Development of sphingosine kinase (SPHK) inhibitors and the role of sphingolipids in adult stem cell proliferation and differentiation

More information about SPHK and S1P functions and applications in some pathological processes and in stem cell research will be addressed in the following sections.. In addition to their

CHAPTER 1 INTRODUCTION

Sphingolipids, components of membrane lipids, have emerged as the sources of several

important signalling molecules (Tay et al., 2005) Sphingolipids are a class of lipids

derived from the aliphatic amino alcohol sphingosine (Figure 1.1) Sphingolipids, such as ceramide and sphingosine-1-phosphate (S1P), belong to a new class of potent bioactive molecules; these sphingolipids have been shown to be involved in a variety of cellular processes, including cell differentiation, apoptosis and proliferation (Spiegel and Merrill,

1996; Hannun et al., 1994; Heller et al., 1994; Kolesnick and Golde, 1994; Wang et al., 1996; Geoffroy et al., 2004)

Figure 1.1 Structure of Sphingosine, D-erythro

Sphingomyelin, the major membrane sphingolipid, is the precursor of the bioactive ceramide, sphingosine and S1P When sphingomyelin is hydrolyzed by sphingomyelinases, ceramide is formed; ceramide can then be hydrolyzed by ceramidases

to produce sphingosine, and sphingosine in turn can be phosphorylated, by sphingosine kinases (SPHKs), to yield S1P; this metabolic process is summarised in Figure 1.2

Ceramide and sphingosine are implicated in diverse stress-related responses, such as

cell-cycle arrest and apoptosis (Kolesnick and Golde, 1994; Hannun et al., 1996; Spiegel and Merrill, 1996) In contrast, S1P has been shown to regulate cell growth (Zhang et al., 1991; Olivera and Spiegel, 1993) and suppress programmed cell death (Cuvillier et al.,

1996 and 1998; Edsall et al., 1997) It has been suggested that the balance between the intracellular levels of ceramide and S1P could determine the cell fate (Cuvillier et al., 1998; Morita et al., 2000; Perez et al., 1997; Xia et al., 1999)

Figure 1.2 The sphingolipid metabolic pathway Sphingomyelin is hydrolyzed by sphingomyelinases to form ceramide Ceramide is metabolized by ceramidase to generate sphingosine SPHK phosphorylates sphingosine into S1P, which is further cleaved by S1P lyase to a fatty aldehyde and ethanolamine phosphates

1.1 SPHK AND S1P

So far, two mammalian SPHKs have been cloned, sequenced and characterized These

kinases are encoded by two genes, SPHK1 (Kohama et al., 1998; Melendez et al., 2000; Pitson et al., 2000), and SPHK2 (Liu et al., 2000) Comparison of the two isoforms of SPHKs is shown in Table 1.1 (modified from Liu et al., 2002) Both SPHKs possess a

conserved kinase catalytic domain which contains the ATP-binding site, as well as five

other conserved domains (Liu et al., 2002; Pitson et al., 2002) These domains may play a role in substrate recognition (Pitson et al., 2002)

Both SPHK1 and SPHK2 are capable of phosphorylating erythro-sphingosine,

dihydrosphingosine and phytosphingosine; however, no other phospholipids appear to be

significantly phosphorylated by these enzymes (Kohama et al., 1998; Melendez et al., 2000; Pitson et al., 2000) Despite the overall homology of the conserved domains and

substrate recognition of SPHK1 and SPHK2, diversity between the gene sequences

implies they did not arise from a simple gene-duplication event (Liu et al., 2002)

Moreover, SPHK1 and SPHK2 have been shown to possess different kinetic properties (SPHK1>SPHK2) and different temporal expression patterns during development (Spiegel and Milstien, 2003) Therefore, it is reasonable to deduce that these two isoforms may have distinct cellular functions and may be regulated by different signalling mechanisms Unfortunately, so far, no 3-dimention structure information is available for SPHK1 and SPHK2, which, to a certain extent, limits functional studies on these two isoforms

Liver, heart>>kidney, brain, testes

DHS (competitive) DMS (noncompetitive) Comparison of SPHK1 and SPHK2

SPHK activity has been shown to be stimulated by several external stimuli including

growth factors such as, platelet derived growth factor (PDGF) (Olivera et al., 1999), nerve growth factor (NGF) (Edsall et al., 1997); cytokines such as tumor necrosis factor-

α (TNFα) (Xia et al., 1999; Liang et al, 2005); phorbol esters (Olivera et al., 1993), antigen receptors such as IgG and IgE receptors (Melendez et al., 1998, 2002); and several other receptors/stimuli (reviewed in Tay et al 2005) All those diverse external

stimuli can activate the endogenous SPHK to generate S1P

S1P has been reported to possess dual functions (Spiegel and Milstien, 2003) It was initially suggested as an intracellular second messenger as some growth factors, like PDGF, NGF and TNFα-, could activate SPHK and increase S1P level in the cells

(Olivera and Spiegel, 1993) In addition, some reports showed that intracellular S1P can

induce cell proliferation and survival (Desai et al., 1992), as well as calcium mobilization (Zhang et al., 1991; Ghosh et al., 1990) However, no potential intracellular receptor(s) that mediate S1P intracellular functions have been identified so far (Kluk et al., 2002)

More recently, researchers found that S1P could function as an extracellular mediator by stimulating S1P receptors, present on the cell surface of the same or nearby cells, in an autocrine or a paracrine manner (Spiegel and Milstien, 2003) S1P receptors (S1PRs) are members of the endothelial differentiation gene (EDG) G-protein-coupled family of receptors So far, EDG1/S1PR1, EDG5/S1PR2, EDG3/S1PR3, EDG6/S1PR4, and EDG8/S1PR5, five members have been identified (Chun et al., 2002) These five

receptors are coupled to different G proteins (for example, S1PR1 and S1PR4 are coupled mainly to Gi; S1PR2 and S1PR3 activate Gi, Gq and G12/13; and S1PR5 is coupled to Gi and G12/13 (Spiegel and Milstien, 2003) Among all five S1PRs, S1PR1, S1PR2 and S1PR3 are most widely expressed, while S1PR4 is mainly found in hematopoietic system, and S1PR5 is predominantly expressed in brain and spleen (MacLennan et al., 1994; McGiffert et al., 2002; Okazaki et al., 1993; Yamaguchi et al., 1996; Zhang et al., 1999; Liu et al., 2000; Kluk and Hla, 2002)

Interaction of S1P with these receptors regulates different cellular processes such as migration, proliferation, cytoskeletal organization, adherens-junction assembly and

morphogenesis (Kluk and Hla, 2002) in vitro; as well as other physiological processes such as blood vessel maturation, cardiac development and angiogenesis in vivo (Liu et al., 2000; Ishii et al., 2002) In particular, S1PR1 showed to affect cell survival/proliferation, migration, cytoskeletal organization (Kluk and Hla, 2002), and it is necessary forvascular

maturation in vivo (Liu et al., 2000) S1PR2 was reported to regulate the role of S1P in

heart cell migration during embryogenesis in zebrafish (Ishii et al., 2002), indicating its

function in cardiovascular system Interestingly, while S1PR1 has been shown to promote cell survival/proliferation, S1PR2 has been shown to activate stress-associated kinases leading to apoptosis (Kluk and Hla, 2002) S1PR3 was found to share most of the signaling aspects of both S1PR1 and S1PR2, and its functions can be substituted by other S1PRs (Kluk and Hla, 2002) Compared to the other three S1PRs, much less is known for the function of S1PR4 and S1PR5 It was found that S1PR4 could mediate activation of

adenylyl cyclase in response to low doses of S1P in mouse embryonic fibroblasts (Ishii et al., 2001) Wang et al (2005) reported that S1PR4 could mediate immunosuppressive effects of S1P by inhibiting proliferation and secretion of cytokines, while enhancing secretion of the suppressive cytokine Interleukin-10 (IL-10) S1PR5 is the most recent

member in S1PRs family and its functions in vitro and in vivo still remain to be

determined There are some reports suggesting that S1PR5 could inhibit adenylyl cyclase

in a pertussis toxin (PTX)-sensitive manner (Im et al., 2000; Malek et al., 2001) Jaillard

et al (2005) reported that S1PR5 is an oligodendroglial receptor with dual functions on process retraction and cell survival It is also interesting to note that S1P, binding to S1PR5, was reported to inhibit serum-induced extracellular signal-regulated kinase (ERK)

1 and 2 activation (Malek et al 2001), suggesting that stimulation of S1PR5 might have anti-proliferative effects

A brief description for S1P intracellular and extracellular functions is shown in Figure 1.3 (modified from Spiegel and Milstien, 2003)

Figure 1.3 S1P signaling as a dual functional factor Various stimuli could activate endogenous SPHK SPHK would phosphorylate its substrate-sphingosine, into S1P S1P has dual functions On the one hand, it could act as an intracellular mediator and trigger numerous cellular events, including intracellular calcium release, and promote cell survival and cell growth On the other hand, it could be secreted and function in an autocrine or a paracrine fashion, to bind with its G-protein-coupled receptors (S1PRs), and trigger other downstream signaling pathways

More information about SPHK and S1P functions and applications in some pathological processes and in stem cell research will be addressed in the following sections

SPHK and S1P have been suggested to be potentially involved in several pathological diseases, including in inflammatory diseases and cancer

1.2.1 SPHK and S1P in Inflammation

The activation of SPHK could exert a proinflammatory effects by promoting neutrophil

chemotaxis (Cummings et al., 2002; Ibrahim et al., 2004), and induced certain proteins

Plasma membrane

important in inflammation, such as cyclooxygenase-2 and monocyte chemoattractant

protein-1 (Pettus et al., 2003; Wu et al., 2004; Chen et al., 2004) Moreover, SPHK is

required for antigen-receptors, on mast cell and monocytes, to trigger acute inflammatory

responses (Choi et al., 1996; Melendez et al., 1998 and 2002; Jolly et al., 2004) Additionally, S1P has been shown to induce eosinophil chemotaxis (Roviezzo et al.,

2004) Therefore, the stimulatory effect of SPHK and its product S1P on immune cells such as monocytes, neutrophils, mast cells and eosinophils, suggest that SPHK and its product S1P may play key roles in inflammation

In humans, the gene for SPHK1 maps to a region on chromosome 17q (Melendez et al.,

2000), which contains several genes involved in autoimmune diseases, including in

multiple sclerosis (Kuokkanen et al., 1997), psoriasis (Nair et al., 1977), and epidermodysplasia verruciformis (Enlund et al., 1999)

Therefore targeting SPHK and/or S1P may have profound therapeutic applications Indeed, recently a novel drug (FTY720, “fingolimod”), a structural analogue of sphingosine, is undergoing clinical trials as a novel therapeutic to treat autoimmune diseases (Mansoor and Melendez, 2008) FTY720 could be phosphorylated by SPHKs

(Brinkmann et al., 2002; Mandala et al., 2002; Paugh et al., 2003; Billich et al., 2003)

and the phosphorylated form of FTY720, FTY720P, is an agonist of all S1P receptors except S1PR2 (Brinkmann et al., 2002; Mandala et al., 2002; Taha et al., 2006) By

preventing S1P binding to S1PRs, FTY-720P induced an increase in lymphocyte homing from the blood to peripheral lymph nodes and peyers patches, as well as an inhibition of

lymphocyte egress from the thymus (Yagi et al., 2000) and lymph nodes into the bloodstream (Chiba et al., 1998) In this way, FTY-720P elicited blood lyphopenia

FTY720 is currently in phase III clinical trials as a mono-therapy for remitting multiple sclerosis (ClinicalTrials.gov 2007-08-20)

relapsing-The way FTY-720P works to suppress S1P implies that immune response triggered by activated SPHK might be, at least in part, due to S1P functions through its membrane receptors, which suggests that inhibiting SPHK activity and/or S1P-receptors could be novel therapeutic strategies for treating inflammatory diseases

1.2.2 SPHK and S1P in Cancer

SPHK1 has been reported to express in higher level in tumor tissues when compared to

normal tissues (Hong et al., 1999, Xia et al., 2000) Moreover, it has been proposed that SPHK1, if up-regulated, could act as an oncogene (Xia et al., 200) Furthermore, it had

been found that the inhibition of SPHK is anti-proliferative and pro-apoptotic for

melanoma cells (French et al., 2003)

It has been suggested, that the involvement of SPHK and S1P in cancer diseases is

partially due to their roles on tumor-associated angiogenesis (Argraves et al., 2004; Hla, 2004; Taha et al., 2006), and partially due to their proliferative roles on the tumor cells

themselves (Ogretmen and Hannun, 2004) S1P could function through S1PRs on the

endothelial cell surface membrane and regulate endothelial cell survival (Limaye et al., 2005), migration (Kimura et al., 2000; Lee et al., 2001), barrier enhancement (Schaphorst

et al., 2003), and blood vessel stabilization via interactions with mural cells (a process requiring N-cadherin) (Paik et al., 2004) All these findings support the proposition that

S1P is closely involved in new blood vessel formation, which is a critical process in tumour establishment and growth

Very recently, a monoclonal antibody against S1P has been developed and shown to bind and neutralize extracellular S1P, at its physiologically relevant concentrations (Visentin

et al., 2006) Moreover, this monoclonal was shown to have potentially therapeutic usage

in reducing tumour growth, invasion, and vessel formation in multiple murine models

(Visentin et al., 2006)

Taken together, all these reports strongly suggest that SPHK1 and S1P are potential novel targets for cancer therapy

1.2.3 SPHK and S1P in Other Diseases

Several groups are providing evidence for a role for SPHK activation in cardiovascular and metabolic pathogenesis, such as atherosclerosis and diabetes

The role of SPHK and S1P in atherogenesis is still controversial, as some studies imply

that the S1P may protect against atherosclerosis (Kimura et al., 2001; Nofer et al., 2004),

while others indicate that S1P may be involved in the onset and/or development of

atherosclerosis (Xia et al., 1998; Auge et al., 2000; Siess et al., 2000; Taha et al., 2006)

S1P has been found to form a complex with high-density lipoproteins (HDL), and density lipoproteins (LDL); with HDL containing more S1P than LDL and very low-

low-density lipoproteins (VLDL) (Xu et al., 2004) Oxidized LDL is a major risk factor for

atherosclerosis, and it can sequentially induce sphingomyelinase, ceramidase and SPHK

in smooth muscle cells, resulting in S1P production and enhanced mitogenesis of these

cells (Auge et al., 1999) Other growth factors, such as basic fibroblast growth factor (bFGF), have been shown to induce hyper-proliferation by activating SPHK (Xu et al., 2002) Xia et al (1998) reported that in endothelial cells, TNFα induced ERK and

nuclear factor κB (NF-κB) activation through SPHK activation, while HDL could inhibit all of these, by inhibiting the SPHK activity triggered by TNFα These findings suggested

an atherogenic role for SPHK activation Interestingly, Nofer et al (2004) reported that

S1P functioned as an anti-atherogenic, hypotensive, and vasoprotective molecule On the

other hand, Deutschman et al (2003) reported that S1P appeared to be more predictive

indicative of atherogenesis in clinical trials than many other well-established risk factors, indicating that the levels of serum S1P correlate with the severity of the disease This would indeed suggest that SPHK/S1P could be a potential therapeutic target for atherosclerosis

The hyper-proliferative role of SPHK and S1P has been proposed to contribute to the

early stages of diabetic nephropathy (Katsuma et al., 2002 and 2003; Geoffroy et al.,

2004) In recent reports, streptozotocin-induced diabetes enhanced neutral ceramidase and SPHK activities, which resulted in increased mesangial proliferation, key events in

the pathogenesis of diabetes (Katsuma et al 2002; 2003; Geoffroy et al 2004)

In summary, SPHK and S1P appear to be involved in various pathological processes SPHK activation generates S1P, which can function as an intracellular mediator, as well

as an extracellular mediator by binding to its receptors to stimulate various downstream signaling pathways, to regulate cell survival, cell proliferation and migration Moreover, these processes are key events in the various pathological conditions, as discussed above Furthermore, the literature discussed also indicates that SPHK and/or S1P are involved in inflammatory diseases, cancer, atherosclerosis and diabetes

In addition to their roles in several pathogenic processes as discussed above, recently, SPHK and S1P have been suggested to be involved in stem cell proliferation and differentiation

1.3 SPHK AND S1P REGULATION IN STEM CELL RESEARCH

As was mentioned above, S1P can act as an intracellular second messenger and as an extracellular ligand for specific cell-surface receptors to regulate cell survival, in either case S1P has been shown to promote cell proliferation, morphological changes and

migration (Spiegel et al., 1998; Payne et al., 2002; Spiegel et al., 2003) Interestingly,

recently, it was shown that S1P could function as a growth factor for stem cell

proliferation Harada et al (2004) showed that S1P induced the proliferation of neural progenitor cells from murine embryos, and Donati et al (2007) reported that S1P

mediated proliferation and survival of murine mesoangioblasts

A more recent study, by Pébay et al (2005), investigated the roles of S1P and SPHK in human stem cells and showed that for human embryonic stem cells, S1P also acted as a

growth inducer They also reported that S1P could work synergetically with PDGF on human embryonic stem cell proliferation, which suggests that S1P could be used as one

of the components in developing novel strategies for human stem cells expansion This elegant report broadened the knowledge of the factors known that could promote human embryonic stem cell proliferation

More interestingly, Pébay et al (2005) also found that the inhibition of the endogenous

SPHK by N,N-dimethylsphingosine (DMS), a widely used broad spectrum SPHK inhibitor, blocked human embryonic stem cell proliferation and reduced the levels of

embryonic-specific cell surface markers, maintained by S1P+PDGF Thus, it indicates the involvement of endogenous SPHK and/or S1P in stem cell proliferation and multipotency maintenance

It is well known that in stem cells, there is a balance between cell proliferation and differentiation, which directs the cells to proliferate in an undifferentiated status, or

differentiate into different lineages of subpopulations Pébay et al.’s findings (2005)

might imply that the direct inhibition of endogenous SPHK, by blocking proliferation, could be a method to induce stem cell differentiation into the subpopulations in a shorter time It will be interesting to know, whether the direct inhibition of SPHK facilitates either embryonic and/or adult stem cell differentiation These questions have motivated

my interest in this research area

A general picture is emerging suggesting roles for SPHK and S1P in several pathological processes, but also their utility in stem cell research On the one hand, up-regulated SPHK and S1P correlate with inflammation, cancer, and other diseases such as allergies, which indicates the inhibition of SPHK/S1P could lead to the development of novel therapeutic strategies to treat these diseases On the other hand, S1P has been shown to be

a proliferative factor for various types of cells, including human embryonic stem cells, suggesting S1P as a potential factor for stem cell studies

My research project was motivated by the potential roles of SPHK and S1P in pathological conditions and in stem cell research It is interesting to know that SPHK and S1P, along a single signaling pathway, could mediate so many processes More studies on

these molecules would undoubtedly provide a better understanding of sphingolipids signaling in both basic and clinical research

Derived from the gaps between current knowledge of SPHK/S1P roles in pathological processes and stem cells research, my research project was developed One of the key research aims is to develop novel compounds as specific inhibitors of SPHKs, which might help to suppress inflammation and as potential anti-cancer therapies Another major aim is to study the roles of SPHK and S1P in stem cell proliferation, stemness maintenance and differentiation In particular, we are interested in the role of S1P in human adult stem cell proliferation and stemness maintenance, and how SPHK inhibition would affect human adult stem cell proliferation and differentiation More details about

the research goals and strategies to achieve them are addressed in Section 1.5 Objectives and significance

The significance for the need to study the potential applications and significance of using SPHK and S1P in stem cells research, derived from the current understanding on human stem cells and the strategies used for stem cell expansion and differentiation, are discussed below

1.4.1 Brief Introduction of Human Stem Cells

Diverse stem cells and their progenitors are found and identified in the embryo (Gearhart

1998; Wilmut et al., 2005), fetal tissues (Shmelkov et al., 2005; Shin et al., 2007),

umbilical cord blood (Prindull et al., 1978; Kogler et al., 2006; Ghen et al., 2006), and some specific tissues including bone marrow (Kohsaki et al., 1981; Huang et al., 1992), liver (Blakolmer et al., 1995; Ruck et al., 1996), brain (Orn 1999; Kukekov et al., 1999;

Johansson et al., 1999), eye (Zieske et al., 1992; Tseng et al., 1996), skin (Radu et al.,

2002; Joannides et al., 2004; Roh et al., 2004; Pisati et al., 2007), heart (Frankish 2001; Messina et al., 2004; Wiehe et al., 2005), and kidney (Bussolati et al., 2005; Dekel et al.,

2006) Stem cells have a multitude of differentiation potentials The embryonic stem cells,

which are derived from the inner cell mass of the blastocyte (Thomson et al., 1998; Hwang et al., 2004), possess the most unlimited self-renewal and broadest differentiation

potential, compared with adult or tissue-specific stem cells Thus, embryonic stem are believed to be a suitable source for cellular therapies, as well as an approachable model for studying early human development

Despite their great potential in both clinical and basic research, human embryonic stem cells are not free from controversy As much as they have attracted interest from scientists, they have also provoked many heated debates due to the ethical dilemmas they pose

Compared with the “totipotent” embryonic stem cells, a large group of tissue-specific stem cells, which are also called adult stem cells, possess less self-renewal ability and more committed differentiation potentials but no controversy, making them a better choice for research experimentation, despite their limitation In my research project, I utilized tissue-specific human adult stem cells to study the roles of SPHK and S1P in proliferation and differentiation More specifically, I used human bone marrow (BM-)

and adipose-derived (AD-) mesenchymal stem cells (MSCs) Details about these two types of MSCs are described below

1.4.2 Human BM- and AD- MSCs

MSCs possess the potential for multiple mesenchymal differentiations, as well as a

reasonable high capacity for self-renewal (Pittenger et al., 1999; Prockop, 1997) The

main source of MSCs in the human body is found in BM, as well as in embryos and cord blood The MSCs harvest procedure from BM is highly invasive, which makes it difficult

to get enough donations from healthy people Also, we know from several reports that an increase in age affects MSC maximum expansion and multilineage differentiation

potential (Mueller et al., 2001; Stenderup et al., 2003)

Another source of MSCs that was more recently reported and characterized is human

adipose tissue (Zuk et al 2001) It has been shown that these cells are able to expand and differentiate in vitro into adipogenic, chondrogenic, myogenic and osteogenic cells under suitable culture conditions (Zuk et al 2001 and 2002; Lee et al., 2004)

1.4.3 Characterization of BM- and AD- MSCs

MSCs from BM provide support to hematopoietic stem cell growth, self-renewal and committed differentiation Thus they are sometimes also called marrow stromal cells

They were initially isolated and expanded by Friedenstein et al (1968), and further characterized by other researchers (Mets et al., 1981; Owen et al., 1998; Clark et al., 1995; Bruder et al., 1997; Zohar et al., 1997; Pitternger et al., 1999) according to their

capability of adhesion to culture flasks MSCs from adipose tissue were first isolated