WO2014006379A1

WO2014006379A1 - Stem cell culture with modulators of the g protein signal transduction pathway

Info

Publication number: WO2014006379A1
Application number: PCT/GB2013/051719
Authority: WO
Inventors: Paul Alexandre DE SOUSA; Robert Peter Millar; Stephen Christopher PELLS; Alexander Ermakov; Mark Bradley
Original assignee: University of Edinburgh
Current assignee: University of Edinburgh
Priority date: 2012-07-04
Filing date: 2013-06-28
Publication date: 2014-01-09
Anticipated expiration: 2015-01-04
Also published as: GB201211873D0

Description

STEM CELL CULTURE WITH MODULATORS OF THE G PROTEIN SIGNAL

TRANSDUCTION PATHWAY

FIELD OF THE INVENTION

The present invention relates to methods, uses, compositions and kits for the renewal and maintenance of stem cells. In particular . the invention provides methods and compositions suitable for maintaining stem cells m a proliferative and or pluri potent state.

BACKGROUND OF THE INVENTION

Maintenance of human plunpoient stem cells (hPSCs). including human embryonic stem cells (hESCs) and human induced piuri potent stem (IPS) cells i an undifferentiated state is realised by a multitude of extrinsic factors whose effects are mediated by overlapping intracellular signal transduction pathways. Those Implicated to date have included pathways stimulated by ligands such as basic Fibroblast Growth Factor (bFGF), Wnt and members of 4ε Transforming Growth Factor Beta (TGFP) saperfamily (Okita & Yamanaka, 2006). Binding of bFGF to its cognate receptor normally results in activation of receptor tyrosine kinase activity and subsequent stimulation of the Ras-aciivated nritogen-associated protein kinase (MAPK), phosphati dy i inositol -3 kinase (M3 .)/Akt/Protem kinase B (PKB) and PhosphoHpase €-gamma (PLC-γ) pathways, the former two of which have been shown to be active in hESCs (Dvorak et a!., 2005; Kang et aL, 2005). In contrast, binding of Wnt proteins to cell surface receptors of the Frizzled (FZD) family activates Dishevelled family proteins which in turn inhibit a complex of proteins (axin GSK3/APC) responsible for the proteolytic degradation of the β-catenin intracellular signalling molecule and transcription co-factor, Stabilising the cytoplasmic pool of β-caienrn by pharmacological inhibition of GSK3 cars promote embryonic stem cell renewal together with oilier supportive factors (Sato, Meyer, Sk ftsouras, Greengard. & Brivaniou, 2004). In addition, TGFp and Nodal, expressed by undifferentiated hESCs, and Activin-A, normally secreted by fibroblast feeders which are supportive of hESC renewal, all work to oppose induction of trophoblast/primitive endoderm differentiation by Bone Morphogenic Protein (BMP}- , another member of the TGFp superfamily (Beattie et s , 2005; Sato et aL 2003). This is mediated by activation of different coll surface receptors (Type II vs I) and downstream transcription factors (5MAD2/3 vs SMAB1/5/8) (James, Levine, Besser, & Hemmati-Brivanlou, 2005; Vaidimarsdottir & Mummery. 2005). Understanding the means by which the diverse pathways governing hESC fate can be modulated by extrinsic factors sad the relative importance of other signal transduction pathways to mediate these effects is of importance for the development of safe and efficacious culture systems. A fundamental starting point with which to advance this understanding is to focus on heterotrimeric GTP-binding (or G) protein signalling, G-proteins, comprised of , β, and γ subumls_s are centra! components of primary mechanisms used by virtually all cells to respond to structurally and chemically diverse extracellular stimuli (e.g. ions, amino acids, peptides, nucleotides, lipids, biogenic amines, light) in diverse contexts including development and cancer (reviewed in (Dorsam & Gutkind, 2007; Malhos, 2005). Signalling via specific subuiiits and isoforms is classically understood to be coupled to hsptahelical or seven transmembrane spanning receptors (also known as G protein-coupled receptors, GPCRs), but can also he coupled with other types of receptor including single transmembrane spanning tyrosine kinase receptors for growth factors such as EGF, PDGF, insulin, and IGFs and extracellular proteins such as integrins (reviewed by (PateL 2004).

G-Protein signal transduction is well established. In the resting state, the G_a subunit is bound to GDP and is associated with the Gp and G₁ subunits. Upon Kgand ¾ind.ing, an associated receptor activates the trirneric G protein that it is coupled to, resulting in GDP to OTP exchange and dissociation of the G_a from Gp and G _γ subunits followed by G_a and βγ subunit binding to and activation of respective effectors. subunits are divided into tour groups based on their sequence homology and their immediate effector function; G_as and G subunits stimulate or inhibit adenyl cyclases and the production of cAMP second messengers, respectively, with certain members of the latter also regulating cGMP-gated NaVCa*^""^" channels. ^'The G subunit stimulates phospholipase C-β (PLC-β) and the production of the second messengers inositol triphosphate (IPS) and diacylglycerol (DAG). The G_aivn subunits stimulate the low-molecular-weight G protein Rlio and its downstream targets. Lastly, βγ subunit complexes appear to activate/associate with a broad range of effectors including Na⁺. K^: and Ga^* ion channels, mitogen-activated protein kinases, adenyl cyclases. PLC-β and ε and small G proteins (reviewed by (McCudden, Hains, Kimple, Siderovski, & Wiiiard, 2005).

Despite the fundamental role for G proteins in signal transduction, their significance to embryonic stem cell renewal remains relatively unknown, in mouse ESCs blocking or stimulation of metaboiroplc glutamate receptors acting via the 0·_;,_: snhunit in medium supplemented with serum and Leukaemia inhibitory Factor inhibits the expression of ploripotency transcription factors Oct4 and Nanog or sustains ESC renewal, respectively (Cappuccio et al, 2005). In hESCs, the mitogerjJc GPCR ligand sphmgosine- 1 -phosphate (SIP) synergistically stimulates self-rene ai with PDGF in serum-free medium conditioned by fibroblasts (Pebay et at, 2005).

Stem cell research is set to enter- a new era defined by translation to industrial and clinical practice. This translation is critically reliant on efficacious systems for stem cell production, specification and deliver}'. Currently most culture systems intended to sustain ste n cell renewal or direct differentiation are challenged by a lack of definition in media components, and a lack of Good Manufacturing Practice (GMP) quality reagents required for transition of stem cell-based therapies from the laboratory to the clinic. Over the last 5 years culture media have become more defined, first rising purified aibumin protein instead of serum, and later have even eliminated the requirement for albumin protein altogether. Similarly, there has been a shift away from rising animal- (eg. blood}- derived components in culture media mid replacing those with recombinant human forms of the proteins, made in plants or bacteria. Yet despite these improvements, the dependency of hPSC maintenance on supplementation of media with exogenous cytokines (either directly or indirectly), in particular bFGF and/or TGPb remains.

These generally:

* Rarely yield pure homogeneous populations of the cell type they are intended to support. This is due to the dynamic instability of stem ceils and the inability of conventional practice to recapitulate and control the niicroenviromnentai factors which determine cell behaviour and specification.

* Are costl , owing to their rel ance on expensive protein growth factors which frequently must be added in excess to compensate for their competitive nonspecific consumption or degradation by contammating cell subpopulations.

® Can constitute a source of potential infection and of unpredictable and poorly- controlled variability to the system, due to the requirement for undefined human- or animaj-sourced reagents.

The aforementioned challenges may be overcome with a shift in cell culture systems away from media containing protein growth factors, to the use of synthetic drugs or other molecules wliicli alone or in combination target the specific pathways required for stem cell growth, self renewal and difFerentiation in the desired fashion. It is amongst the objects of the present invention to provide a chemically defined medium substantially free from growth factors such as bFGF and/or TGFp.

SUMMARY OF THE INVENTION

The present invention is based on the finding that intracellular ealdiim downstream of G-protein signalling plays a role in the maintenance of undifferentiated stem cells - in particular, human stem cells,

In a first aspect, the present invention provides a method of modulating th fete of a stem cell, said method comprising the step of providing one or more G- protein coupled receptor (GPCR) agonists, antagonists and/or compounds which activate one or more GPCRs or calcium signalling events downstream of GPCR signalling to the stem cell, in order to modulate its fate. The method may include modulating the concentration or level of intracellular calcium said stem cell.

As explained in more detail below, references to the modulation of intracellular calcium may be taken to encompass the modulation of the concentration or level of "free" intracellular calcium within the cytoplasm of a cell,

A stem cell may adopt one of three fates. Firstly, a stem cell may opt to self re- new and remain in an undifferentiated, proliferative and/or pluripotent state. Secondly, a stem cell may opt to differentiate (perhaps in response to one or more chemical signals) into one or more other cell types. Thirdly, a stem cell may die and/or undergo programmed cell death (apoptosis). As <; ck the ''fate^'" of a cell may be regarded as a stem cell's decision to maintain an undifferentiated, proliferative and/or pluripotent state, differentiate towards one or more cell types or die apoptose.

The methods described herein may be exploited to promote self-renewal in stem cells by maintaining stem cells in an mulifferentiated, proliferative and/or pluripotent state, and/or opposing differentiation or death. la a preferred embodiment the present invention is intended to maintain stem cells in an undifferentiated states by treatments with chemicals or gene products which may affect GPCR and calcium signal transduction specifically or non-£¾>ecifically, without conventional reliance on the provision of exogenous protein growth factors, supplied directly, or indirectly via supplementation of the cell culture environment with serum, or co-culture or conditioning of the environment with helper cells.

It should be understood that the methods for maintaining stem cells described herein preserve or retain the pluripotent phenotype of stem ceils for a period prolonged period of time and/or over many passages. Additionally, or alternatively, stem cells subjected to aspects of this invention may, throughout the period of maintenance, may retain one or more stem cell markers. As such, the term "maintain" or "maintained" as used herein, may refer to the retention of a pluripotent phenotype in stem cells in culture. Additionally or alternatively, these terms may relate to a stem cell's ability to self renew in culture.

Accordingly, this invention may provide a method of mamtaining stem cells, said method comprising the step of providing one or more GPCR agonists, antagonists and ox compounds which activate GPCRs or calcium signalling events downstream of GPCR signalling to the stem cell and may involve modulating the concentration or level of free intracellular calcium in the cytoplasm of said stem cell.

The reader will appreciate that, as used herein, the term "stem cells'⁵ may be taken to refer to any cell which is able to self renew and indefinitely divide - cells of this type may be described as "immortal". In addition, when cultured under suitable conditions and or contacted with, or exposed to, particular compounds and/or conditions, stem cells may differentiate into one or more of the specialised cell types which form embryonic and/or ad a It tissues.

The term "stem cells" may encompass embryonic, foetal, adult and or induced pluripotent (IPS) stem cells. The term "stem cells" may further encompass progenitor cells of any type. The stem cells described herein may be mammalian cells; for example, the term "stem cells" may be applied to human and non-human stem cells of all types. By wa of example this invention may relate to stem cells derived or obtained from, or provided by, primates, ungulates, ruminants and/or rodents (specifically, sheep, pigs, cattle, goats, horses, rats end mice).

Stem cells may be characterised by the presence of one or more markers selected from the group consisting of: ABCG2; ACE; ALCAM; Alkaline Phosphatase; beta-UI Tubulin; BMP-2; BMPR~1A/ALK~3; B PR-IB/AiJt-6; BMPR-II; E-Cadherin; CCR4; CD9: CD71; CD90; CD90 Thyl; Cripto; CXCR4; DPPA5 ESG1; Endoglin/CD105; FABP^'l; FABP2: FGF-4; FGF R4; FoxD3; FoxP3; Frizzled-9;GAD 1/GAD67; GATA-4; GATA-6; GDF-3; Glutl; HNF-3 beta; Integral alpha 6/CD49f; Integrin beta 1/CD29; Lefty; MAP2; Musashi-l; Nanog; NCA -L1; Nectra-2/CD112; Nestm; NeuroDl; Nodal; Noggin; NF-L; NF-M; Nucleostemm;

Otx2; Oct ¾; PAX6; Podoealyxin; Proa lniii 2: ROB03; Sca-I; SCF R/okit; SHO; SOX2; SOX?; SOX17; SPARC; SSEA-l; SSEA-3; SSFiA-4; STAT3; STRQ-1; TP63/TP73L; Tyrosine Hydroxylase: gaiama-Secretase; a!pha-Seeretase; beta- Secretase; beta- III tubulin; alpha-Fetoprotein; beta-Catenin; Vimentiri and VCAM-1. Collectively, these markers may each be referred to as stem cell markers and references in this specification to one or more "stem cell markers", may therefore encompass one or more of the abovementioned markers.

To identify or detect a stem cell, a cell may be probed (using, for example antibodies or other agents capable of binding one or more of the listed stem cell markers) for the presence of otic or more of the stem celi markers listed above. The presence of one or more of the stem cell markers indicates that the cell is a stem cell.

One of skill will appreciate that those markers characteristic of any given type of stem cell may differ. Markers characteristic of specific stem cell types are described below.

Stem cells may be embryonic stem ceils (ESC), for example, mammalian and/or human embryonic stem cells (hESCh derived from early stage embryos and in particular from the inner cell mass of the developing morula or blastocyst. Embryonic stem cells derived from embryos immediately following conception (and for a short time thereafter) may be totipotent and thus capable of generating a complete viable organism as well as any given specialised eel! type. Embryonic stem cells derived from later stage embryos (e.g. from the inner ceil mass of a developing blastocyst) may be plnripotent and thus not capable of generating a complete viable organism, but capable of differentiating to any specialised cell type. As such, the present invention may be applied to embryonic (i.e. totipotent and/or pluripotent) stem cells.

Markers of ESCs may include, for example. ABCG2, Alkaline Phosphatase, B-Cadherin, CCR4, CD9, Cripto, DFPA5/ESG1, FGF-4, FGF R4, FoxD3, FoxP3, GDF-3, Integra alpha 6 CD49C hitegrin beta 1/CD29. Lefty, anog, Oct-3 4, Podoealyxin, SOX2, SPARC, SSEA-l, SSEA-3, SSEA-4 and STAT3. A stem cell, for example an embryonic stern cell, to be subjected to the methods described herein, may be characterised by the presence of one or more of the above mentioned, stem cell markers. The term "stem cells" may also be taken to refer to the multipoteot cells derived from any of the three primary germ layers (ectoderm, mesoderm and endoderrn) which develop during the process of g&srrulation. Cells derived from these layers may express one or more markers which may he used as a means of identification. By way of example, ectoderm germ layer may express markers, including, for example, 01x2, Nestin, TP63/TP73L, beta-Ill Tubulin, SHH, and PAX6. Ectoderm has the potential to form cell types such as neurons and early neuronal lineage markers include ACE, ALCAM, CD90/Thyl₅ GADI/GA.D67, Gluil , MAP2, NCAM-L1, Necm 2/CDI 12, NeuroDl, F-L, MF-M, ROB03, gamma- Secretase, aipha-Secxetase, beta-Secretase, beta-ΪΠ tubulin. Tyrosine Hydroxylase. Neural stem cell markers include ABCG2, CXCR4, FGF R4, Fnzzied- , MusasM-l, Nestm, Noggin, Nucleostemm, Prommfn 2, SOX2, Vimentirs. Markers of early endodermal cells include, for example, FABPL FABP2. GATA-4, HN^'F-3 beta (collectively referred to as definitive endodermal stem cells markers) as well as those markers for primitive endoderm such as alpha-Fetoprotein, beta-Catenin, GATA-4, SOXI7 and SOX7.

This invention may be further exploited in the maintenance of mesenchymal stem cells or cultures or populations thereof. One of skill in this field will appreciate that mesenchymal stem cells are pluripotent in thai they have fee capacity to form a number of more specialised cell lineages including, for example chondrocytes, osisQcytes, adipocytes, eardiomyocyiss, myoblasts and cells of the connective tissue such as, for example, fibroblasts. In addition, mesenchymal stem cells may he characterised by the fact that they express a number of specific markers including CD7i , CD90, GATA6, Nodal, BMP-2. Furthermore, mesenchymal stem cdls may express one or more of the markers selected from the group consisting of BMPR- IA AL -3; BMPR~iB/ALK-6; BMPR-ΪΪ; Endoglin/CD105; Nucleostemin; Sca~I; SCF : /c-kit; STRO-1 and VCAM-1.

"Adult" stem cells may be obtained from adult animals and or adult (or developed/differentiated) tissue (including adult humans and/or human (adult) tissue). However, it should be understood that the term "adult" also includes stem cells derived -from neonatal, infant juvenile and/or adolescent animals. Adult stem cells .may be soureed fr m any suitable tissue, including bone marrow and/or specialised structures such as, for example hair follicles, skin, teeth and the like. Stem cells to which this invention may be applied may be obtained from a variety of sources including, for example, embryonic animals (including human embryos), said embryos being either aborted or created as part of a fertility program. Alternatively, it may be possible to obtain stem cells ftora established stem cell lines, thus avoiding the use of mammalian, particularly human, embryos. Examples of established stem cell lines may include, for example, RHl, RH6 (Fletcher et a!., 2006) and RCM1 (De Sousa et al., 2009).

One of skill will appreciate that hESCs and other cell lines for use in the methods described heroin may be obtained from an embryo without destruction of the embryo, as described, for example, in Chung et al (Cell Stem Cell, vol 2, issue 2, 113- 117, 2008). Stem cells may also be generated using the methods described by Chung et al, (2006) which methods involve taking a blastomere cell from an early stage embryo prior to formation of the balstocyst (at approximatel the 8-cell stage) and co- culturmg this cell with, established stem cell lines to generate a fully competent stem cell line.

Alternatively, the methods of Meissner & Jaenisch (2006) may be used to obtain stem cells to which this invention may be applied. In. these methods, the cdx2 gene is silenced in the donor nucleus during the process of nuclear transfer to prepare a reconstructed embryo from which a line of embryonic stem cells is derived. The cdxl gene is turned back on in the isolated, blastocyst cell taken from the embryo which is used to prepare the cell line. This is an example of so-called, "Alternative Nuclear Transfer" where the embryo is not capable of implantation but the stem cell line derived therefrom is full competes t

The term "stem cells" may also encompass cells otherwise known as induced pluripotent stem cells (iPS). These are re-programmed adult somatic cells which have been modified to express certain factors (such as transcription regulators) and, as a consequence, become pluripotent and thus capable of differentiating to any other specialised cell type. As such, iPS cells, particularly mammalian, for example human or rodent iPS stem cells, may be cultured and/or maintained using any of the methods and/or compositions described, herein.

Desirably the compounds of the present invention are small molecules, typically less than 100 or 50 kDa, more preferably less than 10. 5 or 1 kDa. in a preferred aspect, the compounds of the present inventio ate intended to allow for the replacement ofbFGF and/or TGFp in the medium in which the stem cell, is supported ami or maintained.

The stem cells of the present invention may include inducible genome integrated or non-integrated nucleic acid constructs encoding G-protem coupled receptor (GPCR)s, which can. then be activated by exposure to GPCR agonists,

Compounds for use m this invention may comprise G-protein coupled receptor (GPCR) agonists, antagonists and/or compounds which, activate GPCRs (eg. Lysophosphatidic Acid, Epinephrine, Sulpiride) or calcium signalling events downstream of GPCR signalling (Bromohexine-HCL, Hydralazine, Prednisolone, Sulfisoxazole, Heparin, Thyroxine, Benzoperoxide, Chloropropamide, Mesalamine, Tolfenamic acid, Menadione, Metfoimin, MetoproloL Chloroquine, Riboflavin).

Compounds for use in this invention may affect heterotrimeric G-protein sabunits (G₀ from Gp and G _y. subunits) or the signal transduction molecules mat heterotimeric G-proteins interact with in order to modulate free intracellular calcium levels/concentration such as for example phosp^'holipase C beta,

Compounds for use in this invention may inhibit sarcoplasmic endoplasmic reticulum calcium ATPases (SERCAs, eg. Thapsigargin, Cyclopiazonic acid, Clotrimazole, C oropromazi e). Compounds of this type (and otliers described herein) can raise free intracellular calcium con.centrat.ioos/leYels, and, without wishing to be bound by theory, downstream molecular signalling promotes self renewal of stem ceils.

Compounds for use in this invention may modulate surface plasma membrane and/or intramembranaus ion channel permeability (eg. Aeetazolamide, ( oropropamide, Cmnarizine, Quinine). Compounds of this type may affect the exchange of calcium between extracellular and intracellular compartments. Suitable compounds may include "small" G proteins (for example between about 20 and about 25 fcDA in size). By way of example, compounds for use in this invention may comprise members of the Ras oncogene family that can activate ion channels.

Compounds suitable for use in this invention may directly stimulate substrates that bind calcium, such as for example Calmodulin, or that mediate calcium signal transduction such as for example the calcium modulated kinase (CaM H).

The term "modulation" as applied to intracellular calcium concentration and/or levels, encompasses any increase and/or decrease in the concentration and/or level of free intracellular calcium relative to, for example, that which is sequestered or bound in membranous organelles in the cytoplasm or nucleus. A modulated level of free intracellular calcium may be compared to a normal, reference or control concentration/level of free intracellular calcium. A normal, reference or control concentration/level of free intracellular calcium may be thai present in a stem cell which has not been subjected to aspects of this invention. Free intracellular calcium concentrations and/or levels may be determined using fluorescent calcium indicators monitored by fluorescence activated cell sorting (FACS) or cytometry, and optical, digital video, confocal laser scanning, and multi-photon microscopy.

The term "free intracellular concentration" or "free intracellular level" of calc um , may relate specifically to the concentration or level of cal ium within the cell cytosol cytoplasm as opposed to calcium which is sequestered or harboured within membranes and/or organelles.

As stated, through modulation of the free intracellular calcium concentration level , it is possible to modulate cell fate^■■■■ that is to say, by increasing free intracellular calcium concentrations and or levels in a stem cell, that stem ceil may be maintained in a proliferative/pluripotent state and/or self renew. Conversely, by decreasing the concentration ievel of free intracellular calcium in a stem ceil, it may be possible to force mat stem cell to differentiate or die.

The methods provided in accordance with this invention may exploit compounds which are capable of modulating free intracellular calcium concentrations levels. Such compounds may be used to maintain stem cells in proliferative and/or pluripotent states and/or to control a stem cell's decision to self- renew, remain proliferative and/or lurixstent, differentiate or die.

The compounds capable of modulating intracellular calcium concentration levels may be compounds which increase the free intracellular concentration/level of calcium.

Compounds suitable for use in the methods described herein may comprise amino acids, peptides, proteins, nucleic acids (RNA DNA), carbohydrates and/or small organic/inorganic molecules.

Suitable examples of compounds for use in the methods described herein may include one or more selected from the group consisting of:

(i) Cyelopiazottic acid (CPA);

in) Thapsigargin;

(iii) Lysophospbatidic Acid (LP A): (iv) Clotrimazole; and

(v) C orpromazine,

Suitable examples of compounds for use in the present invention and in particular to replace and/or minimise tlie need for bFGF and/or TGFp include one or more selected from the group consisting of :

(vi) Acetazolarnide

(vii) Bromhexine HCL

(viii) £piiiephrme

(is) Hydralazine

(x) Prednisolone

(xi) Sulfisoxazole

(xii) Sulpiride

(xiii) Heparin

(xiv) Thyroxine

(xv) Benzoperoxide

( vi) Chloropropamid©

(xvii) C iiarizine

(xvii!) Mesalarriine

(xix) Quinine

{xx} Tolfenamic Acid

(xxi) Menadione

(xxii) Metformin

(xxiii) Metopfoiol

(xxiv) Chiofoquine

(xxv) Riboflavin

The compounds for use in the various methods described herein may be provided at specific concentrations. For example, the compounds individually or in combination maybe used at concentrations of between about O.lnM and about 50μΜ, such as μΜ to 25μΜ_> e.g. ΙΟμΜ. Such values may be appropriate if the compounds are delivered by conventional solubilisation in media. However, lower amounts may be appropriate if the compounds are delivered by an alternative means involving affinity targeting to cells specifically as for example exemplified by Corradetti et al„, 2012, Biomaterials , 33 (28): 6834-43. in some cases, low concentrations of the compounds described herein may be used, in other cases, high concentrations may be used. For example, CPA may be use at a concentration of about 100, 200, 300, 400, 509, 600, 700, 800 or 900 mVL CPA may be used at concentrations of about 1 , 2, 2.5, 3, 4, 5, 6, 7, 8, 9 or ΙΟμΜ. Thapsigargin may be used at a concentration of about 0.3, 1, 5, 10, 15, 20, 25, 30, 35, 40 or 50 riM. LPA may be used at concentrations of about 1, 5, 10, 20, 30, 40, 50, 60, 70, 80, 90, 95 or l OQnM, LPA may be used at concentrations of about 0.5, 1, 1.5, 2, 3, 4, 5 or ! 0μΜ. Clotrimazole and Chlorprornazine may be used at concentrations of between about ΙμΜ a d about 50μΜ, for example 1.5, 2, 3, 5, 10, 20, 25, 30, 35, 40 or 45μ .

in addition, the inventors have discovered that modulation of GPCR-mediaied signalling and/or free intracellular calcium (and hence any of the compounds described herein) may used to compensate for a lack of exogenously supplied bFGF and/or TGFp in stem cell culture. In other words when maintained in culture systems which lack bFGF and/or TGFp, the inventors have discovered that it is possible to maintain stem cells in a plunpotent and/or proliferative state by way of the use of small molecules as taught herein.

As such, a further aspect of this invention provides a method of maintaining stem, cells in the absence of basic fibroblast growth factor (bFGF) and/or ΤΟΡβ, said method comprising the step of maintaining stem cells in the presence of an compound capable of maintaining the tern cell in an undifferentiated state.

The method may comprise the step of maintaining stem cells in the presence of one or more G-protein coupled receptor (GPCR) agonists and/or compounds which activate GPC s or calcium signalling events downstream of GPCR signalling fox example compounds which modulate the concentration/level of intracellular calcium within the stem, cell to he maintained. Suitable compounds may include air of those described herein.

By way of example, may comprise maintaining stem ceils in the presence of one or more compounds selected from the group consisting of:

(i) G~protein coupled receptor (GPCR) agonists (eg. Lysophosphatidic Acid, Epinephrine, Sulpiride); (ii) compounds which activate GPCKs or calcium signalling events downstream of GPCR signalling (Bromohexine-HCL, Hydralazine, Prednisolone, Sdfisoxaxole, Heparin, Thyroxine, Benzoperoxide, Chloropropamide, Mesalarnine, Tolfenamic acid, Menadione, Meifonnin, Metoprolol, CMoroquuie, Riboflavin).

(iii) compounds capable of modulating free intracellular calcium concentration levels (eg. Acetazoiamide, Chloropropamide, Cinnarizine, Thapsigargin, Cyclopiazonic a id, Clotrimazole, CWoropromazine).

(iv) compounds which inhibit sarcoplasmic endoplasmic reticulum calcium (eg. eg. Thapsigargin, Cyclopiazonic acid, Clotrimazole, Chloropromazine).

(v) nucleic acid constructs encoding G-protein coupled receptor (GPCR)s, optionally togetliei^* with a ligand for the receptor.

(vi) compounds which affect heterotrimeric G-protein subunits (G_a ftom Gp and G _r subunits) or the signal transduction molecules that heterotimerie G-proteins interact with.

(vii) compounds which modulate surface plasma membrane and/or intramembranous ion channel permeability (eg. Acetazoiamide, Chloropropamide, Cinnarizine, Quinine). ; and

(viii) compounds which directly stimulate substrates that bind calcium or mediate calcium signal transduction.

Examples of each of compounds (i) - (viii) are provided above.

In a further aspect of this invention, there is provided a method comprising the step of maintaining stem cells in the presence of one or more compounds selected from the group consisting of:

(i) Cyclopiazonic acid (CPA);

(ii) Thapsigargin;

(iii) L sophosphatidic Acid (LP A);

(iv) Clotrimazole; and

(v) C orproinazine.

or

(vi) Acetazoiamide

(vii) Brornhexine BCL

(ixj Hydralazine

(x) Prednisolone (χί) Sulfisoxazole

(xii) Sulpiride

(xiii) Heparin

(xiv) Thyroxine

(xv) Benzoperoxide

( vi) Qi!oroproparnide

ixvii) Cionarizine

(xviii) Mesal amine

(xix) Quinine

(xx) Tolfenamic Acid

(xxi) Menadione

(xxii) Metformin

(xxiii) Metoprolol

(xxiv) CMoroq¾ine

(xxv) Riboflavin

In a further aspect, the present invention provides a composition for maintaining stem ceils, e.g. in an undifferentiated state said composition comprising one or more compounds as discussed herein.

Compositions of the type provided by the invention may otherwise be known as cell culture media. As such, the t! ird aspect of this invention provides a cell culture medium for maintaining stem cells.

The composition and/or medium may not contain bFGF and/or TGFp and preferably does not. contain either

A composition or medium according to the invention may be used to maintain stem cells in a proliferative and/or pluripotent state for a period of time, such as at least. 5. 6, 7, 8, 9, 10 days or longer and/or for a number of passages, by which time stem cells cultured in the absence of bFGF and or TGF-beta will have will have undergone a more pronounced differentiation. For example, the composition or medium provided by this iiiveiitioo may be used to maintain stem cell cultures in an undifferentiated state for period of time (for example a prolonged period of time, over multiple passages and/or substanti lly indefinitely)

A composition or medium provided by this invention may comprise one or more compounds selected from the group consisting of: (ί) G-protein coupled receptor (GPCR) agonists;

(ii) compounds which activate GPCRs or calcium signalling events;

(iii) compounds capable of modulating intracellular calcium concentration/levels;

(iv) compounds which inhibit saKX>piasmic/endoplasmic reticulum calcium

(v) nucleic acid constructs encoding G-protein coupled receptor (GPCR)s, optionally together with a ligaad for the receptor

(vi) compounds which affect heterotrimeric G-protein subunits (G_a from Gg and Gy.. subunite) or the signal transduction molecules that lieierotimeric G-proteins interact with.

(vii) compounds which modulate surface plasma membrane and/or intramembranous ion channel permeability and

Specific compounds which are examples for (i) - (viii) above have been listed above and are applicable here also.

A composition or medium according to the third aspect of this invention .may comprise one or more compounds selected from the group consisting of:

Cyclopiazonk acid (CPA):

(iii) Lysophosphatidic Acid (LP A);

(iv) Clotrimazole; and

(v) Chlorpromazine.

or

(vi) Acetazolamide

(vii) Bromhexine HCL

(viii) Epinephrine

(ix) Hydralazine

(x) Prednisolone

(xi) Suifisoxazole

(xii) Sulpiride

(xi.il) Heparin

(xiv) Thyroxine

(xv) Betrzoperoxide Chloropropamide

(xvii) Qntmriziae

(xviii) Mesalamine

(xk) Quinine

(xx) Tolfenamic Acid

(xxi) Menadione

(xxii) Metformin

(xxiii) Meioproiol

(xxiv) CMoroqdne-

(xxv) Riboflavin

Tills invention may provide & composition or medium for maintaining stems, the composition or medium comprising one or more compounds selected f om group consisting of:

(i) Cyelopiazorde acid (CPA);

(n) Thapsigargln;

(iii) Lysophosphatidie Acid (LP A);

(iv) Clotrimazole; and

(v) Chlorpromazitte;

wherein the medium doss not comprise FGF.

or one or .mors compounds selected from the group consisting of

(i) Cyclopiazonic acid

(ii) Thapsigargin;

(iii) Lysophosphatidie i

(iv) Clotrimazole;

(vi) CWoipromazine;

(vi) Acetazoiamide;

(vii) Bromhexine HC1;

(viii) Epinephrine;

(«) Hydralazine;

(x) Prednisolone;

(xi) SulBsoxazole;

(xii) Sulpiride;

(xiii) Heparin; and (xiv) Thyroxins

(xv) Bertzoperoxide

(xvi) Chioropropamide

(xvii) Cianarizme

(xviii) Mesalarnine

(xix) Qiiinine

(x .) Tolfenamic Acid

(xxi) Menadione

(xxii) Metformin

{xxiii} Metop.ro loi

(xxiv) Cliioroq ioe

(xxv) Riboflavin

'herein the medium does not comprise bFGF and or It should be understood that any of the compositions or media provided by this invention may be further supplemented with one or more factors (for example indicators, sugars, growth factors, cytokines, nutrient compounds, vitamins, antibiotics, cells (perhaps in the form of feeder cell layers), substrates, coated substrates and the like) so as to permit stable culture and/or maintenance of the stem cells.

One of skill will appreciate that the one or more compounds described herein may be applied to existing compositions or cell culture media (including compositions and/or media formulated specifically for use with stem, cells) so as to provide compositions or media which may lack bFGF and/or TGFP and which can be used to maintain stem cells.

In. a further aspect, the present invention provides use of a compound, for mamtaining stem cells and/or for inducing self-renewal in stem cells, for example capable of modulating the concentxation or level of intracellular calcium in a stem cell,

Compounds for use in mamtairring stem cells and/or inducing self-renewal in stem, cells may comprise one or more selected from the group consisting of:

(i) G-protein. coupled, receptor (GPCR) agonists;

(iii) compounds which activate GPCRs or calcium signalling events; (ill) compounds capable of modulating intracellular calcium conceotratiori'levds;

(iv) compounds which inhibit, sarcoplasmic/endoplasmic reticulum calcium ATPases (SERCAs);

(v) nucleic acid constructs encoding G-proteia coupled receptor (GPCR)s.

(vi) compounds which affect beterotrimeric G-proteia subanits (G_a from Gg G y. subunits) or the signal transduction molecules that heieroiimeric G~proteins

(vii) compounds which modulate surface plasma membrane and/or intraraembranous ion channel permeability; and

(ix) Cyclopiazonic add (CPA);

(x) Thapsigargin;

(xi) Lysophosphatidic Acid (LP A);

(xii) Clotrimazole;

(xiii) Chlorpromazine;

(xiv) Acetazoiamide;

(xv) Bromhexine HCi;

(xvi) Epinephrine;

(xvii) Hydralazine;

(xviii) Prednisolone;

(xix) Sulfisoxazole;

(xx) Sulpiride;

(xxi) Heparin: and

(xxii) Thyroxine.

(xxiii) Bexizoperoxide

(xxiv) Chloropropamide

(xxv) Cirinanz ie

(xxvi) Mesalamine

(xxvii) Quinine (xxviii) Tolfenamic Acid

(xxix) Menadione

(xxx) Metformin

( X ) Metoprolol

(xxxii) Ch!oroquine

For use in this invention, the compounds described herein, may be formulated (at any of the concentrations noted herein) together with pharmaceutical or GMP grade carriers, excipients and/or diluents.

In a fifth aspect, the invention provides a kit for use in maintaining stem cells, the i comprising a compound capable of modulating the concentration or level of free intracellular calcium in a stem cell.

Optionally, a kit provided by ibis invention may comprise one or more other components selected from the groups consisting of:

(a) cell culture media (to be mixed with, or pre-mixed with, one or more compounds capable of modulating the concentration or level of free intracellular calcium in a stem cell);

(b) optionally sterile vessels and/or vials in which stem cells may be maintained; and

(c) instructions for use

Compoimds for inclusion in the kits described herein may be selected from the group consisting of:

(i) compounds capable of modulating intracellular calcium concentration levels;

(ii) G-protein coupled receptor (GPCR) agonists;

(ill) compounds which activate GPCRs or calcium signalling events;

(iv) compounds which inhibit sarcoplasmic/endoplasmic reticulum calcium

ATPases (SE CAs);

(v) nucleic acid constructs encoding G-protein coupled receptor (GFCR)s.

(vi) compounds which affect heterouimeric G-protein subunits (G_e from Gp and G »,·. subunits) or the signal transduction molecules that heterotimeric G-proteins interact wife. (vii) compounds which modulate surface plasma membrane and/or iniraraembraao s ion channel permesbilfty, and

(viii) compounds which directly stimulate substrates thai bind calcium or mediate calcium signal transduction.

Specific compouads which are examples for (i) - (viii) above have been listed above and are applicable here also.

Cyelopiaamie acid (CPA);

(x) Thapsigargiii;

(xi) L>¾op¾ospliatidic Acid (LP

(xil) Clotrimazole;

(xiii) CMorpromazirse;

(χχχίν) Aeetazolamide;

(xxxv) Bromhexine HQ;

(xxxvi) Epinephrine;

(xxxvii) Hydralazine;

(xxxviii) Prednisolone;

(xxxix) Suifisoxazole;

(xl) Sulpiride;

(xli) Heparin; and

(xiv) Thyroxine

(xv) Benzoperoxide

(xvi) Chloropropamide

(xvii) Cinnarizine

(xviii) Mesdarnine

(xix) Quinine

(xx) Tolfenamic Acid

(xxi) Menadione

(xiii) Metformin

(xxiii) Metoprolol

(xxiv) Chloroquine

(xxv) Riboflavin The present invention may f n her extend to methods of culturing or enriching specific cell types from mixed populations of cells and/or from tissue/biopsy samples. For example, by maintairung a primary tissue sample in the presence of compounds capable of modulating free intracellular calcium concentrations/levels, it may be possible to selectively maintain stem cells. Without wishing to be bound by theory, it is suggested that when contacted with compounds capable of modulating free intracellular calcium concentrations/levels, stem cells within primary tissue may be induced to self renew and/or proliferate allowing them to be selectively enriched.

The invention may further extend to modified stem cells, the modified cells comprising inducible genome integrated or non-integrated nucleic acid constructs encoding G-protein coupled receptor (GPC )s. The GPCRs encoded by such integrated or non-integrated nucleic acid constructs may be activated by exposure to GPCR agonists. Modified stem cells of this type may be maintained in a proliferative, undifferentiated or piaripotent sate by subjecting the modified stem cells to any of the methods, media and/or compounds described herein.

DETAILED DESCRIPTION

The present invention will now be described in detail with reference to the following figures which show;

Figure. 1 : G-protein signalling pathway modulators and significant outcomes of primary screen. (A) Table showing the list of drug compounds and receptor agon sts used in this study, together with their target molecules, specific pathways and their effects. (B) Summary of replicated significant changes (Δ) in total cells (top panel) and AP+ colony frequency normalised to total cells (bottom panel) ^'between drug treatments and respective vehicle or untreated control. Stars indicate the level of significance (* p<0.05; ** p<0.01; ***p«0.001). A cradfi c indicates toxicity so high that a calculation could not be performed. The signalling pathway of which the drug target is a component is indicated. The range of compound concentrations (shaded triangle) tested were as summarised in Table 1 and did not exceed the IC/EC5G (half-maximal inhibitor/activaiing) concentration for each compound.

Figure 2: Modulation of free intracellular calcium and its effectors. (A) Summary of independently replicated significant changes (Δ) in total cells (top panel) and AP+ colony frequency normalised to total cells (bottom panel) between drug treatments targeting the G_fK1-PLCp^«P C pathway and respective vehicle or untreated control. Stars indicate the level of significance (* p<0J)5; ** p≤0.01 ; ***p«0.0l). Calculation of significjance not applicable (n a) in instances where no AP+ colonies were detectable following treatment (B) Representative AP+ colony staining after 6 days of liESC treatment with vehicle control (i), the SERCA inhibitors cy opiazonic 5 acid (ii) and mapsigargin (iii), the fi¾e-calcium chelator BAPTA (iv), the calcenearin inhibitor cyclosporin (y) and the CAMKII kinase inhibitor N-62 (vi). Bar equals 100 μιη. (C) Relative intracellular Ca⁺⁺ levels determined by Fura2:Fmo3 ratios in SSEA4+ - cells following treatment with (+·) or without (-) Thapsigargin for 24 hr. * denotes statistical significance of p < 0.05, as calculated by a paired student's t-test ί θ No significance is denoted by "ns". The range of compound concentrations (shaded triangle) tested were as summarised in Table 1 and did not exceed the IC EC50 (liaif- maximal inMbitor/activating) concentration fox each compound.

Figure 3: mdependent confirmation that G protein and intracellular free calcium signalling promote undifferentiated hESC maintenance. (A) (i) Rat GnRH

15 receptor transgenic RH1 subclones could be discriminated as responders (clone 4.2) or non-responders (clone 3.2) on the basis of intracellular calcium elevation in response to GnRH ligand, as assessed by Fura2/Fl.uo3 relative fluorescence unit ratio (RFU) over time, (ii) Culture of GriRHR transgenic hESC clones for 7 days in human dennal fibroblast-conditioned medium containing 4 ng ml bFGF (HDF-CM+) or not 0 (HDF-CM-) with GnRH at 20, 200 or 500 nM (solid bars) or vehicle alone (clear bsxs),, r sulted in ¾ si^niiicsnt ( ^O.05) QnJ ii™ ss ci ted incrs&ssi in AJ?H^~ colony frequency normalised to total cells only in the responder clone 4.2 in HDF-CM- (ii). (B) Treatment of RH1 hESCs for 7 days with small molecule SERCA inhibitors (CPA, Thapsigargin, Clotrimazole, and Chloropromazine) or the GPR23 ligand LP A 5 in defined medium (mTESR) lacking {-) bFGF. (i) Representative immunocytochemical staining (one example field of 36 is shown; 9 non-overlapping fields were imaged for each of 4 independent wells per condition) for Oct4 (green), Nanog (red), Oct4 Nanog double positive (yellow) cells and DAPI-stained DNA (nuclei, blue) for increasing concentrations of each treatment (vertical gradient) and 0 control, as compared with primary antibody omitted (i.e. unstained) control cells. See Table 2 for the actual concentrations of the drugs, (ii). Acapella-based automated quantification of fluorescence staining. Graph depicting the relative proportion of Oct4 Nanog positive nuclei with respect to vehicle controls for different concentrations of drugs. We observed significantly elevated proportions of Qct4/Nanog positive cells in a concentration dependent manner relative to respective vehicle controls for all drugs except wapsigargin (non-significant increase) at subcytotoxic levels. Triangles indicate increasing concentration of drug. Asterisks indicate a statistically significant increase in the proportion of Oct4 / Nanog double positive cells (* - p<0.05. ** = p≤0.01); crucifixes indicate that the drag is at a generally toxic level. The range of compound concentrations (shaded triangle) tested were as summarised, in Table 1 and did not exceed the IC/ECSO (half-maximal inhibitor/activating) concentration for each compound.

Figure 4: Schematic of putative G protein signalling pathway involvement in hESC maintenance and growth. Agonists, antagonists (stars coupled to→ or ±, respectively) with independently replicated statistically significant simulatory (†) or inhibitory (j.) effects on undifferentiated hESCs (A) and total cells (B) are depicted in relation to their G protein signalling pathway target with their consequent stimulatory (-→) or inhibitor}'" (±.) effects to both denoted in bold. Our study suggests that inhibition of G« i_¾» G_a^_/n_» PLC β, free ealefum chelation and CAMKH kinase can inhibit undifferentiated hESC maintenance whereas increasing intracellular free calcium concentration by blockage of its sequestration into membranous organelles can promote it Stimulation of PKC can also interfere with hESC maintenance suggesting that this pathway operates in opposition to the caldum-dependent signalling downstream of PLC. In contrast, modulation of all G_a and G_«i¾ subunit and PKC signal transduction and free intracellular calcium via Cakine mi can affect total cell counts, the net effect of altered growth, viability and attachment of undifferentiated and differentiating cells.

Figure 5: Optimisation of hESC Maintenance Screening Assay (Related to Figure IB; 2A,2C; 3; 4Aii): (A) hESCs were dissociated by trituration in one of four regimes consisting o TEDTA for 10 min (♦), TEDTA for 20 min (€}, TEGTA for 10 min ( ^A) arid CoUagenase for 10 nrin (X), and plated at 500, S000 and 10,000 cells per well (surface area 1.9 cm²) of a 24 well plate. After 7 days of incubation in self- renewal culture conditions, cultures were fixed and stained for alkaline phosphatase (AP) and the number of AP÷ colonies per well was counted. Calculation of correlation coefficients ( "^') showed that treatment with TEDTA for 10 min yielded the most linear relationship between seeding density and AP+ colony detection. (B) Using the optimised 10 min TEDTA dissociation regime hESCs were plated in quadruplicate in two independent plates si 1000. 5000, 10,000, 50,000 and 100,000 cells per well of a 24 well plate. After 2.4 hours, hESCs from sister plate were collected and either counted by hemacytometer or fixed in situ and stained for protein content by Sulforhodamine B. Calculation of the correlation coefficient showed a linear relationship between total cell count and optical absorbance of SRB staining of protein (at 540 nm). (C) AP+ colony frequency expressed per 10^'' unit total cells {determined by SRB staining) after 10 min TEDTA dissociation and 7 days of culture revealed no significant differences between seeding densities. (D) Phase light micrograph of representative undifferentiated hESC AP+ colonies. Plotted are mean ± standard error (A, C, &=3; B₃ n= ). Bar = 50 μπι„

Figure 6: Secondary Screen Using Pan-PLC-Inhibitor and Pan-P C- Activators (Related to Figure I B): Significant changes (Δ) in total cells (A) or AP+ colony frequency normalised to total cells (B) between treatments and respective vehicle or untreated control. Stars indicate the level of significance (* p<0.05; ** ρ≤0.0ί; ***p«0.01). The range of compound concentrations (shaded triangle) tested were as summarised in Table 1 commencing with the lowest concentration listed up to the IC/EC50 (half-maximal inhibitor/activating) concentration for each compound.

Figure 7i Rat Gonadotropin Releasing Hormone Receptor Expression Construct and hESC Subclone Morphology (Related to Figure 4): (A) Structure of the construct pcD^'NA3-rGNRHR-L The plasmid is based on the peDNAS mammalian expression plasmid, containing a 2.2 kb fragment encoding me cDNA of the rat GnRH receptor (rat GnRHR (Ϊ)) under the control of the constituti ve cytomegalovirus (CMV) promoter. The plasmid has mammalian (Neo) and bacterial (ampicillin) selectable markers. (B) RHi subclones grew as colonies with normal hESC morphology after transfection with. pcDNA3-GNRHR-I and selection in G418 for stable integration of the plasmid. Clone 3.2 (left) was a non-responder to GnRH, clone 42 was a responder to GnRH (see Figur 4A(i), main body of the paper).

Figure 8; Stressing strategy for identification of SM capable of substituting for bFGF in defined medium, (a) Outline of experimental procedures. hESCs were plated in 96 well Special Optics TC plates on matrigel and cultured for 7 days in the absence of growth, factor(s) and the presence of drug. Weils were then fixed and imnttunostained with antibodies directed against OCT4 and NANOG, counterstadined with D PI and examined in the confocal microscope-based OPERA ^'high-throughput screening system, (b) Example of primary data from OPERA screening. 9 non- overlapping fields were imaged in the red (NANOG), green (OCT4) and blue (DAPi) channels for each well of the TC plate, and are arranged here as a large merge of 86-4 discrete images into one image ordered as the original plate. The triplet arrangement of each drag condition is clearly visible. Also detectable are wells with drugs promoting growth of embryonic stem cells (dense wells with a predominantly yellow appearance due to the presence of many red and green nuclei, e.g. C10-C12), wells with drugs that do not support good dual slairsisg, suggesting the drag downregutates one or both sternness markers (e.g., Fl!-12), wells with toxic- drugs containing e few cells (e.g., B4-6) and wells with moderately toxic drugs (e.g._}€4-6), Highlighted are the positive control (mTeSR, red box).

^'Figure 9: Example of H9 hESC ceil growth in £6 medium (E8 lacking bFGF TGFP) arid response to combinations of 2 selected drugs [ΙΟμΜ]. (a) Wells axe in triplicate for each condition, with positive and negative controls indicated. Note yellow colour of the medium, indicating vigorous cell growth, in the wells for combinations CI, C2 and C3 and compare with the alkaline (pink) colour of the wells in the majority of conditions, including the negative control. By passage 2, there were very few cells remaining in the unsupportive wells. Top right, Area staining of alkaline phosphatase positive cells in an example well (H9 / E8), (b) The area stained in the well (left, real image) was quantified automatically (right image, red fill) and ceroid then analysed statistically by ANGVA.

Figure 10: Representation of hESC colony morphology after 1 week in culture of H9 in EoVS medium with small molecules added at 10 nM. Heparin (H), Epinephrine (E), Sulpiride (S) and Aeetazolamide (A) to substitute for bFGF and TGFp.

Figure 11: Representation of hESC colony morphology after 1 week in culture of H9 in mTESR medium with small molecules add©;! at 10 uM. Heparin (H), Epinephrine (E), Sulpiride (S) and Aeetazolamide (A), io substitute for ^'bFGF(F) and T^'GFp (Ϊ).

Figure 12: Representation of hESC expression of surface markers after 1 week in culture of H9 in E6 medium with small molecules added at 10 uM. Heparin H), Epinephrine (E), Sulpiride (S) and Aeetazolamide (A), to substitute for bFGF (F), TGF-beta (T). FL2 channel shows TRA1 -60 as putative marker of undifierentiated hES ceils and in FL1 SSEA1 as putative marker for early differentiation. Table I : Drug concentration and vehicle requirements and citations (related to Figure 1A),

Table 2% Concentrations of drags and (bracketed) DMSO vehicle controls as used in the experiment shown in Figure 3B.

Table 3: - [1st of additional small molecule drugs evaluated

Table 4: Small molecule drags capable of maintaining GCT4 / NANOG +-ve phenotype in hESCs in the absence of bFGF identified in primary and secondary screens on RCM1 and RHI !iESC lines respectively. Triangular arrows indicate up- or dowtaeguktioii of the indicated target, respectively.

Table 5: Small Molecule drugs capable of maintaining a OCT41 NANOG +ve pheootype in repeated screening of RCM1 and RHI hESCs in the absence of bFGF. (a) Summary table of the statistical analysis of the final bFGF substitution screen carried out in RHI hESCs, showing the drugs included, in the screen and the concentrations, if any, at which they increased the proportion of OCT4 NANOG+ve cells over vehicle control to a statistically significant level. Key: *, P < 0.05; **, P < 0.01 ; ***, P < 0.001; ****, P < 0.0001. (b) Triangular arrows indicate up- or dowrtregulation of the indicated target, respectively. (Shaded are drags subsequently evaluated in pilot combinatorial screen).

Table 6: Drugs identified as capable of maintaining OCT4 N ANOG +ve phenotype in hESCs in the absence of TGFp in two independent screens on RHI and RCM1. (a) Summary table of the statistical analysis of the final TGFp substitution screen carried out in R.CM1 hESCs, showing the drugs which increased the proportion of OCT4 NANOG+ve cells over vehicle control to a statistically significant level. Key: * P < 0.05; ** P < 0.01 ; *·*, P < 0.001; *·**, P < 0.0001, (b) Table showing unique drugs taken forward from the TGFp screen including additional drugs demonstrating a subjective improvement Triangular arrows indicate up- or downregulation of the indicated target, respectively. (Shaded is drug subsequently evaluated in combinatorial screen).

Table 7: Molecular structure and summary of outcome of drugs which independent screens exhibited a statistically significant or subjective improvement in maintenance of a human pluripotent stem cell phenotype above vehicle control lacking bFGF or TGFp Table 8: Drags identified as capable of supporting an AP+ve or NANOG +ve phenotype in hESCs in tie absence of bFGF and. TGFp in E8 or mTeSR, respect vely. Greyed out drugs (heparin, prednisolone and sulpiride) appeared in both screens or were removed as apparently merely promoting vigorous cell growth in the short term (riboflavin).

Table 9: Combinations of drugs apparently supporting growth and self renewal of hESCs in defined media, (a) Examples of RCM1 cells grown in rrtTeSR. without bFGF or TGFp for 2 passages and then stained for alkaline phosphatase. * indicates statistical significance of improvement over vehicle control: *, P < 0.05, P < 0.001. (b) Combinations of drugs selected as supporting hESCs for a period of at least one week.

Materials and meihods

MCSC Culture

RH1, RH6 and RCM1 human embryonic stem cells (hESC) were isolated under a licence held by PDS from the UK Human Fertilisation Embryology Authority and deposited in the UK Stem Cell Bank (Fletcher et al, 2006), (Be Sousa. et al, 2009). Their use in this project was approved by the UK Stem Cell Bank steering committee. Unless otherwise stated, reagents were sourced from Invitrogen (Paisley, UK).

Drug screening experiments were performed in feeder-free conditions on growth factor-reduced Matrigel (MG, Becton Dickinson, Oxford, UK) diluted 1 :30 in "Knockout" Dulbecco's Modified Eagle's Medium ( O-DME ), Medium was either pre-conditioned by human, neonatal foreskin dermal fibroblasts (HDFs; Cascade Biologies, Nottinghaa.shire,UK) as described by Fletcher et al., 2006, based on the method of Xu and colleagues (Xu et at, 2001), or in mTES prepared with (Ludwig et al.. 2006) or without bFGF.

Agent Tre&taeisl

Unless stated otherwise, plating of cells for agent treatments was performed as follows: adherent cultures were treated with Trypsin-EDTA (0.025% Trypsin and 0,01% EDTA in PBS) for 10 min. The trypsin was neutralised and the cells disaggregated by trituration. 10,000 hESCs per well were plated in 24 well plates coated with MG. Agent treatments were started 24 hr after plating and lasted for 6 days. Culture medium (containing treatment or vehicle control) was exchanged daily. Alkal e Phosphatase Staining

Alkaline phosphatase (AP) staining was performed on cells fixed for 2 miri with 4% paraformaldehyde in PBS using an AP Detection Kit (Chemicxm, Hampshire, UK.) according to the manufacturer's instructions. For quantification, a colony consisting of 4 or more closely packed, cells, which were clearly positive for AP, was defined as an AP+ colony (See Figare 2B).

SRB Statam for Protein? Quantification

Fixation was performed in 24 well plates by adding 0.5 ml 25% TCA to adherent ceils covered with 1 ml PBS. The fixative was removed after a 60 mm incubation at 4°C and the cells were washed 10 times with dJ¾0. The cells were air- dried at 40^CC. Once dry cells were tt-eated with SRB solution for 30 minutes at room temperature according to manufacturers recommendations (Sigma Aldrich). This was followed by 4 washes with 1% glacial acetic acid and a second cycle of drying at 40°C. The protein-incorporated dye was solubilized for 60 min in 250 μΐ 10 ητΜ Tris.HCi, pH 1.0.5. 200 μΐ of the resulting solution was transferred to a 96 well plate and the absorbance at 540 inn was read ^"using a Labsystems Mrdtiscan EX plate reader (MTX Lab Systems, Inc., Virginia, USA)

Ftaoresee St&mmg for F!ew Cytometry md [Cs-H-ji M&mmrsmmts

Single cell suspensions of cultured cells were prepared b treatment with TrypLE Select (Irtvitrogen, Paisley, UK) for 5-10 min at 37^RC. 2.5xl0⁵ cells were aliquoted, harvested by centnfugation at 300 g and resuspended in 0.5 nil KO-DMEM each. Cells were stained with S rig rriL of F1uo3 (Molecular Probes, Eugene, USA) and 20 ng mL of Fura2 (Molecular Probes, Eugene, USA), both for 20 min on ice, and collected by centrifagation at 300 g for 5 min. The resuspended pellets were then consecutively incubated with primary (1 :125, anti-mouse SSEA-4, DSHB MC-813- 79. Developmental Studies Hybridoma Bank, Iowa, USA.) and secondary (1 :125, goat anti-mouse IgG3 FITC, Caltag Medsystems, Buckingham, UK) antibodies for 20 min on ice. The cells were harvested by a final centrifugation at 300 g and resuspended in FACS-PBS (0.1% FCS, 0.1% NaN3 in PBS). The samples were analysed using a FACSCatibur instrument (Becton Dickinson, j, USA). Controls for SSEA-4 staining were incubated without primary antibody. The relative level of [Ca⁺⁺]; was estimated on the basis of the mean Fura2:Fluo3 ratio. C¾aenrtie¾ of a Mai GOTadotos iffi Reteasmg Hormoae Receptor Expressing Transgenic hESC. Line

A single cell suspension of 1-1.5 x iO⁶ RHI hESCs was transfected by etectroporation at 32(3 V, 200 Ρ using a Btorad Gene Pulser II electroporaior with 50 pcDNA3-rGNRHR-I linear plasmid DNA in PBS. The plasmid pcDNA3- rGNRHR-I contains the rat GN ilH receptor (rGNRHR) under fee control of the strong constitutive CMV promoter (Figure ?A). Ceils were plated out onto Matrigel-coated 10 cm diameter tissue culture dishes and after 48 hours the medium was supplemented with 100 g ml^"1 G418. G418-resistant colonies with a normal hESC morphological phenotype (Figure 7B) were picked after 14 days, expanded on irradiated HDFs is the absence of G41.8 and after establishment as Independent lines were transferred back to feeder-free conditions. Responsiveness to GNRH was assessed by means of a Fhto-3-based Fl^'oiometric Image Plate Reader assay (FLIPR assay, Molecular Devices, Sunnyvale, California), according to the manufacturer's instructions.

Effect af Ca⁺⁺ ATPsse Inhibitory Drags on hESC Self Renewal by High Co&teat Screening

10⁴ mTeSR-adapted RHI hESCs well^"1 were plated in a 96 well tissue culture plate and the following day were switched to mTeSR lacking bFGF (Stemcell Technologies) but containing eyclopiazosie acid, mapsigargin, LP A, Clotrimazole or Chlorpromazine at one of three different concentrations for 7 days (Table 2). Ail conditions were set up in quadruplicate, with vehicle controls (DMSO for CPA, thapsigargin, Clotrimazole and Chlorpromazine, medium for LP.A) at the same concentration in parallel. After 7 days' culture, with daily medium changes, the wells were washed, fixed in 4% PFA PBS for 20 minutes and permeabiUzed wife 0.2% Igepal PBS for 20 minutes. The cells were immunostained consecutively for Oct4 (Santa Cruz mouse monoclonal IgG), with a secondary antibody conjugated to alexailuor-488 (green, Invitrogen) and Nanog (Goat anti-human Nanog, R&D systems), with a secondary antibody conjugated to alexafluor-555 (red, Invitrogen), and couriterstained with DAPI. The plates were scanned in the red, green and blue channels with a confocal microscope-based Opera high-content screening system (Perkoi-Elmer), imaging 9 non-overlapping fields for each well. There were approximately 800-1000 nuclei nek! ^! in a sub-confluent weil in non-toxic conditions. Counts of nuclei positive for a signal in one or more channels were made using an m- house customised version of the nuclei counting script in Acapella software (Perkin Elmer). Counts of Oct4 and Nanog double positive cells for each treatment were normalised relative to its respective vehicle control.

Statistical Analysis

Statistical analysis was performed using GraphPad Prism. In drag and small molecule screening differences between control and. treated samples were assessed by one-way ANOVA followed by impaired t-tests. A paired student's t-test was used to evaluate the significance of [Ca ^"]j differences observed between SSEA4⁺ and SSEA4^" cells. Statistical sigrificars.ee varied from p<0.001 (denoted by p«0.01) to _Ϊ 0.05. Optimisation of hESC Maintenance Screening Assay

To discriminate between the effects of drugs on maintenance of an undifferentiated hESC phenotype vs cell growth, viability or attachment, we established a assay whereby input cell number could he controlled and the frequency of alkaline phosphatase positive (AP+) colonies (Figure SB) could be normalised to total cells after 7 days of culture (Figure 5). HESCs typically respond poorly to disaggregation to a single cell suspension by differentiating or undergoing apoptosis. We therefore began by optimising the cell dissociation regime and plating density. Four enzymatic dissociation regimes were evaluated, using the RHl hESC line. These consisted of 10 or 20 min with TEDTA (0.025% trypsin / 0.01% EDTA is PBS), 10 min with TEGTA (0.25%Trypsin, .1 mM EGTA in. PBS) and 5 mm with 200 units/ml of Collagenase (in. DMSO), followed by plating of SxlO^lxlCf, 5x10^s and lxlO⁴ cells from the resulting suspension in each well (with a surface area of 1,9 cm²). Wells were pre-coated with Matrigel™ (1:30 dilution), and cells were cultured in self- renewal medium. After 7 days of culture, the number of AP+ colonies per well was counted. Wife this experiment we established linear relationships between seeding density and AP+ colony number, with dissociation with TEDTA for 10 min yielding the most direct linear correlation (Figure SA). Using this regime we evaluated whether total protein content assessed by optical absorbance measurement at 540 ran after Sidfurhodamine B (SRB) staining directly correlated with total cell number. This experiment confirmed a direct linear correlation (Figure SB) so that we could normalize AP+ colony frequency to total cells using optical absorbance measurement of SRB stained fixed ceils as an index of the latter. Next, we applied this SRB staining protocol to fixed cells from the optimal TEDTA (10 min) dissociation regime treatment group in the first experiment to determine the AP+ colony frequency per i;m{ cells at each seeding density (Figure SC). This revealed no significant difference in this ratio between seeding densities, suggesting that any variations in total cell number in our assay would not necessarily affect AP+ colony frequency. We thus concluded that AP-^;- colony number to total cell counts could serve as an informative index with which to assess specific- effects of drugs on maintenance of an undifferentiated hESC phenotype.

Secondary Screens IMag Pun-FLC-lmMbitor md Pan-P C-Activators

For the G_ffi,_qm PLCp/PKC pathway, inhibition and activation of PKC using R013822 and PMA, respectively, reduced total ceil counts at higher concentrations. At these concentrations, PMA also reduced the mean normalised AP^™ colony frequency and at the lowest concentration only AP+ colony frequency was reduced. This was further confirmed in a subsequent evaluation of lower doses (0,3 to I rsM) with no effects at less than or equal to 0.2 nM) (Figure 6). in the first screen, the pan-PLC inhibitor U73122 also only affected normalised AP+ colony frequency at the lowest concentration tested (5 μ¾Ί) (Figure 1), Medium and higher concentrations of U73122 necessitated DMSO vehicle at 0.5% and 2.3% which were cell toxic and did not support AP+ colony formation even in vehicle controls. Although the change in total cell number relative to control at the highest concentration of U73122 was significant, the total cell counts for this treatment (and vehicle control) were only approximately a fifth of the vehicle-free control. To investigate the specificity of PLC inhibition, we evaluated the effect of a weak analogue of U73.122, U73343, at a lower concentration range (I, 2. 5, 7, 10 μΜ; Figure 6). At all concentrations of the analogue and 5 μΜ or higher concentrations of U73122, there were reductions in both total cells and normalised AP÷ colony frequency relative to controls. At 2 μΜ, but not at 1 μΜ. U73122 only reduced A?+ colony frequency,

RESULTS Brag screening implicates C>_;,.;_;¾ aad subraM-Msaked sigsial transduction

®Bdiffere»tktod hESC renewal

To screen for G protein signal transduction pathways of particular importance for undifferentiated hESC maintenance we developed a short term culture assay to assess the effects of pharmacological drags and iigands known to affect these pathways (Figure 1; Table 1_» Flgwe 5). This assay involved the assessment of the proportion of phenotypic hESCs, scored by counting colonies positive for alkaline phosphatase (AP+) relative to total cell content, after 7 days of treatments in feeder- free culture conditions. The objective was to identify agents demonstrating a concentration dependent effect restricted to AP+ colony frequency distinct from effects on total cell content, the latter reflective of overall effects on cell adhesion, growth or death. All agents were tested at least twice in independent experiments on the RHl liESC line used to develop the assay. In the case of (he first primary screen of the G_tiq/PLCp/PK€ pathway, results were also confirmed in an independent cell line, RH6 (Fletcher et al, 2006), Only results on RHl whose statistical significance was independently verified in the same line and RH6 are reported, Drugs were applied at low, medium and high concentrations, with the highest concentrations at the IC EC50 for the drug target, Le. the concentration causing half-maximal inhibition or activation of the target (Table 1). In each experimental trial, drug-free or vehicle controls (e.g. BMSQ or propylene glycol) were evaluated, the latter at concentrations corresponding to those used for drug treatment. Overall, the mean frequency of AP+ colonies per well in experimental controls ranged from 10 to 100, with a mean of 25. This corresponded to an AP+ colony/1000 cell ratio of 0.1 to 0.6. Given variation in absolute and normalised values between experiments, and the necessity to test a range of vehicle concentrations to match drug treatments, results are reported in terms of differences (Delta, A) relative to vehicle-matched control treatments for both total cells and normalised AP+ Colony Frequency (Top vs Bottom panels for Figures IE, 2Aj ¾ad Figure 6).

In the first screen of agents affecting pathways downstream of all three G_a subunits (Gaq, Gas. Gm/o) or agonists of Gaq-mediated GPCRs pathways (MPEP for mGluRS and Kisspeptin tor GPR54), 6 agents induced a significant change (Δ) in either or both Total Cell Counts or Mean Normalised AP+ Colony Frequency as compared to respective controls, in a concentration dependent manner (Figure IB).

For the

pathway, inhibition and activation of PKC using Ro31-8220 and PMA, respectively, reduced total cell counts at higher concentrations. At these concentrations, PMA also reduced the mean normalised AP+ colony frequency, although only the lowest concentration reduced AP+ colony frequency alone. This was further confirmed in a subsequent evaluation of lower doses (0,01 to 1 nM) with no effects at less than or equal to 0.2 nM (Figure 6). In the first screen, the pan-PLC inhibitor U731 2 also only affected normalised Ar- colony frequency alone at the lowest concentration tested (2 μΜ). Medium and higher concentrations of U73122 necessitated DMSO vehicle at 0.5% and 2,3% which was cytotoxic (denoted by crucifix Figure IB) and did not support AP+_. colony formation even in vehicle controls (data not shown), The specificity of the effect on AP-+- colony formation achieved using low concentrations of the Pan - PLC inhibitor were farther evaluated using the weak analogue of U73122, U73343, at a lower concentration range (1, 2, 5, 7, 10 μΜ; Fsgors 6). At all concentrations of the analogue and 5 μΜ or higher concentrations of U73122, there were reductions in both total cells and normalised AP+ colony frequency relative to controls. However, AP+ colony frequency alone was reduced at 2 μΜ U73122, with no effects at 1 μΜ. Neither of the G_¾4-rr.ediated GPCR effectors (MPEP and kisspeptin) had any differential effects versus controls on either total cell number or normalised AP-f- colony frequency. The same was the case for the global G inhibitor Y -254890. This suggests that neither these agonists nor the G_a¾ n subunit is normally implicated in undifferentiated hESC renewal or viability, at least in the culture conditions in which the assay was performed

For agents affecting the Gas-AG-PKA pathway, medium and high. concentrations of forskolin (eAM? activator) and 4C3MQ (PKA inhibitor) decreased and increased total cell numbers, respectively;, without any apparent affects on normalised AF+ colony frequency. Cholera toxin, a G_as activator, ^'had no effects at all concentrations tested, These findings suggested that the G_rK~adenylcyckse-PKA pathway may be of less significance to hESC renewal per so. For the G_a.j_/0 pathway we only evaluated pertussis toxin, a direct inhibitor of the G_a.y₀ subunit. Low and medium concentrations of this agent affected both total cell and normalised AP+ colony frequency, although the high, concentration affecting only the latter.

Involvement of intracellular free eakmm d CAMK kinase in isndifferentia ted hESC maintenance.

A principle observation from the primary screen was the ability to achieve a specific reduction in normalised AP+ colony frequency without concurrent effects on total cells, using a Pan~PLC inhibitor at low concentrations. We thus focused on downstream elements to this pathway in our pursuit of signal transduction of most relevance to hESC renewal. PLC signal transduction occurs via the phosphati.dylinositoi-2-phosphate (PIP2)-mediated cleavage products DAG (which activates PK.C) and IP3 (which increases intracellular free calcium) (Berridge & Irvine, 1984), As treatment with the PKC inhibitor Ro31-8220, did not appear to interfere with AP+ colony formation we focused on signalling mediated by free intracellular calcium.

Thapsigargm and cyclopiazonic acid (CPA) are inhibitors of the sarco/endopksmic reticulum Ca^r+ ATFase (SERCA), and this have the effect of raising intracellular free calcium concentration, [Ca^{+ ,}¾, by inhibiting uptake of Ca^{: <} ions to stores in the ER. Treatment ofhESCs with fhapsigargin and CPA reduced both Total Cells and normalised AP+ colony frequency at most concentrations tested. However, both agents increased AP+ colony f equency at lower concentrations (Figure 2A and B). Treatment with Lysophosphatidic Acid (LP A), an agonist of LP A receptors on surface membranes, whose activation, can increase free intracellular calcium, did not appear to have any effect on hESCs at low and medium concentrations but increased the normalised total cell number at high concentrations. Conversely, treatment with the calcium chelator BAPTA reduced Total Cells and normalised AP+ colony frequency at most concentrations, but. only the latter at lower concentrations (1-2 μΜ; Figure 2A md 3$).

The effects of free intracellular calcium may be mediated by protein phosphorylation which can be controlled by a family of caldum calmodulin- dependeni protein kinases (CAM kinases or CAMK) and phosphatases. Treatment with cyclosporin (CY), an inhibitor of the calcium-dependent protein phosphatase calcineurin, decreased only the normalised number of total cells at all concentrations evaluated. In contrast the CAM H inhibitor KN-62 decreased only the normalised AP-t- colony frequency, suggesting it may be of greater significance to maintainance of an undifferentiated state as assessed by AP+ colony formation (Figure 2A sad ).

If intracellular free calcium signalling is more important to hESC renewal than to different ting ceils we reasoned this might be reflected by a higher

in the self-renewing stem cell fraction of an hESC culture. We tested this by quantifying the relative levels of [Ca^ji by Fura-2:Fluo3 loading of hESCs in stem cell marker SSEA4 +ve versus -ve cells, the latter putatively representing differentiating cells. two independent experiments intracellular free calcium levels were significantly gre ter in SSEA4 positive cells. This difference was lost if hESCs were first treated with 10 nM Thapsigargm for 24 hr prior to analysis (Figure 2C), c ivation of mi exogenousry supplied subunit-ompied GPCR can esm eMS&te for a lack of exoge msly su plied bFGF in hESC renewal.

To experimentally validate a potential role for calcium, signalling in hESC renewal the RH1 hESC line was transfected with an expression construct containing the rat G«q-coupied GnRH receptor (Supplementary infomafion, Figure S3A), which is otherwise not expressed in hESCs (data not shown). G418-resisiant subclones were discriminated as GnRH responders (clone 4.2) and non-responders (clone 3.2) by assessment of agonist induced elevation of intracellular calcium quantified by Fura2 Fhio3 relative fluorescence unit (RJFU) ratio measurement over time (Figure 3Ai). Responder and non-responder hESC clones were subjected to the same assay used in the primary drag screen, in a human dermal fibrobkst-eondiiioned medium either supplemented with 4ng ml^"1 bFGF (HDF-C +) or with no added bFGF (HDF- CM-), and GnRH at 20» 200 or 500 nM (solid bars) or vehicle alone (clear bars) (Figure 3AM). In HDF-CM+, there was no significant difference in normalised AP+ colony frequency between GnRH- or vehicle-treated cells for both clones. The same was the case for me non-responder clone 3,2 in HDF-CM-. However, for the responder clone 4.2 in HDF-CM- (i.e. lacking exogenous bFGF), GnRH treatment significantly increased normalised AP+ colony frequency at all concentrations.

Modulation of Intracellular Free Calcium Obviates the Requirement for bFGF

IJR Defined Medium

A prediction of the hypothesis that raising [Ca⁺⁺]i promotes maintenance of an undifferentiated hESC phenotype is that medically-prescribed drugs which inhibit Ca^"f' sequestration by endoplasmic reticulum membrane-bound Ca^ATPases should promote hESC self-renewal. Two such drags approved for isse in humans are chlorpromazine and clotrimazole (Bartolotnmei et a!., 2006; Hwang, Han, & Han, 200.1). Our initial experiments with the GnRH-R transgenic hESC line suggested that GPCR-mediated signalling could be used to artificially compensate for a lack of exogenously supplied bFGF. However, since the human dermal fibroblast conditioned medium used in that experiment was supplemented with 4 ng ml^"1 bFGF at the time of conditioning and the amount of b^'PGF removed and/or produced during conditioning by the HDFs themselves, if any, is -unknown, the exact bFGF concentration in HDF- CM÷ is also unknown. We therefore decided to test this hypothesis in a defined medium completely lacking bFGF, RH1 hESCs that had been transitioned to an naTeSRl (Ludwig ei at., 2006) (Stem Cell Technologies, Vancouver, Canada)-based culture enviimtraent were cultured in the presence of CPA, thapsigargin, LPA_S dotrimazole or ehlorpromazine for 7 days in mTeSRl lacking bFGF. The cells were then fixed and immunostained for Oct4 and Nanog as a direct indicator of an undifferentiated state, rather than the proxy of AP activity employed previously. The proportion of nuclei staining for both nuclear markers was taken as a measure of the proportion of undifferentiated hESCs present, and thus the level of self renewal in each culture. CPA, LP A, clotrimazole and chlorproraazine all increased the proportion of Oet4 and Nanog-positive cells with respect to vehicle control at statistically significant levels as the concentration of drug increased, until the concentration of drug became toxic Toxicity varied with each drug (indicated by crucifixes in Figure 3B). These results with clotrimazole and c oipromazine were independently confirmed in another hESC line, ECMl (data not shown).

DISCUSSION

Our study evaluated the significance of G protein-coupled, signalling pathways on undifferentiated hESC maintenance under self-renewing conditions, in a primary screen using established pharmacological drugs and alkaline phosphatase staining as a surrogate marker of sternness we implicated downstream elements of the G^i subunit pathway, specifically PLCp, intracellular free calcium,

and CAMKil kinase, as well as the G _α~ι¾ subunit as of particular significance tor maintenance of this phenotype. By contrast, perturbation of the G_ils~A€~PKA pathway had more general affects on total cell growth and/or viability (summarised in .Figure 4). The significance of free intracellular calcium signalling was confirmed in independent experiments demonstrating i) correlation with other sternness markers (SSEA4), ii) the potential to compensate for an absence of exogenously supplied bFGF following agonist-mediated activation of a transgenic hESC line expressing an exogenous G_a-qm subi it-coupled GPCR a d iii) the ability of small molecule drugs known to elevate intracellular free calcium to substitute for bFGF in a defined medium, Our study thus substantiates a previoasly unappreciated role for ca!ti m-mediated signal transduction in undifferentiated hESC maintenance.

Our primary screen involving 6 days of sustained treatment with a spectrum of pharmacological agents served to indicate fee general significance of G protein-linked signal transduction pathways to hESCs. Thus, cholera toxin, an activator of G_SB5s causing activation of AC had no effect on cell growth viability generally or undifferentiated hESC maintenance specifically, at the concentrations evaluated (0.1- 5 pg ml^"1). However, both treatment with foiskolin, which activates AC, and pertussis toxin, which inhibits the AC inhibitor Q_an₀ interfered with cell growth viability. These findings suggest thai in hESCs grown under self-renewal conditions, is not particularly important for self-renewal whereas sustained elevation of cAMP can interfere with cell growtli/viability generally. This was corroborated by the observation that interference with. PKA, whose activation would be stimulated by elevated cAMP, promoted cell growth. Our results contrast with the available evidence highlighting a role for the G_ffia~AC~P A pathway in. mouse ESC renewal. In mouse ESCs, cAMP and PKA have been implicated in self-renewal in the absence of LIF signalling (Fahcrty, Fitzgerald, Keohan, & Quinlan, 2007). Stimulation of Go* protein by cholera toxin during differentiation in embryoid bodies increased ceil proliferation and prevented the time-related decline in pluripotency transcription factors (Layden, Newman, Chen, Fisher, & Lowe, 2010). Forskolin has also been reported as a key factor in stabilising a LIF-dependent mESC ike phenotype in hESCs following transduction with Oct4, Sox2 and Klf4, This effect was executed at least partly through induction of Klf2 and K!f4 expression (Kasna el <:<·./., 2010).

The inhibitory effect of pertussis toxin on cell viability was apparent at low and intermediate concentrations (100 and 200 ng ml^*') whereas at high concentration (500 ng ml^"4), it negatively affected undifferentiated hESC maintenance only. Pertussis toxin catalyses the ADP-ribosyiation of G«,¾ thereby precluding GTP exchange and dissociation of the G_a subunit from the Gp and G_y subunits which normally follows GPCR-mediated activation. Pertussis toxin has previously been shown to antagonise Sphingosine-1 -phosphate (SIP) and PDGF stimulation of hESC self-renewal in seram-free culture, i e former being a ligai d of G_fli_¾-coupled receptors (Pebay et cd._t 2005). In self-renewal culture conditions equivalent to those used in our study but involving only treatments over 2 days (not 6 as in our study), it has been reported that treatment of hESCs ami induced pluripolent stem cells (iPSCs) with pertussis toxin altered colony morphology and organisation but did not affect cell proliferation, apoptosis, expression of pluripotency markers or ability to differentiate into germinal lineages in vitro (Nakamura, Salomonis, Tomoda, Yamanaka, & CtonkJin, 2009). In the same study, effects on colony morphology were not apparent following treatment with the Gas activator, cholera toxin, which was also consistent with our findings. In mouse, SIP stimulates mESC renewal via activation of extracellular signal-regulated E 1 and ERK2. This can be inhibited by both pertussis toxin, affecting Gs».¾_> as well as by inhibition ofPKC, downstream of the€¾.

.3 pathway (Rodgers t ai , 2009).

In our evaluation of the G_oq-PLCp-PKC pathway, treatment with agonist (kisspeptin), or a negative allosteric modulator (MPEP) of specific G_a._fJ, n -coupled GPCRs had no significant effects on either total cell growth/viability or undifferentiated hESC renewal. Subsequent evaluation of hESC transcriptome data sets suggested that the targets of these drags are not normally expressed in hESCs (data not shown). However, a general inhibitor of G^u itself (YM-254890) also had no effects, suggesting that like 0_∞ , GPCR-mediated stimalation of this G protein is not normally involved in undifferentiated hESCs or differentiating hESC growth/viability, at least in the specific context of the conditioned medium culture system used in our screen. In contrast, downstream pan-inhibition of PLC by U73122 or activation of PKC by PMA impaired both. To our knowledge, our study is the first to implicate opposing roles for PLC and PKC in undifferentiated hESC maintenance, PKC is a heterogeneous family of phospholipid-dependent kinases, subdivided into three groups: classical isofbrms (α,β,γ) containing domains which confer regulation by diacylglyceroi (DAG) and Ca^"**, novel isoforms (δ, ε, θ, η) which are Ca⁺⁺ insensitive, and atypical isofbrms (ζ, λ 11) which do not respond to either, in mESCs, inhibitors of PKC-S sustain self renewal under hypoxic conditions which promote a limited early differentiation in the presence of LIF (H. J. Lee, Jeong, Cha, & Kim, 2010; Powers, Mfllman, Huang, & Colton, 2008). PLC is a class of enzymes that livdroiyse phosphopholipids yielding DAG and inositol triphosphate, the latter of which in turn stimulates the release of€a^+i' from intracellular stores, in mammals, there are 13 types of PLC classified into six classes (β, γ, δ, ε, ζ, η) according to structure, of which the predominant forms are PLC-β and -y. To confirm the pan- PLC-direeted effect of U73122, we tested its "inactive" analogue, U73343. With the exception of lowest concentration of U73122 that only affected hESC maintenance, the "inactive" analogue exerted the same effects. I mouse microglial cells TJ73122, but not the U73343 analogue, inhibited transient intracellular Ca^{" '} release from intracellular stores normally mediated by PLC activation. However, both drugs inhibited a sustained increase in intracellular Ca^"""1' prior to or after activation of ATP- gated surface membrane ion channels (Takenouchi, Ogihar , Sato, & Kitani, 2005). The differential and common effects of these drags may reflect effects on the availability of intracellular versus extracellular Ca" .

Treatments affecting intracellular free C ^ concentration affected total cell growth/viability generally and undifferentiated hESC maintenance specifically, the latter only at the lowest concentrations. Thapsigargin and cyciopiazonic acid, both blockers of sarco/endoplasmic reticulum Ca^{^}-ATPases which pump free Ca*^1" into intracellular membranous compartments, and BAPTA, a calcium chelator, increased and decreased the prevalence of an undifferentiated hESC phenotype, respectively. The former two drugs can be expected to increase [Ca⁺⁺]i by inhibiting uptake of Ca*⁺ to intracellular stores, and fee latter can be expected to decrease [Ca⁴^]; by chelation. Thus, calcium signalling is likely to be of particular importance for undifferentiated hESC renewal. Consistent with this hypothesis, the undifferentiated (SSEA4+) cells in hESC cultures were found to have significantly elevated [Ca^""4']; compared to the differentiated (SSEA4-) cells in the culture. In mouse ESCs, intracellular fee calcium oscillations are predominantly confined to and important for the Gl/S cell cycle transition (Kapur, Mignery, & Banach, 2007). This could also be the case for hESCs, Lysophosphatidic acid (LPA) is a GPCR ligand which can stimulate free intracellular Ca ^' oscillations via G_K »¾ and Go, and also elevates cAMP via Q (C. W. Lee, Rivera, Dubin, & Chun, 2007). There are multiple receptors for LPA of which one, GPR23 (ie. LPA₄) is specifically expressed in hESCs as determined by meta-analysis of the hESC transcriptome (Assou et aL_t 2007), and our own affynietrix data set on 3 hESC lines (data not shown). In our screen of drug treatments, LPA at the highest concentration tested promoted total cell growth. This is consistent with mouse ESCs where LPA induces Ca.^{" "} mobilisation and c-myc expression via PLC, presumably mediated by (Todorova, Fuentes, Soria, Nadal, & Quesada, 2009). Differential outcomes following interference with edekeurin vs CAMKII kinase suggest that general and specific roles for Cst * in cell growth viability vs undifferentiated hESC renewal may be discriniisiatabk at this juncture, although further work is needed to dissect the network of relevant effectors., in addition to its SERCA inhibitory activity, ch!orpromazine also has some CAMKII inhibitory activity, possibly explaining why this drag only promotes self renewal at low concentrations. CAMKII kinases are multi-functional regulators of transcription factors, glycogen metabolism and neurotransmitter secretion. They are distinct in possessing an autocatalytic activity which serves as a molecular memory device whose effect persists long after its activation (Braun & Schuiman, 1995). This raises the interesting possibility that stimulation of hESC renewal might be achieved by periodic activation of CAMKII signalling,

CONCLUSION

It has only been in the last few years that it has been possible to control pluripotent stem cell renewal and differentiation using chemical modulators of intracellular signalling pathways (reviewed by (Ng & Surani, 2011)) In mESCs independence from LIF can be achieved ^'using inhibitors of the FGF/.E and glycogen synthetase kinase-3 (GSK3) signalling pathways (Nichols & Smith, 2009). Our study implicates modulation of intracellular calcium as a means by which to obviate bFGF signalling in the context of hESC renewal, which according to the available literature is distinct from signal transduction in mESCs. This constitutes a novel role for this second messenger providing new opportunities to develop chemically-defined systems supportive of human pluripotent stem cell renewal. This may be of value to facilitate the transfer of pluripotent stem cell-based therapies from fce laboratory to the clinic.

FURTHER METHODS AND RESULTS

Screeatog Method & ©otJi∞

Based on the aforementioned studies, we sought to identify additional small molecule chemical drugs with a capacity to specifically or non-specifically affect GPCR and calcium signalling and obviate dependence of pluripotent stem cells on an exogenous supply of bFGF and/or TGF-beta. To do so we implemented a high- content screening strategy, as outlined in Figure 8. Briefly, hESCs were plated as a single cell suspension on Matrigel®~coat©d 96-well Coming Special Optics tissue culture plates (flat clear polystyrene bottom; black walls) at a density of 3xlif - ! i cells well^"1 in the presence of antibiotics and ROCK inhibitor. The following day, the medium was changed from standard defined mTeSR medium. (Siemcell Technologies) + ROCK inhibitor and antibiotics to test conditions. Triplicate wells were treated for each, condition, consisting of either complete mTeSR medium (positive controls), or roTeSR lacking a required growth factor (e.g. bFGF or TGFp) but containing either 1 μΜ, 10 μΜ or 100 μΜ of a small molecule drug approved .for use in humans. (See Table 3 for the list of drugs employed in this study). In all experiments, ihe experiment was effectively ^c¾lind" as the experimenters did not know the locations of individual drugs. The wells were cultured under standard conditions (37°C, 5% C02 ^•in air) for ? days, refeeding with new median, containing drug and antibiotics every day, after which they were fixed with 4% PFA PBS. The fixed material was inimunostained for Oct4 (SC-5279, Alexafluor 488) and Nanog (R&D AF1997, Alexafluor 555), and coiinterstained with DAPI. Plates were scanned using the confe-cal-based OPERA system to determine cell numbers in 9 independent fields in each well (by D API-stained nuclei count) and the percentage of ES cells that had undergone self-renewal (defined as the percentage of cells in the well positive for both Oct4 and Nanog).

Mean and standard deviation for both cell numbers (DAPI counts) and self- renewal (% cells positive for both Nanog and Gci4) were calculated in Microsoft Excel. Within, each plate, values for the drugs were compared by single factor ANOVA (Dunnett) with both the positive (mTeSR with growth factors) and negative (vehicle, lacking factor) controls to identify drugs promoting either or both an increase in cell number or % [Nanog & Oet'4 ÷ve] cells.

Sereesmg for Small iVfeiecoJe substitutes for bFGF

A primary screen with RCM1 cells in mTeSR medium produced 61 drags which were supportive at the P<0.05 level at 1 or more concentrations of drag. This is many more than was expected. Another finding was that m.TeSR is obviously a very supportive, "rich" medium, and even cells grown in vehicle control conditions (lacking bFGF) have a high proportion of Oct4/Nanog +ve cells. Therefore a second round screen was performed, using the same method as above, on another hESC line ( H1), to remove spurious but statistically significant results from the first round screen. This screen identified a smaller set of drags which when cross-referenced with the results of the first-round screen identified a list of 18 drugs, which to a statistical significance level of P<0.05 apparently promoted the proportion of cells which were OCT4 / NANOG+ve over vehicle control in both hESC lines (Table 4).

A tertiary screen was performed with die aforementioned 18 drugs to which were added two SERCA inhibitors (clotrimazole and chlorpromazine) fttan the library, which had already been shown to promote the presence of OCT4 NANOG H-ve cells in the absence of bFGF as described above, and, Phenformin, a GSK-3 inhibitor, an activity which has been reported to support hESC renewal (Oilman et al., 2008), These 21 drugs which were tested at 6 different concentrations in approximately 0.5-log concentration increments as described (300nM, 1 μΜ, 3 μΜ, 10 μΜ, 30 μΜ and 100 μΜ). Fifteen drugs advanced from this screen (Figure 10) which exhibited significant increases in the proportions of OCT4 / NANOG+ve over vehicle control in both hESC lines at one or more concentrations (Table 5).

Screeffl g for Small Molecule substitutes for TGFp

After the final set of drags identified as confirmed candidates for bFGF substitutes was identified (see above), we screened for substitutes for TGFp, the other major growth factor present in mTeSR medium required to support piuripotent stem v il renewal.

Using the RH1 hESC line we rescreened all of the drugs in the library, now for th eir capacity to substitute for TGFP drugs using the same screening design as for the bFGF studies, ie. mTeSRI lacking ΤΟΡβ on matrigel, in 96 well plates for 7 days. Data were again acquired with the OPERA high content screening platform, and analysed by one-way ANOVA / Dunnett using Graphpad statistical software. However, in this experiment we employed NANOG xmanunostaining as our phiripotency phenotype readout instead of NANOG and OCT4 as previously, since it had become apparent in previous experiments that OCT4 has a longer half-fife of detectability, such that there is a negligible attenuation of expression over the short timecourse of the screening experiments. Thus, %[NANOG +ve cells] ~ %[OCT4 & NANOG +ve Cells]. This identified 46 drugs that appeared in the first screen as apparently promoting a higher proportion ofNANOG +ve hESCs over vehicle control in the absence of TGFpi

This group of 46 drugs was rescreeaed using the same system against RCMl hESCs, to eliminate false positives and identify SMs which were effective on independent cell lines, Only three drugs, tolfenamic acid, menadione and hydralazine were statistically significantly better than vehicle control in this experiment. Several drugs, however, were either statistically significantly better than mTeSRl itself, appeared noticeably better than vehicle controls upon visual inspection of the original images, or were almost but not quite statistically significantly better than controls at more than one concentration, or a combination of these criteria (classified as subjective improvement). Therefore, in total 8 drugs were advanced from the second ΤΟΡβ screen as unique candidate substitutes ). An additional 3 drugs (sulpiride, brombexine HQ and hydralazine ) were identified which had also appeared in the bFGF screen and had thus already been selected for combination screening (Table 6).

A table identifying all of the drugs identified from the bFGF and TGFp substitution screens, and those selected for further analysis based on subjective criteria is shown below in Table 7. Subjective criteria are those such as improving the proportion of phenotypic hESCs to levels approaching but sot quite reaching statistical significance at multiple dreg concentrations, or those drugs for which microscopic examination of the wells suggested a beneficial effect even if the effect was not statistically significant

Testing of combin&rioii ©f Small Molecule substitutes for bFGF & TGFp

Screening m both bFGF- and TGFp-deficieni conditions had thus identified relatively small groups of drugs that might support growth and self-renewal of hESCs in place of these factors, in RH1 and RCMl hESCs. We then sought to combine drugs to identify conditions which could be both completely defined and also lacking any non-approved (for use in humans) factors. As mentioned previously, mTcSR is a very supportive medium and the proportion of piuiipotent stem cells, as assayed by inmiunostaining for transcription factors such as OCT4 and NANOG, remains high even in the absence of necessary growth factors, reducing the signal rnoise ratio of an experiment. We therefore tested the candidate combinations in "E6 medium", which we defined as E8 medium lacking bFGF and ΤΟΡβ, E8 is a recently-described defined, serum-free medium developed from mTeSR with a much simplified list of components (Chen et al., 20.11). As such, we anticipated that differentiation to a non- pluripotent phenotype in the absence of the necessary support from cytokines or truly- active drags would be more rapid, and hence produce a wider range of readout between positive and negative (vehicle) controls. Furthermore, to minimise the chance of selecting drugs that were cell line-specific, the first combination screen with the above drugs was performed with H9 hESCs (derived in Wisconsin) rather than RH1 or IlCMl, derived at Roslin. Finally, to accelerate the acquisition of data over the OPERA high-content screening system described above, we used alkaline phosphatase staining rather than NANOG/OCT4 immimostaining, after 4 days. This permitted the use of a Zeiss Observer phase contrast-based scanning system instead of OPERA, which proved to have a much faster turnaround time for the generation of data. Scripts were written in the Fiji distribution of Image! to quantify stained areas in each well of a 96 well plate from 36 "stitched" images covering almost the entire well area, permitting statistical analysis of proportions of pluripotent cells as defined by macro-level AP expression (Figure 9). .

the E6 system, with H9 cells we evaluated (15x8=) 120 combinations of

SM candidates for bFGF and ΤΟΡβ, respectively at a single concentration, 10 μΜ. This constituted a practical number which could be processed manually, i parallel, the same experiment was conducted with H9 and CM! cells in mT^'eSR medium using NANOG +ve cells as a readout, as previously. Cross-referencing the behaviour of the drugs in different cell lines and media permitted the selection of a smaller group of drugs for further evaluation in combination over longer periods by the elimination of spurious results and toxic combinations (Table 8. ). Tins final group of 9 drugs was men tested for longer term promotion of growth and self renewal over two successive passages against all of the other 8 in both the mTeSR and E8 systemsln this way, 6 supportive combinations were identified, shown in Table 9.

Ev& atleffl of drug sombmatseas.

As a pilot experiment combinations of 2 and 4 of small molecules (all at a concentration of 10 μΜ) were evaluated, as substitutes for or in addition to bFGF and ΤΟΡβ in mTESR and E8 media formulations on H9 hESC. This included Heparin (H), Epinephrine (E), or these plus Sulpiride (S) and Acetazolamide (A).

After 1 week of culture, combinations of 2 (H&E) and 4 (HESA) small molecules appeared to promote maintenance of undifferentiated colonies as compared with bFGF and TGFbeta-free media for both formulations, with the benefits of drugs more pronounced in the E8 system (Figure 10, 11). In the E8 system, flow cytometry for Tral-60 ami SSEA1 as putative markers of an undifferentiated ami differentiating phenotypes, suggested that drags might help preserve the proportion of SSEA1 ceils to levels to found in ES, which mcrease in E6. However, drags did not preserve proportions ofTra-1-όΟ positive cells (Figure 15). This benefit was not apparent in the raTESR system (Figure 12).

DISCUSSION

We selected a library of 200 small molecule drugs approved for use in humans. The selection criteria included interactions with known members of intracellular signalling pathways established as relevant to stem ceil growth and self renewal, interactions with cell surface molecules known to be expressed in hESCs, most notably GPC s, capacity to modulate calcium signal transduction directly or indirectly, and practical characteristics such as experimental practicability, a wide variety of biochemical functions and low cost. We screened this library against three different human ES cell lines (RH1, H9 and RCM1) in two different media (mTeSRl and E8) to identify molecules that apparently supported the growth and self renewal of hESCs in the absence ofbFGF and/or TGFj^'3 in short-term culture, 7 days, by which time the absence of these factors under our experimental culture conditions, resulted in cell differentiation. Using expression of either ES cell-specific transcription factors (OCT4 and/or NANOG) or the enzyme marker alkaline phosphatase, we have identified small groups of drugs that maintain an undifferentiated hESC phenotype in the short term in the absence of these factors being supplied either directly or indirectly, such as through supplementation of the culture environment with serum or co-culture or conditioning of the culture mviroranent by other cell types. We have done limited testing of combinations of these drugs as substitutes for both factors and provide evidence of significant effects.

Drug Cssrsc&iitratiors Tasted Ve eete Rs¾rer¾c®

rsaS CaiisertiratSofi)

MPEP (2~met yl-6- 2rM, 100 nM, 1M None (Csppuccio ei 2005: (phenylethynyl)pyriding) Gaspanni el al., 1999) isspeptin i , SOOnSVi, 1μΜ 20% Propylene Glycol (Kotani et al.,2001)

(2x10^4to"2%)

Y -254890 «0, 100, 250 rs&? DMSO (0.01-0.2%) (Takasaki et al., 2004)

U73122 1,2,5,7,10, 20,200 μΜ DMSO (0.02-2.3%) (Bieasdale et aL 1980)

U73343 1,2, 5, 7, 10 μ DMSO (0.02 - 0.2%) (Bieascla!e etal, 1390)

Rs31-B220 20.100, 500 μ None iBiidworih and Gescher,

1995)

PMA(phorbol 12- 0.01 , 0.05?., 0.03, 0.1, 0.2, DMSO (2x10^"s,o'1%) (Casiagna et al., 1982) rrtyrlstate 13-acataie) 0.3, 1,2,5, 100ηΜ,2μΜ

Cholera toxin (CTX) 0. , 1,5 pg/frsl Ncrce (Caasei and Pfeuffer, 1978;

Gill and Coburn, 1887)

Forskoiin 2, 5, 10 μΜ DMSO (0.02-0,1%) (Seaman et al., 1981)

4C3MQ (4-cyano-3~ 2, 5, 0 μΜ DMSO (0.0003-0.1%) {Luetal., 1996) meth iisoquin ilne)

Pertussis toxin (PT) 100,200,S00ngm! None (Abood et al., 1S87)

Thspsigatgin 0.3, 1,10, 30, 50, 100 ηΜ, DMSO (0.0015-0.1%) (Rogers et al., 1995}

1 μΜ

Cyelopiazonic acid SOOnM, 2.5 \Μ, 10μΜ DMSO (0.02-0.2%) (Goegeretai., 1988; Sediee (CPA) eial., 1S89)

BAPTA(1,2~bis(o- 0.25, 0.5.1,2,2.5,5, 10, DMSO (0.005-0.6%) (Tsien, 1980) amirtof! 8nox )sthane- 30 μΜ

NA.M^:. ^"-!i:;!"iiH¾if!c

acid)

Lysophospbatidic acid 10ηΜ, 1Q0nM, 1μ None (Anetai., 1898; Meyer Zu (LPA) Heringdort 2004)

Clotrimazole 3, 10,30μΜ D!vlSO (7.5x10 ^{' -ltt "} (Bsrto!ommei et al., 2006)

¾)

Chlorpromazine 3, 10,30μΜ DMSO (7.5x10"^,ic" (Hwang et al.,2001)

3%)

Cyclosporin (CY) 2, 10,40 πΜ DMSO (0.01-0.2%) (Liu et al., 1991; S anson et al., 1892)

N-82 (1-[N,0-B;s(5- 1,5, 20 μΜ DMSO (0.01-0.2%) (Tokurctftsu eta!., 1990) isoquinoiineeulforsylJ-N- mefiyl-L-tyrosyi]-4- p enylpiperazine) Table 2 Concentrations of drugs and (bracketed) DMSO vekMe controls as used in the experiment shown m Figure 3B.

€®n% t tmn

Low Medium High

Cyciopsazonic Acid 500 nM 2.5 μΜ 10 μΜ

(0.005%) (0.025%) (0.1 %)

Thapsigargin 1 nM 3 10 nM

(0.0001 %) (0.0003%) (0.001 %)

LPA 10 n 100 nM 1 μΜ

(H₂0) (H₂0) (H₂0)

Clotrimazole 3 μΜ 10 \iM 30 \M

(0.00075%) (0.0025%) (0,0075%)

Chlorpro asane 3 μ 10 μ 30 μΜ

(0,003%) (0.01%) (0.03%)

Table 3 ~ List of additional small molecules? evaluated

Menadione Aminothiazoie j

Table 4: Additional small molecule drugs capable of maintaining OCT4 / A OC3 +ve phenotype in hESCs in the absence of bFGF identified in primary and secondary screens on RCMl and RHl hESC lines respectively. Triangular arrows indicate up- or downreguiation of the indicated target, respectively.

Table 5: Candidate non-nutrient SMs capable of maintaining a OCT4 / NANOG +ve pfeenotype in repeated screening of CM1 and RH 1 hESCs in the absence of bFGF.

(a) Summary table of the statistical analysis of the final bFGF substitution screes carried out in REE I hESCs, showing the drags included, in the screen and the concentrations, if any, at which they increased the proportion of QCT4/NANOG+ve cells over vehicle coatroi to a statistically significant level. Key: *, P < 0.05; **, P <

0.01 ; P < 0.001; ****, P < 0.0001. (b) Triangular arrows indicate tip- or dowsreguiaiion of the indicated target, respectively. (S aded are drugs subsequently

Table 6: Dmgs identified as capable of mamtaining OCT4 / NANOG +ve plienoiype in hESCs i fee absence of ΤΟΡβ in two independent screens on RH1 and RCML (a) Summary table of the statistical analysis of the final TGFp substitution screen carried out in RCM1 hESCs, showing the drags which increased the proportion of OCT4 NANOG -ve cells over vehicle control to a statistically significant level. Key: *, P < 0.05; **, P < 0.01; ***, P < 0.001; **»*, P < 0.0001. (b) Table showing unique drags taken forward from the TGFp screen including additional drags demonstrating a subjective improvement. Triangular arrows indicate up- or downregulation of the indicated target, respectively, (Shaded is drag subsequently evaluated in combinatorial screen).

Table 7 Molecular structure and summary of outcome of drugs which independent screens exhibited a statistically significant or subjective improvement in maintenance of a human piuripotent stem cell phenotype above vehide control lacking bFGF or TGFp

Table 8: Drugs identified as capable of supporting an AP+vc or NANOG +ve phenotype in hESCs in die absence of bFGF and TGFp in E8 or niTeSR, respectively. Greyed out drugs (heparin, prednisolone and sulpiride) appeared in both screens or were removed as apparently merely promoting vigorous cell growth in the short, term (riboflavin).

Table 9, Combinations of drags apparently supporting growtii and self renewal of hESCs in defined media, for a. period of at least one week,

REFERENCES

Assou, S., Le Carrour, ^'!^'., Tondeur, S., Strom, $., Gabeile, A., Marty, S._s Nadal, L.

Paniesco. Heme, T., Hugnot, J. P., Gasca, S„, Kovatta, O., Hamamah, S., Klein. B., & De Vos, J. (2007). A meta-analysis of human embryonic stem cells transcripto e integrated into a web-based expression atlas. Stem Cells, 25(4), 961-973.

Bartolommei G., Tadim-Buomusegni, F., Hua, S., Monoelli, M. R., Inesi, G., &

Guidelli, R, (2006). Clotrimazole inhibits the Ca2+-ATPase (SERCA) by interfering with Ca2+ binding and favoring the E2 conformation. J Biol Ckem,

2*7(14), 9547-9551.

Seattle, G. M„ Lopez, A. D., Bucay, N., Hinton, A., Firpo, , T., King, C. C, &

Hayek, A. (20(35). Activin A maintains pluripotency of human embryonic stem cells in the absence of feeder layers. Stem Ceils, 25(4). 489-495.

Berridge, M. J., & Irvine, . F. (] 984), Inositol trisphosphate, a novel second

messenger in cellular signal transduction. Nature, 3 J 2(5992), 315-321,

Braun, A, P., & Schulman, H. (1 95). The multifimctional calcium caimoduliii- dependent protein kinase: from form to function. Ar m Rev Physiol 57, 417-

445.

Cappuccio, I, Spinsanti. P., Porcellini, A., Dssiderati, F., De Vita, T., Storto, M.,

Capobianco, L., Battaglia, G.. MeoletiL P. , & Mekhiorri D. (2005).

Endogenous activation of mGiuS metaboiroplc glutainate receptors supports self-renewal of cultured mouse embryonic stem cells. Neuropharmacology, 49 S ppl 196-205.

Chen, G. et al. (2011). Chemically defined conditions for human iPSC derivation and culture. Nature Methods 8 424-429,

De Soasa, P. A., Gardner, J„ Sneddon, S., Pells, S„ Tye, B. J, Band P., Collins, D.

M< Stewart, ¥..., Shaw. L., Przyborski, S., Cooke, M., McLaughlin, K. J.,

Kimber, S. J,, lieberman, B. A.. Wilmut, L, & Brison, D. R. (2009). Clinically failed eggs as a source of normal human embryo stem cells. Stem Ceil Res,

2(3), 188-197,

Dorsam, R, T,, & Cmtkind, J. S. (2007), G -protein-coupled receptors and cancer. Nat

Rev Cancer, 7(2), 79-94. Dvorak, P., Dvorakova, D., Koskova, S,, Vodinska, M., Najvirtova, .. Krekac, 0.. & Bampl, A. (2005). Expression and potential role of fibroblast growth factor 2 and its receptors in human embryonic stem cells. Stem Cells, 25(8), 1200- 1211.

Faheriy, S., Fitzgerald, A., Keohan, ., & QrMan, L, it. (2007). Self-renewal nd differentiation of mouse embryonic stem cells as measured by Oct4 expression: the role of fee cAMP/PKA pathway. /« Pi^'i o€¾// Z¾v BiolAnim, 43(1), 37-47.

Fletcher, I. M., Ferrier, P, M., Gardner, J. O., Harkness, L., Dhanjal, S,, Serhal, P., Harper, J., Dclhanty. J., Brownstesn, D. ⁽.)... Prasad, Y. R., Lebkuwski, J.,

Mandal&rn, R.. Wilmut, I., & De So«sa, P. A. (2006). Variations in humanized and defined culture conditions supporting derivation of new human embryonic stem cell lines. Cloning Stem Cells, 8(4), 319-334.

Hanaa, J., Cheng, A. W., Safea, K., m, I., Lengner, C, J., Soldner, F., Cassady, J. P., Muffat, J,, Carey, B, W., & Jaenisch, R. (2010). Human embryonic stem cells with biological and epigenevic characteristics similar to those of mouse ESCs, Proc Natl Acad Sci U S A, 707(20), 9222-9227,

Hwang, . H._s Han, Y. N., & Han, B. H. (2001). Inhibition of calmodulin-dependent calcium-ATPass and phosphodiesterase by various eyclopeptides and peptide alkaloids from the Zizyphus species, Arch Ph r Res, 24Q% 202-206,

James, D.₅ Levine, A. J., Besser, D., & Hemmati-BrivanJoa, A. (2005).

TGF%eta/aetivin/nodal signaling is necessary for the maintenance of pluripoteney in human embryonic stern cells, Development, 132(6% 1273- 1282,

Kaag, H. B., Kim, J. S., Kwon, H. J., Nam. K. H., Yours, H, S., Sok, D. E., & Lee, Y, (2005). Basic fibroblast growth factor activates ERK and induces c-fos in human embryonic stem cell line Mizb.ES 1. Stem Cells Dev, 14(4), 395-401.

Kap r. N., Mignery, G. A., & Banach, K. (2007). Cell cycle-dependent calcium

oscillations in mouse embryonic stem cells. Am J Physiol Cell Physiol, 292(4), CI 510-1518.

Layden, B. T., Newman, M., Chen, F., Fisher, A., &. Lowe, W. L., Jr. (2010). G

protein coupled receptors in embryonic stem cells: a role for Gs-alpha signaling, PLoS One, 5(2), ©9105. Lee,€ W., Rivera, R,, Dubiri, A, £., & Chun, J, (2007). LPA(4)/GPR23 is a lysophosphatidic acid (LPA) receptor utilizing G(s)-, G(q)/G(i)-mediated calcium signaling and G(12/13)-mediated R o activation. J Biol Chan, 282(1% 4310-4317.

Lee, H. J., Jeong, C. I , Cha, J, R, & Kim, . W. (2010). P C-delta inhibitors

sustain self-renewal of mouse embryonic stem cells under hypoxi a i.u vilro.

Exp Mol Med, 42(β), 294-301.

Ludwig, T.E. et al, (2006). Derivation of human embryonic stem cells in defined.

conditions. Nature Biotech. 24 185-187

L d ig, T. £., Bergendahl. V., Levensteia, M. E., Yu, J._} Probasco, M, D., &

Thomson, J. A. (2006). Feeder-independent culture of h m n embryonic stem cells. Nat Methods, 3(8), 637-646.

Malbon, C. C (2005). G proteins in development. Nat Rev Mol Cell Biol, 6(9% 689-

701.

cCuddcn, C. R., Mains. M. D,, KImple, R. J., Siderovski, I). P„, & Willard, F. S.

(2005), G-protein signaling: hack to the future. Cell Mol Life Sci, 62(5), 551- 577.

Nakamura, K._s Salomonis, N., Tomoda, K., Yamanaka, S., & Conklin, B. R. (2009).

G(i)-eoypied GPCR signaling controls the formation and organization of human pluripoient colonies. PLoS One, 4(11 ), e7780.

Ng, H. H., & Surani, M. A. (201 1). The transcriptional and signalling networks of pluripotency, Nat Cell Biol, 13(5% 490-496.

Nichols, L, & Smith, A. (2009). Naive and primed pluripoient states. Cell Stem Cell,

4(6% 487-492.

Okita, K., & Yamanaka, S. (2006), intracellular signaling pathways regulating

pluri potency of embryonic stem cells. Curr Stem Ceil Res Ther_s 1(1), 103-1 1 ί , Patel, T. B. (2004). Single transmembrane spanning heterotrimeric g protein-coupled receptors and their signaling cascades. Pharmacol Rev, 56(3), 371-385, Febay, A., Wong, R. C, Pitson, S. M._s Wolveiang, E. J ., Pell, G. 5., Fiiipczyk, A., Koh, K. L., Tellis, L, Nguyen, L. T., & Pera, M. F. (2005). Essential roles of sphingosine-1 -phosphate and platelet-derived growth factor in the maintenance ofhuman embryonic stem cells. Stem Cells, 23(10), 1541-1548, Powers, D. E., Millman, J. R., Huang, R. B,, & Colton, C. , (2008). Effects of oxygen on mouse embryonic stem cell growth, phenotype retention, and cellular energetics. Biotechno! Bioeng, 101(2), 241 -254.

ilodgers, A., Mormeneo, D„, Long, J. $„, Delgado, A., Pyne, N. J., & Pyne, S. (2009).

Sphingosine 1 -phosphate regulation of extracellular signal-regulated iiase-

1/2 in embryonic stem cells. _¾e/« C¾/fe Z5ev, I<¥(9), 1319-1330.

Sato, N., Meyer, L., Skaltsounis. L,, Greengard, P., & Brivanlou, A. H. (2004),

Maintenance of pluripotency in human and mouse embryonic stem cells through, activation of Wnt signaling by a pharmacological GSK -3 -specific inhibitor. _Vef Med, 10(1 ), 55-63.

Sato, N., Sanjnan, L M. Heke, M._s UcMda, M,, Naef, F._s & Brivanlou, A. H. (2003).

Molecular signature of human embryonic stem cells and its comparison with the mouse. D v Biol, 260(2), 404-413.

Takenouchi, T., Ogihara, K,, Sato, M., & Kitani, H. (2005). Inhibitory effects of U73122 and U73343 on Ca2+ influx and pore formation induced by the activation of P2X7 nucleotide receptors in mouse microglial cell line. Biockim

Biophys Acta, 1726(2 177-186.

Todorova, hi. G., Fuenies, E„ Soria, B„ Nadal A., & Qiiesada, I. (2009).

Lysophosphatidic acid induces Ca2+ mobilization and c-Myc expression in mouse embryonic stem cells via the phospholipase€ pathway. Cell Signal

21(4), 523-528.

Uluiiann U, et al. (2008) GSK-3-specdfic inhibitor-supplesrnenied hESC medium

prevents the epithelial-mesenchymal transition process and the up-rcgulat on of matrix metal!oproteinases in hESCs cultured in. feeder-free conditions, Mol Hum Reprod. 14 (3): 169-79.

Vaidimarsdottir, G,, & umme y, C. (200S), Functions of the TGFbeta superfarnily in human embryonic stem cells. Apmis, 113(11-12), 773-789.

Xu, C._s InokxttTia, M. S._s Denham, J., Golds, K,_? undu, P., Gold, J. D.. & Carpenter, M. K. (2001). Feeder-free growth of undifferenti ated human embryonic stem cells. Nat Biotechno!, 19(10), 971 -974. Abood, MB., Lee, N.M,, and Loh, H.H. (1987). Modification of opioid agonist binding by pertussis toxin. Brain research 417, 70-74,

An, S., Bleu, T,, Zheng, Y„ and Goetzl, E.J. (1998). Recombinant human G protein- coupled iysophosphatidic acid receptors mediate intracellular calcium mobilization. Molecular pharmacology 54, 881-888.

Bartolommei. G., Tadjiii-Buomnsegni, F., Hua, S., Moncelli, M.R., inesi, G., and Guidelli, R. (2006). Clotrimazole inhibits the Ca2+~ATPase (SERCA) by interfering with Ca2+ binding and favoring the E2 conformation. The Journal of biological chemistry 28L 9547-9551.

Bleasdale, I.E., Tbak ix, N.R., Gremban, R.S., Bundy, G.L., Fiizpatrick, F.A., Smith, ILL, and Bunting, S. (1 90). Selective inhibition of receptor-coupled phospholipase C-dependent processes in human platelets and polymorphonuclear neutrophils. The Journal of pharmacology and experimental therapeutics 255, 756-768,

Budworth, J., and Geschec, A. (1995). Differential inhibition of cytosolic and membrane-derived protein kinase€ activity by staurosporine and other kinase inhibitors, FEBS letters 362, 139-142.

Capp ccio, L Spmsanti, P., PoreeltinL A., DesideratL F., Do Vita, T.₅ Storto, M., Capobianco, L., Batiaglia, G., Nicoietti, F., and MelcMorri, D. (2005). Endogenous activation of mGl.u5 metahotropic g!utaraate receptors supports self-renewal of cultured mouse embryonic stem cells. Neuropharmacology 49 Suppi 1, 196-205.

Cassel, D., and Ffeuffer, T. (1978). Mechanism of cholera toxin action: covalent modification of the guanyi nucleotide-binding protein of the adenylate cyclase system. Proceedings of the National Academy of Sciences of the United States of America 75, 2669-2673.

Castagna, M., Takai, aibuchi, .₉ Sano, K._s Kikkawa, U., and Nishizuka, Y.

(1982). Direct activation of calcium-activated, phosphoh^'pid-dependerii protein kinase by tumor-promoting phorbol esters. The Journal of biological chemistry 257, 7847- 7851 ,

Gasparini, F., lingeahohl, K., Stoehr, N., Flor, P.J., Heimich, M., Vranesic. I., Biollaz, M., Allgeier, H., Heckeiidoro, R., Urwyler, S., et el. (1999). 2-Methyl-6- (phenylethynyf)-pyridine ( PEF), a potent, selective and systemically active mGJu5 receptor antagonist. Neuropharmacology 38, 1493- Ϊ 503.

Gill, D.M., and Cobuni, I. (1987). ADP-ribosylation by cholera toxin: functional analysis of a cellular system that stimulates the enzymic activity of cholera toxin fragment Al, Biochemistry 26, 6364-6371 ,

Goeger, D.E., Riley, R.T., Domer, J.W., and Cole, RJ. (1988). Cyc piazonic acid inhibition of the Ca2-i- -transport ATPase in rat skeletal muscle sarcoplasmic reticulum vesicles. Biochemical pharmacology 37, 978-981

.

Hwang, K.H., Han, Y.N., and Han, B.H. (2001), Inhibition of calmodulin-dependent calcium-ATPass and phosphodiesterase by various cyclopeptides and peptide alkaloids from the Zizyphus species, Archives of pharmacal research 24, 202-206 Kotani, M., Detheux, M,, Vandenbogaecde, A., Communi, D., Vanderwinden, J.M., Le PouL E.. Brezillon, S., Tyldesley, R., Saarez-Huerta, N., Vaadeput, P., el aL

(2001). The metastasis suppressor gene KiSS-1 encodes kisspeptins, the natural ligands of the orphan G protein-coupled receptor GP 54. The Journal of biological chemistry 276, 34631-34636.

Lin, J., Farmer, J.D., Jr., Lane, W.S., Friedman, J., Weissman, I., and Schreiber, S.L. (1991), Calcineurin is a common target of cyciopliilm-cyclosporin A and F BP- F 506 complexes. Cell 66, 807-815. La, Z.X., Quazi, N.H., Deady, L.W., and Polya, G.M. (1996). Selective inhibition of cyclic AMP-dependent protein kinase by isoquinoline derivatives. Biological chemistry Hoppe-Seyler 377, 373-384. Meyer Zu Heringdorf, D, (2004). Lysophospholipid receptor-dependent and ~ independent calcium signaling. Journal of cellular biochemistry 92, 937-948.

Rogers, T.B., inesi, G., Wade, R , and Lederer, W.J. (1995), Use of thapsigargin to study Ca2+ homeostasis in cardiac cells. Bioscience reports 15, 341-349.

Seamon, K.B., Padgett W,, and Daly, I,W. (1.981), Forskolin: unique diterpene activator of adenylate cyclase in membranes and in intact cells. Proceedings of the National Academy of Sciences of the United States of America 78, 3363-3367. Seidler, ., Jona, L, Vegh, M,₅ and Martonosi, A. (1989). Cyelopiazonie- acid is a specific inhibitor of the Ca2+-ATPase of sarcoplasmic reticulum. The Journal of biological chemistry 26 17816- 17823.

Swanson, S.K., Bom, T., Z dowsky, L.D., Cho. H., Chang, H.Y., Walsh, C.T., and Rusiiak, F. (1992). Cyelosporin-rnediated inhibition of bovine caleineurin by cyclophil s A and B. Proceedings of the National Academy of Sciences of the United States of America 89, 3741-3745.

Takasaki, J., Saito, T,, TanigucM, M., Kawasaki, T., Moritani, Y., Bayashi, K.. and. Kobori, M. (2004), A novel Galphaq/11 -selective inhibitor. The Journal of biological chemistry 27» 47438-47445.

Takumitsu, H.. Cbijiwa, T., Hagiwara, M., Mlzutani, A., Terasa a, M., and Hidaka. H. (1990). fCN -6 , l-fN^O-bisCS-isoquinoliiiesulfony^-N-melhyl-I^tyrosyl]^- phenylpiperazi ne, a specific inhibitor of Ca2÷ calniCKlulin-dq>endcait protein kinase II The Journal of biological chemistry 265, 4315-4320.

Tsien, R..Y. (1980), New calcium indicators and buffers with high seleciraiy against magnesium and protons: design, synthesis, and. properties of prototype structures. Biochemistry 19, 2396-2404.

Claims

c a; ins

1. A method of modulating the fate of a stem cell, said method comprising the step of providing one or more G-preteks coupled receptor (GPCR) agonists, antagonists and/or amipoisnds which, activate one or more GPCRs or calcium signalling events downstream of GPCR signalling to the stem cell, in order to modulate its fate.

2. The method according to claim 1 wherein said one or more compo¾nd(s) effects the heterotrimeric G-protein subunits (G« from and G _γ- subunits) or the signal transduction molecules that, heterotrimeric G-protems interact with in order to modulate free intracellular calcium levels/concentration such as for example phospholipase C beta,

3. The method according to claim 1 wherein at least one of said one or more compounds modulates the concentration or level of intracellular- calcium within said stem cell.

4. The method according to claim 3 wherein said one of more compowd(s) inhibits sarcoplasmic/endoplasmic reticulum calcium ATPases (SERCAs).

5. The method according to claim 3 wherein said one or more compo ad(s) modulates surface plasma membrane and/or mtramembtanous ion channel permeability.

6. The method according to claim 3 wherein said one or more compound(s) directly stimulates substrates that bind calcium, such as for example Calmodulin, or um signal transduction such as for example the calcium modulated

7. The method according to claims 1- 6 wherein said one or more compounds is/are selected from the group consisting of: (i) Cyclopiazoj ic acid (CPA);

(;;) Thapsigargin;

(iii) Lysopfiosphaiidic Acid (LP A);

(iv) Clotrimazole;

(v) CMoxprornazims;

8, The method according to any of claims 1 - 6 to maintain stem cells in an undifferentiated state by treatment with one or more chemicals or gmc products which may a Heel GPCR and calcium si nal transduction specifically or non~ specifically, without conventional reliance on the provision of exogenous protein growth factors, such as bFBF and/or TGFp supplied directly, or indirectly via supplementation of the cell culture environment with serum, or co-culture or conditioning of the environment with helper cells.

9. The method according to claim 8 wherein said one or more compounds

ienzopen

(xi) CMoropropamide

(xii) Ciimarizrine

(xiii) Mesalaniine

(xiv) Quinine

(xv) Tolfenamic Acid; and

(xvi) Menadione; further optionally including

(xvii) Metformin (xviii) Metoprolol

(xix) C oroquine

fxx) Riboflavin

10. ^'The method according to my of claims 8 or 9 wherein said one or more compounds is/are selected from the groups consisting of:

(i) heparin and Epinephrine; and

(ii) heparin, epinephrine, sulpiride and acetazo!amide.

11. The method according to any preceding claim, wherdn said one or more compounds are capable of modulating the fate of the stern cell for at least 5, 6, 7, 8, 9, 10 or more days.

12. The method according to my preceding claim wherein the method is intended to promote self-renewal in stem cells and/or to maintain stem cells in an undifferentiated, proliferative and or plunpotent state,

13. The method according to claim 12 wherein at least one of said one or more compounds is intended to maintain the stem cell in an undifferentiated state in the absence of bFGF and/or TGFbeta.

14. The method according to any preceding claim wherein the stem cell is an embryonic, foetal, adult and/or induced phmpo eot (IPS) stem ceil.

15. A composition for maintaining a stem cell in a proliferative and/or pluripotent state, e.g. in an ^'undifferentiated state, said composition comprising one or more compounds as defined in any of claims 1 - 10 and wherein the composition is substantially free of bFGF and/or ΤΟΡβ,

16. A culture medium for maintenance of stem cells which contains any of the compounds listed in claims 1-10, in a stem cell culture medium, such as an ES culture medium wi hout cytokine supplementation.

17, A medium according to claim 1 wherein the culture medium is selected from mTeS ™! , TeSR™2, TeSR™-E8™ and modified versions thereof.

18, A medium according to claim 15 wherein the culture medium is free of protein and cytokines tor pluripotency.

1 , A culture comprising the composition or cuiture medium according to any of claims 15 - 18 together with one or more stern cells.

20. Use of one or more compounds as defined in any of claims 1 - 10 , for maintaining stem cells and or for inducing self-renewal in stem cells, such as maintaining stem cells in an undifferentiated state.

21. A kit comprising one or more compounds as defined in any of claims 1 - 10, together with one or more other components selected from the groups consisting of:

(a) cell culture media (to he mixed with, or pre-mixed with said one or more compounds);

(c) instructions for use.

22. A method of c ltaring or enriching specific cell types from mixed populations of cells and/or from tissue/biopsy samples comprising contacting said mixed population of cells with one or more compounds as defined in any of claims 1 - 10,