Studying the effects of Muscone on PANC-1 in vitro
Bachelor Thesis Project in Biomedicine 30 ECTS
Spring term 2020
Kyriaki Gerontaki
a17kyrge@student.his.se
Supervisors: Ferenc Szekeres & Heléne Lindholm ferenc.szekeres@his.se & heléne.lindholm@his.se
Examiner: Anna Benrick anna.benrick@his.se
School of Health and Education University of Skövde
Address
Högskolevägen 1 PO Box 408 541 28 Skövde
Abstract
In this project cancer was viewed as a metabolic disorder. The study focused on the effect of synthetic muscone on PANC-1 cells. Synthetic muscone is a ketone. Some cancers lack the potential to metabolise ketone bodies, mainly due to mitochondrial dysfunction. The Warburg effect is a biochemical phenomenon in which cancer cells favour metabolism through glycolysis instead of oxidative phosphorylation to produce ATP. The cells were treated with various concentrations of muscone for 24 and 48 hours. Three different assays were chosen to assess the effects of the treatment. These included an MTS-Assay to estimate the cell-viability, a Lactate-Glo kit to assess the lactate production after the treatment and Caspase 3/7 assay for the apoptotic effects of the treatment. Lactate is the end product of glycolysis, thus its decrease could be an indication of an effective treatment. Instead, the lactate produced exhibited an increasing tendency. Caspase activity was increased for concentration 500μΜ, whilst no effect was noted for the remaining concentrations tested. Expression of genes involved in glycolysis (HK2, GAPDH & PKM2), apoptotic pathways (CYCS), mitochondrial function (TFAM) and ketolytic activities (BDH1 & OXCT1) were chosen to be tested for a better understanding of the results derived from the assays. The expression of CYCS was in accordance with the findings of the MTS and Caspase assays for concentration 500μΜ, making its investigation promising for future research. Due to time restrictions most experiments were conducted once, thus no safe conclusions could be reached. However, we expect this pilot study to provide valuable insight for future researchers who aim to approach cancer treatment from a metabolic perspective.
Popular scientific summary
This project aimed to study the effects of synthetic muscone (ketone) on pancreatic cancer cells. When the body does not have enough carbohydrates for fuel, it can burn fatty acids for energy. Ketones are the by-product of this process. The ketogenic diet forces the body to burn fat instead of carbohydrates as a source of energy. Glycolysis is the biochemical process of sugar breakdown (carbohydrates) and especially glucose, hence its name. Glycolysis is the first stage of cellular respiration, regardless of the presence or absence of oxygen, which means that glycolysis takes place in both aerobic and anaerobic respiration of cells. When glucose supply stops for an extended period of time and the body reaches a level of glycogen depletion all the energy is provided from fatty acids (ketones).
The relationship between ketogenic diets and cancer, both in terms of prevention and treatment, is a controversial topic in genetics. Nevertheless, it awakens much interest. The ketogenic diet in cancer is a metabolic therapy aiming at one of the most special features of the cancer cell: Its metabolism. Some cancers lack the potential to metabolise ketone bodies mainly due to mitochondrial dysfunction and down- regulation of enzymes necessary for ketones. The peculiarity that cancer cells have is that they tend to favour anaerobic glycolysis over oxidative phosphorylation. This phenomenon is known as the Warburg Effect. Ketones are not used in glycolysis but in oxidative phosphorylation. As a result, providing cancer cells with ketones as the sole energy source could lead to the cancer cells’ death. More specifically, the pancreatic cancer cells will be treated with a substance called muscone. Muscone is derived from the gland of the musk deer and is an aromatic substance. It is the very same substance that is used in perfumes with the distinctive smell of “musk”. In this project, synthetic muscone was used which has a fair percentage of ketone in its composition. Different concentrations of the substance were administered to the cells.
One of the most important experiments was measuring the cell viability after a 24- and 48-hour treatment.
This would show whether muscone could decrease the number of cells if the number of the treated cancer cells was decreased in comparison to the untreated. Another experiment conducted was one where lactate production was measured. Lactate is usually the end product of glycolysis which is the metabolic activity cancer cells favour. According to our theory, if cancer cells were to be affected by the muscone, lactate production would be decreased since the cells would not be able to use the muscone as an energy source and consequently die. Another experiment conducted would provide a better understanding on the apoptotic activity of muscone on the pancreatic cancer cells. A cell’s programmed death is a phenomenon called apoptosis. The results indeed exhibited an increased apoptotic activity at 500μΜ muscone (caspase 3/7 experiment), indicating that the muscone probably affects the pancreatic cancer cells in that specific pathway. To further assess the effects of muscone, expressions of genes with important metabolic functions were investigated in order to see the effect of the muscone on the metabolism of the cell. For instance, the Hexokinase-2 gene that is involved in glycolysis, had an upregulated tendency indicating a potential increased glycolytic rate after the treatment. This project offers a new perspective in viewing cancer as a metabolic disease and could give rise to new therapeutic strategies that will mainly target the energy supply of cells. Given the project’s duration, the use of
muscone has not provided significant results concerning pancreatic cancer. However, it is very likely that this study will encourage both ordinary readers and researchers to think outside of the box.
Table of content
Table of Contents
Pancreatic cancer
...1
Cancer as a metabolic disease
...1
Dietary energy restriction
...2
Synthetic Muscone/Musk
...2
Genes encoding key metabolic enzymes
...3
Hypothesis and aim of the study ... 5
Materials and Methods ... 6
Cell line and culture
...6
Cell Proliferation Assay-MTS
...6
Lactate-Glo™ Assay
...7
Caspase 3/7 Assay
...8
RNA extraction and cDNA synthesis
...8
Quantitative Real Time Polymerase Chain Reaction (qRT-PCR)
...9
Primers ... 9
TaqMan ... 9
SYBRGreen ... 9
Statistical Analysis
... 10
Student’s T-test ... 10
Kruskal-Wallis test ... 10
Mann Whitney U-test ...Error! Bookmark not defined. Results ... 11
MTS-Assay
... 11
Lactate Dehydrogenase
... 12
Caspase 3/7 activity
... 13
qRT-PCR
... 14
A. TaqMan Gene Expression Assay...Error! Bookmark not defined. B. SYBRGreen ...Error! Bookmark not defined. Discussion ... 17
Conclusions ... 21
Ethical aspects and impact of the project on the society ... 22
Acknowledgements ... 23
References ... 24
Appendices ... 31
Appendix 1
... 31
Appendix 2
... 33
Appendix 3
... 34
List of Abbreviations
βHB 3--hydroxybutyrate
AcAc Acetoacetate
ATP Adenosine triphosphate
cDNA Complementary deoxyribonucleic acid
GAPDH Glyceraldehyde 3-phosphate dehydrogenase
ds double stranded
KD Ketogenic diet
LDH Lactate dehydrogenase
MTS 3-(4,5-dimethylthiazol-2-yl)-5- (3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium
OXPHOS Oxidative Phosphorylation
PPP Pentose Phosphate Pathway
qRT-PCR Quantitative Real-Time Polymerase Chain Reaction
RNA Ribonucleic acid
SEM Standard Error Mean
SPSS Statistical Package for Social Science
1 Introduction
Recently, there have been conflicting views on the origin of cancer. There is strong evidence challenging the somatic mutation theory of cancer. Is cancer a nuclear genetic disease or is it a mitochondrial metabolic disease? This project aimed to provide evidence for the latter through a series of experiments related to metabolic functions in PANC-1 cells.
Pancreatic cancer
Pancreatic cancer is a relatively common type of cancer caused be several risk factors. It is the 12th most common cancer in men and the 11th most common cancer in women worldwide. Pancreatic cancer is also the 7th leading cause of cancer-related deaths worldwide (Rawla, Sunkara, & Gaduputi, 2019) and has a five-year survival rate of 9%. Infiltrating ductal adenocarcinoma is the most common tumour type (Wild, Weiderpass, & Stewart, 2020). The leading identified cause of pancreatic cancer appears to be cigarette smoking while certain risk factors have been identified, such as obesity, alcohol abuse, genetic factors and Helicobacter pylori infections (Rawla, Sunkara,& Gaduputi, 2019; Wild, Weiderpass, & Stewart, 2020).
Cancer as a metabolic disease
Emerging evidence suggests that cancer is originally a metabolic disease that results from the disruption of energy production through the processes of respiration and fermentation. The cancer cells’ genomic mutability as well as the hallmarks of cancer are considered to be secondary effects of the primary disorder of cellular energy metabolism (Fouad, & Aanei, 2017; Seyfried et al., 2013). These findings are opposed to the dogma that cancer is a genetic disease (Wishart, 2015). Normal cells obtain the majority of their functional energy through oxidative phosphorylation. Conversely, studies show that most cancer cells are dependent on substrate-level phosphorylation (Chinopoulos, & Seyfried, 2018). Substrate-level phosphorylation refers to the formation of ATP from ADP and a phosphorylated intermediate, rather than from ADP and inorganic phosphate, Pi, as is occurs in OXPHOS (Chinopoulos et al., 2010). The ATP generated by glycolysis is rather low, yet anaerobic glycolysis is relatively rapid and occurs in an environment that lacks oxygen.
The cancer cells undergo anaerobic glycolysis (Warburg effect) in order to speedily produce ATP (Tan-Shalaby, 2017). The reason for this is that cells have dysfunctional mitochondria and maybe electron transport chain defects, which disrupt normal ATP production from the mitochondria. This results in cancer cells becoming greatly dependent on ATP coming from the less efficient process of glycolysis (Tan-Shalaby, 2017). A general hypothesis currently under the scope is that all hallmarks of cancer, including anaerobic glycolysis -the Warburg effect-, can be linked to the damaged mitochondrial function and energy metabolism (Seyfried et al., 2013). If cancer is to be viewed as a metabolic disease, it could impact approaches to novel cancer therapy and prevention tools.
2 Dietary carbohydrate restriction
Dietary energy restriction provides the cell with energy derived from the catabolism of fats. Ketogenic Diet (KD) increases the circulating levels of fatty acids and ketone bodies shifting the focus from breaking down carbohydrates to breaking down fats for energy (Mahoney, Denny, & Seyfried, 2006). Fatty acids, and especially ketone bodies β-hydroxybutyrate (bHB) and acetoacetate (AcAc) derived from fatty acids, can replace glucose as a primary metabolic fuel under caloric restriction (Cahill, 1970). During fasting and starvation, ketones are produced from fatty acids in the liver. The ketones can be used as an alternative form of energy for normal cells. On the contrary, most cancer cells cannot benefit from the products of oxidative phosphorylation because of mitochondrial dysfunction (Chinopoulos et al., 2010). Once the ketones βΗΒ, AcAc and acetone are released into the bloodstream, they are taken up by the brain and other organs. They are then transferred into the mitochondria and used as fuel. Excess βΗΒ and acetoacetate are excreted in urine, while acetone is breathed out. Meanwhile, the glucose levels in the blood remain physiologically normal due to gluconeogenesis (Tan- Shalaby, 2017). KDs target the Warburg effect. The Warburg effect is a biochemical phenomenon in which cancer cells favour metabolism through glycolysis instead of oxidative phosphorylation to produce ATP (Alfarouk et al., 2014). Some cancers lack the potential to metabolise ketone bodies, mainly due to mitochondrial dysfunction and down-regulation of enzymes necessary for ketone exploitation (Tan-Shalaby, 2017). The levels of circulating ketones differ between populations of normal individuals. Most researchers agree that normal serum levels of ketone bodies can be defined as <0.5 mM. Levels higher than 1.0 mM are in indication of light nutritional ketosis (hyperketonemia) and optimal nutritional ketosis can be achieved at levels 3.0 mM to 5.00mM (Stralfors, Olsson, & Belfrage, 1987).
Synthetic Muscone/Musk
Over the years, muscone has been used in traditional Chinese medicine and in some clinical practices (Xu,
& Cao, 2014). There is evidence that various types of cancer are sensitive to musk treatment (Xu, & Cao, 2014; Qi et al., 2020). Muscone is a ketone and is derived from the gland capsule of a musk deer. The effects of native musk and synthetic musk in cancer cells are the same, thus the use of synthetic muscone is preferred (Xu, & Cao, 2014). A most recent study reveals that muscone could be potentially used as an anticancer drug (Qi et al., 2020). The study of gene expression can be related not only to a genetic background but to a metabolic one as well. One of the key ketolytic enzymes that could be affected in cancer cells is the one encoded by OXCT1 since it is involved in ketone catabolism by transferring the CoA moiety from succinate to acetoacetate. Ideally, all the genes encoding the key enzymes for the glycolytic steps would be tested, but all the pathways are interconnected, hence the choice of different genes from various pathways instead of the investigation of all the genes involved.
3 Genes encoding key metabolic enzymes
Shown in Figure 1, a schematic diagram of the metabolic pathways that allows us to envision all the genes investigated and their potential involvement with each other.
Figure 1. Glycolysis and fatty acid metabolism. The green rectangles indicate the genes encoding for the enzymes responsible for the corresponding pathways.
Hexokinase–2 (encoded by HK2) catalyses the first step of the glycolysis where the glucose ring gets phosphorylated and produces glucose-6-phosphate by adding a phosphate group to a molecule derived from ATP (Robey, &Hay, 2006). Moreover, deletion of HK2 has been shown to decrease cancer cell proliferation without particular side effects in animal models, which suggests that targeting HK2 is a viable strategy for cancer therapy (Garcia, Guedes, & Marques, 2019). On the other hand, overexpression of HK2 is an indication of an increased glycolytic rate in cancer cells (Mathupala, Ko, & Pedersen, 2006).
Interactions once thought to be nessecary between HK2 and the mitochondria are now crucial for tumour survival. Mitochondrial-bound HK2 is considered to act as gatekeeper and facilitator of the malignant condition in many cancer cells (Mathupala, Ko, & Pedersen, 2006). HK2 appeared to be highly expressed in pancreatic cancer metastases, proposing a link between HK2 as well as the aggressive tumour biology
4
and a possible correlation to the PANC-1 cell line related to this project (Chang, Olson, & Schwartz, 2013;
Anderson et al., 2017).
Pyruvate kinase M2 isoform (encoded by PKM2 gene) is a cytosolic glycolytic enzyme which catalyses the phosphoenolpyruvate conversion to pyruvate. It is one of the most important rate-limiting enzymes of glycolysis together with HK2 and has been proven to be over-expressed in many kinds of cancers (Altenberg, & Greulich, 2004).
TFAM is encoded by the TFAM gene. It is a mitochondrial transcription factor A thus a key activator of mitochondrial transcription as well as a participant in mitochondrial genome replication. This suggests that it could be indicative of the number of mtDNA copies thus its downregulation would indicate fewer mtDNA copies (Tan-Shalaby, 2017). TFAM is not prognostic in pancreatic cancer but it is indicative of the mitochondrial function and can provide valuable insight on energy production and how it is affected after the ketone treatment (Uhlén et al, 2015).
Cytochrome C (encoded by CYCS) is one of the three mitochondrial DNA encoded subunits of the complex IV in the electron transport chain. The electron transport chain is part of the pathway for synthesis of ATP.
Cytochrome C plays an important role not only in the electron transport chain but is also involved in apoptosis by regulating the suppression of anti-apoptotic members or activation of some pro-apoptotic members that lead to altered mitochondrial membrane permeability resulting in the release of cytochrome c into the cytosol (Tafani et al., 2002; Kim et al., 2000).
D-beta-hydroxybutyrate dehydrogenase (BDH1), encoded by the BDH1 gene in human, is a mitochondrial enzyme that catalyses the interconversion of acetoacetate and (R)-3-hydroxybutyrate, the two major ketone bodies produced during fatty acid catabolism (Chang, Olson, & Schwartz, 2013b). AcAc and pyruvate are reduced, using BDH1 and LDH in the presence of excess NADH, to beta-hydroxybutyrate and lactate, respectively (Nuwayhid, Johnson, & Feld, 1989). This gene is broadly expressed in liver, colon and small intestine tissue, but it usually appears to be highly expressed in glioma cells (Fagerberg et al., 2014;
Maurer et al., 2011).
Succinyl-CoA:3-ketoacid-coenzyme A transferase, encoded by OXCT1. Key enzyme required for oxidation of ketone bodies (catabolism). The expression of the gene is indicative of the activity in the ketolytic pathway and the rate of ketolysis (Cotter et al., 2013).
Initially, GAPDH was the housekeeping gene of choice. Recent studies suggest that GAPDH is involved in glycolysis and has been localised in the mitochondria (Tristan et al., 2011) with a context-dependent function. Thus, it was not used as a reference gene but was assessed for its participation in the treatment.
GAPDH can bind with the voltage-dependent anion channel (VDAC), which may promote the release of Cytochrome C causing apoptosis. A decrease in mitochondrial membrane potential leads to caspase- independent cell death when the cell is under stressed conditions (Wang, & Youle, 2009). In this regard, GAPDH can inhibit cell death by simultaneously boosting glycolytic activity leading to increased ATP production and stimulating autophagy-mediated clearance of permeabilised mitochondria (Tristan et al., 2011).
Reference gene PMM1: As a housekeeping gene PMM1 was chosen. Housekeeping genes are expressed in all cells under almost any conditions. The key to their choice is their low variability under experimental and control circumstances (Greer et al., 2010).
5
Hypothesis and aim of the study
The study of muscone is an innovative approach to treating cancer. This project aimed to study the effects of synthetic muscone (ketone) on pancreatic cancer cells. The effects were studied on a molecular as well as on a cellular level with an emphasis on apoptosis, gene expression, cancer mitochondrial dysfunction related to glycolysis and provision of energy. More specifically, the cells were treated with muscone and the cell viability was investigated with an MTS assay. Furthermore, glucose metabolism was assessed via a lactate production assay. A caspase assay was used to investigate the apoptotic activity on the cells of caspases 3/7 since caspase 3 is considered to be involved in most apoptotic pathways. Finally, the gene expression for GAPDH, CYCS, TFAM, BDH1, OXCT1, HK2 and PKM2 were investigated because these genes are key activators in glycolysis, apoptosis and mitochondrial functions. We hypothesised that the administration of muscone to pancreatic cancer cells would have a cytotoxic effect due to their inability to use ketones as a source of energy which can be attributed to their mitochondrial dysfunction.
6 Materials and Methods
Cell line and culture
The PANC-1 cell line was derived from a 56 year-old Caucasian man who had pancreatic carcinoma. The cell line originated from ductal cells and they are adherent in cell culture (#87092802, Sigma-Aldrich, USA).
Cells were grown in Dulbecco Modified Eagle Medium - DMEM (#D5796, Sigma Aldrich, UK) with 10% fetal bovine serum (#D5796, Sigma Aldrich, UK), 1% penicillin, streptomycin (Sigma Aldrich). In order for the cells to multiply they had to be split into different culture flasks. They were washed with 3ml phosphate buffered solution (#7059, Sigma-Aldrich, USA) and were added 3.5ml Trypsin (#T4049, Sigma-Aldrich, USA) to dissociate the adherent cells from the cultivating vessel. A brief incubation at 37°C followed until the cells were floating. To inactivate the trypsin, 3.5ml of DMEM were added and the cells were centrifuged at 1000rpm for five minutes. The cell pellet was resuspended with 1ml of DMEM. 1ml of the resuspended culture was added to two flasks containing 25ml of fresh medium. To seed the cells, cell count was performed. 40μl of the resuspended culture was mixed and stained with trypan blue solution 0.4%
(#15250061, GibcoTM, Thermo Fisher Scientific, US) to perform manual cell count. 5μl of this mixture were used to fill each chamber of the microscope slide and the cells were observed under 10X magnification.
The cells were cultivated to expand the number of cells and split only when confluency was equal to/or more than 80%. Thus, change of media and/or passaging occurred three times a week. All cells were incubated at 37°C and 5% CO2.
Cell Proliferation Assay-MTS
For checking the viability of the cells the CellTiter 96® AQueous One Solution Cell Proliferation Assay-MTS (#G3582, Promega, USA) was used. The MTS reagent contains a tetrazolium compound. Principle is that the cells utilise the yellow tetrazolium salt which is metabolized by mitochondrial succinic dehydrogenase activity of proliferating cells to yield a purple formazan product by the mitochondria of the viable cells.
The amount of formazan product as measured by the amount of 490nm absorbance is directly proportional to the number of alive PANC-1 cells in the culture. For the seeding, a clear 96-well plate was seeded with 100μl media and 5.000 cells/well. The treatment started the day after the initial seeding, to provide the cells with time to grow and adjust to the new environment. The plate was incubated for 24hrs at 37°C. 24hour and 48hour treatment followed. The concentrations of muscone tested were 0.1, 0.5, 1, 1.5, 2.5, 5, 10, 50, 100, 250 and 500 (μΜ). The muscone was dissolved in pure ethanol, so the concentrations of ethanol affecting the cells in the higher muscone concentrations were assessed. These concentrations were referred to in percentage (%) of their effect in the corresponding muscone concentration (i.e for muscone concentration 250μΜ, 0.25% of that concentration was pure ethanol) (Appendix 1, Figures A, B). For all the muscone concentrations tested, two columns (sixteen wells) were used as technical replicates and one column of eight technical replicates with untreated cells (control). In
7
that way, all wells contained the same media and different concentrations of muscone. Any differences observed would be attributed to the muscone since the media factor remained constant (controlled experiment). For the experiments regarding the ethanol concentrations, eight wells/concentration were used and eight wells/corresponding muscone concentration /plate. After 24/48hours of treatment, 20μl of MTS were added to each well and 4 wells were used as negative control containing only DMEM used by cells and MTS (#G3582, Promega, USA). The negative controls were needed for measuring the background signal to background ratios. One hour of incubation at 37 °C followed before recording absorbance at 490nm with the 96-well plate reader FLUOstar Omega (BMG Labtech). Two measurements of each plate were performed, after one and two hours. For the results only the first measurement was included.
Lactate-Glo™ Assay
The Lactate-Glo™ assay (#J5021, Promega, USA) is a bioluminescent assay for measuring the enzymatic activity of L-Lactate. Bioluminescence is a form of chemiluminescence where light energy is released by a chemical reaction (Vacher et al., 2018). In this case, the chemical reaction occurring starts with the lactate dehydrogenase catalysing the oxidation of lactate producing pyruvate, following the reduction of NAD+ to NADH. In the presence of NADH, a pro luciferin reductase substrate is converted by Reductase to Luciferin.
Luciferin is detected in a luciferase reaction using Ultra-GloTM rLuciferase and ATP which is causing the light emission. The luminescent signal is proportional to the lactate present in the sample. Lactate is the end product of glycolysis and the pentose phosphate pathway (Rogatzki et al., 2014). A large amount of lactate is an indication of increased glycolytic activity. 5.000 cells/well were seeded in a 96-well plate for most of the different concentrations of muscone with two replicates/concentration. 5μl from every concentration were retrieved after 24, 36 and 48 hours of treatment. They were diluted 20times in PBS to a final volume of 100μl and stored at -22°C. 5 μl from already diluted samples were further diluted 10times in PBS to a final volume of 50 μl. In the end, they were added to a final volume of 100μl in the wells of a white 96-well plate (2-fold dilution), 1:1 ratio with the Lactate Reagent (Luciferin detection solution and to a final dilution of 400 of the initial sample was performed. The use of a white 96-well plate was because it offers maximum light reflection and augments the luminescent signal. The plate was incubated at room temperature for 60mins before luminescence was recorded with the 96-well plate reader FLUOstar Omega (BMG Labtech).
A standard Lactate curve was created to establish the relationship between various lactate concentrations and their luminescence. The concentrations chosen for the creation of the standard curve were in the range 0-200μΜ (12.5, 25, 50, 100, 150, 200 μΜ). 200μΜ was proposed as the maximum according to the protocol. For these dilutions, a 2-fold serial dilution was performed and an extra dilution for the 150μΜ concentration which was needed for more cohesion between the concentrations 200μΜ and 100μΜ. For negative controls, the wells contained only buffer and the Lactate reagent. The negative controls were needed for measuring the background signal and calculating signal to signal background ratios.
8 Caspase 3/7 Assay
Caspase 3 is activated in an apoptotic cell by extrinsic and intrinsic (mitochondrial) pathways. The precursor of caspase7 is cleaved by caspase 3 and activated upon cell death stimuli inducing apoptosis.
Caspase 3 is important in most apoptotic pathways and its activation is a crucial event in apoptosis, hence the choice of this assay (Katunuma et al., 2001). The assay detects solely the activity of caspases 3/7 while other ways inducing cell death such as necrosis, autophagy or apoptosis triggered by the activation of other caspases still remain unexplored.
The caspase 3/7 activities were measured with CellEvent™ Caspase - 3/7 Green Detection Reagent (#C10423, Invitrogen, Thermo Fisher Scientific, USA). CellEvent Caspase-3/7 Green Detection Reagent is intrinsically non-fluorescent. For that DEVD peptide is responsible by inhibiting the ability of the dye to bind to DNA. However, after activation of caspase-3/7 in apoptotic cells, the DEVD peptide is cleaved, enabling the dye to bind to DNA and produce a fluorescence signal. The fluorescent emission of the dye when bound to DNA is ~530 nm and can be observed using a standard FITC filter set. For this experiment excitation was measured at 530nm and emission at 503nm.
For this Assay 5.000 cells/well were seeded in a black 96-well plate with three technical replicates for each concentration. Two wells were used for negative control containing just uncoloured RPMI-1640 Medium (#7509, Sigma- Aldrich, Sweden). The black well plates were used because they are not transparent and therefore do not let any light spread within the walls of the wells which otherwise would be interfering with the fluorescent signal. They also reduce background noise and crosstalk. After a 48 hour treatment, the diluted caspase detection reagent (5 μM) was added to three replicates and the plate was incubated for 60 minutes at 37°C. Fluorescence was recorded at 530nm (excitation) and 503nm (emission) with the 96-well plate reader FLUOstar Omega (BMG Labtech). Apoptotic cells with activated caspase-3/7 will have bright green nuclei and will be emitting a strong fluorescent signal.
RNA extraction and cDNA synthesis
For the RNA extraction a 6-well plate was seeded with 150.000 cells per well to a final volume of 2.5ml and incubated at 37 °C in 5% CO2. After 24 hours the cells were treated with the concentrations 1, 10, 250, 500μΜ of muscone and one well was left untreated (including only new media). The 6-well plate was incubated for 48hours at 37 °C. The RNA from the supernatant was extracted following the RNeasy Plus Mini Kit manual (#74134, Qiagen, Germany). RNA quantification purity was measured through spectrophotometric analysis of absorbance at 260 and 280 nm, using the ND-1000 Spectrophotometer. A ratio of ~2.0 is generally accepted as pure for RNA samples (Thermo Fisher Corporation). If the ratio is noticeably lower, it may suggest the presence of protein, organic compounds or other contaminants that absorb strongly near 280 nm. The purity values are presented in Appendix 3. The RNA samples were all diluted to the same final concentration of 50ng/μl but some concentrations contained more than 50ng/μl as observed from the reference gene expression. Nonetheless, the same RNA was used for all the qPCR experiments and all concentrations were normalised to the reference gene. RT-PCR was performed following the protocol for the High Capacity cDNA Reverse Transcript Kit (#4368814, Applied Biosystems™, Thermo Fisher Scientific, Lithuania). For the cDNA samples a final concentration of 25 ng/ul was created including 10μl of Master Mix and 10μl of RNA. The cDNA samples were further diluted to a final
9
concentration 5ng/μl and were used for all the gene expression experiments. All samples were stored at -22°C.
Quantitative Real Time Polymerase Chain Reaction (qRT-PCR)
Both TaqMan and SYBR Green have advantages and disadvantages. TaqMan provides higher specificity, reproducibility and levels of quantification than SYBR Green, whereas SYBR Green is relatively cost efficient and easy to use (TaqMan vs. SYBR Chemistry for Real-Time PCR, Thermo scientific Corporation). SYBR Green is technically based on binding the fluorescent dye to double-stranded deoxyribonucleic acid (dsDNA). The TaqMan method is more expensive and is based on dual labelled oligonucleotide and exonuclease activity of Taq polymerase enzyme. Specificity is the most important concern with the usage of any nonspecific dsDNA-binding Dyes (SYBR Green) but through optimization of primers and design, its performance and quality could be comparable to TaqMan method (Tajadini, Panjehpour, & Javanmard, 2014).
Primers
The primers were chosen according to base composition (GC-content 40-60%), melting temperature (less than 5°C between the forward and reverse primers) and the fact that 3 ́terminal sequences should not be complementary to the other primer (avoiding self-complementary sequences that result in primer dimer).
For primer sequences see (Appendix 3)
TaqMan
For the genes PMM1, GAPDH, Cytochrome C & TFAM validated TaqMan probes were used (Applied Biosystems Corporation). All probes were composed by quenchers in the 3’ and 5’ ends of the oligonucleotides and the fluorescence emitted was the one quantified. Probe was diluted to 1:40 and 2μl/
well were used. The probe evaporated when placed in an incubator at 40°C for approximately one hour to keep the volume to 2μl. The TaqMan™ Gene Expression Master Mix used was from Applied Biosystems,
#4369016. Final volume in each well was 2μl (1μl Master Mix & 1μl cDNA 5ng/μl). Samples were run in triplicates. The qPCR plate (PikoReal, Thermo scientific Corporation) was also vortexed and centrifuged for 1 min at 1500rpm. For the relative quantification of gene expression Livak’s method was used (Livak, &
Schmittgen, 2001).
SYBRGreen
Primers for the genes HK2, PKM2, BDH1 & OXCT1 were diluted to a final concentration of 200nΜ in each reaction. cDNA of 5ng/μl and 2.5μl of Master Mix (2X) were eventually added to a final volume of 5μl/
reaction. The negative controls (NTC) did not contain any cDNA.
10 Statistical Analysis
Student’s T-test
IBM, SPSS Statistics, Version 25 was used for all the data gathered from the Cell proliferation MTS-Assay.
The concentrations were assessed individually for their normal distribution. For the assessment of significance, the two tailed Student’s T-Test was applied with a significance set at α= 0.05. A p-value less than 0.05 (typically ≤ 0.05) was set to be statistically significant. Because multiple comparisons took place, the Bonferroni correction was performed. The statistical significance was set at α’=0.05/ (No. of comparisons) = 0.05/10= 0.005. The null hypothesis stated that the samples did not differ significantly and any difference noted was generated by chance.
Inferential error bars (Standard Error SE)
For the Lactate-Glo Assay, the Caspase Assay 3/7 and the gene expressions, assessment of normality could not be performed because of the small sample size(n=1). Non parametric tests were also not performed because of this reason. The experiments were repeated only once due to time restrictions. The SE can be used to understand if there is a difference between groups when the sample size is very small (Cumming, Fidler, & Vaux, 2007). The mean of the data, M, with SE can indicate the region where the mean of the whole possible set of results, or the whole population, μ, can be expected to lie. The space between the SEM can define the values that are most plausible for the whole population, μ, but cannot return significant values (Cumming, Fidler, & Vaux, 2007).
11 Results
MTS-Assay
In Figure 2a concentration 50μΜ exhibited a significant decrease in cell viability after 24 hours of treatment. Concentration 2.5μΜ had no effect and the rest of the concentrations tested showed a tendency to decrease the cell viability without, however, significant effects.
In Figure 2b concentrations 5μΜ and 500μΜ appeared to significantly affect the cell viability whereas the rest of the concentrations appeared to not affect the cells.
By observing the two panels, it can be said that the cell viability was boosted after 48 hours and followed the effect the concentrations seemed to have in panel a. No general descending effect was observed, which was initially hypothesised (the higher the muscone concentration, the higher the effect).
Figure 2. Cell viability after treating PANC-1 cells with different concentrations of muscone for 24 hours (a) and 48 hours (b). The results are expressed as the relative viability of the cells compared to the control. All values are presented as the mean of the data gathered. Unpaired T-Test was performed and Bonferroni correction followed.
Significance was set at α=0.005 (*p< 0.005, ***p< 0.0001.) Cell counts were performed and are shown as mean
± SEM.
12 Lactate Dehydrogenase
A general tendency for increase of lactate production is noticed as the treatment time passes. The raw data were normalised to the cell viability derived from the MTS-Assay results presented in Figure 2. The data gathered from the Lactate-Glo Assay with n=1 were not adequate to perform an accurate non- parametric test.
For the 24 hour treatment (Figure 3, orange colour), the only cells that exhibited a slightly lower tendency to produce lactate, when compared to the untreated cells, were the 2.5μΜ and the 5μΜ. The rest of the cells either produced the same lactate as the untreated (0.1μΜ & 1 μΜ) or the lactate production tended to be boosted after the 24hour treatment (0.5μΜ, 10μΜ, 100μΜ, 250μΜ & 500μΜ).
For the 36 hour treatment (Figure 3, green colour), the lactate production appeared to present the average of the results presented in 24hours and 48hours of treatment. Strangely enough, the cells treated with muscone concentration 5μΜ tended to produce less lactate after 24hours and later on a considerable inclination towards increased lactate production was observed that continued throughout the 48hour assessment as well. Concentration 100μΜ exhibited a tendency for increase at 36hours that did not follow a specific trend when compared to the lactate production at 24 and 48hours.
For the 48hour treatment (Figure 3, black colour), the cells treated with concentration 100μΜ exhibited a lower tendency to produce lactate in comparison to the lactate produced by untreated cells. The cells treated with the rest of the concentrations tended to produce more lactate than the untreated ones.
Unfortunately, no general pattern could be discerned making the inkling of the existence of type II errors even more likely. On the other hand, concentrations 5μΜ and 500μΜ exhibited a high inclination towards increased production of lactate after 48hours.
Figure 3. Lactate production of the PANC-1 cells after having been treated with different concentrations of muscone for 24, 36 and 48 hours respectively. The results are presenting the lactate production of the treated cells compared
13
to the untreated. The values presented are the mean of this outcome (n=1/sample concentration normalised to the MTS values) + SEM.
Caspase 3/7 activity
Figure 4 presented the relative apoptotic activity triggered of the treated cells compared to the control/untreated. The Assay was run with three technical replicates (triplicates) per concentration and was performed once (n=1). The raw data were normalised to the cell population derived from the MTS-Assay results presented in Figure 2b (& Appendix 1, Figure B). The results showed that the caspases 3 and 7 were most probably activated for concentration 500μM. The rest of the concentrations exhibited no effect. This information is in agreement with the results exhibited in Figure2b showing that concentration 500μM had a significant effect on the cell viability.
No statistical test was performed due to the small sample size. Any comparison was accomplished through the interpretation of the SEM according to Cumming, Fidler and Vaux (2007).
Figure 4. Caspase 3/7 activity of the PANC-1 cells after having been treated with various concentrations of muscone for 48 hours. The values presented are the mean of the technical replicates (n=1) normalised to the MTS values + SEM.
14 qRT-PCR
No statistical test was performed due to the small sample size. For every gene expression assessed, n=1 with three technical replicates for every concentration tested. Any comparison performed was accomplished through the interpretation of the Inferential Error Bars, SEM, according to Cumming, Fidler and Vaux (2007).
A. TaqMan Gene Expression Assay
For the gene expression assessment of Figure 5, a TaqMan Gene expression Assay was used. The values presented are the mean of the average Ct (n=1)/sample concentration and include + SEM. PMM1 appeared not to be as affected from the treatment but concentration 500μΜ tended to be slightly upregulated in comparison to the control. This could be due to excess amount of cDNA while performing the dilutions.
Nonetheless, PMM1 was chosen to be the reference gene because of its low variability and its easy and quick access in the lab.
Figure 5. Average Ct values of PMM1 + SEM of the PANC-1 cells after having been treated with various concentrations of muscone (1, 10, 250 and 500 μΜ) for 48 hours.
15
In Figure 6, GAPDH in panel a and Cytochrome C in panel b tended to be mostly downregulated. The obvious SEM interval of concentrations 1, 10 and 500 μΜ indicated the possibility for a strong difference between these concentrations and the controls in both panels. TFAM in panel c exhibited a general tendency for downregulation at all concentrations with intervals between the SEM of 250 and 500 μΜ and the SE of the control, indicating that this ought to be the most probable region where the rest of the mean of the future data could probably lie.
Figure 6. Fold change of GAPDH (a), Cytochrome C (b) and TFAM (c) of the PANC-1 cells after having been treated with different concentrations of muscone (1, 10, 250 and 500μΜ) for 48 hours and detected with TaqMan Gene expression assay. The values presented are the mean of the three technical replicates for each sample concentration (n=1) and include + SEM.
16
B. SYBRGreen
Figure 7 showed the expression of four genes, detected with SYBRGreen. (a) HK2 appeared not to follow a specific pattern of upregulation or downregulation. Interestingly, the highest concentration (500μΜ) tended to have a completely different expression from the lowest (1μΜ). (b) PKM2 and BDH1 (c) tended to be generally less expressed than the control with concentration 500μΜ (panel b) looking very promising for further research. OXCT1 (d) exhibited a tendency for upregulation at concentration 250μΜ where the SEM of this concentration seemed to be deviating a lot from the SEM of the control.
Figure 7. Fold change of four genes HK2 (a), PKM2 (b) and BDH1 (c) and OXCT1 (d) of the PANC-1 cells after having been treated with various concentrations of muscone (1, 10, 250 and 500μΜ) for 48 hours. All gene expressions are relative to the corresponding control group. The values presented are the mean of the three technical replicates for each sample concentration (n=1) and include + SEM.
17 Discussion
Considering the cell viability assay; while trying to infer which concentrations of muscone were best to use for the treatment, it was concluded that the effect of any set of ascending concentrations used was dose dependent (the higher the concentration – the higher the effect). Surprisingly, Figure 2 (cell viability) suggested that the muscone treatment tended to be more biphasic than dose-dependent. Muscone treatment has been characterised biphasic in the past while treating CNS diseases (But, & Chang, 2001). A major drawback when a treatment appears to be biphasic is the relative lack of flexibility when it comes to administration. Since not all concentrations provided a significant outcome, it could not be claimed that the nature of muscone was biphasic in tumour cells. However, it would be an interesting research question in the future.
Following the most recent study of the Qi et al. (2020), the initial concentrations chosen to treat the cells were very low (0.1 μΜ – 2.5 μΜ) since they seemed promising to give the desired effect. Assessing the cell viability of PANC-1 cancer cells through the MTS-Assay suggested otherwise. The treatment had almost no effect but for concentrations 0.1 μΜ – 2.5 μΜ a descending tendency was observed rather than the ascending effect presented in the study. This was a very striking contrast that could be attributed to the small sample size. For concentrations higher than 10μΜ muscone appeared to not follow a linear trend.
Hence, concentrations up to 1mM were used in order to visualise the differences depicted in Figure 2 for both treatment times (24- and 48-hour treatments respectively). With the choice of the higher concentrations, an additional test had to be conducted to assess if any tendency on the cells was a result of the synthetic muscone or the ethanol the muscone was dissolved in. The results proved that for the 24hour treatment, the tendency presented was due to the muscone treatment because the concentrations of ethanol tested had no effect on the cells and followed the trend of the untreated ones (Figure A, Appendix 1).
Although Qi W. et al. conducted the study on two different hepatic cell lines and this study was on the PANC-1 cell line, both cell lines appear to have some similarities. PANC-1 cells are known to have an epithelial morphology and are adherent in cell culture flasks and also have a slower doubling time (52hours) than HepG2 (Deer et al., 2010; Knowles, & Aden, 1983). HepG2 cells also exhibit an epithelial- like morphology, have adherent properties and grow as monolayers in small aggregates, just like PANC-1 cells (Knowles, & Aden, 1983).
Further verification of the cells’ phenotype was also performed regularly by observation under the microscope (Figure C, Appendix 1) -when high concentrations were involved- to assess the cells’ viability and potential to offer correct results. That is to say, for the highest concentrations used, 750 &1000 uM (data not presented), the cells died completely and could not generate results.
Figure 2b presented the cell viability relative to the control after 48hours of treatment with muscone.
There were no available studies comparing an extensive 48hour treatment with muscone to compare the results with. Expectation would suggest a decreased cell viability after 48 hours. However, the population
18
doubling time is 52 hours which could be interfering with the expected results regardless of the normalisation (Deer et al., 2010). That being said, the results presented in Figure 2b were not very promising and exhibited an inclination towards increased cell viability. Figure 4 results were also derived from samples treated for 48hours. Figure 4 exhibited only a possible caspase 3/7 activation at 500μΜ that seemed to agree with the presented results in Figure 2b for the corresponding concentration. The apoptotic effect of concentration 500 μΜ though, appeared to be because of the ethanol rather than the muscone (Figure B, Appendix 1). Overall, the MTS-assay showed a steady decrease in cell viability for the concentrations 5 and 500 μΜ which was verified by the probable increased apoptotic tendency assessed with the caspase-assay only for concentration 500μΜ. Concentration 5μM was not assessed for its caspase 3/7 activity because at the time it was not considered of importance. On the other hand, concentration 250μΜ seemed ineffective after the 48-hour treatment and this was in agreement with the results from the caspase, where the activity was most probably decreased considering the SEM interval. Additionally, the lactate production was tended to remain the same as the lactate produced from the untreated cells (Figure 3).
Concerning the rest of the experiments conducted, statistical tests were not performed due to the small sample size (n=1). Nevertheless, studies prove that SEM can be used to assess probable significance if the experiments were to be repeated (Cumming, Fidler, & Vaux, 2007). Additionally, other studies have claimed that while reporting significance is important, it does not have to be correlated to biological importance and vice versa (Alderson, 2003; Hewitt, Mitchell, & Torgerson, 2008; Guyatt et al., 2011). In other words, statistical significance should not be confused with the size or importance of an effect. Since statistical significance was not provided due to the small sample size, the discussion focused on the importance of the results and the reason behind their occurrence as well as their potential to offer insight in future research.
The Lactate-Glo™ Assay performed measured the extracellular activity of L-Lactate. As described in the methods part, the luminescent signal emitted was proportional to the lactate present in the sample up to a certain concentration (200uM). Lactate is the end product of glycolysis and the pentose phosphate pathway (Rogatzki et al., 2014). Thus, a large amount of lactate would indicate an increased glycolytic activity, which is what should be observed in cancer cells, since they tend to favour glycolysis over oxidative phosphorylation (Warburg effect). If the amount of lactate produced was significantly reduced, it could possibly mean that the cancer cells were dying because they were unable to use glycolysis for energy provision. However, a novel approach concerning the Warburg effect suggests that tumour cells can actually switch to OXPHOS due to lactic acidosis, where low glycolytic activity can be observed (Wu, Ying, & Hu, 2016). Expression of genes involved in both glycolytic and ketolytic pathways was assessed to provide a broader understanding of how the lactate production was affected from the muscone treatment.
Figure 3 showed a general tendency for increased lactate production with increased treatment time. A problem with the experiment conducted was the fact that the samples were run in duplicates and not triplicates. Furthermore, it was conducted only once due to its high cost. These would imply that the experiment was underpowered statistically with type II errors indicating that false conclusions could be drawn. For instance; assuming that the lactate production was high when in reality it was not and vice versa. Because of the small sample size, the data were not assessed with a non-parametric test. To provide as reliable results as possible only the standard error bars were used for group comparisons. Since the cell
19
viability was decreased for concentrations 5 and 500 μΜ, the lactate production should also appeared to be reduced if muscone affected the PANC-1 cells in a negative way (de la Cruz-López et al., 2019; Le et al., 2010). A possible reason behind the tendency for increased lactate production in Figure 3 is discussed below.
The effect of the 500μΜ concentration was of interest considering the cell viability presented in Figure 2a and Figure 2b and lactate production in Figure 3 for the 24hours and 48hours. In Figure 2, 500μΜ exhibited apoptotic tendency on the PANC-1 cells with a significantly greater apoptotic effect after 48hours. As discussed above, the cytotoxic effect concerning the 48hour treatment was largely due to the effect of the ethanol rather than that of the muscone (Figure B, Appendix 1). Additionally, this was probably the reason behind the extremely increased trend in the apoptotic activity in Figure 4 (500μΜ). In Figure 3, the extracellular lactate accumulation seemed to have an increasing tendency with the hours. According to the Warburg effect, if the cancer cells’ glycolysis was to be affected by the treatment, there would be less production of lactate. In that event, Figure 2 suggested fewer viable cells for concentrations 5 and 500 μΜ with a tendency for increased caspase 3/7 activity (Figure 4, 500μΜ), but Figure 3 presented a very probable increase in lactate production. From the results derived, the muscone (ketone) did not seem to affect the lactate production negatively and to further affect the rate of glycolysis in a negative way. A possible explanation behind this could be that the lactate increase was due to ketones taking over the process of lactate oxidation (Pan et al., 2000). Oxidised lactate results in higher levels of pyruvate which then enters the mitochondrion and is metabolised in the TCA-cycle (Figure 1). In return, more lactate is produced without changing the rate in glycolysis (Pan et al., 2000). In light of these, because the study from Pan et al. (2000) was generally about energy provision on healthy cells on KD, further explanation for the lactate production had to be associated with the PANC-1 cells. This was achieved by assessing gene expression through qRT-PCR. Some of the genes assessed were involved in activities such as glycolysis and mitochondrial regulation because these activities are indicative in tumour cells (San-Millán, & Brooks, 2017).
Generally it should be taken into consideration that every gene expression measured the mRNA of the gene tested and not the actual protein levels. The correlation between mRNA and protein was estimated at 40% depending on all the procedures between transcription and translation such as transcriptional regulation, RNA splicing, RNA in correlation to protein stability and protein modification (Vogel, &
Marcotte, 2012; Arvas et al., 2011).
Figure 6a presented the gene expression of GAPDH and Figure 7a & 7b presented the fold change in HK2 and PKM2 respectively. Recent studies suggest GAPDH localisation in the mitochondria with a context- dependent function (Tristan et al., 2011). GAPDH has been associated with the release of Cytochrome C, causing apoptosis but it also encodes for an important rate limiting enzyme in glycolysis as shown in Figure 2, together with the HK2 and PKM2. All three genes are encoding key enzymes in glycolysis. HK2 is responsible for the catalysation of the first step in glycolysis (Robey, & Hay, 2006). HK2 general tendency for upregulation (Figure 7a) provided strong indications towards increased glycolysis resulting in extracellular lactate accumulation as the results from the Lactate-Glo reveal.
According to Berg, Tymoczko, & Stryer, (2002) increased ethanol leads to accumulation of NADH which in turn will lead to the prevention of lactate oxidation to pyruvate. The increased concentration of NADH will cause the reverse reaction to dominate and will eventually lead to lactate accumulation. That could explain
20
the high tendency for upregulation of HK2 for concentration 500μΜ since its effects are mostly due to the presence of ethanol. HK2, also appeared to be upregulated in PDAC in a similar study of Wang et. al. with ketone treatment (2020). The general tendency for upregulation could occur because the cancer cells were trying to make up for the energy deprivation by boosting the activity of HK2 to catalyse the glycolysis and thus provide the cells with lactate and subsequently with energy. On the other hand, PKM2 and GAPDH tended to be generally downregulated which would indicate a probable effect of the muscone on the glycolysis pathway in the final steps of glycolysis. These results could also suggest that the cells were not in need of extra protein production. Once again, it was observed in Figures 6a, 7a and 7b that the expression of concentration 500μΜ was the one that tended to be more affected when compared to the rest, especially considering the HK2 expression (Figure 7a). These inclinations strongly indicate that ethanol was mainly responsible for the gene expressions for this concentration.
TFAM can provide valuable information on energy production and how it is affected after the ketone treatment (Uhlén et al, 2015). Downregulation of TFAM has previously indicated increased apoptotic activity according to Xie et al. (2016). Cytochrome C is part of the electron transport chain responsible for cytochrome oxidase complex in the mitochondria. TFAM and Cytochrome C (Figure 6c & Figure 6b respectively) tended to be downregulated which could be an indication of less mRNA and thus downregulation of the mitochondrial function. Mitochondria are the powerhouse of the cell and decreased mitochondrial function could lead to less energy production. With less energy provided, the PANC-1 cells’ viability would appear to be affected.
Figures 7c & 7d presented the expression of BDH1 and OXCT1 respectively. The presence of two enzymes that are encoded by BDH1 and OXCT1 are essential for the utilisation of ketones as an energy source (Zhang et al., 2018). With OXCT1 crucial for ketone body utilization and BDH1 responsible for ketolysis, their general tendency to be downregulated presented in Figure 7 could only indicate inability of the PANC-1 cells to break down the muscone and use it as an energy source (Chang, Olson, & Schwartz, 2013b; Tan- Shalaby, 2017). Whether cancer cells could utilise these ketone bodies effectively was of grave importance to the cell viability and for proving our initial hypothesis. These findings could also shed light on why the glycolytic rate was not negatively affected being the sole energy provider. The tendency for lactate accumulation revealed from the lactate assay was the outcome of the increased glycolytic rate to ensure the cells’ survival in the presence of muscone. Conversely, concentration 250μΜ exhibited an upregulation in OXCT1 indicating that, in this specific concentration, the cancer cells were able to use the muscone by oxidising it to pyruvate to gain energy from (Pan et al., 2000). These results were also in line with all the rest of the experiments conducted for the 250μΜ concentration. The caspase assay indicated that the caspases did not exhibit any tendency for activation, the lactate production tended to be slightly higher than the one from the untreated cells, implying that the cells did not create excess of lactate because they could utilise different pathways for energy since they were not affected by the muscone. The MTS (Figure 2b) showed no effect in cell viability with no significance detected. A similar rationale applied for concentrations 0.5, 1, 10, 50, 100 μΜ.
21 Conclusions
Overall, the treatment with muscone on pancreatic cancer cells was interesting but returned mixed results.
Concentration 500μΜ of muscone seemed to work in the experiments. The PANC-1 cells were significantly reduced (5μΜ, 500μΜ) however, the levels of caspase activity were not affected with the exception of concentration 500μΜ providing strong inclination for the cells’ death. Unfortunately, this observation was largely due to the ethanol the muscone was dissolved in. Additionally, the lactate production exhibited a tendency for increase and yet that did not prevent the cells from dying (500 μΜ). Moreover, all the gene expressions for the aforementioned treatments tested were in line with the rest of the experiments. They appeared to be connected with the findings of the MTS, lactate and caspase assays. For the OXCT1 and BDH1 expressions, inclinations from their tendency for downregulation provided strong suspicions towards the inability of the cells’ to oxidise the muscone and use it as energy source. The tendency for downregulation of TFAM and CYCS also indicated a mitochondrial dysfunction leading to less production of energy. Finally, HK2 tended to be upregulated suggesting a probable increased glycolytic activity because of the increased need for energy. Taking everything into account, the gene expressions were promising but did not provide any proof to verify out hypothesis.
Under a ketogenic diet, anyone can achieve “light nutritional ketosis” at the levels of 1.0 mmol/L - 3.0 mmol/L. From there, they can further continue to achieve “optimal ketosis”, which is when the ketone levels are between 3.0 mmol/L - 5.0 mmol/L. This suggests that a treatment with a concentration of 0.01 up to 0.5mM muscone, could be accepted by the human body without posing a threat to it physiologically. A good start would be to investigate the gene and protein expression at 500μΜ of all the genes involved in glycolysis and especially the ones with ketolytic properties and assess how much of the expression is attributed to the ethanol effect. Another suggestion would be to dissolve the muscone in another solvent (i.e DMSO). A common suspicion verified through this study was that everything works synergistically in the cell (normal or tumour). This makes it extremely difficult to trace the source of the problem -assuming it is one problem- responsible for the energy provision.
Due to the probable biphasic nature of muscone, it would be better to emphasise the future research on specific concentrations that work, which in this case appear to be concentrations 5μΜ and 500μΜ. Then again, not all the concentrations were assessed in all the different assays used, making it possible for the concentrations not mentioned here also to be effective. Nonetheless, previous studies focused on very small concentrations and their effect after 24hours, which is something we suggest future researchers take into consideration. Finally, hepatic cell lines might be of particular interest, since previous studies proved that muscone worked in treating hepatic cancer cells. Investigating hepatic cell lines can also shed light on whether the muscone treatment is directly related to the physiology of the cell line because glycolysis occurs in the liver. Consequently, the use of muscone might be more promising for treating hepatic cancer instead.
22 Ethical aspects and impact of the project on the society
The cells used in the project were bought from a reputable company that is enrolled with the European Collection of Authenticated Cell Cultures operated by the Public Health of England. This contributed to the mitigation of cross-contamination of the cell line and the risk for misidentification. The purchase of a cell line from a cell-bank presupposes that there has been informed consent of the volunteer patient/participant. That is to say that the cell donor was aware of the future and potential use of the cells and their contribution to cancer research (Millum, & Bromwich, 2013). The donor is always provided with full disclosure by the cell-bank (Millum, & Bromwich, 2013) which also covers the part of privacy and confidentiality according to the Declaration of Helsinki (World Medical Association, 2018). The cells were handled with extreme care and were inspected daily under the microscope. Additionally, the cells were utilised to the maximum and never underwent splitting without serving a cause. With these methods followed, very little space was allowed for mistakes and cells’ misuse. Standard operating procedures had been established for all routine procedures from the beginning of the project. On top of these, media preparation and cell handling occurred completely isolated in the sterile hood. The media and reagents were purchased from trustworthy sources.
This project was in line with the three ethical Rs. Despite the fact that there was no direct contact with animals, synthetic muscone was used over native muscone (which is derived from the musk deer) for the treating of the cells (refinement).
The evaluation of the theory that cancer may be a metabolic disorder can lead to the rise of new therapeutic strategies that will shift the standard chemotherapeutic approach to a metabolic approach.
With this new approach, therapies would involve mitochondrial enhancement, nutritional ketosis and cancer metabolic drugs. Furthermore, this pilot study can prepare a promising ground for the development of effective anticancer drugs derived from natural compounds. However, native musk is a very rare and precious natural drug. On the other hand, synthetic musk ketone can substitute for native musk to treat cancer patients˙ and this study provided results indicating that synthetic musk ketone may be a promising anti-cancer drug for concentrations of 0.5mM if diluted in another solvent.
There is an important downside if cancer is to be viewed as a metabolic disorder which is the financial aspect of the approach. Seyfried et.al has previously stated that by increasing the levels of ketones in the body, cancer can be eliminated significantly (2014). Such a thing could affect the pharmaceutical companies immensely since chemotherapy options would be considered as secondary. Seyfried and his team have though claimed that chemotherapy should not be paused when trying to increase the ketone levels in the body, but also that the genetic mutations of the cancers are secondary causes (epiphenomena). Ideally, most of the cancer patients could start by changing their diet to ketogenic and then use chemotherapeutic drugs that target the same pathways as the KD (i.e. monoclonal antibodies).
This would obviously have a grave effect on the finances of the pharmaceuticals. According to a recent study concerning one of the biggest pharmaceuticals in the United States of America, the average cost of developing a cancer drug is estimated at $720 million (Prasad, & Mailankody, 2017). The average annual revenue is about $2.7 billion. Within just one year, the annual revenue of the top five drugs —which treat
23
lymphoma, prostate, leukemia, and colorectal cancers—covers the total costs of research and development (Prasad, & Mailankody, 2017). On the other hand, future research in this field could benefit individuals since cancer management can start by ketone supplements at a cost of around 30$ and if taken according to the recommended daily dose, no danger is imminent.
Acknowledgements
I would like to thank my supervisors Ferenc Szekeres and Heléne Lindholm for their guidance, support and useful insight into my work.
24 References
Alderson P. (2003). Survey of claims of no effect in abstracts of Cochrane reviews. BMJ, 326 (475)
Alfarouk, K. O., Verduzco, D., Rauch, C., Muddathir, A. K., Adil, H. H., Elhassan, G. O., Harguindey, S. (2014).
Glycolysis, tumour metabolism, cancer growth and dissemination. A new pH-based etiopathogenic perspective and therapeutic approach to an old cancer question. Oncoscience, 1(12), pp. 777–802.
Altenberg B, Greulich KO. (2004). Genes of glycolysis are ubiquitously overexpressed in 24 cancer classes.
Genomics, 84(6), pp.1014–1020.
Anderson, M., Marayati, R., Moffitt, R., & Yeh, J. J. (2017). Hexokinase 2 promotes tumor growth and metastasis by regulating lactate production in pancreatic cancer. Oncotarget, 8(34), pp. 56081-56094.
Arvas, M., Pakula, T., Smit, B., Rautio, J., Koivistoinen, H., Jouhten, P., Lindfors, E., Wiebe, M., Penttilä, M.,
& Saloheimo, M. (2011). Correlation of gene expression and protein production rate - a system wide study.
BMC genomics, 12, pp.616.
Berg MJ, Tymoczko JL, Stryer L., (2002). Biochemistry New York: W H Freeman. ISBN-10: 0-7167-3051-0
But, P. and Chang, H., (2001). Pharmacology and Applications of Chinese Materia Medica. New Jersey:
World Scientific.
Cahill G. F., Starvation in man. (1970). the New England Journal of Medicine. 282(12), pp. 668-675.
Chinopoulos, C., Gerencser A. A., Mandi, M., Mathe, M., Töröcsik, B., Doczi, J., Turiak, L., Kiss, G., Konràd, C., Vajda, S., Vereczki, V., Oh, R. J., Adam-Vizi, V. (2010). Forward operation of adenine nucleotide translocase during F0F1-ATPase reversal: critical role of matrix substrate-level phosphorylation. The FASEB Journal. 24(7).
Chinopoulos, C., & Seyfried, T. N. (2018). Mitochondrial Substrate-Level Phosphorylation as Energy Source for Glioblastoma: Review and Hypothesis. ASN neuro, 10 (1-27).
25
Chang, H. T., Olson, L. K., & Schwartz, K. A. (2013b). Ketolytic and glycolytic enzymatic expression profiles in malignant gliomas: Implication for ketogenic diet therapy. Nutrition and Metabolism, 10(1), p.47.
Cotter, D. G., Schugar, R. C., Wentz, A. E., d'Avignon, D. A., & Crawford, P. A. (2013). Successful adaptation to ketosis by mice with tissue-specific deficiency of ketone body oxidation. American journal of physiology.
Endocrinology and metabolism, 304(4), pp. E363–E374.
Cumming, G., Fidler, F., Vaux, D.L. (2007). Error bars in experimental biology. J Cell Biol. 177 (1), pp. 7–11.
de la Cruz-López KG, Castro-Muñoz LJ, Reyes-Hernández DO, García-Carrancá A, Manzo-Merino J. (2019) Lactate in the Regulation of Tumor Microenvironment and Therapeutic Approaches. Frontiers in Oncology.
9, pp.1143.
Deer, E. L., González-Hernández, J., Coursen, J. D., Shea, J. E., Ngatia, J., Scaife, C. L., Firpo, M. A., &
Mulvihill, S. J. (2010). Phenotype and genotype of pancreatic cancer cell lines. Pancreas, 39(4), pp. 425–
435.
Fagerberg, L., Hallström, B. M., Oksvold, P., Kampf, C., Djureinovic, D., Odeberg, J., Habuka, M., Tahmasebpoor, S., Danielsson, A., Edlund, K., Asplund, A., Sjöstedt, E., Lundberg, E., Szigyarto, C. A., Skogs, M., Takanen, J. O., Berling, H., Tegel, H., Mulder, J., Nilsson, P., … Uhlén, M. (2014). Analysis of the human tissue-specific expression by genome-wide integration of transcriptomics and antibody-based proteomics.
Molecular & cellular proteomics : MCP, 13(2), 397–406. https://doi.org/10.1074/mcp.M113.035600
Fath MA, Simons AL, Erickson J, Anderson ME, Spitz DR. Enhancement of cancer therapy using ketogenic diet. In: Spitz DR, Dornfeld KJ, Krishnan K, Guis D, editors. (2012) Oxidative Stress in Cancer Biology and Therapy. Humana Press. pp. 47–58
Fouad, Y. A., & Aanei, C. (2017). Revisiting the hallmarks of cancer. American journal of cancer research, 7(5), pp. 1016–1036.
Garcia SN, Guedes RC, Marques MM. (2019) Unlocking the Potential of HK2 in Cancer Metabolism and Therapeutics. Curr Med Chem.26(41), pp.7285-7322.
Greer S., Honeywell R., Geletu M., Arulanandam R., Raptis L., (2010) Housekeeping genes; expression levels may change with density of cultured cells, Journal of Immunological Methods,355(1-2),pp 76-79
