0.01 (One-Way ANOVA and Turkeys multiple comparison) compared to respective control group. C) GST activity measured in cells treated with TLK-117 or control. D) Measurement of lactate in cells treated with TLK-117 or control medium (with 0.02%
DMSO). Insert: *ONE-Way ANOVA p<0.05 compared to respective control group of untreated cells.
A
B
A549 H460 H292 H1437 HLCB1
GSTP1
β-‐ Actin
C
D
0 2×10-5 4×10-5 6×10-5 8×10-5 1×10-4
GST activity (µM/µg/min)
*
A549 H460 H292 H1437 HLCB1
TLK117 (50µM) Cntrl
0 10 20 30 40 50
Int ra ce llula r la ct at e c onc ent ra tion ( uM /µ g)
A549 H460 H292 H1437 HLCB1
*
Cntr TLK-117
(50µM) TLK-117
(100µM) 0.00
0.05 0.10 0.15
Activity (µM/µg/min)
*
***
rec GSTP1
4.2 Pharmacological GSTP1 inhibition in NSCLC cell lines
To determine the effect of TLK-117 on cancer cell migration, cells were treated with 50 µM TLK- 117 every 24 hours during a time course of 72 hours. Migration was measured with a scratch wound assay by applying scratches to cell monolayers of A549, H460, H292, H1437 and HLCB1 cells.
Scratches were photographed and measured to assess cell migration. In A549 and H292 cells as shown in figure 7 A/C and F, migration was significantly reduced by TLK-117. In H460, H1437 and HLCB1 cells TLK-117 cell migration was not affected. Respective migration of NSCLC cells is quantified in figure 7F.
4.3 Protein S-
glutathionylation (PSSG) in a NSCLC cell line
In order to determine the potential role of GSTP1 in catalyzing PSSG in KRAS induced tumorigenesis we isolated total protein and RNA from A549 cells, a human epithelial lung cancer cell line, with a KRASG12D mutation following pharmacological inhibition of GSTP1.
Cells were exposed to 10, 50 and 100 µM TLK-117 for 24 hours. To determine the effect of TLK-117 on overall S- glutathionylation of proteins in A549 cells, SDS–PAGE analysis was performed under non-reducing conditions and GSH-conjugated proteins were assessed using an anti-GSH antibody. We observed a decrease in total S-glutathionylation of
D
Control TLK-‐117
A549
H460
H292
H1437 A
B
C
HLCB 1
E
F
Figure 7: Impact of TLK-117 treatment on cell migration. A) A549, B) H460, C) Human cells obtained from a patient with NSCLC, D) H1437 cells, and E) H292 cells treated with 50µM TLK-117 or vehicle control. F) Scratch assay carried out with A549, H460, H292, H1437 and primary Human patient cells treated with TLK-117 or control (cell culture medium with 0.2%
DMSO). Cell migration was quantitatively evaluated by measuring the distance between the scratch edges. A549 (migration of control is 56.5% and TLK-117 treated cells is 27.03%), H460 ((migration of control is 56.5% and TLK-117 treated cells is 27.03%). Insert: *Student T test p<0.05 compared to respective control group of untreated cells.
0 20 40 60 80
Migration (%) *** CntrTLK-117
(50µM)
A549 H460 H292 H1437 HLCB1
**
proteins in A549 cell when treated with 50µM TLK-117 (fig. 8A). In addition following immunoprecipitation of GSH we observed a decrease in S-glutathionylation of PKM2. (fig. 8B). GSTP1 protein abundance was not affected by TLK-117 treatment. Actin was examined to ensure equal loading.
4.4 KRAS
G12Dinduced tumorigenisis
To further investigate the role of KRAS induced tumorigenesis, the role of GSTP1 on its involvement in the Warburg effect and metabolism we developed a triple transgenic mouse expressing CCSP-rtTA/tetON-Cre/LSL-KRASG12D. Mice were fed doxycycline over a time course of 6 weeks (fig.
11A). Following doxycycline administration H&E staining showed numerous tumor formations in the lung of the KRASG12D mice. In control mice no tumors were observed (fig 9B). An important cancer hallmark is the production of lactate. We assessed lactate production in mice expressing the oncogene KRASG12D and compared them to mice lacking the oncogene. Lactate production was significantly increased in KRASG12D mice compared to KRASWt mice (fig. 9C). Next we evaluated the influence of KRASG12D on gene expression
of
metabolism-associatedgenes. Quantitative analysis of gene expression demonstrated that PKM2, HKII, HKIII, Glut2 and PFKFB3 were significantly up regulated in KRASG12D mice (fig. 9 D/E). Only LDHA gene expression was significantly down regulated. Evaluation of protein expression by Western blot revealed an increase of proteinsFigure 8 S-glutathionylated proteins in A549 cells. Each lane represents 100 µg total protein extracted. (A) Western blot analysis with a GSH antibody. (B) A549 cell lysates were immune-precipitated using anti- GSH antibody. Blots were probed with antibodies against PKM2, GSTP1 and actin.
associated with metabolism (PKM2, HKII, PFKFB3, G6PD, PD and GSTP1), in mice expressing KRASG12G could (fig. 9F).
4.5 Pharmacological GSTP1 inhibition in KRAS
G12Dinduced tumorigenesis
The earlier presented data suggests that pharmacological GSTP1 manipulation with TLK-117 effects GST activity, lactate production and migration in human cancer cells, which possibly is regulated by s- glutathionylation of PKM2.
To determine the effect of TLK-117 on protein expression KRASWt and KRASG12D mice where treated with 50mg/kg TLK- 117 or vehicle (1.M Tris-HCl, 0.2% DMSO), every three days for the duration of 4 weeks (fig.10A). TLK-117 administration in AdCre activated KRASG12D mice did not change the expression of metabolic associated proteins (fig. S1). There were no significant differences in protein expression of GSTP1, Pyruvate Kinase M2 (PKM2), Pyruvate Dehydrogenase (PD), 6-phosphofructo-2 kinase/fructose-2,6- biphosphatase 3 (PDKFB3) and Glucose- 6-poshpate dehydrogenase (G6PD), all important mediators in the Glycolysis and Pentose Phosphate Pathway. We have previously shown that following ablation Figure 9 KRASG12 induced tumorigenesis in C57BL6 mice. A) Schematics
showing the experimental set up after DOX administration. B) Histopathological analysis of lung section from KRASWT and KRASG12D mice by H&E staining (magnification: x10). C) Analysis of lactate expression in BALF, obtained from KRASWT and KRASG12D mice. D/E. Quantification of mRNA levels for metabolic genes (PKM2, H6PD, G6PDX, HKII, HKIII, Glut1, Glut2, Glut3, PFKF, PFKL, LDHA, LDHB PDHKI and PFKFB3) in lung tissue homogenates by q-PCR.
Results are expressed as fold change compared with WT and reflect SEM from 1 independent experiment (n = 7 Wt, 3 KRASG12D). F. Protein analysis for GSTP1, PKM2, PD, HKII, G6PD and PFKFB3 by SDS-Page and Western blot, for equal amounts of protein (20 µg). Actin was used as a loading control.
Insert: *p < 0.05 (Student T test) compared with respective controls.
Day 0 W 1 W 2 W 3 W 4 W 5 W 6 Day 42
A Dox
KrasWT KrasG12D 0.0
0.5 1.0 1.5 2.0 2.5
Lactate concentration from BALF (mM)
****
D C
B
E
0 1 2 3 4
Relative Expression
PKM2 H6PD G6PDX HKII HKIII Glut1 Glut2
**
***
*
*
0 1 2 3 50 100
Relative Expression
Wt KrasG12D
Glut3 PFK PFKL LDHA LDHB PDHK1 PFKFB3
***
****
GSTP1
KRAS
WtKRAS
G12D10x
HKII
Actin GSTP 1 PKM2 PD
PFKFB 3
G6PD
KRAS
WtKRAS
G12DF
of GSTP1, A549 cells (harbor a KRASG12D mutation) expressed less G6PD, PD and PFKP (fig. S2). In Adcre induced KRASG12D we did not observe a decrease in G6PD and PD protein levels. Next we evaluated tumor formation by analyzing lung histopathology, using the Hematoxylin and Eosin staining to visualize physiological differences in KRASWt (control group) and KRASG12D mice. Histological evaluation of the lung sections demonstrated that KRASG12D mice treated with TLK-117 showed less tumors tissue compared to control treated animals. No tumor burden was observed in control mice (fig 10B). In addition to these findings we investigate the effect of TLK-117 on lactate production and GST-activity in vivio. As previously mentioned,
TLK-117 decreased lactate production and GST-activity in human lung cancer cell lines.
However evaluation of GST activity in lung lysates of TLK treated animals did not show differences (fig10C).
4.6 GSTP1 in NSCLC
It is reported that GSTP1 expression is significantly elevated in human tumors. We created tissue micro arrays (TMA) of 106 patients with adenocarcinomas. Tumor tissue and adjacent normal tissue
Figure 10: Intratracheal delivery of TLK-117 in KRASG12D induced tumorigenesis. A) Schematics showing the TLK-117 treatment regime given after AdCre delivery. Mice were harvested after 6 weeks (n=3). B) Mouse lung sections from wild type and KRASG12D mice, instilled with TLK-117 or the vehicle, after staining by Hematoxylin and Eosin (magnification: x10). C) GST activity measured form mouse lungs from animals (KRASWT, KRASG12D) treated with TLK-117 or vehicle.
A AdCre
Day 0
Treatment every 3 days with 25mg/kg TLK-‐117
W 1 W 2 W 3 W 4 W W TLK-‐117
Day 14 Day 32
B
TLK-‐117 Control
KRAS
G12D10X C
KRAS
Wt0.0000 0.0001 0.0002 0.0003
GST activity (µM/µg/min)
KrasWT KrasG12D
Cntr TLK-117 (50µM)
PSSG or GSTP1 expression and PSSG. TMAs representing the 106 individual patients with NSCLC were visually scored for GSTP1 expression. A representative picture is shown in figure 11A of a negative and a positive patient for GSTP1 expression and PSSGH. Scoring the TMAs revealed that 41% of the tumor tissue was positive GSTP1 expression (fig.11C). In 47 % of the cases, tumor tissue was scored negative for GSTP1 expression. 12% of the tumor tissue stained intermediately for GSTP1. In normal tissue only 5% stained positive for GSTP1 (fig.11B). Interestingly PSSG correlated with GSTP1 status of the tumor tissue (fig. 11A).
C
Total=106
4.72% GSTP1 positive 86.79% GSTP1 negative 8.49% GSTP1 intermediate
Normal tissue
Total=106
40.57% GSTP1 positive 47.17% GSTP1 negative 12.26% GSTP1 intermediate
Tumor tissue
Normal tissue
B
Tumor tissue
GSTP1
A
HLCPS1 HLCPS2PSSG DAPI
HLCPS1 HLCPS2
Figure 11: Non-Small cell lung carcinoma tissue micro array. A) IHC staining for GSTP1 expression (left). Fluorescence staining for protein s-glutathionylation (PSSG) (right). B) Percentages of normal tissue from patients, positive or negative for GSTP1 expression (n=106). C) Percentages of tumor tissue, positive or negative for GSTP1 expression (n=106).
5. Discussion and Conclusion
The aim of this study was to examine the role of GSTP1 in KRAS induced tumorigenesis and its involvement in cancer metabolism. Here we demonstrated that GSTP1 protein levels are differentially expressed between several human lung cancer cell lines. Interestingly there appears to be an association between the levels of GSTP1 expression and the sensitivity to TLK-117, a GSTP1 inhibitor. In cells with a higher abundance of GSTP1, TLK-117 reduced the lactate production and migratory capacity, in contrast, cells with low GSTP1 expression were not sensitive to TLK-117 treatment. Conflicting results where found by Zhang et al., which described that genetic ablation of GSTP1, resulted in an increase in myeloproliferation and migration [36]. Similarly it has been observed that TLK199 stimulated cytokine-induced myeloproliferation in wild type mice but not in GSTπ–/– mice [37]. A possible explanation is that GSTP1 has distinctive roles in specific cell types. Our findings suggest that GSTP1 inhibition and the use of TLK-117 could be applicable in reducing cell migration, cancer progression and the formation of metastases. GSTP is a member of a larger family of Glutathione S-transferases, and is well known for its role in phase II drug metabolism, as it catalyzes conjugation of GSH to electrophiles. GSTP is overexpressed in a variety of cancers [38,39], where it has been associated with resistance to chemotherapeutic drugs [40]. GSTP1 can also act as a modulator of signal transduction and has recently been discovered to catalyze protein s- glutathionylation (PSSG) reactions. Our data suggest that TLK-117 inhibits overall glutathionylation in cells with high GSTP1 expression and more specifically TLK inhibits PKM2 glutathionylation in A549 cells. Since PKM2 is an important regulator of the Warburg effect, the Pentose Phosphate Pathway and PKM2 appears to be a potential target for GSTP1 induced PSSG, the findings in this study highlight the possible role and contribution of GSTP1 and PKM2, in glycolysis and involvement in human cancers [26,27,34]. In addition to its role as a catalyst of PSSG, GSTP1 also acts as a ligandin [41,42]. GSTP1 can bind to a number of proteins to alter their function. For example, GSTP1 can bind to c-Jun-N-terminal kinase (JNK) and the adaptor protein, TRAF2 to inhibit their activation [12]. This can be an additional potential mechanism to elucidate the effect observed with TLK-117 but was beyond the scope of this study.
Lung cancer is a disease that encompasses many different cell types, and is thus impossible to entirely recapitulate in cell culture models. In order to elucidate a connection between these two
processes, we will need to utilize the whole mouse as a biological system to investigate if genetic manipulation is a possible form of lung cancer treatment. There are several ways to induce cancer and specific NSCLC in experimental animals. Cho et al. described this inducible mouse model where they established that mice carrying KRAS mutations were highly susceptible to early onset lung cancer.
They established that bronchiolar Club cells are the origin of cells and tumorigenesis for KRASG12D- induced neoplasia in lungs. [43] Correspondingly in approximately 20-30 % KRAS is frequently mutated in human tumors of the lung [44,45,46,47]. To further investigate the relevance of GSTP1 in KRAS induced NSCLC we bred a triple transgenic mouse that expresses the KRASG12G oncogene. In comparison to a xenograft model, in which human lung cancer cell lines have been subcutaneously implanted in immunodeficient mice, we utilized an inducible model of KRAS driven tumorigenesis [48].
Food containing doxycycline, led to a controlled activation of the oncogene KRASG12D, inducing tumor formation originated from CCSP positive lung epithelial cells. An immense advantage of this model is that it can recapitulate spontaneous cancer formation due to a mutation and activation of an oncogene. We investigated KRAS induced tumor formation by evaluating physiological differences of lung sections that were obtained from the CCSP-rtTA/tetO-Cre/LSL-KRASG12D. After 6 weeks, KRASG12D mice revealed the development of epithelial hyperplasia and tumor formation along the left lung lobe. After evaluating protein, mRNA expression and lactate production, we confirmed that KRASG12D expressing mice produced an excessive amount of lactate compared to KRASWt mice.
Protein expression of PKM2, PD, G6PD, HKII and PFKFB3 was increased as well as mRNA expression of PKM2, HKII, HKIII, Glut2 and PFKFB3. All these are biomarkers for cancer and are known to be elevated in human cancers [49,50,51,52,53,54]. These findings exposes the CCSP- rtTA/tetO-Cre/LSL-KRASG12D mouse model as a suitable preclinical model for intervention studies that make it possible to assess therapy responses and evaluate new therapeutic agents.
In the next part of this study we determined whether pharmacological inhibition of GSTP1 attenuated KRAS induced tumorigenesis in vivo. To examine this, we used adenovirus expressing Cre recombinase (AdCre), which induces tumor formation by activating the KRASG12D in adeno-infected cell. KRASG12D expressing mice and wild type mice were exposed to TLK-117, a selective GSTP1 inhibitor [55]. Preliminary data suggested that TLK-117 attenuated tumor growth and cancer progression in mutant KRASG12D induced mice. In addition, as shown in previous studies, epithelial hyperplasia was observed in AdCre induced KRASG12D mice, and TLK treatment appeared to
