The History About Leukaemia

Published Date: 02 Nov 2017

Leukaemia is a general term describing cancers of white blood cells and bone marrow. There are four main types of leukaemia: acute myeloid (AML), acute lymphoblastic (ALL), chronic myeloid (CML) and chronic lymphocytic (CLL). Cancer Research UK organisation revealed that leukaemia is the twelfth most common cancer in the UK, accounting for more than 2% of all cancers.(1) Leukaemia Research Foundation has published estimated statistics regarding the new and death cases for leukemia and lymphoma.(2) The table below suggests that a significant although not greatest proportion of cancers affecting immune system are attributable to leukaemia.

2012 Statistics

Estimated new cases

Estimated deaths

Hodgkin's lymphoma

9,060

1,190

Non-Hodgkin's lymphoma

70,130

18,940

Total lymphomas

79,190

20,130

Acute lymphocytic leukemia

6,050

1,440

Chronic lymphocytic leukemia

16,060

4,580

Acute myeloid leukemia

13,780

10,200

Chronic myeloid leukemia

5,430

610

Other leukemia

5,830

6,710

Total leukemias

47,150

23,540

TOTALS

126,340

43,670

Figure 1. Estimated leukaemia and lymphomas statistics for 2012. (Figure adapted from (2))

1.1.1Acute Lymphoblastic Leukaemia (ALL)

From the statistic (Figure 2) published by Cancer Research UK, the highest incidence of leukaemia in children is at the age group of 0-4 and most of them were diagnosed with ALL.(3)

Figure 2. Leukaemia: Average Number of New Cases per Year and Age-Specific Incidence Rates, UK, 2006-2008. Incidence of leukaemia varies with age. The risk increases soon after birth and children in the 0-4 age group have highest incidence; mainly consisting ALL. Rates then decline with increasing age and peak again at age between 70 and 79 for male and over 85 for female. Overall, male have higher incidence rate than women at all age group. (Figure adapted from (3) )

ALL is the most common type of lymphomas malignancies in children; approximately 1 in 2000 children develops this disease.(4). Accounting for more than 80% of childhood leukaemia in UK, ALL arises from mutation of progenitor cells which would normally differentiate to different types of white blood cells (WBCs) such as B lymphocytes, T lymphocytes and natural killer cells.(4, 5) WBCs are valuable in our immune system to fight infections. In ALL, there is a growth of abnormal lymphoblasts which do not develop into mature cells. These immature cells do not work like normal WBCs in producing antibodies to foreign antigens and fighting viral/bacterial infections. They would enter the circulatory system, interfering with the production of normal blood cells leading to infection, bleeding, anaemia and other complications.(5)

1.1.1.1Incidence

In the UK, according to NHS, about 7600 people are diagnosed with leukaemia annually.(6) Of these, about 9% leukaemic patients have ALL.(7) In UK, approximately there are 370 children diagnosed with ALL every year. (8) The incidence varies with age, being most common in early childhood. The risk increases rapidly after birth, peaking between 2 and 5 years of age.(9) In general, males are more prone to develop ALL than females. Children with Cancer UK organization concludes that "for every 3 girls diagnosed with ALL, 4 boys are diagnosed". (5)

Comparing to prostate cancer which mostly affecting male aged over 65 years old, ALL is more prevalent in younger age group.(7) Although the 5 year survival rate for children with ALL has improved from 10% to 88% owing to the development of more effective drug treatments, highest number of death has been observed in patient aged under 20s.(10, 11) These young patients have died before having the opportunities to realize their full potential and become productive members of the society. This has led to an immeasurable loss to our society in all aspects. Thus, it has created a strong incentive to invest money and time in investigating and improve clinical treatment available so that the overall prognosis of ALL could improve.

1.1.1.2 Treatments

A wide range of anti-leukaemic agents have been used in the treatment regimen in ALL patients. A combination therapy may have to be used in order to improve treatment outcomes and avoid resistance. The exact regimens may vary from person to person depending on their age, total white cell counts and side effects that are tolerable by individual patient. In general, ALL treatment is divided into three phases: remission-induction phase, consolidation of therapy and remission maintenance.

Remission Induction Phase

Remission-induction phase usually takes about 4 weeks. It starts with the use of Vincristine and an anthracycline drug alongside with Dexamethasone (Dex) or Prednisone.(9, 12, 13) This combination therapy aims to eliminate lymphoblast and restore the function of the bone marrow. Vincristine and anthracycline drugs work by interfering with the proliferation of rapidly growing cancer cells. However, they are often associated with some off target effects such as hair loss and nausea and vomiting as hair cells and gastro-intestinal lining cells are fast reproducing. Glucocorticoid (GC) is a drug of choice in the treatment of ALL due to its ability to trigger apoptosis.(14-16) Dexamethasone is preferred to other synthetic corticosteroids e.g. Prednisone due to its desirable pharmacokinetic properties such as extended plasma half life and better central nervous system penetration. (17)However, prolonged use of Dex leads to the emergence of resistant ALL cells.(14) Extensive research has been carried out over the past few years to investigate the molecular mechanisms of resistance/sensitivity of ALL cells to GCs treatment with the aims of reversing the resistance and improve the prognosis of patient.(13, 18, 19)

Consolidation phase

Remission of ALL is maintained by additional therapies. Chemotherapy used in the consolidation phase includes those used in the remission induction phase, frequent pulses of vincristine and corticosteroids. (9, 13)Other than that, high dose methotrexate plus mercaptopurine are frequently used in preventing relapse.(9) Unlike remission-induction phase, consolidation phase usually last for several months. This is to avoid any residual undetectable lymphoblast to cause relapse.

Remission maintenance phase/Continuation phase

It is important for ALL patients to have continuation treatment to prolong their survival after remission. Daily mercaptopurine and weekly methotrexate doses form the basis of continuation treatment. (9) Alternatively, thioguanine which produce more intense antileukaemic response compared to mercaptopurine can also be used in remission maintenance phase. However, more profound and severe side effects associated with thioguanine has rationalized the use of mercaptopurine. (9, 13, 20)

Basically, the most important phase in ALL patient treatment would be the remission induction phase. A success in remission induction phase is essential in preventing disease progression and preventing relapse. However, traditional leukaemia treatment does not always work like what have been anticipated as complications often arise from chronic ALL treatment with GCs. Looking at the role of GCs in inducing remission in the initial stage of ALL treatment and the problems faced with chronic treatment, we can predict that greater returns would be generated if time and money are spent wisely in investigating the differential response of ALL cells upon GCs treatment. If we can successfully target the resistance of ALL cells to glucocorticoids-induced apoptosis at molecular level, obviously we can prolong the life of children diagnosed with ALL who often die at young age, improving their quality of life and perhaps return them to the society.

1.2 Pharmacological effect of Glucocorticoids (GCs)

Glucocorticoid (GC) is a steroid hormone; depending on nuclear receptor for its transcriptional activity. Synthesis of GCs in adrenal cortex is self-regulated by negative feedback loop.(21) Adrenocorticotropic hormone secreted from anterior pituitary gland in response to corticotrophin-releasing hormone would regulate GCs secretion.(21) GC has a number of physiological roles; including regulation of salt and water contents, metabolisms of glucose, fat and amino acids and anti-inflammatory and immunosuppressant. The rationale of using GCs in induction therapy of ALL is due to their ability in inducing apoptosis i.e. programmed cell death in cancer cells.

1.2.1 General Mechanisms of actions of Glucocorticoids

The way GCs regulate various functions are studied at molecular level. GCs achieve theirs effects by binding to cytosolic glucocorticoid receptors (GRs) in target cells. In unbound state, GRs are located predominantly in cytoplasm in an inactive form. It is stabilized by forming complexes with several chaperone proteins such as heat shock proteins (hsp): hsp90 and hsp70.(13, 19, 22) Binding of ligands onto the GR induces conformational changes, causing GR to dissociate from the protein complexes and homodimerize. The dimerised GR-GR complexes translocate into nucleus and bind to specific DNA sequences in the promoter region of target gene which are known as Glucocorticord Responsive Elements (GREs). (13, 23-26)

The activation domain and DNA binding domain of the homodimerised GR-GR complex would bind to palindromic GREs: AGAACANNNTGTTCT. (26, 27) Activation of GREs located upstream from the transcription initiation site in target genes would either induce or repress the transcriptional activity of target genes. (26, 28) Binding of GR to positive GREs would up-regulate gene expression whilst binding to negative GREs (nGREs) would repress gene expression.(13, 28)

fig2

Figure 2. Mechanisms of glucocorticoid action.

GCs passively diffused through cell membrane and bind to GR alpha. Associated heat-shock proteins (hsp) are released. Ligand-bound receptor translocates into the nucleus. A) A GR homodimer binds to GC responsive elements (GRE) on the promoter region of target genes and activate gene transcription. B) Binding of GR alpha to a negative GRE (nGRE) leads to repression. C) Interactions between GR alpha and transcription factors, such as NF-kappa B and AP-1, repress the transcription of pro-inflammatory genes. D) GR alpha can alter the mRNA or protein stability of inflammatory mediators. IL: interleukin, MKP-1: mitogen-activated protein kinase phosphatase-1, POMC: proopiomelanocortin, COX-2: cyclooxygenase-2, VEGF: vascular endothelial growth factor, TNF alpha: tumor necrosis factor alpha, TF-RE: transcription factor-response element. (Figure adapted from (29))

1.3 Glucocorticoids-induced apoptosis

GCs induce apoptosis by both genomic and non-transcriptional pathways. GCs modulate apoptosis by regulating the balance of pro-apoptotic and anti-apoptotic members of BCL-2 family via GRs. This involves nuclear translocation and binding on specific response elements, thus leading to regulation of target gene expression

GCs exert its genomic effects by regulating the transactivation/transrepression of target gene. A variety of genes that are involved in the initiation of apoptosis contains GREs which depends upon binding of GCs to GR for transactivation/transrepression. For instance, transactivation of Bim (a BH3 domain only pro-apoptotic gene) would result in increased activation of Bax and Bak; priming the cells for apoptosis via intrinsic pathway.

1.3.1 Intrinsic and Extrinsic pathway

In response to stress, a BH3-only protein, Bim is transactivated. Up-regulation of apoptosis activator, Bim results in activation of another two pro-apoptotic genes, Bcl 2-associated X (Bax) and Bcl-2 homologous killer (Bak) which mediate the process of Mitochondrial Outer Membrane Permeabilisation(MOMP). (13, 30)Disruption of mitochondrial membrane potential releases pro-apoptotic factors including cytochrome C and Second-mitochondria derived activator of caspases (SMAC) into cytosol. Released cytochrome C would bind to its adaptor apoptotic protease activating factor (Apaf-1), allowing oligomerisation of Apaf-1. (31)

Oligomeric Apaf-1 recruits pro-caspase 9 through exposed Caspases Recruitment Domain (CARD), forming apoptosome.(32) Assembly of apoptosome would activate caspase 9 which would then further activating other executioner caspases such as caspase 3 and 7; eventually leads to programmed cell death. Also, loss of mitochondrial membrane potential leads to inhibition of calcium influx and generation of ROS. These rapid cytoplasmic effects have augmented the effect of GC-induced apoptosis.(13, 31, 33)

Role of intrinsic pathway especially BCL-2 family in GCs-induced apoptosis has been evidenced.(34-36) For example, thymocytes obtained from Apaf-1 and caspase 9 deficient mice are resistant to GCs-induced apoptosis. (13, 37) BCL-2 family members are one of the GR transcription targets. Several studies have shown that relative ratio of pro- and anti-apoptotic BCl-2 protein but not their expression that plays a vital role in determining cellular response upon GCs treatment.(38) When the balance between the two proteins is pro-apoptotic, apoptosis is initiated and vice versa.

Although ability of GCs to inhibit induction of Fas-Ligand (Fas-L) has confirmed the role of GCs in extrinsic pathways, extrinsic pathway is not prerequisite in GCs-induced apoptosis.(39) Roles of extrinsic pathway in GCs-induced apoptosis still need to be clarified.(40, 41) In extrinsic pathway, GC-induced apoptosis is stimulated upon activation of membrane death receptor which is the member of tumour necrosis factor receptor superfamily. (13, 42) Binding of extracellular signalling molecule such as death ligand to its corresponding transmembrane death receptor transduces intracellular signals which would lead to the activation of inducer caspase 8. Binding of Fas-L onto complementary Fas receptor induces conformational change, allowing it to interact with death domain of Fas Associated Death Domain (FADD) containing protein. This subsequently leads to the formation of Death Inducing Signalling Complex (DISC), activating the initiator caspases, caspase 8. DISC transduces a downstream signal directly activating caspase 3 resulting in apoptosis.(30, 33)

In some cells, caspase 8 cannot directly activate caspase 3 as XIAP can prevent Fas induced apoptosis by inhibiting caspase 3 activation.(33) Fas mediated caspase 8 activation needs to activate the mitochondrial pathway of apoptosis by cleaving a pro-apoptotic BH3-only protein, Bid.(33) The C-terminal truncated Bid (t-Bid) translocates to the mitochondria and activates Bax and Bak on mitochondria.(30) Subsequent MOMP release SMAC which would then release the inhibition of XIAP on caspase 3.

1.3.2 Non-genomic effect of Glucocorticoids

Despite the genomic effects, novel non-transcriptional mechanisms of signal transduction through steroid hormone receptors have been identified.(13, 43) Steroid-induced modulation of cytoplasmic or membrane-bound regulatory proteins plays a major role in the rapid "non-genomic" effects. The non-genomic effects happen fairly quickly because it does not involve the gene transcription or protein synthesis.(44)

Non-genomic effects of GCs play a vital role in inducing apoptosis. Mutated GRs lacking ligand binding domain (mainly in cytosol) could still cause apoptosis in leukaemic cell although it lacks of ligand binding domain and the subsequent ligand activation.(14, 45)

Another example is GCs interfere with reactive oxygen species (ROS) level by down-regulating level of enzymes that are involved in ROS protective mechanisms.(13, 46, 47) Study has shown that proteins such as glutathione peroxidase, catalase, superoxide dismutase and others were down-regulated by GCs at the mRNA level.(13, 14, 46, 47) This suggested that GCs-induced apoptosis can be achieved through breaking cellular antioxidant defense so that cells unable to detoxify efficiently. Eventually, the subsequent cellular damage caused by reactive radicals lead to apoptosis. Besides, GCs also induce apoptosis via regulation of intracellular Calcium level. (13, 14, 48) This would then affect translocation of GRs to mitochondria and its ability in triggering downstream signaling pathway to induce apoptosis.(15) This may have indirect effect on cells energy production pathways and affects the ROS level. The non-genomic effects of GCs in inducing apoptosis are undeniable. However, this study aims to focus on the genomic pathway of GCs in inducing apoptosis and its associating resistance issue at molecular level.

1.4 Mechanisms of Resistance/Sensitivity to Glucocorticoids-induced apoptosis

Under normal conditions, mutated cells or damaged cells would undergo cell cycle arrest to repair damaged DNA or if the damage is irreparable, the cells would receive apoptotic signals and undergo well regulated cell death to avoid vertical transmission of mutated/damaged traits during mitosis. However, cancer cells normally omit the apoptotic signal due to various adaptive mutations including p53 protein. GCs used in ALL treatment aims to trigger apoptosis in these malignant cells and therefore controlling the malignancy. However, resistance to GCs-induced apoptosis arises from repeated GCs treatment has become a great obstacle in achieving full remission.(13, 14) Despite the acquired glucocorticoid resistance, a significant number of ALL patients were found to be insensitive to GCs treatment.(14, 49) Glucocorticoid resistance describes the phenomenon where cancer cells unable to initiate apoptosis in response to GR activation. This resistance issue has reduced ALL cure rates and thus it is definitely worthwhile to study in depth the resistance/sensitivity of GCs-induced apoptosis at molecular level. (13, 50)

Several theories have been proposed to explain glucocorticoid resistance. Some of them have been described in the following sections.

1.4.1 Existence of Glucocortocoid Receptor isoforms

GR is a member of steroid nuclear receptor superfamily.(13, 24, 28) It is a ligand dependent transcriptional factor which regulates various metabolic processes via transactivation (up-regulation of gene expression) and transrepression (repress transcription of target genes) in a cell specific manner. There are two major isoforms abbreviated as GR α and GR β as a result of alternative splicing at exon 9. (13, 51)

GRα is the major isoform; transcriptionally active and localized in the cytoplasm in the absence of hormone. Glucocorticoid resistance due to mutations on GRα possibly caused by abnormal interaction with ligands, target DNA sequence and co-regulators proteins and deviant nucleocytoplasmic trafficking.(13, 52)

Unlike GRα, GRβ does not bind to its ligand therefore it does no exert direct intervention on transcriptional activity induced by GCs.(50) It is found to regulate GRα-induced gene expression by exerting dominant negative effect.(51) Transactivation of target genes requires dimerization of GRs while transrepression can occur with monomeric GR. However, heterodimerisation of GR α-GR β is functionally impaired, attributing to the dominant negative effects of GRβ.(51) Besides, in the absence of ligand, GRβ exhibits greater binding capacity to GRE-containing DNA compared to GRα.(51) This means ligands enhance interaction of GRα and specific DNA sequence and thus a single mutation on GRα may lead to various outcomes including resistance issue. (51)

Basal level of GRα is greater than GRβ in normal tissue.(49, 51, 53) Some studies have shown that resistant cells have higher ratio of GRβ compared to GRα. This has suggested that the dominant negative effect exerted by GRβ played a role in causing glucocorticoid resistance. (49, 54, 55) However, relationship between GRβ level and glucocorticoid resistance need to be elucidated further as there are controversial findings from previous work.(38, 56) Since GCs-induced apoptosis is mediated by GRs, presence of different levels and isoforms of GRs certainly confers resistance/sensitivity of ALL cells to GCs treatment.

1.4.2 Post-translational modifications-phosphorylation

Post-translational modifications of GRs are of vital importance in determining its stability, interactions pattern with cofactor and transcriptional activity of GR.(57) The common post-translational modifications of GRs include phosphorylation, glycosylation, acetylation and many others. Inevitably, phosphorylation is one of the most important modifications on protein; regulating protein activities by switching it on and off. A protein such as GR can be switched on by attaching one or more phosphate group on specific amino acid. Enzymatic de-phosphorylation of GR would turn off its activity. A range of kinases are involved in phosphorylation of different GR isoforms and it varies from one to another. ref

Phosphorylation of GRs usually happens after ligand binding although basal phosphorylated GRs have been identified.(58) It has been suggested that there are additional phosphorylation when agonist binds to GRs.(58) GR being a multiphosphorylated protein molecule; numerous kinases are involved and they are responsible for activation and inactivation of GR. Examples of kinases involved in the phosphorylation of GRα both in vitro and in vivo include: p38 MAPK, CNS-specific Cyclin dependent kinase 5 (Cdk5), glycogen kinase synthase 3β and c-Jun N-terminal kinase (JNK).(13, 57)

Mitogen-activated protein kinases (MAPKinases) and Cdks are the major kinases involved in the phosphorylation of GRs. However, they exert opposite effects on the transcriptional activity of GRs.(58) GRs can be phosphorylated at different serine residues within its N-terminus: S203, S211 and S226. (13, 19)Differential phosphorylation of GRs may cause differential subcellular trafficking, interactions with co-regulators, target promoter specificity and many other processes that may affect GRs’ activity.(57) Different GREs would be targeted when different phospho-isoforms of GRs are activated, causing differential transcriptional activity, differential gene expression and hence leads to either survival or death of ALL cells.

Previous study has revealed that the transcriptional activity of GRs increased when GRs are predominantly phosphorylated at serine 211 (S211) position.(13, 58) Also, there are greater amount of GR-S211 than that of GR-S226 in sensitive CEM-C7-14 cell lines and the opposite is true for the resistant cell line.(58, 59) Furthermore, GR-S211 has been shown by ChIP assays to be a an active form of the receptor because of its ability to occupy several GRE-containing promoters (60, 61)Replacement of S226 with an alanine amino acid has dramatically increased transcriptional activity of GRs also suggested that GR-S226 is inactive.(19, 62) Together, these studies further confirmed the hypothesis that GR-S211 is active GR isoform whilst GR-S226 is inactive. However, unlike S203; both forms of GRs can be efficiently recruited by GREs-containing promoters.(58, 60)

GR-S211 has been shown to have a more favourable conformation, predisposing it for interactions with cofactors and hence leading to elevated transcriptional activity.(60, 62) Cyclin/Cdks complexes and p38 MAPK are responsible for full GR transcriptional enhancement as they both phosphorylate GRs at serine 211 position.(63) GCs induce activity of p38 MAPK which in turn promote phosphorylation of GRs at S211, making it trancriptionally active. This pathway is crucial in GCs-induced apoptosis as use of p38 MAPK inhibitors suppressed GCs-induced apoptosis.(60) JNK is the major kinase involved in the phosphorylation of GRs at serine 226 position.(60) JNK-mediated phosphorylation of human GR-S226 was proclaimed to exhibit reduced transcriptional activity due to enhanced nuclear export. (60)

In conclusion, differential phosphorylation of GRs would lead to different fates when ALL cells treated with Dex. Phosphorylation of GRs at different sites renders GRs being in active or inactive state. Active GRs would be able to induce apoptosis via regulation of gene expression and involvement in both extrinsic and intrinsic pathway of apoptosis whilst inactive GRs are incapable of generating both genomic and non-genomic effects. Thus, inactive GRs cannot stimulate cell death by activating the apoptosis cascades. This provides an explanation for the existence of resistant cells.

1.4.3 Mitochondrial Glucocorticoid Receptor

A study observed translocation of GRs into mitochondria when treated with Dex in GCs sensitive cells but not in GCs resistant cells.(64) The increased GC-dependent mitochondrial translocation in sensitive cell suggested that mitochondrial GRs involve in inducing apoptosis.(64) A positive correlation between mitochondrial GR translocation and sensitivity of cell to GCs treatment has been established. (64) However, the same study also revealed that mitochondrial localization of GRs alone is inadequate to elicit apoptosis. Combination treatment containing GCs inhibitor and Dex has promoted mitochondrial translocation; but its failure in inducing apoptosis has suggested the importance of GR conformation in GCs-induced apoptosis.(64)

Moreover, presence of GRs and GRE-like domains in mitochondria in some cell types suggests the transcriptional activity of GCs is not limited to nuclear genome. (65, 66) Some oxidative phosphorylation (OXPHOS) subunits were encoded by mitochondrial genome although vast majority of them are encoded in nuclear DNA.(32) This may suggest the role of mitochondrial GRs in regulating OXPHOS.(67)

A study evidenced that greater mitochondrial Cytochrome C Oxidase (COX) subunits were expressed in CEM-C7-14 relatively to CEM-C1-15.(32) This could be explained by an observation where transactivation and mitocohondrial translocation (68) of GR-S211 allow it to interact with mitochondrial GREs-like domains thereby exerting transcriptional regulation on mitochondrial genome. On the other hand, GRs in GCs-resistant cells have relatively fewer or none mitochondrial translocations indicates that GRs in resistant cells unable to regulate transcription of mitochondrial genome such as genes encoded for COX subunits. This makes OXPHOS pathway become inaccessible to resistant cells to meet their high energy demand for cell proliferation. (19, 32)

Current literature also revealed that higher level of COX assembly was seen in sensitive cells.(32) This indicates sensitive cells have higher dependence on OXPHOS relatively to resistant cells. Taking the privilege of utilizing OXPHOS as major energy source, sensitive cells which contain greater level of GR-S211 exhibits increased transactivation function of GR to mediate GCs-induced apoptosis. Since GCs-induced apoptosis is ATP dependent process, GCs-sensitive cell lines clearly have an advantage over GCs-resistant cell line in mediating GCs-induced apoptosis.

1.5 Reactive Oxygen Species (ROS) level in sensitive and resistant cells

Reactive Oxygen Species (ROS) refers to chemically reactive molecules containing oxygen. A broad range of diseases/ toxicity exert their harmful effects via generation of ROS. Examples of ROS are superoxide radicals, hydroxyl radicals, hydrogen peroxide, singlet oxygen and reactive nitrogen species. These radicals are highly reactive and short lived by reacting with and damaging inter/intracellular components such as lipid, protein, DNA, amino acids and nucleotides.

ROS do have physiological functions such as signal transduction and destructing invading microorganisms. However, the shift in redox balance of cells would lead to increased oxidative damage and increased cell death by different mechanisms such as apoptosis, necrosis and autophagy. Damaging effects of ROS are implicated in many diseases and degenerative conditions such as type II diabetes mellitus, cardiovascular diseases, Alzheimer’s disease and also cancers. There are various sources of ROS in human body: mitochondrial electron transport chains (which involved in oxidative phosphorylation (OXPHOS)), microsomal electron transport chains, peroxisomes, NADPH oxidases, uncoupled nitric oxide synthase and photochemistry.

ROS in sensitive cell line, CEM-C7-14 is in such a level that it favours the activation of p38MAPK and the cyclins/Cdks complexes. Thus, a greater proportion of GR-S211 was seen in sensitive CEM-C7-14 cell lines. This observation is corresponding to the fact that sensitive cells are susceptible to GCs-induced apoptosis due to presence of more active form of GRs. (59)Unlike CEM-C7-14, ROS level in CEM-C1-15 favour the activation of JNK hence greater proportion of GRs is phosphorylated at GR-S226. GR-S226 is an inactive form of GR therefore cannot mediate transcriptional activity induced by GCs. GCs signaling pathway is not activated and making the lymphoid cells able to survive. This may explain the reason why CEM-C1-15 cell lines are resistant to GCs treatment.

1.6 Use of 2-deoxy-D-glucose

2-deoxy-D-glucose is used in this study to investigate the ability of reversing the resistance of ALL cells to Dex treatment. It inhibits glycolysis by acting as a false substrate to glucose hexokinase. Replacement of 2-hydroxy group with hydrogen makes it not able to undergo further glycolysis.

1.6.1 Cellular Energy Pathways

Research shows that cancer cells have different energy metabolism pathways as to normal cells. (69-71) Cancer cells have high proliferation rate and often associated with angiogenesis i.e. malformed blood vessels. This makes distant cells unable to obtain sufficient nutrient and oxygen to support its high turnover rate. Under hypoxic condition, normal cells may receive signals to undergo apoptosis/necrosis/autophagy in order to maintain homeostasis. However, hypoxic cancer cells would switch from OXPHOS to anaerobic glycolysis pathway for energy generation due to inaccessible oxygen molecule. This change in metabolism is well known as the Warburg effect and it has been recognized as a new hallmark in cancer cells for many years.(69, 70) Warburg effect is also observed in cancer cells even under aerobic conditions as this certainly gives a survival advantage for cancer cells over normal cells for proliferation.

ALL cells have shown to express elevated level of GLUT-1 which is a membrane spanning receptor mediating cellular glucose uptake. (18) Also, the expression of genes encoding OXPHOS enzyme i.e. COX subunits were down regulated in resistant ALL cells compared to sensitive one. Collectively, this indicates that the Warburg effect comes into play in ALL cells. (18, 69, 71)They switch from the more efficient energy production pathway, OXPHOS to glycolysis which produces fewer adenosine triphosphates (ATP) even under aerobic condition. This adaptive change is probably due to mitochondrial damage, hypoxia and loss of OXPHOS capability.(71)

GCs modulate OXPHOS via mitochondrial GRs by increasing the expression of OXPHOS subunits. (67) When comparing the two cell lines, resistant cells have relatively lower number of mitochondria compared to sensitive cells. These evidences may suggest that resistant cells do not rely on OXPHOS for energy production. In contrast, glycolytic pathway is more extensively used in resistant cells for energy production. (32, 71)

Another prominent issue that worth mentioning is that GRs in resistant cells are mostly phosphorylated at serine 226 position which is shown to have increased nuclear export, transcriptionally less active and incapable of translocating into mitochondria. All these properties have made OXPHOS become unfavourable in resistant cells and therefore making them more rely on glycolysis. Additionally, previous studies suggested that up-regulated glycolysis impedes apoptosis while OXPHOS exerts both pro- and anti-apoptotic functions in cells.(72, 73) Moreover, OXPHOS is essential to activate pro-apoptotic genes. Therefore, it is important to disrupt glycolysis pathway in resistant cells that use glycolysis in preference to OXPHOS so that apoptosis can be triggered. (73)

Since CEM-C1-15 is more relying on glycolysis, it is not illogical for us to relate its ROS level to their adaptive cellular energy pathway i.e. glycolysis. Glycolysis synthesizes ATP in both aerobic and anaerobic condition. Unlike OXPHOS, glycolysis pathway generates relatively lower amount of ROS compared to OXPHOS. We are using glycolytic inhibitor hoping that it would be able to interfere with the ROS level by forcing resistant cells which have relatively lower number of mitochondria to switch their energy metabolism pathways to OXPHOS. OXPHOS may possibly generate ROS in such an extent that favour the activation of kinases that responsible for GR phosphorylation at Ser-211 and do not favour the activation of JNK and hence phosphorylation at S226 position. This forms our hypothesis: glycolysis inhibitor would alter ROS in resistant cells which would reverse their resistance to Dex treatment.

1.7 NF-E2 related factor (Nrf-2)

NF-E2 related factor (Nrf-2) plays an important role in protecting cells from oxidative insult. Nrf-2 is a transcriptional factor which is activated under oxidative stress.(74) Activation of Nrf-2 would up-regulates antioxidant-response elements(ARE) containing genes such as those encoded for antioxidants, phase II enzymes and relevant transporters.(74) In response to oxidative stress, transcription and translation of enzymes such as inducible nitric oxide synthase, catalase and gluthatione-S-transferase are up-regulated by Nrf-2 signalling pathway.(75) Basal level of Nrf-2 is maintained by Kelch-like ECH-associated protein (Keap1) mediated ubiquitination and the subsequent protein degradation. (74, 76) Cysteine residues within Keap1 making Keap1 eligible to function as "redox sensor" and hence transducing signal down the cascade.

An external file that holds a picture, illustration, etc. Object name is nihms84142f1.jpg Object name is nihms84142f1.jpg

Figure 3. Schematic model of Nrf-2 regulation by Keap-1

Keap1 regulates Nrf-2 signalling pathway by turning on and off the Nrf-2 mediated anti-oxidant response. Under basal conditions, Keap1 acts as E3 ubiquitin ligase, catalysing ubiquitination of Nrf-2 and targeting for degradation. Therefore, Nrf-2 level is kept at minimum. In response to oxidative stress, expression of Nrf-2 enhanced due to reduced ubiquitination and degradation. Nrf-2 translocates into nucleus and activates downstream genes containing AREs. After achieving homeostasis, Keap1 translocates into nucleus and dissociates Nrf-2 from AREs by forming a complex. Keap1 and Nrf-2 complex is exported into cytosol and associates with Cul-3-Rbx1 core ubiquitin machinery, resulting in degradation of Nrf-2. ( Figure adapted from (74) )

1.7.1 Mechanism of action of Nrf-2

Under physiological condition, ROS is constantly being generated as waste products of essential cellular processes such as OXPHOS. These highly reactive and damaging radical species impose oxidative stress when the redox balance is disrupted. The low level of ROS may activate Nrf-2 signalling pathway, leading to the transcriptional activation of downstream genes and therefore achieving homeostasis. Keap1 sense first sign of redox disturbance then it undergoes conformational change and subsequently releasing Nrf-2.(74) Nrf-2 accumulates and translocates into nucleus, binding to AREs of target genes and promotes its transcription. After restoring redox balance, Nrf-2 would be target by Keap 1 for ubiquitination and proteasomal degradation.

1.7.1.1 Glucocorticoids and Nrf-2

Recent study evidenced that GCs would suppress Nrf-2 dependent anti-oxidant response in hepatic cells. (77) Besides, previous studies have shown that GCs reduced mRNA of antioxidant enzymes such as glutathione peroxidase in rat hippocampal region, leading to accumulation of ROS. Accumulation of ROS is associated with oncogenic disease progressions and mitochondrial DNA mutation in cancers including leukaemia.(78) The relationship between oxidative stress and disease progression propose the importance of determining the anti-oxidant Nrf-2 protein levels in CEM-C7-14 and CEM-C1-15 so that we can correlate it with cells susceptibility to GCs-induced apoptosis.

1.7.1.2 Relationships between Glycolysis and Nrf-2

Induction of Nrf-2 production in response to oxidative stress generated from glycolysis is to a smaller extent compared to OXPHOS. In other words, cells that rely heavily on glycolysis for energy production would express lower level of Nrf-2 compared to the one depending on OXPHOS. By inhibiting glycolysis, cells would be forced to utilize OXPHOS instead and express higher level of Nrf-2 in response to oxidative stress brought from the ROS generated during OXPHOS.

1.7.2 Controversial findings of Nrf-2

Increased production of Nrf-2 dependent proteins would elicit protective mechanisms to eliminate toxicants or damaging radicals and therefore suppressing carcinogenesis.(74, 75) This beneficial effect has increasingly become an interest in the area of cancer prevention. A wide variety of natural products and synthetic compounds that have been commonly recognized as chemopreventive substances such as curcumin, wasabi, resveratrol (red wine), oltipraz (a substituted 1, 2-dithiole-3-thione) and many more have been shown that are able to boost Nrf-2 level when consumed. (74) (75) Again, this has proposed a relationship between Nrf-2 and tumour suppression. A tumour suppression effect mediated by Nrf-2 has been shown both in vivo and in vitro in many cell types such as lung, breast tissue, bladder and colon.(74-76, 79)

However, recent studies also revealed that cell lines resistant to GCs treatment show higher level of Nrf-2 which indicates that Nrf-2 also plays a vital role in survival of cancer cells. An in vitro study has come out with a conclusion stating that reduced expression of Keap1 in a tumor-mimicking microenvironment has increased nuclear Nrf2 translocation and activation of Nrf-2. Enhanced activation of Nrf-2 elevated the transcription of its downstream genes; increasing expression of peroxiredoxin 1 (Prx 1) which plays a significant role in generating redox balance in cells. This results in removal of ROS and hence protects cancer cells from oxidative damage. (74, 80) Indeed, many independent studies suggested that over expression of Nrf-2 bring both clinically and physiologically important effects on cells. It makes cells more susceptible to carcinogenesis, promoting chemoresistance, metastasis and survival of cancer.(74, 75, 81, 82) These downsides of Nrf-2 are achieved by several mechanisms: scavenging ROS which often generated by chemotherapy to kill cancer cells, detoxifying xenobiotics via up-regulated phase II enzymes, transporting drugs out of target cells via induction of multidrug resistance transporter proteins. (75, 79)

However, these contradicting effector functions of Nrf-2 may be dose or time dependent. Nrf-2 inducers may suppress carcinogenesis when used at low dose or at early stages of cancers. On the other hand, when used at high dose or during malignancy, Nrf-2 would protect cancer cells from ROS/chemotherapies, conferring a growth advantage to cancer cells and therefore leads to poor prognosis in cancer patient.(75)

1.8 Mitogen-Activated Protein Kinases (MAPKinases)

MAPKinases are serine or threonine specific kinases mediating diverse cellular processes range from gene expression, cell survival, cell growth, cell differentiation and cell death. (83) The most commonly recognized MAPKinases are: extracellular signal-regulated kinases 1 and 2 (ERK1/2), JNK1-3, p38 (α,β,ϒ and δ) and ERK5 families. Examples of atypical MAPKinases are ERK3/4, ERK7/8, Nemo-like kinase (NLK).(84-86) MAPKinases are activated by dual-phosphorylations at activation loop; one at tyrosine and one at threonine. (86) Prior to activation of MAPKinases, MAPK kinase kinase interacts with a membrane bound protein Ras which is normally inactivated by complexing with guanosine diphosphate (GDP).Exchange of GDP against guanosie triphosphate (GTP) activates Ras protein. Activated Ras binds to several effector proteins including B-Raf. For instance, activated B-Raf phosphorylates one of the MAPK kinase, MEK 1/ 2 which in turn phosphorylate MAPKinases, ERK1/ 2. Unlike conventional MAPKinases, atypical MAPKinases are not activated via this classic three-tier cascade.(84) Collectively, typical and atypical MAPKinases coordinate and regulate various intracellular proteins to generate response to extracellular stimuli.

Figure 4. Schematic diagram of MAPKinase signaling pathway.

Binding of ligand onto membrane spanning receptor e.g. binding of epidermal growth factor (EGF) to the extracellular domain of EGF receptor would lead to dimerisation of two subunits of the receptor tyrosine kinases (RTK). This leads to subsequent transphosphorylation on both subunits. Then, adaptor proteins Grb2 binds to phosphorylated RTK. Binding of SOS to Grb2 mediates guanosine exchange activity of Ras. (i) Membrane bound inactive Ras is complex with GDP. (ii) Replacement of Ras GDP by GTP activates Ras and allows it to binds to several effector proteins including B-Raf. (iii) Active B-Raf phosphorylates MEK1/2 which in turn phosphorylates ERK1/2. (iv) ERK 1/2 phosphorylates transcriptional factors of activator protein 1 (AP-1) family such as Jun and Fos. Phosphorylated Jun and Fos translocates into nucleus and binds to the AP-1 motif of DNA. This leads to transcription of downstream genes, resulting in cell growth proliferation. (v) Activated Ras being inactivated shortly after its activation. GTPase activating protein (GAP) binds to Ras-GTP and enhances Ras intrinsic GTPase activity. Bound GTP is hydrolysed to GDP. Ras-GDP is not active and cannot bind to B-Raf thereby turning off the MAPKinase signaling pathway.

One of the proteins regulated by MAPKinases is termed MAPK-activated protein kinases (MAPKAPKs). Members of MAPKAPKs family include the p90 ribosomal S6 kinase (RSK), the mitogen- and stress-activated kinase (MSK), the MAPK-interacting kinase (MNK), the MAPK-activated protein kinase 2/3 and MK5.(84) Despite its widespread physiological roles, MAPKinases signaling pathway has immensely become a subject of interest and investigations on its role in mediating GCs-induced apoptosis.(63, 87) A recent study has reported the role of MAPK kinase 3 in conferring GCs sensitivity. Induction of MAPK Kinase 3 has activated p38 MAPKinase which in turn phosphorylates GR to its active form and leads to transactivation of apoptosis-inducing genes.(16, 63) Other MAPKinases such as ERK and JNK are associated with resistance to GCs-induced apoptosis possibly through enhanced phosphorylation at S226. Inhibition of either of them enhanced GCs-induced apoptosis. (63) Besides, other study also demonstrated that enhanced ERK mediate co-stimulation induced GCs resistance via increasing synthesis of c-fos and the subsequent AP-1 as shown in figure 4.(88)Therefore, it is worth to explore the expression of these proteins in sensitive CEM-C7-14 treated with Dex and leukaemic patient’s tissue sample cultivated in conditioned medium (CM) and their relations to the molecular basis of resistance/sensitivity to GCs-induced apoptosis.

2. Aims ,Objectives and Hypothesis

2.1 Aims and Hypothesis of the study

The main aim of this study is to understand the molecular basis of the ALL cells’ resistance/sensitivity to glucocorticoid treatment. We have recently acquired evidence in our laboratory supporting the notion that resistant CEM-C1-15 and sensitive CEM-C7-14 ALL cells produce energy using different energy production pathways. In particular CEM-C1-15 cells generate energy through glycolysis whereas CEM-C7-14 cells through OXPHOS.

Under the light of these observations, we hypothesize that in the presence of the glycolytic inhibitor 2DG, CEM-C1-15 cells would switch their energy production pathway to OXPHOS, thereby affecting the intracellular ROS levels and become sensitive to GCs-induced apoptosis in a manner similar to that of CEM-C7-14 cells. Changes in the intracellular ROS levels as a consequence of the reversion of the energy production pathway from glycolysis to OXPHOS would activate kinases of the MAPKinase family and thus impact on the ratio of S211 and S226 GR phosphorylation. GR phosphorylation at S211 occurs as a consequence of Cdk5 and p38 kinase activity and results in induction of GR transcriptional activity whereas GR phosphorylation at S226 occurs as a consequence of JNK activation and has repressive effect of GR transcriptional activity.

We have recently reported that the S211/S226 GR phosphorylation ratio is higher in the sensitive CEM-C7-14 cells compared to that detected in the resistant CEM-C1-15 cells implying that by increasing the S211/S226 GR phosphorylation ratio in the resistant cells it might be possible to reverse glucocorticoid resistance in these cells.

2.2 Objectives

To investigate the potential of a glycolytic inhibitor in reversing glucocorticoid resistance, the relative ratio of the GR S211/S226 phosphoisoforms in the presence of dexamethasone, glycolytic inhibitor or combination of the two will be followed. The potential of 2DG to change the ratio of GR phosphorylation at S211/S226 in CEM-C1-15 will be tested in these cells in the presence or absence of dexamethasone or 2DG or combination of the two. In addition, the correlation between the ratio of GR phosphorylation at S211/S226 and the levels of proteins anti-oxidant properties such as Nrf-2 will be studied. Furthermore whether the changes in the ratio of GR phosphorylation at S211/S226 are a consequence of alterations occurring in components of the MAPKinase pathway will be investigated by analyzing the microarray data obtained from CEM-C7-14 cells under conditions simulating inflammatory or normal state.