OSTEOGENIC DIFFERENTIATION FROM MOUSE EMBRYONIC STEM CELLS AND THE ROLE OF CALRETICULIN

by

Yanhong Yu

A thesis submitted in conformity with the requirements for the degree of Master of Science in the Graduate Department of Laboratory Medicine and Pathobiology, in the

University of Toronto

© Copyright by Yanhong Yu 2013

ii

OSTEOGENIC DIFFERENTIATION FROM MOUSE EMBRYONIC STEM CELLS AND THE ROLE OF CALRETICULIN

Yanhong Yu M.Sc., 2013

Department of Laboratory Medicine and Pathobiology, University of Toronto

Abstract

Calreticulin, an endoplasmic reticulum (ER)-resident protein, is a calcium

buffering chaperone. In this study, with an optimized differentiation protocol from mouse

R1 ES cells, we demonstrate a novel role of calreticulin in osteogenic commitment and

differentiation. To enhance the efficacy of the method, we manipulated cell density and

examined the addition of retinoic acid, dexamethasone and peroxisome proliferator-

activated receptor γ. The regimen consisting of seeding 250 cells per embryoid body,

with the addition of RA (from day 3 to 5) and Dex (from day 10 to 21) gave the most

efficacious output. Using this optimized protocol, we investigated the potential

involvement of calreticulin in osteogenesis. Calreticulin knock-out cells displayed

impaired osteogenesis compared to wild-type cells. In particular, the nuclear

translocation of the runt-domain related transcription factor 2 and Osterix, were impaired

in the absence of calreticulin. The stimulatory effect of calreticulin on osteogenesis was

mediated by its calcium buffering function.

iii

Acknowledgements First, I would like to thank my family and friends for their support and

encouragement.

I am grateful to my supervisor, Dr. Michal Opas, who has given me my start in

research. I really appreciate your knowledge, patience, wisdom, and your one of a kind

personality.

I would like to thank my committee members, Dr. Craig Simmons and Dr. Jane

Mitchell, for their invaluable expert advice and scientific input in the past two years. I am

also grateful to the Graduate coordinator, Dr. Harry Elsholtz, who had helped me

tremendously throughout my undergraduate and graduate studies in LMP.

I would like to thank former and current members of the Opas lab. Ewa, Layla,

Peter, Carlos, thank you for all that you have taught me and for your support. Without

you guys, this journey would not have been possible.

Finally, I would like to thank everyone else who has helped me along the way.

iv

Table of Contents Abstract ii Acknowledgements iii List of Figures and Tables vii List of Abbreviations viii INTRODUCTION Skeletal functions and development 1 Stages of osteogenic differentiation 4 Transcription factors in skeletogenesis 4 Embryonic stem cells 6 Embryoid bodies: an in vitro model of ES cell differentiation 7 Calreticulin 11 Calreticulin and its Ca2+ buffering function 15 Calreticulin and embryonic development 16 Calreticulin affects gene expression and cell differentiation 17 Calreticulin and skeletogenesis 18 HYPOTHESIS 19 METHODS AND MATERIALS Cell culture 19 Treatment with putative osteogenic factors 20 Ca2+ manipulation studies 23 Mineralization assays 23 Alcian blue staining 26

v

Real-time PCR 26 Immunofluorescence staining 28 Transmission electron microscopy 29 Statistical analysis 29 RESULTS

Aim 1: To enhance the efficacy of mouse ES cell osteogenic differentiation protocol

-The effect of cell number on osteogenic differentiation 31 -The effect of Dex 31 -The effect of RA on osteogenesis 34 -The combined effects of RA and Dex 34 -The effects of PPARγ 37 -Confirmation of osteogenic differentiation from ES cells 37 Aim 2: The role of calreticulin in osteogenesis

-Calreticulin expression during mouse ES cell osteogenic 44 differentiation -Calreticulin deficiency impairs osteogenesis 45 -Chondrogenic potential of calreticulin-null cells 48 -Nuclear localization of Runx2 and OSX is impaired in the 48 absence of calreticulin -Calreticulin affects osteogenic differentiation via its Ca2+ 53 buffering activity -The involvement of calreticulin’s functional modules 58 in its regulation of osteogenesis

DISCUSSION Different cell lines have different osteogenic potential 58

vi

The effect of cell density on ES cell osteogenic differentiation 60 The pro-ostegenic effects of B-Gly and AA 61 The role of Dex in osteogenesis 63 RA and osteogenesis 64

PPARγ: a positive regulator of adipogenesis and negative 65 regulator of osteogenesis

Roles of calreticulin throughout differentiation 67 Osteogenic marker expression 68 Increased chondrogenic potential in calreticulin-deficient cells 69 Impaired nuclear translocation of Runx2 and OSX in KO cells: 69

implications Calreticulin in the regulation of transcription factors 71 FUTURE DIRECTIONS

The involvement of calcineurin-NFAT pathway downstream 72 of calreticulin in its regulation of osteogenesis STAT1: a cytosolic attenuator of Runx2 73 Calreticulin and OSX nuclear localization 76

REFERENCES 80 APPENDIX 92

vii

List of Figures and Tables Table 1. A summary of the studies which investigated the effects 10 of different factors promoting osteogenesis from ES cells. Figure 1. Skeletal maintenance and development. 3 Figure 2.Structural and functional domains of calreticulin. 14 Figure 3. The effect of cell number on osteogenic differentiation 22 from mouse ES cells. Figure 4. The effect of soluble factors on osteogenic differentiation. 25 Figure 5. The lowest cell number and/or EB size favoured 33 ostegenic differentiation. Figure 6. Mineralization assays and PCR analysis revealed that the 36 RA and later addition of Dex treatment resulted in the highest level of osteogenic differentiation. Figure 7. Immunofluorescence staining of the osteogenic markers, 39 Runx2 and OSX. Figure 8. Transmission electron microscopy images of differentiated 41 day 21 nodules obtained from the RA and Dex (late) protocol. Figure 9. The most efficacious culture condition for osteogenesis 35 from mouse R1 cells. Figure 10. The expression of calreticulin throughout osteogenic 47 differentiation. Figure 11. Detection of osteogenic differentiation in calreticulin- 50 containing and deficient cells. Figure 12. Increased chondrogenic potential in calreticulin-deficient 52 cells. Figure 13. Immunofluorescence staining of the osteogenic markers, 55 Runx2 and OSX in calreticulin wild type and KO cells. Figure 14. The effect of changes in cytosolic Ca2+ levels on 57 osteogenic differentiation in calreticulin-containing and deficient cells.

viii

Figure 15. The involvement of calcineurin-NFAT pathway downstream 75 of calreticulin in osteogenesis. Figure 16. Calreticulin regulates the subcellular localization of STAT1. 78

Abbreviations AA - ascorbic acid ALP - alkaline phosphatase ATF4 – activating transcription factor 4 BMP-2 - bone morphogenetic protein-2 BSP - bone sialoprotein B-Gly - β-Glycerophosphate Ca2+ - calcium C/EBPα - CCAAT-enhancer-binding protein alpha Col 1- type 1 collagen Col 10 - type 10 collagen Dex - dexamethasone EB - embryoid body EGTA - ethylene glycol tetraacetic acid ER - endoplasmic reticulum ES - embryonic stem FITC - fluorescein isothiocyanate-conjugated GFP - Green Fluorescent protein InsP3R - inositol 1,4,5-triphosphate receptor KO - knock-out LIF - leukemia inhibitory factor MEF2C - myocyte-specific enhancer factor 2C MSX2 – msh homeobox 2 NFAT - nuclear factor of activated T-cells OCN - osteocalcin OSX - osterix PBS - phosphate buffered saline PPARγ - peroxisome proliferator-activated receptor PI - propidium iodide RA - retinoic acid Runx2 - runt-related transcription factor 2 SERCA - sarcoplasmic/endoplasmic reticulum Ca2+-ATPase SOX9 - sex determining region Y-box 9 STAT1 - signal transducer and activator of transcription 1!

1

INTRODUCTION

Skeletal functions and development

The skeleton of mammals is maintained through the bone-forming

activities of osteoblasts and bone-resorbing activities of osteoclasts (Fig. 1A).

Proper maintenance of the mature skeleton is crucial for its functions such as

mechanical support, muscle attachment, storage of calcium (Ca2+) and

phosphorus, as well as regulation of serum phosphorus levels (Long, 2012).

Osteoblasts are cells specialized in the production of extracellular matrix (mainly

collagen type 1 (Col 1)) and the mineralization process in bone (Aubin, 2001).

During vertebrate embryogenesis, bones are formed from two different

processes, endochondral ossification and intramembranous ossification (Fig. 1B).

These two mechanisms are responsible for forming different types of bones in

the vertebrate skeleton. For instance, bones in the craniofacial regions and the

clavicles are produced directly from osteoblasts via the process of

intramembranous ossification, while the rest of the skeleton such as the axial

components and the limbs utilizes a cartilage framework formed by chondrocytes

which is replaced by bone and bone marrow by the process of endochondral

ossification (Olsen et al. 2000). In this process, the mesenchymal progenitors

condense and differentiate into chondrocytes and periochondrial cells. These

chondrocytes eventually become hypertrophic and secrete factors to initiate the

differentiation of perichondrial cells into osteoblasts (Long, 2012). Furthermore,

different parts of the vertebrate skeleton have different embryonic origins. Cranial

neural crest cells give rise to the craniofacial skeleton, the limb bones are formed

2

Figure 1. Skeletal maintenance and development. A. The bone formation activities of osteoblasts and bone resorping activites of osteoclasts are crucial for the proper maintenance of the skeleton. B. During embryogenesis, bone is formed from either intramembranous or endochondral ossification.

4

from cells from the lateral plate mesoderm, whereas somites (paraxial

mesoderm) give rise to the axial skeleton (Olsen et al. 2000).

Stages of osteogenic differentiation

Osteogenic lineage cells are a population of cells that include

mesenchymal progenitors, preosteoblasts, osteoblasts, osteocytes, and bone-

lining cells. The identities of cells at different stages of differentiation are not well

defined. Preosteoblasts are heterogeneous since they comprise all cells

differentiating from progenitors to mature osteoblasts; they express the

transcription factor the runt-related transcription factor 2 (Runx2), or both Runx2

and Osterix (OSX) at a later stage (Long, 2012). Osteoblasts express unique

combination of extracellular proteins such as osteocalcin (OCN), alkaline

phosphatase (ALP), and Col 1 (Long, 2012). Col 1, which is abundant in the

extracellular matrix of bone, is first synthesized by osteoblasts; Ca2+ phosphate is

then deposited on this matrix in the form of hydroxyapatite. Some populations of

osteoblasts become osteocytes when they are trapped in the bone matrix that

they produced. Others can undergo apoptosis or become inactive bone-lining

cells (Bonewald, 2011). Osteocytes are important in responding to mechanical

and hormonal signals, playing major roles in bone remodeling (Bellido et al.

2005; Chen et al. 2010).

Transcription factors in skeletogenesis

Several transcription factors are required at different stages during

osteogenic differentiation. All osteoblast precursors express msh homeobox 2

(MSX2), TWIST, and sex determining region Y-box 9 (SOX9), Since MSX2 and

5

TWIST are osteogenic repressors, their expression must be downregulated in

osteoblast precursors for osteogenesis to proceed. The role of SOX9 in

osteogenesis is not well understood. Conditional knock-out (KO) of SOX9 in the

limb bud mesenchyme resulted in abolition of formation of both chondrocytes and

osteoblasts (Akiyama et al. 2002). In contrast, the absence of SOX9 in the

neural crest cells caused cells that would normally differentiate into chondrocytes

to express osteoblastic markers instead (Long, 2012; Mori-Akiyama et al. 2003).

These data may be due to the fact that SOX9 is required for both

chondrogenesis and osteogenesis in endochondral ossification; while during

intramembranous ossification, SOX9 is a negative regulator of Runx2 and OSX

in the bipotential osteochondral progenitor cells. Runx2 is indispensable for

osteoblastic differentiation for both endochondral and intramembranous

ossification. Runx2-null mice were found to have an abnormal cartilaginous

skeleton. In these mice, osteoblast differentiation was arrested at early stages

(Long, 2012). Haploinsufficiency of Runx2 in mice or in humans resulted in

hypoplastic clavicles and delayed closure of the fontanelles (which are the

spaces between cranial bones held together by fibrous tissues), both phenotypes

are characteristic of cleidocranial dysplasia in humans (Mundlos et al. 1997; Otto

et al. 1997). In the process of endochondral ossification, higher Runx2

expression induces chondrocyte hypertrophy and osteoblast differentiation.

Furthermore, Runx2 expression is also upregulated at early embryonic

development in the osteo-chondroprogenitor cell during mesenchymal

condensation (Ducy et al. 1997; Otto et al. 1997).

6

OSX is another transcription factor required for the regulation of osteoblast

differentiation and bone formation. OSX-deficient mice displayed abnormal bone

formation, suggesting a major role of OSX in osteoblast differentiation (Long,

2012). Interestingly, OSX-null osteoblast precursors still express Runx2,

suggesting OSX acts downstream of Runx2 (Long, 2012). Activating transcription

factor 4 (ATF4) is required in more mature osteogenic lineage cells. It directly

regulates the expression of OCN. It is also promotes amino acid import for

protein synthesis by osteoblasts (ref 76 long)

Embryonic stem (ES) cells

ES cells are primitive cells with the potential to give rise to all types of cells

in the body. Thus, they have tremendous potential in clinical applications and

they have been extensively studied in the past few decades. The first isolation of

mouse ES cells was reported 25 years ago by two independent groups (Evans

and Kaufman, 1981; Martin, 1981), which has revolutionized studies in genetics,

developmental biology, and regenerative medicine. ES cells are derived from the

inner cell mass of pre-implanted blastocysts. They have the potential of self-

renewal in vitro and they can also differentiate under certain conditions and give

rise to cells of all three germ layers (endoderm, mesoderm, and ectoderm).

These ES cells are commonly grown and maintained on a feeder layer of

division-incompetent mouse fibroblasts, which are known to be important to

maintain ES cells in their undifferentiated state (Khillan and Chen, 2010).To

further prevent the spontaneous in vitro differentiation of ES cells, the addition of

a soluble factor, leukemia inhibitory factor (LIF) is required. Subsequent LIF

7

withdrawal and growth in suspension leads to embryoid body (EB) formation and

differentiation (Niwa et al. 2009).

EBs: an in vitro model of ES cell differentiation

EBs, which are three-dimensional aggregates of stem cells, are a unique

in vitro model that mimics the development of embryos during early gastrulation.

They represent a unique tool to investigate developmental processes and the

generation of the three germ layers (endoderm, ectoderm and mesoderm)

(Boheler et al. 2005) . The endodermal layer can give rise to adult tissues such

as the pancreas, gastrointestinal organs, and the lungs; the ectodermal layer

gives rise to skin, connective tissue of the head and face, neurons, etc. The

mesenchymal progenitor cell can differentiate and commit to several different cell

fates such as adipocytes, cardiomyocytes, chondrocytes, and osteoblasts (Long,

2012).

Several methods that are widely used to generate the EB include: liquid

suspension culture in bacteriological dishes, culture in methylcellulose semisolid

media, hanging drops, suspension culture in low-adherence vessels, and spinner

flask/bioreactor techniques (Kurosawa, 2007). Suspension culture in bacterial-

grade dishes relies on the hydrophobicity of the nontreated polystyrene dish to

prevent the attachment of the seeded ES cells (Doetschman et al. 1985). The

ES cells thus form aggregates of EBs in the absence of shaking. This method of

EB induction has been utilized to produce a variety of mature cell types such as

neural progenitors (Plachta et al. 2004), hepatic cells (Jones et al. 2002),

cardiomyocytes (Klug et al. 1996), and germ cells (Toyooka et al. 2008) from

8

murine ES cells. A disadvantage of this method is that because ES cells form

aggregates spontaneously, the number of cells in each EB would vary. Thus, the

size and shape of the EBs would be heterogeneous (Wartenberg et al. 1998).

This is problematic since cell number/density, as illustrated later in Results, is an

important determinant of the efficacy of the differentiation protocol. It can further

affect the lineage commitment of progenitor cells.

In addition to bacterial-grade dishes, there are also other suspension

culture methods using low-adherence surfaces. For instance, dishes coated with

proteoglycan (Shinji et al. 1988) or uncoated dish with positively charged

surfaces are commonly used to form spherical aggregates of rat hepatocytes

(Kurosawa, 2007). Round-bottomed, 96-well plates with low-adherence surfaces

are also commonly used to generate EBs of uniform initial seeding numbers.

These plates are first coated with reagents such as a commercially available

nonionic detergent (F-127 from Sigma) or a phospholipid polymer, 2-methyl-

acryloyloxyethyl phosphorylcholine (MPC) that will prevent cell adhesion to the

well surfaces (Kurosawa, 2007).

Culturing in methylcellulose matrix has been widely used for

hematopoietic differentiation of ES cells (Keller et al. 1993; Wiles and Keller,

1991). Since the media is semisolid, the seeded ES cells remain isolated from

each other, and these single ES cells will develop into aggregates of EBs.

However, handling of semisolid solution by pipettes can be quite difficult and the

methylcellulose matrix can disturb the distribution of soluble factors added during

differentiation (Kurosawa, 2007).

9

The hanging drop culture has been widely used for ES cell differentiation

to a variety of cell types such as neuronal cells, cardiomyocytes, gametes and

osteoblasts (Keller, 1995). Twenty to 30 μL drops containing ES cells are

pipetted on the lid of Petri dishes. The lids are then inverted and placed over the

bottom of the Petri dish. Gravity causes ES cells to aggregate at the bottom of

the hanging drop and form EBs. The number of ES cells per EB can be

controlled by varying the number of ES cells in the initial cell suspension.

The EB model has been used previously to elucidate important factors

and mechanisms underlining the differentiation of mouse ES cells to osteoblasts

(Kawaguchi et al. 2005; Zur Nieden et al. 2003). Several studies used the

formation of EBs to generate osteoblasts in vitro within 3-4 weeks in the

presence of osteogenic factors, while others have also succeeded in generating

osteoblasts directly from ES cells in 10 days, without the formation of EBs

(Duplomb et al. 2007b; Hwang et al. 2006).

In all of these studies, the formation of osteoblasts and bone tissue have

been characterized by the capacity of the differentiated cells to mineralize and to

express elevated mRNA levels of osteoblastic marker genes such as OCN,

Runx2, ALP, OSX and bone sialoprotein (BSP). BSP and OCN are both

extracellular matrix proteins. They are markers for late osteoblastic differentiation

and are required for bone mineralization (Malaval et al. 2008; Martin, 1981).

There are several protocols in the literature attempting to enhance

osteogenic differentiation from mouse ES cells. The addition of soluble factors

such as β-glycerophosphate (B-Gly), ascorbic acid (AA), dexamethasone (Dex),

10

retinoic acid (RA) and even peroxisome proliferator-activated receptor γ (PPARγ)

inhibitors have been shown to increase the efficacy of osteogenesis in vitro from

ES cells (Akiyama et al. 2002; Buttery et al. 2001; Phillips et al. 2003;

Yamashita et al. 2006). Table 1 provides a summary of the different factors

which have been studied to enhance osteogenic differentiation from ES cells.

Table 1. A summary of the studies investigated the effects of different factors promoting osteogenesis from ES cells.

References Cell Line Used Factors Studied

(Phillips et al. 2001)

CGR8 (mouse) Β-Gly, AA, RA, BMP-2, Compactin.

(Kuske et al. 2011)

D3 (mouse) Β-Gly, AA, BMP-2, vitamin D3.

(Kawaguchi et al. 2005)

CGR8, E14Tg2a (both mouse)

Β-Gly, RA, AA, RA, Dex.

(Buttery et al. 2001)

CEE (mouse) Β-Gly, AA, RA, Dex (both early and late addition).

(Bourne et al. 2004)

CEE (mouse) Β-Gly, AA, Dex.

(Woll and Bronson,

2006)

HM1 (mouse) Β-Gly, AA

(Yamashita et al. 2006)

SV6 (mouse) PPARγ siRNA.

(Rose et al. 2012)

D3 (mouse) Basic fibroblast growth factor.

(Yamashita et al. 2005)

CMS-A-2 (monkey) Β-Gly, AA, BMP-2, Dex.

These studies have either used mouse or monkey ES cells. Abbreviations: B-Gly: beta-glycerophosphate; AA: ascorbic acid; RA: retinoic acid; Dex: dexamethasone, PPARγ: peroxisome proliferator-activated receptor-gamma; BMP-2: bone morphogenetic protein-2.

11

Importantly, it also has been demonstrated that the initial cell number comprising

the EBs and/or their size affects the end product of differentiation (McBeath et al.

2004; Park et al. 2007). This parameter has not been optimized in osteogenic

differentiation protocols. Thus, before further studies of molecular mechanisms

controlling osteogenesis, it is necessary to optimize culture conditions and to use

the most efficacious differentiation protocol.

Calreticulin

The endoplasmic reticulum (ER) has crucial roles in many cellular

processes such as Ca2+ storage, lipid and protein synthesis, protein folding and

post-translational modifications. These diverse functions are carried out by many

proteins that reside in the ER lumen. Calreticulin is a 46 kDa ER luminal Ca2+-

binding protein and molecular chaperone. It has numerous important roles in

biological processes such as apoptosis, immunity, cancer, angiogenesis, wound

healing, cardiac development, stem cell differentiation and central nervous

system development (Michalak et al. 2009). Calreticulin was first isolated in

1974 by Ostwald and MacLennan (1974). Genes encoding calreticulin have been

isolated from several vertebrates, invertebrates, and higher plants (Michalak et al.

1999). There is no calreticulin gene in the genomes of yeast and prokaryotes.

The human gene for calreticulin is located on chromosome 19 at locus p13.3-

13.2 and in the mouse on chromosome 8 (Gelebart et al. 2005). The nucleotide

sequences of the mouse and human gene displays greater than 70% identity,

suggesting a strong evolutionary conservation of the gene (Gelebart et al. 2005).

12

A 2002 study identified a second isoform of calreticulin (Persson et al. 2002);

however, the physiological function of this newly identified protein is unclear, but

it is abundantly expressed in the testes.

The presence of a N-terminal cleavable signal sequence directs

calreticulin to the ER, and it also contains an ER KDEL (Lys-Asp-Glu-Leu)

retention/retrieval signal. Calreticulin, together with calnexin and ERp57 (ER

protein of 57 kDa), is involved in the chaperoning of nascent polypeptides that

traverse through the ER (Hebert and Molinari, 2007). It has three domains, the N

domain, P domain and the C domain (Fig. 2). Structural studies revealed that the

N-domain is a globular structure comprised of eight anti-parallel β-strands

(Michalak et al. 1999). The highly conserved N-domain has been shown to

interact with the DNA-binding domain of the glucocorticoid receptor in vitro

(Burns et al. 1994), and with protein disulphide-isomerase (PDI) and ER protein

57 (ERp57) (Baksh et al. 1995; Corbett et al. 1999). Interestingly, these protein-

protein interactions have been shown to be dependent on Ca2+ binding to the C-

domain of calreticulin (Corbett et al. 1999). The proline-rich P-domain of

calreticulin is critical for the high-affinity Ca2+ binding and lectin-like chaperone

activity of calreticulin (Tjoelker et al. 1994; Vassilakos et al. 1998). The C-

domain binds to over 25 mol of Ca2+/mol protein (Baksh and Michalak, 1991) and

binds to blood-clotting factors (Kuwabara et al. 1995).

Molecular chaperones prevent the aggregation of partially folded proteins,

and increase the amount of correctly folded proteins. It has been widely

documented that calreticulin functions as a lectin-like molecular chaperone

13

Figure 2. The structural and functional domains of calreticulin. Calreticulin has three domains: the N, P and C domains. Its N and P domains are responsible for its chaperoning function. The C domain carries out its high capacity low affinity Ca2+-buffering activity.

15

(Michalak et al. 1999). Its chaperone activity is carried out by the N and P

domains; while its high capacity Ca2+ binding activity is found in its C domain.

Calreticulin and calnexin, an integral ER membrane protein, are both highly

versatile proteins that are involved in “quality control” during the synthesis of a

variety of molecules (Coe and Michalak, 2009) such as ion channels, surface

receptors, and transporters (Michalak et al. 1999; Michalak et al. 2009).

Calreticulin and Ca2+ buffering function

Ca2+ is a universal signalling molecule that influences many

developmental and cellular processes such as muscle contraction, cell

proliferation and cell death, neurotransmitter release, etc. The majority of the

intracellular Ca2+ is stored in the ER lumen. Extracellular Ca2+ level is 2mM, total

ER Ca2+ concentration is up to 1mM, free ER Ca2+concentration is approximately

200 µM, and finally, the free cytoplasmic Ca2+ concentration is only 100 nM

(Michalak et al. 2009). Ca2+ release from the ER via inositol 1,4,5-trisphosphate

receptor (InsP3R) and ER Ca2+ uptake by the sarcoplasmic/endoplasmic

reticulum Ca2+-ATPase (SERCA) are crucial in maintaining ER Ca2+ homeostasis

and signalling. The C-domain of calreticulin contains many negatively charged

residues that are responsible for the Ca2+-buffering function of the protein

(Michalak et al. 2009). It binds over 50% of ER luminal Ca2+ with high capacity

(25 mol of Ca2+ per mol of protein) and low affinity (Kd=2 mM) (Nakamura et al.

2001). Although the P-domain of calreticulin also binds Ca2+, it does so with high

affinity (Kd= 1 μM), but low capacity (1 mol of Ca2+ per mol of protein) (Baksh and

Michalak, 1991). Considering the high concentration of Ca2+ within the lumen of

16

the ER, the high affinity Ca2+-binding site of calreticulin has been predicted to be

permanently occupied by Ca2+ and probably serves a structural role only.

Consequently, the Ca2+ buffering function of calreticulin is mainly carried out by

its C-domain. Calreticulin is responsible for binding over 50% of the ER free Ca2+

(Michalak et al. 2009). This critical role of calreticulin as a regulator of

Ca2+homeostasis is further demonstrated in loss-of-function and gain-of-function

experiments. Upregulation of calreticulin lead to increased Ca2+ capacity of the

ER, increased free Ca2+ concentration in the ER lumen, and a larger pool of

releasable Ca2+ from the ER, where as calreticulin deficiency leads to reduced

Ca2+ storage capacity in the ER and impaired Ca2+ release from the ER

(Arnaudeau et al. 2002; Mery et al. 1996).

Calreticulin and embryonic development

Since calreticulin is involved in diverse basic cellular functions, it was

expected that calreticulin-null mice would not be viable due to early embryonic

lethal phenotype. In fact, calreticulin deficiency is embryonic lethal between day

12.5 and 14.5 due to abnormal cardiac development. Experiments using

calreticulin-null mice and ES cells showed that the absence of calreticulin lead to

impaired myofibrillogenesis (Li et al. 2002; Lozyk et al. 2006) and deficient

intercalated disc formation in the hearts. Further investigations revealed that

calreticulin affects transcription factors involved in cardiac development such as

the Nuclear factor of activated T-cells (NFAT) and Myocyte-specific enhancer

factor 2C (MEF2C) via its Ca2+ buffering activity (Michalak et al. 2009). In

addition to heart defects, calreticulin-null embryos also had disrupted cranial

17

neural tube closure and displayed exencephaly (Rauch et al. 2000). This is not

surprising since calreticulin is abundant in the developing brain (Mesaeli et al.

1999). Further studies demonstrated that slower cell migration and decreased

cell adhesion caused by calreticulin deficiency could have led to this phenotype

(Rauch et al. 2000). The failure of the ventral body wall closure was also

observed in the calreticulin-deficient embryos. The mechanism by which this is

regulated by calreticulin is unclear; however, it is highly plausible that

calreticulin’s effects in cell migration and attachment may have a role (Rauch et

al. 2000).

Calreticulin affects gene expression and cell differentiation

During cardiomyogenic differentiation, the expression of calreticulin in the

developing heart is high and then it declines (Mesaeli et al. 1999). Cardiac

transcription factors Nkx2.5 and MEF2C activate the calreticulin gene (Guo et al.

2001). Nkx2.5 then combines with GATA-4 transcription factor to further enhance

their activity (Nemer and Nemer, 2001). Furthermore, the combined activities of

the GATA-4 and NFAT transcription factors are also critical for cardiomyogenesis

(Nemer and Nemer, 2001). NFAT must be dephosphorylated by calcineurin, a

Ca2+- and calmodulin-dependent protein phosphatase (Crabtree, 2001) in order

to translocate to the nucleus. Calreticulin was found to regulate the

transcriptional activity of the GATA-4/NFAT complex by affecting calcineurin

activity through the control of ER Ca2+ release (Lynch and Michalak, 2003). Thus,

the absence of calreticulin leads to inactivation of the critical cardiac transcription

factors, leading to impaired cardiomyogenesis.

18

In addition to its involvement in cardiomyogenic development, it was

discovered that calreticulin also affects mesenchymal stem cell commitment

towards the adipogenic lineage. Calreticulin was found to act as a Ca2+-

dependent molecular switch that inhibits the commitment of ES cells to the

adipogenic lineage. PPARγ2 and CCAAT-enhancer-binding protein alpha

(C/EBPα) are both crucial transcription factors during adipocyte development

(Rosen and Spiegelman, 2000). Binding of PPARγ2 to the calreticulin gene

promoter enhances the expression of calreticulin. This increase in expression of

calreticulin downregulates the expression and transcriptional activation of

PPARγ2 and C/EBPα, forming a negative feedback loop (Szabo et al. 2008).

Calreticulin and Skeletogenesis

There are limited data available on calreticulin expression or function in

osteogenic cells in vivo or in vitro. It is well known in the literature that

calcineurin, a serine-theronine phosphatase that is regulated by calreticulin (Guo

et al. 2002; Lynch et al. 2005), plays an important role in chondrogenesis

(Reinhold et al. 2004) and osteogenesis (Sun et al. 2005). Hence, it is highly

plausible that calreticulin may be affecting osteocyte and chondrocyte

differentiation via the calcineurin pathway.

Previous studies have shown calreticulin’s critical roles in adipogenesis

and cardiomyogenesis (Lynch et al. 2005; Szabo et al. 2008). However, its

function in osteogenesis remains largely unknown; thus, the role of calreticulin in

osteogenic differentiation from mouse ES cells is examined in this study.

19

HYPOTHESIS

Calreticulin has a critical role in the osteogenic differentiation from mouse

ES cells.

Aims:

1) To optimize culture conditions for osteogenic differentiation from

mouse ES cells

2) To determine whether calreticulin has a role in osteogenic

differentiation

METHODS AND MATERIALS

Cell Culture

Osteogenic lineage differentiation was performed from G45 calreticulin KO

ES cells, CGR8 ES cells (both cell lines derived from J1 129/SV mice and were

kindly provided by Dr. Marek Michalak, Department of Biochemistry of University

of Alberta) or R1 mouse ES cells (derived from J1 129/Sv mice, purchased from

Dr. Janet Rossant, Mount Sinai Hospital, Toronto, Canada) (Nagy et al. 1993) by

the hanging drop method. The ES cells were maintained on mitomycin C-treated

mouse embryonic fibroblast feeder cells on gelatin-coated plates. At passage 2,

cells were trypsinized and collected for hanging drops. To test the effect of cell

number and/or EB size on the osteogenic lineage differentiation, varying

numbers of ES cells per 25 µL drops were placed on the lids of tissue culture

dishes for 3 days (Fig. 3). The medium consisted of: high glucose Dulbecco

modified Eagle’s medium with sodium pyruvate and L-glutamine (Multicell,

20

Wisent), 20% Fetal Bovine Serum (Multicell, Wisent), nonessential amino acids,

and β-mercaptoethanol. After 3 days in hanging drop, these ES cells aggregated

and formed EBs and then were floated in differentiation medium for 2 days. On

day 5, the EBs were then plated in tissue culture dishes coated with 0.1% gelatin

(in phosphate-buffered saline [PBS]) in differentiation medium with 50 µg/mL AA

and 10 mM B-Gly for commitment to the osteogenic lineage. The medium was

changed every 2 days for the entire 21 days of differentiation.

Treatment with putative osteogenic factors

Dex, RA and PPARγ inhibitor (GW9662) were selected over other soluble

factors because they were more frequently utilized in the literature. Although

bone morphogenetic protein 2 (BMP2) was also commonly used, it has also been

reported to induce chondrogenic differentiation. Another frequently used factor is

vitamin D. It was not examined in this study since it has been shown that Dex

can substitute its pro-osteogenic effects in the culture. The effects of Dex, RA,

and PPARγ inhibition on osteogenic differentiation from ES cells were examined

in six different protocols: A) Control: only AA and B-Gly addition; B) 0.1 µM Dex

(Sigma-Aldrich) early addition, added at day 6 and kept throughout

differentiation; C) 0.1 µM Dex late addition, added at day 10 and kept throughout

differentiation; D) PPARγ inhibitor: 1 µM GW9662 (Sigma-Aldrich) is added at

day 3 and kept throughout differentiation; 0.1 µM Dex is added at day 10 to

further enhance mineralization. E) 0.1 µM RA (Sigma-Aldrich) added during the

floating stage (from day 3 and kept until day 5), media was changed every day;

F) RA added from day 3 and kept until day 5 + Dex early addition (added at day 6

21

Figure 3. The effect of cell number on osteogenic differentiation from mouse ES cells. The hanging drop method was used to differentiate mouse ES cells into the osteogenic lineage. Aggregates of stem cells form embryoid bodies or EBs in the first 2 days. To test the effect of cell number and/or EB size on the osteogenic lineage differentiation, varying numbers of ES cells per 25 µL drops were placed on the lids of tissue culture dishes. These EBs then go into the “floating stage” in which they grow in suspension for another 2 days. On day 5, the EBs then are attached on gelatin-coated dishes until day 21. B-Gly and AA are added into the culture media starting at day 6 and kept throughout differentiation.

23

until day 21); G) RA added from day 3 until day 5 + late addition of Dex (added at

day 10 until day 21). Protocols B) to G) also contain B-Gly and AA. A schematic

diagram representing the above protocols and the timeline of differentiation can

be found in Fig. 4.

Ca2+ Manipulation Studies

The ionomycin and 1,2-Bis(2-aminophenoxy)ethane-N,N,N′,N′-tetraacetic

acid tetrakis(acetoxymethyl ester) (abbreviated BAPTA-AM) treatments were

performed as previously described by Szabo et al. (2008). To chelate

cytoplasmic Ca2+, floating EBs were incubated with 500 nM BAPTA-AM for 30

min each day from days 3 to 5. To increase cytosolic Ca2+, 1 μM ionomycin was

added to the media for 2 hr each day from days 3 to 5.

Mineralization Assays

von Kossa staining is a routine method for detecting Ca2+ deposits in

culture. Osteoblasts are known to produce a matrix of osteoid, which is mainly

composed of Col 1. Ca2+ and phosphate deposition takes place later by

osteoblasts, resulting in the formation of hydroxyapatite crystals (Field et al.

1974). von Kossa staining is based on the visualization of reduced silver ions that

replaced Ca2+ ions which were bound to phosphate in tissue deposits (Rungby et

al. 1993). von Kossa staining was performed on differentiated day 21 nodules to

detect mineralization. Before staining, cells were washed three times with PBS,

fixed in 10% neutral formalin buffer for 2 hr, and then washed 3 times with

distilled water. Nodules were then stained with 2.5% silver nitrate solution for 30

min. Finally, they were washed 3 times with distilled water and were examined

24

Figure 4. The effect of different soluble factors on osteogenic differentiation. The hanging drop method was used to study the effects of different soluble factors on osteogenesis from mouse ES cells. The different combinations of osteogenic factors tested are: A. control with AA and B-Gly addition; B. 0.1 µM Dex Early, added at day 6 and kept throughout differentiation; C. 0.1 µM Dex Late, added at day 10. D. 1 µM PPARγ inhibitor GW9662 is added at day 3 and kept throughout differentiation; 0.1 µM Dex is added at day 10 to further enhance mineralization. E. 0.1 µM RA added from day 3 and kept until day 5; F. 0.1 µM RA added from day 3 and kept until day 5 + Dex Early (added at day 6 until day 21.) G. 1 µM RA added from day 3 until day 5 + Dex Late (added at day 10 until day 21).

26

for the presence of mineralization. For a quantitative measure of the amount of

Ca2+ in the mineralized nodules, the Arsenazo III method ( a Ca2+-sensitive dye)

was used (Zayzafoon et al. 2004). Briefly, the differentiated bone nodules at day

21 were collected and incubated with 3 M HCl at 80°C for 1 to 2 hr. the resulting

suspension was neutralized to pH 5 with 3 M Trizma base, and mixed with

Arsenazo III (100 µM), after 5 min incubation the intensity of resulting colour were

measured at 595 nm. The total Ca2+ amount was calculated by dividing it by the

total protein determined from a duplicate well. The resulting values were

expressed as mmol of Ca2+/µg of total protein.

Alcian blue staining

Differentiated day 21 nodules were rinsed three times with PBS, and fixed

in 10% neutral formalin buffer for at least 15 min or overnight. After rinsing 3

times with distilled water, nodules were left in 2 changes of 0.5 N HCl, 3 min

each. Cells were then stained with 0.25% Alcian Blue overnight. Immediately

following the removal of the stain, cells were rinsed with 0.5 HCl 3 times, 3 min

each. Finally, cells were rinsed with slowly running tap water and then were

immediately observed.

Real-time Polymerase Chain Reaction (PCR)

Total RNA was extracted from day 21 nodules using the Qiagen RNeasy

Mini Kit according to the manufacturer’s instructions. 4 μg of RNA was reverse

transcribed to cDNA using Superscript II (Invitrogen) in a total reaction volume of

48 μL. To examine the mRNA expression of osteogenic markers, cDNA was

amplified using Real-time PCR. Real-time PCR analysis was performed in Bio-

27

Rad’s CFX384 TouchTM detection system. The accumulation of PCR products

was monitored by measuring the fluorescence caused by the binding of SYBR

Green (Applied Biosystems) to double-stranded DNA. The cDNA levels were

normalized using ribosomal protein L32 (a housekeeping gene). The primer

sequences that were used are as follows:

for L32, forward primer 5’-CATGGCTGCCCTTCGGCCTC-3’ and reverse primer

5’-CATTCTCTTCGCTGCGTAGCC-3’;

for Runx2, forward primer 5’-CCTCTGACTTCTGCCTCTGG-3’ and reverse

primer 5’-TAAAGGTGGCTGGGTAGTGC-3’;

for OSX, forward primer 5′-GAAAGGAGGCACAAAGAAG-3′ and reverse primer

5′-CACCAAGGAGTAGGTGTGTT-3′;

for calreticulin, forward primer 5’-AAGTTCTACGGTGACGAGGAG-3’ and

reverse primer 5’-GTCGATGTTCTGCTCATGTTTC-3’;

for Col 2, forward primer 5’-CACACTGGTAAGTGGCGCAAGACCG-3’ and

reverse primer 5’-GGATTGTGTTGTTTCACGGTTCGGG-3’;

for aggrecan, forward primer 5’-GGTGATGATCTGGCATGAGAGA-3’ and

reverse primer 5’-GGTGCCCTTTTTACACGTGAAG-3’;

for BSP, forward primer 5’-ACCCCAAGCACAGACTTTTGA-3’ and reverse

primer 5’-CTTTCTGCATCTCCAGCCTTCT-3’;

and for OCN, forward primer 5’-AAGTCCCACACAGCAGCTG-3’ and reverse

primer 5’-AGCCGAGCTGCCAGAGTTTG-3’.

28

Immunofluorescence Staining

Day 21 differentiated nodules grown on coverslips were rinsed 3 times

with PBS. They were fixed in 3.7% formaldehyde for 15 min. After washing

(three times for 5 min each time) in PBS, the cells were permeabilized with 0.1%

Triton X-100 in buffered containing 100 mM 1,4-piperazinediethanesulfonic acid,

1 mM ethylene glycol tetraacetic acid (EGTA), and 4% (wt/vol) polyethylene

glycol 8000 (pH 6.9) for 2 min. Nodules were then rinsed 3 times in PBS for 5

min and they were incubated with primary antibodies (dilution of 1:50) for 30 min

at room temperature. The primary antibodies used include: anti-Runx2 (Abcam)

and anti-OSX (Abcam). After washing in PBS (3 times for 5 min), the cells were

stained with the secondary antibody (fluorescein isothiocyanate-conjugated

[FITC] donkey anti-mouse, or anti-rabbit; diluted 1:50) for 30 min at room

temperature. After, nodules were washed 3 times in PBS for 5 min. Nodules were

incubated with 100 ug/mL DNase-free RNase in PBS for 30 min at 37ºC, followed

by rinsing in PBS 3 times, 1 min each. For nuclear staining, 500 nM Propidium

Iodide (PI) solution was added to cover the nodules for 5 min. After final rinsing in

PBS (3 times for 1 min each), the slides were mounted in fluorescent mounting

medium (Dako). A confocal fluorescence microscope (MRC 600; Bio-Rad

Laboratories) equipped with a 60/1.40 planapochromatic oil immersion objective

(Nikon) and krypton-argon laser was used for fluorescence imaging. To avoid

bleed-through of fluorescence between channels, we imaged FITC and PI using

separate, single excitation filters and single fluorochrome filter/mirror blocks: 10

nm 488 nm bandpass excitation filter and BHS block with 510 nm longpass

29

dichroic mirror and 515 nm long pass OG filter versus 10 nm 514 nm bandpass

excitation filter and YHS block with 540 nm longpass dichroic mirror and 550 nm

long pass OG filter, which prevented any spillover of fluorescent signal.

Transmission Electron Microscopy

Samples for hydroxyapatite imaging by transmission electron microscopy

(TEM) were fixed in 2% glutaraldehyde in 0.1 M sodium cacodylate buffer

followed by rinsing in the same buffer. They were then fixed in 1% osmium

teroxide in buffer, dehydrated in a graded ethanol series followed by propylene

oxide, and embedded in Quetol-Spurr resin. Sections 100 nm thick were cut on

an RMC MT6000 ultramicrotome, stained with uranyl acetate and lead citrate and

viewed in an FEI Tecnai 20 TEM.

Statistical Analysis

All statistical analyses were performed using two-way analysis of variance

(ANOVA; with post hoc Bonferroni test) or independent t-test (IBM SPSS,

version 21). Experiments were independently repeated at least three times, each

repeated in triplicate unless otherwise stated. Values are expressed as the mean

± SD. Differences between mean values for different treatments were considered

to be significant at P<0.05 and P<0.01.

RESULTS

Aim 1: To enhance the efficacy of mouse ES cell osteogenic differentiation

protocol

The hanging drop protocol which had been previously utilized in our

laboratory for cardiomyogenic and adipogenic differentiation was modified for

30

osteogenic differentiation from mouse ES cells (Papp et al. 2009; Szabo et al.

2009; Szabo et al. 2008). Mouse ES cells were placed on the lids of tissue

culture dish for 3 days in differentiation media. On day 3, the formed EBs were

transferred to bacteriological dishes in which they grew in suspension for 2 days.

On day 5, the EBs were then plated on gelatinized tissue culture dishes. To

enhance osteogenic differentiation, the medium was supplemented with 50

µg/mL ascorbic acid and 10 mM β-glycerophosphate on day 5. The mouse cell

line used initially was CGR8, a widely used ES cell line developed by Austin

Smith (Dirks et al. 1995). However, this protocol did not yield a high output of

osteogenic lineage cells from the CGR8 cells. von Kossa staining of

differentiated day 21 nodules did not show high levels of mineralization, although

PCR analysis revealed that these nodules did express Runx2, OSX, OCN, and

BSP. The experiment was repeated 3 times to confirm the efficacy of the protocol.

Additional experiments confirmed the low mineralization results of the initial trial.

Furthermore, the expression levels of osteogenic markers were inconsistent

across experiments. Thus, I employed a different ES cell line, R1 (one of the

most commonly used mouse ES cell line that has been isolated by Andras Nagy

(Nagy et al. 1993)) to determine whether more consistent results would arise. R1

cells had higher level of mineral deposition and expressed higher levels of

osteogenic markers than CGR8 cells. Experiments using R1 cells had better

reproducibility than CGR8 cells. Thus, the remainder of the current project

utilized the R1 cell line instead. Before I commenced my study on the role of

calreticulin in osteogenesis, I proceeded to explore different culture conditions

31

with the aim to further increase the efficacy of our osteogenic differentiation

protocol. The results are presented below.

The effect of cell number on osteogenic differentiation

The generation of the EB is a critical step in ES cell differentiation. It has

been shown that the cell number forming these EBs and their size has crucial

effects on the downstream product of ES cell differentiation (McBeath et al.

2004; Park et al. 2007). To determine the effects of cell number/density on

osteogenesis, EBs consisting of 250, 500, or 1000 ES cells were allowed to grow

under osteogenic condition using the hanging drop method (Fig. 3). Detection of

osteogenic differentiation in the day 21 nodules was performed by mineralization

assays and Real-time PCR for the expression of osteogenic markers. Both von

Kossa staining and Arsenazo III revealed that nodules resulting from EBs of 250

cells gave the highest level of mineralization compared with the other two cell

numbers (Figs. 5A and B). Furthermore, nodules resulting from EBs of 250 cells

showed the highest expression level of osteogenic markers (Runx2, OSX, and

OCN) compared to those derived from other cell numbers (Fig. 5C). To correct

for loading errors, all mRNA expression values were normalized against the

housekeeping gene L32. For all real-time PCR raw data, please see Appendix.

The effects of Dex on osteogenesis

Dex has been shown to have stimulatory effects on osteogenic

differentiation which include increased ALP activity and mRNA expression

(Ogston et al. 2002). To determine the effect of Dex on mouse ES cell

differentiation 0.1 µM of Dex was added in the media either starting at day 6 or

32

Figure 5. The lowest cell number and/or EB size favoured ostegenic differentiation. A and B. At day 21 of differentiation, nodules derived from EBs containing either 250, 500, or 1000 cells were examined for the level of mineralization using von Kossa staining and the Arsenazo II stain. Both methods revealed that nodules derived from the initial seeding number of 250 cells resulted in the highest level of mineralization. C. Upon analysis using Real-time PCR, nodules differentiated from EBs containing 250 cells expressed the highest level of osteogenic markers (Runx2, OCN, and OSX). Conditions were performed in triplicates, data are presented as means ± SD of three independent experiments (* represents p < 0.05 and ** represents p < 0.01).

34

day 10 (Fig. 4). Dex addition in the media enhanced mineralization (Figs. 6A and

B) and the transcription of the critical osteogenic transcription factor Runx2 was

upregulated (Fig. 6C); the effect of late addition of Dex at day 10 lead to more

profound results compared to early addition at day 6. Early addition of Dex gave

only a 2-fold increase in osteogenic marker OCN mRNA; while there was 4-

fold increase with late Dex addition (Fig. 6C). Similarly, early addition of Dex

enhanced the mRNA expression of Runx2 by 12-fold compared to the control;

while in the cells treated with Dex later in differentiation, there was a 22-fold

increase in Runx2 mRNA expression. Fig. 6C shows that both early and late

addition of Dex increased the expression of OSX compared to their control.

The effects of RA on osteogenesis

The effect of RA treatment during the floating stage on the commitment

and differentiation of ES cells to the osteogenic lineage can be seen by

enhanced mineralization (Figs. 6A and B). Furthermore, the osteogenic markers,

Runx2 and OCN, also had a 33-fold and 5.5-fold increase in their mRNA

expression, respectively (Fig. 6C). The expression of OSX was increased by

22.3-fold compared to the non-treated control (Fig. 6C)

The combined effects of RA and Dex on osteogenesis

Thus far, the treatment with either RA or Dex alone enhanced osteogenic

differentiation from ES cells. When they were both added together into the media,

this further enriched the ES cells commitment and differentiation to the

osteogenic lineage than when the factors were added separately (Fig. 6). The

addition of both RA and Dex (early addition at day 6) lead to a 46.7-fold increase

35

Figure 6. Mineralization assays and PCR analysis revealed that the RA and later addition of Dex treatment when combined, resulted in the highest level of osteogenic differentiation. A. von Kossa staining was used to detect mineralization in day 21 nodules treated with different combinations of osteogenic factors. Nodules were fixed in formaldehyde and then stained with 2.5% silver nitrate solution. The combination of RA and late addition of Dex gives the most mineralization compared to other groups upon von Kossa staining. B. Quantification of extracellular Ca2+ content by Arsenazo III kit confirmed the von Kossa results. C. Real-time PCR analysis revealed that RA and later addition of Dex treatment resulted in the highest level of expression of osteogenic markers. Conditions were performed in triplicates, data are presented as means ± SD of 3 independent experiments. (Abbreviations: E: early, L: late).

37

in Runx2 mRNA expression and a 7.3-fold increase in OCN mRNA expression

compared to control (Fig. 6C). The mRNA expression of OSX was increased

43.3-fold in the treatment regimen of RA and late Dex addition (Fig. 6C). The

combination of RA and Dex (late addition at day 10) gave the most mineralization

and highest increase in osteogenic markers expression compared to all other

protocols.

The effects of PPARγ on osteogenesis

Upon treatment of the PPARγ inhibitor, GW9662 (1 µM), mineralization

was enhanced (Figs. 6A and B). The mRNA expression of Runx2 and OCN

increased as well. There was a 26.9-fold increase in Runx2 mRNA transcript; a

5.6-fold increase in the OCN mRNA expression; and a 5.8-fold increase in OSX

mRNA expression (Fig. 6C). The increase in the osteogenic induction by

GW9662-treated nodules was not as significant compared to the addition of RA

and (late) Dex.

Confirmation of osteogenic differentiation from ES cells

Thus far, we found that the differentiation protocol using 250 cells per EB

with the addition of RA and late Dex resulted in the highest enrichment of

osteogenic lineage cells compared to other treatments examined in this study. To

further confirm the osteogenic lineage differentiation of the cells obtained from

this protocol, we performed immunofluorescence staining and transmission

electron microscopy (TEM) on the differentiated nodules. Immunofluorescence

staining revealed the nuclear localization of the osteogenic transcription factors

Runx2 and OSX (Fig. 7). Negative controls for the primary antibodies were

38

Figure 7. Immunofluorescence staining of the osteogenic markers, Runx2 and OSX. A. At day 21 of differentiation, nodules were fixed in 3.7% formaldehyde, permeabilized, and labelled for the transcription factors Runx2 or OSX. PI was used to identify the nuclei. Both Runx2 and OSX were found to be nuclear in most cells. B. Negative controls for the primary antibodies were accomplished by labelling MDCK cells with anti-Runx2 or anti-OSX antibodies. C. Negative controls for the secondary antibodies were performed by omitting primary antibodies in the staining procedure. Scale bar division represents 10 µm.

40

Figure 8. Transmission electron microscopy images of differentiated day 21 nodules in RA and Dex (late) protocol. The presence of hydroxyapatite deposition is detected in the extracellular matrix of the nodules. Classic modes such as bright field were used to image the texture of the deposit in different nodules. A-D shows representative pictures of the nodules examined. E shows the diffraction pattern of a selected area identifying the crystalline material as biological hydroxyapatite.

42

Figure 9. The most efficacious culture condition for osteogenesis from mouse R1 cells. In the different culture conditions we tested, the regimen of RA and late addition of Dex using an initial seeding number of 250 cells per EB, yielded the highest level of osteogenic induction from mouse R1 ES cells.

44

accomplished by labelling the non-osteogenic cell line, Madin-Darby canine

kidney (MDCK) cells with anti-Runx2 or anti-OSX antibodies. Negative controls

for the secondary antibodies were performed by omitting primary antibodies in

the procedure. As shown in Figure 7, no staining was detected in all negative

controls.

Although von Kossa staining is routinely used to detect the level of

mineralization in osteogenic cultures, this technique is not specific since it

detects the presence of Ca2+ which can be induced due to other biological

processes such as apoptosis. Thus, it is imperative to confirm that the improved

protocol (250 cells per EB, RA and late Dex treatment) does not induce

nonphysiological mineral deposition using more sensitive methods such as TEM

for hydroxyapatite imaging. Figure 8 shows the transmission electron

micrographs illustrating the presence of hydroxyapatite crystals in the

extracellular matrix which has been further confirmed with the crystalline

microdiffraction pattern obtained with 1 nm electron probe (Fig. 8C). Thus, the

high levels of mineral deposition in the improved protocol observed by von Kossa

staining was hydroxyapatite and not any other nonphysiological forms. In

summary, the optimized osteogenic differentiation protocol using initial seeding

number of 250 cells per EB and RA and Dex late treatment (Fig. 9) results in the

highest enrichment of osteogenic lineage cells and no nonphysiological

mineralization.

Aim 2: The role of calreticulin in osteogenic differentiation

Calreticulin expression during mouse ES cell osteogenic differentiation

45

Real-time PCR analysis was performed in order to determine the

expression profile of calreticulin throughout the 21-day osteogenic differentiation.

RNA was extracted from differentiated day 21 R1 and calreticulin KO nodules; it

was then reverse transcribed to cDNA, which was then analyzed using real-time

PCR. Figure 10 shows the calreticulin expression profile in the 21 days of

osteogenic differentiation. In calreticulin-containing R1 cells, the expression of

calreticulin is low in pluripotent cells and it becomes more abundant and peaks at

day 14. The expression remains relatively abundant for the rest of differentiation.

Not surprisingly, in KO cells, there is no calreticulin expression throughout the

differentiation process.

Calreticulin deficiency impairs osteogenic differentiation from mouse ES

cells in vitro

To characterize the role of calreticulin in the osteogenic differentiation of

mouse ES cells, we allowed the EBs from wild type (R1) and calreticulin KO ES

cells to differentiate under osteogenic conditions with the additions of osteogenic

factors B-Gly, AA, RA and Dex. The level of extracellular mineralization was

examined at day 21, and the expression of osteogenic molecular markers, Runx2,

OSX, and BSP were examined throughout differentiation. von Kossa staining for

the presence of mineral deposition showed that the absence of calreticulin

significantly reduced the level of mineralization in KO cells compared to their wild

type counterparts (Fig. 11A). The Arsenazo III method was used to measure Ca2+

deposition in the matrix. It showed that wild type cells had 7-fold increase in the

concentration of extracellular Ca2+ compared to that present in

46

Figure 10. The expression of calreticulin throughout osteogenic differentiation. Real-time PCR analysis revealed that calreticulin expression is low in pluripotent (plu) R1 stem cells. As differentiation commences, its mRNA expression steadily increases and peaks at day 14. Its expression remains abundant later in differentiation. In the calreticulin KO cells, no calreticulin mRNA expression is detected. Conditions were performed in triplicates, data are presented as means ± SD (* represents p < 0.05 vs day 14 value).

48

the KO nodules (Fig. 11B). Real-time PCR analysis revealed that the mRNA

expression of the osteogenic markers were higher in wild type cells compared to

KO cells (Fig. 11C). This suggests that osteogenic differentiation from mouse ES

cells is impaired in the absence of calreticulin.

Chondrogenic potential of calreticulin KO cells

Thus far, we found that calreticulin deficiency leads to a reduction in osteogenic

differentiation. Then, the question arises as to whether these calreticulin KO cells

have an increase potential to differentiate towards other lineages that are also

derived from mesenchymal origin such as chondrocytes. To address this

question, the expression of chondrogenic markers, Aggrecan and collagen 2 (Col

2) were examined in these cells. Col 2 is the principle component of cartilage;

while aggrecan is the predominant proteoglycan of cartilage (Fraser et al. 2003).

Figure 12 shows that differentiated day 21 nodules from calreticulin KO cells

expressed higher levels of Col 2 and Aggrecan compared to their calreticulin wild

type counterpart. Moreover, Alcian blue staining revealed that calreticulin KO

cells exhibit a stronger blue staining for cartilage extracellular matrix than R1

cells. Thus, this data suggest that the absence of calreticulin leads to higher

chondrogenic potential in ES cells.

Nuclear localization of Runx2 and OSX is impaired in the absence of

calreticulin.

Since Runx2 and OSX are critical transcription factors required for the regulation

of osteogenic differentiation and early ES cell commitment to the osteoblastic

lineage, their nuclear localization is crucial for their functions. Thus,

49

Figure 11. Detection of osteogenic differentiation in calreticulin-containing and deficient cells. Calreticulin-expressing mouse ES cells (R1) and calreticulin KO cells were allowed to differentiate for 3 weeks in the presence of RA, B-Gly, AA, and Dex. A. The resulting nodules at day 21 were then fixed in formaldehyde and mineralized (dark) nodules were detected by von Kossa staining. B. Arsenazo III results showed a 7-fold decrease in the extracellular Ca2+ content in the KO cells compared to WT cells (* represents p < 0.05). C. The relative mRNA expression of osteogenic markers (Runx2, OSX, BSP) was detected by Real-time PCR throughout the process of differentiation. All of the three markers were found to have lower expression levels in KO cells compared to wild type cells (** p < 0.01 for all R1 values vs KO values). Conditions were performed in triplicates, data are presented as means ± SD of three independent experiments.

51

Figure 12. Increased chondrogenic potential in calreticulin-deficient cells. A. Differentiated day 21 wild type and KO cells were stained with Alcian blue. KO cells had more intense blue staining compared to that in WT cells. B. Real-time PCR analysis revealed that KO cells also expressed higher levels of chondrogenic markers (Col 2 and aggrecan) than WT cells. Conditions were performed in triplicates, data are presented as means ± SD (* represents p < 0.05).

53

we performed immunofluorescence staining for Runx2 and OSX in the

differentiating R1 and calreticulin-null cells at days 7, 14, and 21. Interestingly,

most of Runx2 and OSX are found to be cytosolic in calreticulin KO cells,

whereas they were nuclear in wild type cells (Fig. 13). This suggests a prominent

role of calreticulin in the nuclear localization of these transcription factors in

normal osteogenic differentiation.

Calreticulin affects osteogenic differentiation via its Ca2+ buffering activity

Calreticulin is important for its role in intracellular Ca2+ homeostasis through its

Ca2+ buffering function. Studies have shown that the absence of calreticulin leads

to a reduced Ca2+ store in the ER and thus a smaller releasable pool of Ca2+

from the ER (Arnaudeau et al. 2002; Mery et al. 1996). To examine whether

calreticulin exerts its effect in osteogenesis through this function, the intracellular

Ca2+ level was manipulated using either Ionomycin (a Ca2+ ionophore) to

increase the cytosolic Ca2+ level in the calreticulin-deficient cells; or BAPTA-AM

(Ca2+ chelator) to decrease it in the wild type cells. In the case of the KO cells,

the increase in the cytosolic Ca2+ enhanced the level of extracellular

mineralization and increased the mRNA expression of osteogenic markers

(Runx2, OSX and BSP) (Figs. 14A and B). Furthermore, the increase in Ca2+

level restored the nuclear localization of Runx2 and OSX in calreticulin KO cells

(Fig. 14C). In the case of the wild type cells treated with the Ca2+ chelator

BAPTA-AM, the level of extracellular mineralization and the expression of

osteogenic markers were lowered compared to the non-treated controls (Figs.

14A and B). Lower cytosolic Ca2+ disrupted the nuclear localization of Runx2 and

54

Figure 13. Immunofluorescence staining of the osteogenic markers, Runx2 and OSX in calreticulin wild type and KO cells. At days 7, 14 and 21 of differentiation, nodules were fixed in 3.7% formaldehyde, permeabilized, and labelled for the transcription factors Runx2 (A) or OSX (B). PI was used to identify the nuclei. Both Runx2 and OSX were found to be mainly localized in the cytosol in the KO cells, while in the wild type cells Runx2 and OSX were nuclear. Scale bar division represents 10 µm.

56

Figure 14. The effect of changes in cytosolic Ca2+ levels on osteogenic differentiation in calreticulin- containing and deficient cells. BAPTA-AM (Ca2+ chelator) or ionomycin (Ca2+ ionophore) was used to either lower or increase the cytosolic Ca2+ concentration. A. von Kossa staining for mineralized nodules showed an increase in mineralization level in KO cells treated with ionomycin; while decreasing the cytosolic Ca2+ concentration in R1 cells by treatment with BAPTA-AM decreased the level of mineralization compared with their non-treated controls. B. Arsenazo III staining revealed that BAPTA-AM treatment reduced the extracellular Ca2+ contents by more than 2-fold. Calreticulin KO cells treated with ionomycin had more than 4-fold increase in their extracellular Ca2+ contents compared to their non-treated controls. C. The relative mRNA levels of osteogenic markers were increased with increase in cytosolic Ca2+ in KO ionomycin-treated cells, while their expression was decreased in the R1 BAPTA-treated cells. D. Immunofluorescence staining for Runx2 and OSX localization after cytosolic Ca2+ concentration manipulation showed restoration of the nuclear localization of both transcription factors in KO ionomycin-treated cells, while in R1 BAPTA-treated cells, their nuclear translocation was largely impaired. Scale bar division represents 10 µm. Conditions were performed in triplicates, data are presented as means ± SD of three independent experiments (* represents p < 0.05, ** p< 0.01).

58

OSX; in BAPTA-AM-treated wild type cells, the localization of these osteogenic

transcription factors became cytosolic (Fig. 14C). These data suggest that

calreticulin exerts its effect on osteogenic differentiation from mouse ES cells via

its Ca2+ buffering function.

The involvement of calreticulin’s functional modules in osteogenesis

Calreticulin has three structural domains (N, P and C) that comprise two

functional modules, one responsible for its chaperoning activity ([N+P]) and the

other ([P+C]) that carries out the Ca2+ buffering function of the protein (Michalak

et al. 1999; Michalak et al. 2009; Nakamura et al. 2001). In order to elucidate

which structural domain may be involved in calreticulin’s modulation of

osteogenesis, we employed two cell lines: calreticulin KO ES cells expressing

either its chaperoning [N+P] functional module or its Ca2+ buffering [P+C]

functional module). These two cell lines were previously utilized in our laboratory

to study the involvement of calreticulin’s functional modules in adipogenesis

(Szabo et al. 2008). However, experiments using these two cell lines in the

present study did not yield consistent results. It appears that both cell lines are

needed to be re-established/recloned; hence, this work has been deemed

beyond the scope of this thesis.

DISCUSSION

Different cell lines have different osteogenic potential

The hanging drop protocol (outlined in Figs. 3 and 4) was previously used

in our laboratory for ES cell differentiation into the cardiomyogenic and

59

adipogenic lineage (Papp et al. 2009; Szabo et al. 2009; Szabo et al. 2008).

The first cell line we utilized in our osteogenic protocol was mouse CGR8 cells,

one of the most common stem cell lines utilized when feederless conditions are

desired. Nevertheless, with only the addition of pro-osteogenic molecules, B-Gly

and AA, the hanging drop differentiation protocol had low efficiency in inducing

the desired osteogenic cell population from mouse ES CGR8 cells. Although

differentiated cells did express osteogenic markers such as Runx2, OSX, OCN,

and BSP, functional assay by using von Kossa staining revealed that the nodules

had little to no mineral deposition, which is a characteristic of functional

osteoblasts. The first step in achieving higher efficacy of the differentiation

protocol was to utilize another mouse ES cell line, R1, that is also one of the

most extensively used stem cell lines isolated by Andras Nagy (Nagy et al.

1993). Our rationale in choosing another cell line is because different cell lines

may have different propensity to differentiate to different mature cell types

(Cohen and Melton, 2011). A study by Osafune et al. compared the

differentiation potential of 17 human ES cell lines and found that the propensity to

differentiate into different cell lines are so markedly different that there was often

greater than 100-fold difference in the lineage-specific gene expression in the

final differentiated cells (Osafune et al. 2008). This is because more evidence is

pointing to the inherent differences between different cell lines (Allegrucci and

Young, 2007), for instance, there have been reports on the different genetic and

epigenetic stability and growth rate observed in long-term culture (Cowan et al.

2004; Hoffman and Carpenter, 2005). These variations could be due to a variety

60

of reasons, including culture-related effects, inherent genetic variation,

differences in genetic and epigenetic stability, and methodological variations

(Allegrucci and Young, 2007). Indeed, in our study, differentiated R1 nodules had

higher levels of mineral deposition and expressed higher levels of osteogenic

markers compared to those obtained with CGR8 cell line. In addition,

experiments using R1 cells also achieved better reproducibility and more

consistent results. The difference between CGR8 and R1 could be due to any of

the inherent differences mentioned above, thus resulting in differing potentials for

osteogenic differentiation.

The effect of cell density on ES cell osteogenic differentiation

To further enrich osteogenic cell populations in the differentiated nodules,

I proceeded to manipulate additional aspects of culture conditions. Previous

studies have suggested that cell density affects stem cell commitment and

differentiation to different lineages (Choi et al. 2010; Hwang et al. 2009; Lorincz,

2006; McBeath et al. 2004; Park et al. 2007). Our study found that lower cell

density and smaller EB size enhanced osteogenic differentiation, as evident in

the higher expression of osteogenic markers and higher mineral deposition in

these nodules. This finding is in accordance with a 2004 study conducted by

McBeath et al., in which they discovered that lower cell density allows more cell

adhesion and spreading, which leads to a more flattened cell shape and higher

RhoA activity, induces osteogenic differentiation (McBeath et al. 2004). One

unexpected observation in the present study is that EBs consisting of 1000 cells

lead to higher osteogenic differentiation than those of 500 cells (Fig. 5).

61

According to the findings of McBeath et al. (2004), cell density should inversely

correlate with the propensity to differentiate into osteogenic cells. One possible

explanation for this phenomenon is that the higher levels of hypoxia in the 1000

cell-EBs could be inducing ES cell to differentiate into the osteogenic lineage. It

has been reported that hypoxia, by activating the hypoxia-inducible factor 1-alpha

(HIF-1α) pathway, has stimulatory effect on osteoblast development (Wang et al.

2007). However, this is only a speculation since there have also been reports of

opposite, inhibitory effects of hypoxia on osteogenesis (D'Ippolito et al. 2006;

Park et al. 2002; Salim et al. 2004). Thus, although it is known that oxygen

tension is a potent regulator of skeletal mass, whether hypoxia is anabolic or

catabolic to the skeletal system remains inconclusive.

The pro-osteogenic effects of B-Gly and AA

While protocols for the differentiation of cardiomyocytes, neurons, or

adipocytes from ES cells have been well established for many years (Okabe et al.

1996; Wobus et al. 1991), only recently has their differentiation into osteogenic

lineage cells been reported (Phillips et al. 2001; Zur Nieden et al. 2003).

Osteoblasts differentiate from a common mesenchymal progenitor that can also

give rise to adipocytes, chondrocytes and cardiomyocytes. Evidence exists for

the presence of cells belonging to non-osteogenic lineages in a variety of

cultures (Zur Nieden et al. 2003). Therefore several studies, including the

present one, attempted to enhance osteogenesis in vitro by the use of different

combinations of various factors in order to increase the efficacy of ES cell

62

commitment to the osteoblastic lineage (Table 1). Among these pro-osteoblastic

factors, B-Gly figures prominently as it is required as a source of organic

phosphate for proper matrix mineralization. B-Gly induces differentiation and

decreases proliferation in the osteoblastic cell population (Coelho and Fernandes,

2000). In spite of several studies, which questioned whether or not mineralization

induced by B-Gly is an exclusive characteristic of osteoblasts1, B-Gly is routinely

added to the differentiation media at the concentration of 10 mM with AA

(Bellows et al. 1991). Prior to differentiation, osteoblasts secrete a collagenous

matrix. AA was found to be essential for osteoblast differentiation in vitro and

collagen synthesis (Coelho and Fernandes, 2000). Hydroxyapatite-containing

mineral is then deposited on this collagenous matrix if a source of organic

phosphate such as B-Gly is provided. Furthermore, the addition of AA and B-Gly

upregulated several osteogenic markers such as ALP and OCN mRNA

expression (Franceschi and Iyer, 1992). The transcription of OCN was triggered

by the addition of AA, which was found to have a major role in the stabilization of

Runx2 binding to the OCN gene promoter (Xiao et al. 1997). Due to their critical

roles in osteogenic differentiation from ES cells, B-Gly and AA were both

included in all the tested osteogenic protocols.

1 Chung et al. (1992) have shown that the addition of B-Gly induces mineralization that is nonapatitic (Chung et al. 1992). Other studies have also questioned whether exogenous addition of B-Gly induces physiological mineralization (Beresford et al. 1993; Khouja et al. 1990). However, as illustrated in Results, we have confirmed using TEM that mineralization in our culture is apatitic, which is characteristic of those found in bone.

63

The role of Dex in skeletogenesis

We examined the effect of several factors reported to enhance osteogenic

differentiation: RA, Dex, and PPARγ inhibitor GW9662. We studied their effects

alone, as well as in combination. We also looked at the effect of differential timing

of addition of Dex. Dex is a glucocorticoid known for its anti-inflammatory

properties and thus it is often included in the treatment regimens of inflammatory

diseases (Lian et al. 1997). Although some studies have reported its stimulatory

effects in osteogenic differentiation (Leboy et al. 1991; Ogston et al. 2002),

others have also pointed to its negative effects in osteogenesis (Smith and

Frenkel, 2005; Pereira et al. 2002; Shi et al. 2000). Critical factors that may

influence such effects of Dex is the maturity and density of the differentiating

cells in culture (Mikami et al. 2011). It has been shown that Dex can enhance

adipogenesis from mesenchymal stem cells by inducing the expression of the

critical adipogenic transcriptional factor, PPARγ and repressing the expression of

the osteogenic transcription factor, Runx2 (Pereira et al. 2002; Shi et al. 2000).

On the other hand, Dex can have stimulatory effects on osteogenic differentiation

by increasing the expression as well as activity of the enzyme ALP (Ogston et al.

2002). It is also interesting to note that Dex is required to be added in the media

continuously to induce maximal osteogenesis; removal of Dex from the culture

may cause osteogenic lineage cells to dedifferentiate (Porter et al. 2003). In our

study, Dex did indeed enhance osteogenic differentiation as demonstrated by

higher levels of mineralization and expression of osteogenic markers. However,

in our hands, late addition led to better results than early addition of Dex. This is

64

because early addition of Dex was found to promote more ES cells to commit to

the adipogenic lineage instead of osteogenic lineage, which is achieved through

differential temporal regulation of OSX by Dex. This has been demonstrated by

Mikami et al. (2011) ;. In their study, early Dex treatment was found to promote

adipogenesis, while late addition of Dex promoted osteogenesis in the same

cultures (Mikami et al. 2011).

RA and osteogenesis

During ES cell differentiation, RA has also been implicated as an

important factor to be added to the cultures shortly after EB formation but before

further differentiation; it has been shown to induce cells to undergo adipogenesis

and neurectoderm differentiation (Duester, 2008). RA signalling is vital for

organogenesis in utero and the maintenance of adult organs. (Duester, 2008;

Niederreither and Dolle, 2008). RA is a steroid hormone and it exerts its effects

on cell differentiation and metabolism by binding to its nuclear receptors, the

retinoic X and the RA receptors (Chawla et al. 2001; Mark et al. 2006). It has

been recently demonstrated that RA signalling plays an important role in

regulating myogenic and hepatic progenitor cell proliferation (Huang et al. 2009;

Zhu et al. 2009). It remains controversial regarding the effects of RA on

chondrogenesis and osteogenesis (Wan et al. 2006; Wang et al. 2008; Weston

et al. 2000). RA has been shown to positively regulate chondrogenesis by

increasing the activity of bone morphogenetic protein 2 (BMP2) and inducing the

expression of collagen X (Col X) (Zhang et al. 2010). It has also been suggested

65

that RA promotes osteogenesis by stimulating the expression of Runx2 (Dingwall

et al. 2011). This is accomplished by RA interacting with Smad1 and thus

enhancing BMP2 signalling (Li et al. 2003). In addition, RA is important in the

development of hypertrophic chondrocytes (Drissi et al. 2003). In our study, the

addition of RA during the floating stage of differentiation did in fact enhance

mineralization as well as the expression of Runx2 and OCN. RA is important for

early osteogenic lineage commitment by upregulating Runx2 expression and

inhibiting PPARγ (Shao and Lazar, 1997) and Dex enhances later osteoblast

differentiation and maturation through its regulation of ALP and OSX activity and

expression. Thus, the combined treatment of RA and Dex leads to higher

enrichment of osteogenic lineage cells than either addition alone.

PPARγ: a positive regulator of adipogenesis and negative regulator of

osteogenesis

Several studies have indicated the inhibitory role of PPARγ in osteogenic

differentiation (Jeon et al. 2003; Kawai and Rosen, 2010; Lecka-Czernik et al.

2002; Takada et al. 2007), reviewed in (Lecka-Czernik and Suva, 2006). PPARγ

is a critical transcription factor that positively regulates adipocyte differentiation

and negatively regulates osteoblast formation. In vivo, the decreased number of

osteoblasts and rate of bone formation that accompanies aging inversely

correlates with an increase in number of adipocytes and fat in the bone marrow

(Lecka-Czernik and Suva, 2006). This inverse relationship between osteoblast

and adipocyte differentiations and their common mesenchymal origin makes

PPARγ an obvious candidate to inhibit in order to enhance osteogenic

66

differentiation from human mesenchymal stem cells (Krause et al. 2010).

Inhibitions of this transcription factor have been done using either the potent

inhibitor GW9662 or siRNA-mediated inhibition in ES cells (Yamashita et al.

2006). In our study, we also observed an enhancement in osteogenic

differentiation from ES cells with the addition of 1 µM of GW9662; although this

was not as prominent as the osteogenic enhancement achieved with the addition

of RA and (late) Dex. Furthermore, PPARγ is a transcription factor that affects

the transcription of several other genes in addition to adipogenic markers. Due to

this pleiotropic effect, its inhibition may have other nonspecific downstream

effects.

Of all of the 6 tested osteogenic protocols (Fig. 4), consisting of

osteogenic factors alone and in combination, with varying timing of addition, RA

and late addition of Dex yielded the most pronounced enhancement of

osteogenic differentiation based on the results of mineralization assays and the

expression of osteogenic markers. Thus, our findings provide valuable guidelines

into fine-tuning of culture condition to differentiate mouse R1 ES cells to

osteoblastic cells in vitro.

Roles of calreticulin throughout differentiation

The second goal of the present work was to determine whether the ER

luminal protein, calreticulin, has a role in osteogenic differentiation from mouse

ES cells in vitro. There is limited data on the expression and function of

calreticulin in osteogenic differentiation. Previous work from our laboratory using

transgenic mice containing a Green Fluorescent Protein (GFP) reporter gene

67

under the control of the calreticulin promoter found that the calreticulin gene is

activated in the developing bone (unpublished data, Opas lab). These data

suggest an unknown and possibly important role of calreticulin in skeletal

development. In the current study, the expression profile of calreticulin in the

current cell model gives insight into its functions in osteogenic differentiation. The

upregulation of calreticulin from day 0 to day 14 suggests that it has a role in

early steps of differentiation such as ES cell commitment to the osteogenic

lineage. Indeed, it was found that the Ca2+ buffering function of calreticulin is

needed for the proper nuclear translocation for critical osteogenic transcription

factors Runx2 and OSX. The nuclear translocation is observed as early as day 7

in our system and it is well known in the literature (Duplomb et al. 2007a; Long,

2012) that the activation of these two transcription factors are required early in

differentiation. The abundance of calreticulin expression near day 21 suggests its

further role in later differentiation such as mineralization. As demonstrated by

Somogyi et al. (2002), calreticulin was found to be present in the extracellular

matrix of developing odontoblasts in rats (Somogyi et al. 2003). It is thought that

calreticulin’s Ca2+ buffering function may be important for the process of

mineralization. Others have also found that calreticulin is upregulated during the

mineral deposition phase for ameloblasts, during which these cells require a

large amount of Ca2+ (Hubbard, 1996).

Osteogenic marker expression

The expression patterns of osteogenic markers Runx2, OSX, and BSP in

both calreticulin wild type and KO cells are characteristic of those observed in

68

vitro (Chou et al. 2005; Duplomb et al. 2007a), albeit calreticulin KO cells

displayed significantly lower osteogenic potential. The expression patterns of

these osteogenic markers can be explained by their functions in osteogenesis.

The osteo-chondroprogenitor cell becomes committed to the osteoblastic cell

lineage under the influence of Runx2 and OSX (Karsenty, 2008; Komori, 2008;

Long, 2012). Runx2 is a master regulatory gene required to control the

expression or activity of several key osteogenic transcription factor (OSX among

others) or proteins such as BSP, throughout the process of ES cell osteogenic

differentiation (Ducy et al. 1997). OSX is also required during differentiation to

activate a repertoire of genes such as Col 1, OCN, osteonectin, osteopontin and

BSP (Long, 2012). Since Runx2 and OSX are transcription factors that are

needed for the ES cell commitment to the osteogenic lineage as well as

activation of the transcription of important extracellular matrix proteins; thus, they

are upregulated earlier in the differentiation process. While BSP is a component

of the mineralized tissue, it is expressed later in the differentiation sequence.

Increased chondrogenic potential in calreticulin deficient cells

Since the absence of calreticulin leads to decreased osteogenic potential

in ES cells, what cell lineage(s) are these cells differentiating into alternatively?

Chondrogenic lineage cells are obvious candidates to examine since they are

derived from common progenitor cells with osteoblasts. Alcian blue staining and

PCR analysis revealed that the absence of calreticulin leads to the higher

69

populations of chondrocytes. It is widely known that Runx2 has an important role

for chondrocyte maturation (Komori, 2003), a process that is essential not only in

skeletal development but also in bone repair. In the absence of proper

chondrocyte maturation, endochondral ossification cannot properly proceed

(Mackie et al. 2011). This is because matured chondrocytes secrete factors that

are needed for vascular invasion and osteoblast development. In the absence of

calreticulin, the nuclear translocation of Runx2 is impaired, thus, it cannot

properly carry out its functions in chondrocyte maturation. This could explain why

osteogenic differentiation is arrested, as demonstrated by the lower expression of

osteogenic markers, and the higher presence of chondrocytes in calreticulin KO

cells.

Impaired nuclear translocation of Runx2 and OSX in KO cells: implications

Normal osteogenic differentiation of mouse ES cells involves a number of

signalling molecules: Runx2, Wnt, insulin/phosphatidyinositol 3-kinase/Akt, bone

morphogenic proteins/Smads, OSX, and many others (Karsenty, 2008). Among

them, Runx2 and OSX are indispensable for osteoblastic differentiation; mice

with Runx2 or OSX deficiency have a complete lack of bone formation (Komori et

al. 1997). Runx2 is a major transcription factor needed for the various stages of

osteogenesis such as lineage commitment, proliferation and differentiation

(Stewart et al. 1997; Vaillant et al. 2002). Osx is also required for osteoblast

70

differentiation and bone formation during embryonic development and in

postnatal bone growth and homeostasis (Zhou et al. 2010). Thus, the nuclear

localization of these two transcription factors is indispensable for their function;

and conversely, their cytosolic retention, in the absence of calreticulin, may have

prevented their transcriptional activity.

Other studies have also observed cytosolic localizations of these two

transcription factors and given insight into how this may impact osteoblast

differentiation and maturation. STAT1, a member of the Signal Transducers and

Activators of Transcription family, was found to be a negative regulator of

osteoblastogenesis. Differentiation of osteoblasts and subsequent bone

formation can be prevented by cytosolic retention of Runx2 by STAT1 (Long,

2012). Huang et al. have described that frameshift mutations, which caused at

least partially impaired nuclear translocation of Runx2, lead to the phenotype of

cleidocranial dysplasia in patients (Huang et al. 2013). Another study by Deepak

et al. showed that Tbox3, a previously reported negative regulator of

osteoblastogenesis, interferes with Runx2 transcriptional activity. In addition,

overexpression of Tbox3 leads to the mis-localization of Runx2 in the cytosol

(Deepak et al. 2011). Although it is less known about what may affect the

subcellular localization of OSX, Tai et al. showed that OSX nuclear localization

and activation is required for early osteoblast development (mesenchymal stem

cell and pre-osteoblast stages) and remains nuclear in differentiated osteoblasts;

while in non-osteogenic cells such as fibroblasts, OSX remains inactive in the

cytosol (Tai et al. 2005).

71

Calreticulin and the regulation of transcription factors

The finding that calreticulin is required for the proper subcellular location

of critical osteogenic factors, Runx2 and OSX is not surprising. This is because

calreticulin has previously been shown to be involved in the regulation of

subcellular localization of critical transcription factors in other cell lineages. For

instance, our laboratory has shown that calreticulin impairs the nuclear

translocation of PPARγ, a critical transcription factor in adipogenesis (Szabo et al.

2008). In vitro, calreticulin was found to bind to several transcription factors that

contain the KxFFKR amino acid sequence, including PPARγ and the vitamin D

receptor (Gurland and Gundersen, 1995; Wheeler et al. 1995). In

cardiomyogenesis, calreticulin is needed for the proper nuclear translocation of

MEF2C and NFAT. Calreticulin affects calcineurin activity through the control of

ER Ca2+ release (Lynch and Michalak, 2003). Thus, the nuclear import of NFAT

is impaired in calreticulin null cells and this can be restored by re-expression of

calreticulin (Mesaeli et al. 1999). Another critical transcription factor for

cardiomyogenesis, MEF2C (Lin et al. 1997), does not translocate to the nucleus

in the absence of calreticulin either. This is also through the effect of ER Ca2+

release and calcineurin activity (Li et al. 2002; Lynch et al. 2005).

FUTURE DIRECTIONS

The involvement of calcineurin-NFAT pathway downstream of calreticulin

in its regulation of osteogenesis

72

Thus far, we discovered a novel role of calreticulin in the osteogenic

differentiation from mouse ES cells via its Ca2+ buffering function. The nuclear

translocation of Runx2 and OSX in the differentiating nodules was affected by

manipulation of intracellular Ca2+ concentrations. Calreticulin affects several Ca2+

signaling pathways; one in particular, the calcineurin-NFAT pathway, has been

shown to be critical in osteogenesis. The expression of constitutively nuclear

NFATc1 variant in osteoblasts causes the development of high bone mass,

massive osteoblast overgrowth, and enhanced osteoblast proliferation (Winslow

et al. 2006). Sun et al. (2005) discovered that calcineurin deficient mice show

diminished bone formation and severe osteoporosis; while calcineurin inhibition

reduces osteoblast differentiation (Sun et al. 2005). Previous findings also

showed that calreticulin is an upstream regulator of calcineurin activity (Lynch

and Michalak, 2003). To further investigate whether calcineurin is downstream of

calreticulin in its regulatory role in osteogenesis, we utilized the non-osteogenic

calreticulin-null cells stably transfected with constitutively active calcineurin (cell

line denoted “CN”). When differentiated under osteogenic conditions, preliminary

data showed that CN cells displayed osteogenic potential comparable to that of

calreticulin-containing WT cells (Fig. 15).

Calcineurin, a serine-threonine phosphatase activated by Ca2+, while not

considered to have transcriptional activity per se (Hogan et al. 2003; Hyatt et al.

1990), dephosphorylates NFAT proteins (Crabtree, 1999). Dephosphorylated

NFATs, with the exposed nuclear localization sequences, then rapidly enter the

nucleus (Beals et al. 1997) and activate transcription of their target genes. To

73

firmly establish the involvement of the calcineurin-NFAT pathway in calreticulin’s

effects in osteogenesis, we inhibited the nuclear translocation of NFATc using

the chemical A-285222. A-285222 specifically inhibits the Ca2+-induced

dephosphorylation of NFAT by calcineurin and its nuclear translocation without

inhibiting calcineurin activity (Djuric et al. 2000; Trevillyan et al. 2001). Inhibition

of either calcineurin or NFAT in both R1 and CN cells lines decreased their ability

to undergo osteogenesis as measured by a decrease in mineralization and

downregulation of osteogenic markers (Fig. 15). Further work would be required

to decipher how calreticulin, via affecting the calcineurin-NFAT pathway,

regulates osteogenesis. It is very plausible that this signalling cascade is

responsible for the decreased expression of osteogenic markers, Runx2, OSX,

and BSP in the calreticulin-null cells. Furthermore, an interesting report by Li et al

(2011) demonstrated that calcineurin-NFAT signalling affects c-Src expression,

which would have a negative effect on cell adhesive properties that are important

in osteogenesis.

STAT1: cytosolic attenuator of Runx2

Since the absence of calreticulin impairs the nuclear translocation of

Runx2, we further investigated how calreticulin, via its Ca2+ buffering activity,

affects this phenomenon. STAT1 is a signalling molecule that, when

phosphorylated at Y701, dimerizes and translocates to the nucleus. However,

74

Figure 15. The involvement of calcineurin-NFAT pathway downstream of calreticulin in osteogenesis. A. Non-osteogenic calreticulin-null cells stably transfected with constitutively active calcineurin (cell line denoted “CN”) showed increased mineral deposition, comparable to that in wild type cells. Arsenazo III assay revealed that CN cells had increased extracellular Ca2+ content. B. In addition to increased mineralization, CN cells also had upregulated expression of osteogenic markers (Runx2, OSX, BSP). A and B. Wild type and CN cells treated with either calcineurin inhibitor (cyclosporin, or CsA) or NFAT inhibitor (A-285222) had lowered mineralization and expression of osteogenic markers, further confirming the involvement of calcineurin and NFAT downstream of calreticulin. Conditions were performed in triplicates, data are presented as means ± SD of three independent experiments (* represents p < 0.05, ** p< 0.01).

76

when the Y701 residue of STAT1 is not phosphorylated, the protein resides in

the cytosol (Darnell, Jr. et al. 1994) where it binds Runx2, and the complex is

retained in the cytosol (Kim et al. 2003). Differentiation of osteoblasts and

subsequent bone formation can be prevented by cytosolic retention of Runx2 by

STAT1 (Kim et al. 2003; Long, 2012; Xiao et al. 2004). Thus, it is possible that

calreticulin affects localization of STAT1. Immunostaining revealed that the

majority of STAT1 was localized to the nuclei in calreticulin-containing R1 cells,

while it was predominantly cytosolic in non-osteogenic calreticulin KO cells (Fig.

16). Moreover, changes in intracellular [Ca2+] altered STAT1 localization: in R1

cells, after depleting cytosolic Ca2+ with BAPTA-AM, STAT1 resided in the

cytoplasm. In contrast, after ionomycin treatment to increase intracellular [Ca2+]

in KO cells, STAT1 nuclear translocation was restored (Fig. 16).

Future work could investigate how the intracellular Ca2+, modulated by

calreticulin, has a role in STAT1 localization. It is highly plausible that calcineurin

may also have a role in modulation of STAT1 subcellular localization, possibly by

affecting its phosphorylation status (Y701) indirectly.

Calreticulin and OSX nuclear localization

In contrast to Runx2, less is known about the regulation of subcellular localization

of OSX. The present study uncovered a novel regulator, calreticulin, in OSX

nuclear translocation. Calreticulin has previously been shown to have a role in

nuclear trafficking of various proteins (Coppari et al. 2002; Holaska et al. 2002).

A study by Holaska et al. (2002) illustrated the ability of calreticulin to directly

bind to the DNA binding domain of the glucocorticoid receptor and

77

Figure 16. Calreticulin regulates the subcellular localization of STAT1. STAT1, a cytosolic attenuator of Runx2 when unphosphorylated at Y701, was found to be affected by the presence or absence of calreticulin. Immunofluorescence staining on day 21 differentiated nodules revealed that in the presence of calreticulin, majority of STAT1 is localized in the nucleus. Calreticulin deficient cells was found to have the majority of STAT1 in the cytosol instead. This observation was Ca2+-dependent: increasing intracellular Ca2+ by ionomycin in calreticulin KO cells restored the nuclear localization of STAT1; while decreasing Ca2+ concentration by BAPTA-AM led to cytosolic retention of STAT1 in calreticulin containing R1 cells. Scale bar division represents 10 µm.

79

facilitate its nuclear export. This event was found to be Ca2+ dependent.

Furthermore, it was uncovered that the nuclear import of protein disulfide

isomerase, ERp57, may be facilitated by calreticulin and STAT3 (Coppari et al.

2002). Thus, it is highly plausible that calreticulin affects OSX subcellular

localization by directly binding to it and facilitating its nuclear import. This

phenomenon could be addressed in future studies by performing

immunoprecipitation with OSX to investigate whether calreticulin physically

associates with it.

80

References

Akiyama, H, Chaboissier, MC, Martin, JF, Schedl, A, and de Crombrugghe, B. 2002. The transcription factor Sox9 has essential roles in successive steps of the chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6. Genes Dev. 16: 2813-2828.

Allegrucci, C and Young, LE. 2007. Differences between human embryonic stem cell lines. Hum. Reprod. Update. 13: 103-120.

Arnaudeau, S, Frieden, M, Nakamura, K, Castelbou, C, Michalak, M, and Demaurex, N. 2002. Calreticulin differentially modulates calcium uptake and release in the endoplasmic reticulum and mitochondria. J. Biol. Chem. 277: 46696-46705.

Aubin, JE. 2001. Regulation of osteoblast formation and function. Rev. Endocr. Metab Disord. 2: 81-94.

Baksh, S, Burns, K, Andrin, C, and Michalak, M. 1995. Interaction of calreticulin with protein disulfide isomerase. J. Biol. Chem. 270: 31338-31344.

Baksh, S and Michalak, M. 1991. Expression of calreticulin in Escherichia coli and identification of its Ca2+ binding domains. J. Biol. Chem. 266: 21458-21465.

Beals, CR, Clipstone, NA, Ho, SN, and Crabtree, GR. 1997. Nuclear localization of NF-ATc by a calcineurin-dependent, cyclosporin-sensitive intramolecular interaction. Genes Dev. 11: 824-834.

Bellido, T, Ali, AA, Gubrij, I, Plotkin, LI, Fu, Q, O'Brien, CA, Manolagas, SC, and Jilka, RL. 2005. Chronic elevation of parathyroid hormone in mice reduces expression of sclerostin by osteocytes: a novel mechanism for hormonal control of osteoblastogenesis. Endocrinology 146: 4577-4583.

Bellows, CG, Aubin, JE, and Heersche, JN. 1991. Initiation and progression of mineralization of bone nodules formed in vitro: the role of alkaline phosphatase and organic phosphate. Bone Miner. 14: 27-40.

Beresford, JN, Graves, SE, and Smoothy, CA. 1993. Formation of mineralized nodules by bone derived cells in vitro: a model of bone formation? Am. J. Med. Genet. 45: 163-178.

Boheler, KR, Crider, DG, Tarasova, Y, and Maltsev, VA. 2005. Cardiomyocytes derived from embryonic stem cells. Methods Mol. Med. 108: 417-435.

Bonewald, LF. 2011. The amazing osteocyte. J. Bone Miner. Res. 26: 229-238. Bourne, S, Polak, JM, Hughes, SP, and Buttery, LD. 2004. Osteogenic

differentiation of mouse embryonic stem cells: differential gene expression analysis by cDNA microarray and purification of osteoblasts by cadherin-11 magnetically activated cell sorting. Tissue Eng. 10: 796-806.

Burns, K, Duggan, B, Atkinson, EA, Famulski, KS, Nemer, M, Bleackley, RC, and Michalak, M. 1994. Modulation of gene expression by calreticulin binding to the glucocorticoid receptor. Nature 367: 476-480.

Buttery, LD, Bourne, S, Xynos, JD, Wood, H, Hughes, FJ, Hughes, SP, Episkopou, V, and Polak, JM. 2001. Differentiation of osteoblasts and in vitro bone formation from murine embryonic stem cells. Tissue Eng. 7: 89-99.

81

Chawla, A, Repa, JJ, Evans, RM, and Mangelsdorf, DJ. 2001. Nuclear receptors and lipid physiology: opening the X-files. Science 294: 1866-1870.

Chen, JH, Liu, C, You, L, and Simmons, CA. 2010. Boning up on Wolff's Law: mechanical regulation of the cells that make and maintain bone. J. Biomech. 43: 108-118.

Choi, YY, Chung, BG, Lee, DH, Khademhosseini, A, Kim, JH, and Lee, SH. 2010. Controlled-size embryoid body formation in concave microwell arrays. Biomaterials 31: 4296-4303.

Chou, YF, Huang, W, Dunn, JC, Miller, TA, and Wu, BM. 2005. The effect of biomimetic apatite structure on osteoblast viability, proliferation, and gene expression. Biomaterials 26: 285-295.

Chung, CH, Golub, EE, Forbes, E, Tokuoka, T, and Shapiro, IM. 1992. Mechanism of action of beta-glycerophosphate on bone cell mineralization. Calcif. Tissue Int. 51: 305-311.

Coe, H and Michalak, M. 2009. Calcium binding chaperones of the endoplasmic reticulum. Gen. Physiol Biophys. 28 Spec No Focus: F96-F103.

Coelho, MJ and Fernandes, MH. 2000. Human bone cell cultures in biocompatibility testing. Part II: effect of ascorbic acid, beta-glycerophosphate and dexamethasone on osteoblastic differentiation. Biomaterials 21: 1095-1102.

Cohen, DE and Melton, D. 2011. Turning straw into gold: directing cell fate for regenerative medicine. Nat Rev. Genet. 12: 243-252.

Coppari, S, Altieri, F, Ferraro, A, Chichiarelli, S, Eufemi, M, and Turano, C. 2002. Nuclear localization and DNA interaction of protein disulfide isomerase ERp57 in mammalian cells. J. Cell Biochem. 85: 325-333.

Corbett, EF, Oikawa, K, Francois, P, Tessier, DC, Kay, C, Bergeron, JJM, Thomas, DY, Krause, KH, and Michalak, M. 1999. Ca2+ regulation of interactions between endoplasmic reticulum chaperones. J. Biol. Chem. 274: 6203-6211.

Cowan, CA, Klimanskaya, I, McMahon, J, Atienza, J, Witmyer, J, Zucker, JP, Wang, S, Morton, CC, McMahon, AP, Powers, D, and Melton, DA. 2004. Derivation of embryonic stem-cell lines from human blastocysts. N. Engl. J. Med. 350: 1353-1356.

Crabtree, GR. 1999. Generic signals and specific outcomes: signaling through Ca2+, calcineurin, and NF-AT. Cell 96: 611-614.

Crabtree, GR. 2001. Calcium, calcineurin, and the control of transcription. J. Biol. Chem. 276: 2313-2316.

D'Ippolito, G, Diabira, S, Howard, GA, Roos, BA, and Schiller, PC. 2006. Low oxygen tension inhibits osteogenic differentiation and enhances stemness of human MIAMI cells. Bone 39: 513-522.

Darnell, JE, Jr., Kerr, IM, and Stark, GR. 1994. Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264: 1415-1421.

Deepak, V, Zhang, Z, Meng, L, Zeng, X, and Liu, W. 2011. Reduced activity and cytoplasmic localization of Runx2 is observed in C3h10t1/2 cells overexpressing Tbx3. Cell Biochem. Funct. 29: 348-350.

82

Dingwall, M, Marchildon, F, Gunanayagam, A, Louis, CS, and Wiper-Bergeron, N. 2011. Retinoic acid-induced Smad3 expression is required for the induction of osteoblastogenesis of mesenchymal stem cells. Differentiation 82: 57-65.

Dirks, RW, Daniël, KC, and Raap, AK. 1995. RNAs radiate from gene to cytoplasm as revealed by fluorescence in situ hybridization. J. Cell Sci. 108: 2565-2572.

Djuric, SW, BaMaung, NY, Basha, A, Liu, H, Luly, JR, Madar, DJ, Sciotti, RJ, Tu, NP, Wagenaar, FL, Wiedeman, PE, Zhou, X, Ballaron, S, Bauch, J, Chen, YW, Chiou, XG, Fey, T, Gauvin, D, Gubbins, E, Hsieh, GC, Marsh, KC, Mollison, KW, Pong, M, Shaughnessy, TK, Sheets, MP, Smith, M, Trevillyan, JM, Warrior, U, Wegner, CD, and Carter, GW. 2000. 3,5-Bis(trifluoromethyl)pyrazoles: a novel class of NFAT transcription factor regulator. J. Med. Chem. 43: 2975-2981.

Doetschman, TC, Eistetter, H, Katz, M, Schmidt, W, and Kemler, R. 1985. The in vitro development of blastocyst-derived embryonic stem cell lines: formation of visceral yolk sac, blood islands and myocardium. J. Embryol. Exp. Morphol. 87: 27-45.

Drissi, MH, Li, X, Sheu, TJ, Zuscik, MJ, Schwarz, EM, Puzas, JE, Rosier, RN, and O'Keefe, RJ. 2003. Runx2/Cbfa1 stimulation by retinoic acid is potentiated by BMP2 signaling through interaction with Smad1 on the collagen X promoter in chondrocytes. J. Cell. Biochem. 90: 1287-1298.

Ducy, P, Zhang, R, Geoffroy, V, Ridall, AL, and Karsenty, G. 1997. Osf2/Cbfa1: a transcriptional activator of osteoblast differentiation. Cell 89: 747-754.

Duester, G. 2008. Retinoic acid synthesis and signaling during early organogenesis. Cell 134: 921-931.

Duplomb, L, Dagouassat, M, Jourdon, P, and Heymann, D. 2007a. Concise review: embryonic stem cells: a new tool to study osteoblast and osteoclast differentiation. Stem Cells 25: 544-552.

Duplomb, L, Dagouassat, M, Jourdon, P, and Heymann, D. 2007b. Differentiation of osteoblasts from mouse embryonic stem cells without generation of embryoid body. In Vitro Cell Dev. Biol. Anim. 43: 21-24.

Evans, MJ and Kaufman, MH. 1981. Establishment in culture of pluripotential cells from mouse embryos. Nature 292: 154-156.

Field, RA, Riley, ML, Mello, FC, Corbridge, MH, and Kotula, AW. 1974. Bone composition in cattle, pigs, sheep and poultry. J. Anim Sci. 39: 493-499.

Franceschi, RT and Iyer, BS. 1992. Relationship between collagen synthesis and expression of the osteoblast phenotype in MC3T3-E1 cells. J. Bone Miner. Res. 7: 235-246.

Fraser, A, Fearon, U, Billinghurst, RC, Ionescu, M, Reece, R, Barwick, T, Emery, P, Poole, AR, and Veale, DJ. 2003. Turnover of type II collagen and aggrecan in cartilage matrix at the onset of inflammatory arthritis in humans: relationship to mediators of systemic and local inflammation. Arthritis Rheum. 48: 3085-3095.

Gelebart, P, Opas, M, and Michalak, M. 2005. Calreticulin, a Ca(2+)-binding chaperone of the endoplasmic reticulum. Int. J. Biochem. Cell Biol. 37: 260-266.

83

Guo, L, Lynch, J, Nakamura, K, Fliegel, L, Kasahara, H, Izumo, S, Komuro, I, Agellon, LB, and Michalak, M. 2001. COUP-TF1 antagonizes Nkx2.5-mediated activation of the calreticulin gene during cardiac development. J. Biol. Chem. 276: 2797-2801.

Guo, L, Nakamura, K, Lynch, J, Opas, M, Olson, EN, Agellon, LB, and Michalak, M. 2002. Cardiac-specific expression of calcineurin reverses embryonic lethality in calreticulin-deficient mouse. J. Biol. Chem. 277: 50776-50779.

Gurland, G and Gundersen, GG. 1995. Stable, detyrosinated microtubules function to localize vimentin intermediate filaments in fibroblasts. J. Cell Biol. 131: 1275-1290.

Hebert, DN and Molinari, M. 2007. In and out of the ER: protein folding, quality control, degradation, and related human diseases. Physiol Rev. 87: 1377-1408.

Hoffman, LM and Carpenter, MK. 2005. Human embryonic stem cell stability. Stem Cell Rev. 1: 139-144.

Hogan, PG, Chen, L, Nardone, J, and Rao, A. 2003. Transcriptional regulation by calcium, calcineurin, and NFAT. Genes Dev. 17: 2205-2232.

Holaska, JM, Black, BE, Rastinejad, F, and Paschal, BM. 2002. Ca2+-dependent nuclear export mediated by calreticulin. Mol. Cell. Biol. 22: 6286-6297.

Huang, J, Bi, Y, Zhu, GH, He, Y, Su, Y, He, BC, Wang, Y, Kang, Q, Chen, L, Zuo, GW, Luo, Q, Shi, Q, Zhang, BQ, Huang, A, Zhou, L, Feng, T, Luu, HH, Haydon, RC, He, TC, and Tang, N. 2009. Retinoic acid signalling induces the differentiation of mouse fetal liver-derived hepatic progenitor cells. Liver Int. 29: 1569-1581.

Huang, Y, Song, Y, Zhang, C, Chen, G, Wang, S, and Bian, Z. 2013. Novel RUNX2 frameshift mutations in Chinese patients with cleidocranial dysplasia. Eur. J. Oral Sci. 121: 142-147.

Hubbard, MJ. 1996. Abundant calcium homeostasis machinery in rat dental enamel cells - Up-regulation of calcium store proteins during enamel mineralization implicates the endoplasmic reticulum in calcium transcytosis. Eur. J. Biochem. 239: 611-623.

Hwang, YS, Chung, BG, Ortmann, D, Hattori, N, Moeller, HC, and Khademhosseini, A. 2009. Microwell-mediated control of embryoid body size regulates embryonic stem cell fate via differential expression of WNT5a and WNT11. Proc. Natl. Acad. Sci. U. S. A. 106: 16978-16983.

Hwang, YS, Randle, WL, Bielby, RC, Polak, JM, and Mantalaris, A. 2006. Enhanced derivation of osteogenic cells from murine embryonic stem cells after treatment with HepG2-conditioned medium and modulation of the embryoid body formation period: application to skeletal tissue engineering. Tissue Eng 12: 1381-1392.

Hyatt, SL, Klauck, T, and Jaken, S. 1990. Protein kinase C is localized in focal contacts of normal but not transformed fibroblasts. Mol. Carcinog. 3: 45-53.

Jeon, MJ, Kim, JA, Kwon, SH, Kim, SW, Park, KS, Park, SW, Kim, SY, and Shin, CS. 2003. Activation of peroxisome proliferator-activated receptor-gamma inhibits the Runx2-mediated transcription of osteocalcin in osteoblasts. J. Biol. Chem. 278: 23270-23277.

84

Jones, EA, Tosh, D, Wilson, DI, Lindsay, S, and Forrester, LM. 2002. Hepatic differentiation of murine embryonic stem cells. Exp. Cell Res. 272: 15-22.

Karsenty, G. 2008. Transcriptional control of skeletogenesis. Annu. Rev. Genomics Hum. Genet. 9: 183-196.

Kawaguchi, J, Mee, PJ, and Smith, AG. 2005. Osteogenic and chondrogenic differentiation of embryonic stem cells in response to specific growth factors. Bone 36: 758-769.

Kawai, M and Rosen, CJ. 2010. PPARgamma: a circadian transcription factor in adipogenesis and osteogenesis. Nat. Rev. Endocrinol. 6: 629-636.

Keller, G, Kennedy, M, Papayannopoulou, T, and Wiles, MV. 1993. Hematopoietic commitment during embryonic stem cell differentiation in culture. Mol. Cell Biol. 13: 473-486.

Keller, GM. 1995. In vitro differentiation of embryonic stem cells. Curr. Opin. Cell Biol. 7: 862-869.

Khillan, JS and Chen, L. 2010. Feeder-independent culture of mouse embryonic stem cells using vitamin A/retinol. Methods Mol. Biol. 652: 75-83.

Khouja, HI, Bevington, A, Kemp, GJ, and Russell, RG. 1990. Calcium and orthophosphate deposits in vitro do not imply osteoblast-mediated mineralization: mineralization by betaglycerophosphate in the absence of osteoblasts. Bone 11: 385-391.

Kim, S, Koga, T, Isobe, M, Kern, BE, Yokochi, T, Chin, YE, Karsenty, G, Taniguchi, T, and Takayanagi, H. 2003. Stat1 functions as a cytoplasmic attenuator of Runx2 in the transcriptional program of osteoblast differentiation. Genes Dev. 17: 1979-1991.

Klug, MG, Soonpaa, MH, Koh, GY, and Field, LJ. 1996. Genetically selected cardiomyocytes from differentiating embronic stem cells form stable intracardiac grafts. J. Clin. Invest. 98: 216-224.

Komori, T. 2003. Requisite roles of Runx2 and Cbfb in skeletal development. J. Bone Miner. Metab 21: 193-197.

Komori, T. 2008. Regulation of bone development and maintenance by Runx2. Front. Biosci. 13: 898-903.

Komori, T, Yagi, H, Nomura, S, Yamaguchi, A, Sasaki, K, Deguchi, K, Shimizu, Y, Bronson, RT, Gao, YH, Inada, M, Sato, M, Okamoto, R, Kitamura, Y, Yoshiki, S, and Kishimoto, T. 1997. Targeted disruption of Cbfa1 results in a complete lack of bone formation owing to maturational arrest of osteoblasts. Cell 89: 755-764.

Krause, U, Harris, S, Green, A, Ylostalo, J, Zeitouni, S, Lee, N, and Gregory, CA. 2010. Pharmaceutical modulation of canonical Wnt signaling in multipotent stromal cells for improved osteoinductive therapy. Proc. Natl. Acad. Sci. U. S. A. 107: 4147-4152.

Kurosawa, H. 2007. Methods for inducing embryoid body formation: in vitro differentiation system of embryonic stem cells. J. Biosci. Bioeng. 103: 389-398.

Kuske, B, Savkovic, V, and Zur Nieden, NI. 2011. Improved media compositions for the differentiation of embryonic stem cells into osteoblasts and chondrocytes. Methods Mol. Biol. 690: 195-215.

85

Kuwabara, K, Pinsky, DJ, Schmidt, AM, Benedict, C, Brett, J, Ogawa, S, Broekman, MJ, Marcus, AJ, Sciacca, RR, Michalak, M, Wang, F, Pan, YC, Grunfeld, S, Patton, S, Malinski, T, Stern, DM, and Ryan, J. 1995. Calreticulin, an antithrombotic agent which binds to vitamin K-dependent coagulation factors, stimulates endothelial nitric oxide production, and limits thrombosis in canine coronary arteries. J. Biol. Chem. 270: 8179-8187.

Leboy, PS, Beresford, JN, Devlin, C, and Owen, ME. 1991. Dexamethasone induction of osteoblast mRNAs in rat marrow stromal cell cultures. J. Cell Physiol 146: 370-378.

Lecka-Czernik, B, Moerman, EJ, Grant, DF, Lehmann, JM, Manolagas, SC, and Jilka, RL. 2002. Divergent effects of selective peroxisome proliferator-activated receptor-gamma 2 ligands on adipocyte versus osteoblast differentiation. Endocrinology 143: 2376-2384.

Lecka-Czernik, B and Suva, LJ. 2006. Resolving the Two "Bony" Faces of PPAR-gamma. PPAR. Res. 2006: 27489.

Li, J, Puceat, M, Perez-Terzic, C, Mery, A, Nakamura, K, Michalak, M, Krause, KH, and Jaconi, ME. 2002. Calreticulin reveals a critical Ca2+ checkpoint in cardiac myofibrillogenesis. J. Cell Biol. 158: 103-113.

Li, X, Schwarz, EM, Zuscik, MJ, Rosier, RN, Ionescu, AM, Puzas, JE, Drissi, H, Sheu, TJ, and O'Keefe, RJ. 2003. Retinoic acid stimulates chondrocyte differentiation and enhances bone morphogenetic protein effects through induction of Smad1 and Smad5. Endocrinology 144: 2514-2523.

Li, X, Zhu, L, Yang, A, Lin, J, Tang, F, Jin, S, Wei, Z, Li, J, and Jin, Y. 2011. Calcineurin-NFAT signaling critically regulates early lineage specification in mouse embryonic stem cells and embryos. Cell Stem Cell 8: 46-58.

Lian, JB, Shalhoub, V, Aslam, F, Frenkel, B, Green, J, Hamrah, M, Stein, GS, and Stein, JL. 1997. Species-specific glucocorticoid and 1,25-dihydroxyvitamin D responsiveness in mouse MC3T3-E1 osteoblasts: dexamethasone inhibits osteoblast differentiation and vitamin D down-regulates osteocalcin gene expression. Endocrinology 138: 2117-2127.

Lin, Q, Schwarz, J, Bucana, C, and Olson, EN. 1997. Control of mouse cardiac morphogenesis and myogenesis by transcription factor MEF2C. Science 276: 1404-1407.

Long, F. 2012. Building strong bones: molecular regulation of the osteoblast lineage. Nat. Rev. Mol. Cell Biol. 13: 27-38.

Lorincz, MT. 2006. Optimized neuronal differentiation of murine embryonic stem cells: role of cell density. Methods Mol. Biol. 330: 55-69.

Lozyk, MD, Papp, S, Zhang, X, Nakamura, K, Michalak, M, and Opas, M. 2006. Ultrastructural analysis of development of myocardium in calreticulin-deficient mice. BMC Dev. Biol. 6: 54.

Lynch, J, Guo, L, Gelebart, P, Chilibeck, K, Xu, J, Molkentin, JD, Agellon, LB, and Michalak, M. 2005. Calreticulin signals upstream of calcineurin and MEF2C in a critical Ca2+-dependent signaling cascade. J. Cell Biol. 170: 37-47.

Lynch, J and Michalak, M. 2003. Calreticulin is an upstream regulator of calcineurin. Biochem. Biophys. Res. Commun. 311: 1173-1179.

86

Mackie, EJ, Tatarczuch, L, and Mirams, M. 2011. The skeleton: a multi-functional complex organ: the growth plate chondrocyte and endochondral ossification. J. Endocrinol. 211: 109-121.

Malaval, L, Wade-Gueye, NM, Boudiffa, M, Fei, J, Zirngibl, R, Chen, F, Laroche, N, Roux, JP, Burt-Pichat, B, Duboeuf, F, Boivin, G, Jurdic, P, Lafage-Proust, MH, Amedee, J, Vico, L, Rossant, J, and Aubin, JE. 2008. Bone sialoprotein plays a functional role in bone formation and osteoclastogenesis. J. Exp. Med. 205: 1145-1153.

Mark, M, Ghyselinck, NB, and Chambon, P. 2006. Function of retinoid nuclear receptors: lessons from genetic and pharmacological dissections of the retinoic acid signaling pathway during mouse embryogenesis. Annu. Rev. Pharmacol. Toxicol. 46: 451-480.

Martin, GR. 1981. Isolation of a pluripotent cell line from early mouse embryos cultured in medium conditioned by teratocarcinoma stem cells. Proc. Natl. Acad. Sci. U. S. A 78: 7634-7638.

McBeath, R, Pirone, DM, Nelson, CM, Bhadriraju, K, and Chen, CS. 2004. Cell shape, cytoskeletal tension, and RhoA regulate stem cell lineage commitment. Dev. Cell 6: 483-495.

Mery, L, Mesaeli, N, Michalak, M, Opas, M, Lew, DP, and Krause, K-H. 1996. Overexpression of calreticulin increases intracellular Ca2+-storage and decreases store-operated Ca2+ influx. J. Biol. Chem. 271: 9332-9339.

Mesaeli, N, Nakamura, K, Zvaritch, E, Dickie, P, Dziak, E, Krause, KH, Opas, M, MacLennan, DH, and Michalak, M. 1999. Calreticulin is essential for cardiac development. J. Cell Biol. 144: 857-868.

Michalak, M, Corbett, EF, Mesaeli, N, Nakamura, K, and Opas, M. 1999. Calreticulin: one protein, one gene and many functions. Biochem. J. 344: 281-292.

Michalak, M, Groenendyk, J, Szabo, E, Gold, LI, and Opas, M. 2009. Calreticulin, a multi-process calcium-buffering chaperone of the endoplasmic reticulum. Biochem. J. 417: 651-666.

Mikami, Y, Lee, M, Irie, S, and Honda, MJ. 2011. Dexamethasone modulates osteogenesis and adipogenesis with regulation of osterix expression in rat calvaria-derived cells. J. Cell Physiol 226: 739-748.

Mori-Akiyama, Y, Akiyama, H, Rowitch, DH, and de Crombrugghe, B. 2003. Sox9 is required for determination of the chondrogenic cell lineage in the cranial neural crest. Proc. Natl. Acad. Sci. U. S. A 100: 9360-9365.

Mundlos, S, Otto, F, Mundlos, C, Mulliken, JB, Aylsworth, AS, Albright, S, Lindhout, D, Cole, WG, Henn, W, Knoll, JH, Owen, MJ, Mertelsmann, R, Zabel, BU, and Olsen, BR. 1997. Mutations involving the transcription factor CBFA1 cause cleidocranial dysplasia [see comments]. Cell 89: 773-779.

Nagy, A, Rossant, J, Nagy, R, Abramow-Newerly, W, and Roder, JC. 1993. Derivation of completely cell culture-derived mice from early- passage embryonic stem cells. Proc. Natl. Acad. Sci. USA 90: 8424-8428.

Nakamura, K, Zuppini, A, Arnaudeau, S, Lynch, J, Ahsan, I, Krause, R, Papp, S, De Smedt, H, Parys, JB, Muller-Esterl, W, Lew, DP, Krause, KH, Demaurex,

87

N, Opas, M, and Michalak, M. 2001. Functional specialization of calreticulin domains. J. Cell Biol. 154: 961-972.

Nemer, G and Nemer, M. 2001. Regulation of heart development and function through combinatorial interactions of transcription factors. Ann. Med. 33: 604-610.

Niederreither, K and Dolle, P. 2008. Retinoic acid in development: towards an integrated view. Nat Rev. Genet. 9: 541-553.

Niwa, H, Ogawa, K, Shimosato, D, and Adachi, K. 2009. A parallel circuit of LIF signalling pathways maintains pluripotency of mouse ES cells. Nature 460: 118-122.

Ogston, N, Harrison, AJ, Cheung, HF, Ashton, BA, and Hampson, G. 2002. Dexamethasone and retinoic acid differentially regulate growth and differentiation in an immortalised human clonal bone marrow stromal cell line with osteoblastic characteristics. Steroids 67: 895-906.

Okabe, S, Forsberg-Nilsson, K, Spiro, AC, Segal, M, and McKay, RD. 1996. Development of neuronal precursor cells and functional postmitotic neurons from embryonic stem cells in vitro. Mech. Dev. 59: 89-102.

Olsen, BR, Reginato, AM, and Wang, W. 2000. Bone development. Annu. Rev. Cell Dev. Biol. 16: 191-220.

Osafune, K, Caron, L, Borowiak, M, Martinez, RJ, Fitz-Gerald, CS, Sato, Y, Cowan, CA, Chien, KR, and Melton, DA. 2008. Marked differences in differentiation propensity among human embryonic stem cell lines. Nat Biotechnol. 26: 313-315.

Ostwald, TJ and MacLennan, DH. 1974. Isolation of a high affinity calcium binding protein from sarcoplasmic reticulum. J. Biol. Chem. 249: 974-979.

Otto, F, Thornell, AP, Crompton, T, Denzel, A, Gilmour, KC, Rosewell, IR, Stamp, GW, Beddington, RS, Mundlos, S, Olsen, BR, Selby, PB, and Owen, MJ. 1997. Cbfa1, a candidate gene for cleidocranial dysplasia syndrome, is essential for osteoblast differentiation and bone development [see comments]. Cell 89: 765-771.

Papp, S, Dziak, E, Kabir, G, Backx, P, Clement, S, and Opas, M. 2009. Evidence for cardiac hypertrophy attenuation by calreticulin induction using pressure overload and soluble agonists. Am. J. Pathol. 176: 1113-1121.

Park, J, Cho, CH, Parashurama, N, Li, Y, Berthiaume, F, Toner, M, Tilles, AW, and Yarmush, ML. 2007. Microfabrication-based modulation of embryonic stem cell differentiation. Lab Chip. 7: 1018-1028.

Park, JH, Park, BH, Kim, HK, Park, TS, and Baek, HS. 2002. Hypoxia decreases Runx2/Cbfa1 expression in human osteoblast-like cells. Mol. Cell Endocrinol. 192: 197-203.

Pereira, RC, Delany, AM, and Canalis, E. 2002. Effects of cortisol and bone morphogenetic protein-2 on stromal cell differentiation: correlation with CCAAT-enhancer binding protein expression. Bone 30: 685-691.

Persson, S, Rosenquist, M, and Sommarin, M. 2002. Identification of a novel calreticulin isoform (Crt2) in human and mouse. Gene 297: 151.

88

Phillips, BW, Belmonte, N, Vernochet, C, Ailhaud, G, and Dani, C. 2001. Compactin enhances osteogenesis in murine embryonic stem cells. Biochem. Biophys. Res. Commun. 284: 478-484.

Phillips, BW, Vernochet, C, and Dani, C. 2003. Differentiation of embryonic stem cells for pharmacological studies on adipose cells. Pharmacol. Res. 47: 263-268.

Plachta, N, Bibel, M, Tucker, KL, and Barde, YA. 2004. Developmental potential of defined neural progenitors derived from mouse embryonic stem cells. Development 131: 5449-5456.

Porter, RM, Huckle, WR, and Goldstein, AS. 2003. Effect of dexamethasone withdrawal on osteoblastic differentiation of bone marrow stromal cells. J. Cell. Biochem. 90: 13-22.

Rauch, F, Prud'homme, J, Arabian, A, Dedhar, S, and St-Arnaud, R. 2000. Heart, brain, and body wall defects in mice lacking calreticulin. Exp. Cell Res. 256: 105-111.

Reinhold, MI, Abe, M, Kapadia, RM, Liao, Z, and Naski, MC. 2004. FGF18 represses noggin expression and is induced by calcineurin. J. Biol. Chem. 279: 38209-38219.

Rose, LC, Fitzsimmons, R, Lee, P, Krawetz, R, Rancourt, DE, and Uludag, H. 2012. Effect of basic fibroblast growth factor in mouse embryonic stem cell culture and osteogenic differentiation. J. Tissue Eng. Regen. Med. 7: 371-382.

Rosen, ED and Spiegelman, BM. 2000. Molecular regulation of adipogenesis. Annu. Rev. Cell Dev. Biol. 16: 145-171.

Rungby, J, Kassem, M, Eriksen, EF, and Danscher, G. 1993. The von Kossa reaction for calcium deposits: silver lactate staining increases sensitivity and reduces background. Histochem. J. 25: 446-451.

Salim, A, Nacamuli, RP, Morgan, EF, Giaccia, AJ, and Longaker, MT. 2004. Transient changes in oxygen tension inhibit osteogenic differentiation and Runx2 expression in osteoblasts. J. Biol. Chem. 279: 40007-40016.

Shao, D and Lazar, MA. 1997. Peroxisome proliferator activated receptor gamma, CCAAT/enhancer-binding protein alpha, and cell cycle status regulate the commitment to adipocyte differentiation. J. Biol. Chem. 272: 21473-21478.

Shi, XM, Blair, HC, Yang, X, McDonald, JM, and Cao, X. 2000. Tandem repeat of C/EBP binding sites mediates PPARgamma2 gene transcription in glucocorticoid-induced adipocyte differentiation. J. Cell. Biochem. 76: 518-527.

Shinji, T, Koide, N, and Tsuji, T. 1988. Glycosaminoglycans partially substitute for proteoglycans in spheroid formation of adult rat hepatocytes in primary culture

1. Cell Struct. Funct. 13: 179-188. Smith, E and Frenkel, B. 2005. Glucocorticoids inhibit the transcriptional activity

of LEF/TCF in differentiating osteoblasts in a glycogen synthase kinase-3beta-dependent and -independent manner. J. Biol. Chem. 280: 2388-2394.

89

Somogyi, E, Petersson, U, Hultenby, K, and Wendel, M. 2003. Calreticulin-an endoplasmic reticulum protein with calcium-binding activity is also found in the extracellular matrix. Matrix Biol. 22: 179-191.

Stewart, M, Terry, A, Hu, M, O'Hara, M, Blyth, K, Baxter, E, Cameron, E, Onions, DE, and Neil, JC. 1997. Proviral insertions induce the expression of bone-specific isoforms of PEBP2alphaA (CBFA1): evidence for a new myc collaborating oncogene. Proc. Natl. Acad. Sci. U. S. A. 94: 8646-8651.

Sun, L, Blair, HC, Peng, Y, Zaidi, N, Adebanjo, OA, Wu, XB, Wu, XY, Iqbal, J, Epstein, S, Abe, E, Moonga, BS, and Zaidi, M. 2005. Calcineurin regulates bone formation by the osteoblast. Proc. Natl. Acad. Sci. USA 102: 17130-17135.

Szabo, E, Feng, T, Dziak, E, and Opas, M. 2009. Cell adhesion and spreading affect adipogenesis from ES cells: the role of calreticulin. Stem Cells 27: 2092-2102.

Szabo, E, Qiu, Y, Baksh, S, Michalak, M, and Opas, M. 2008. Calreticulin inhibits commitment to adipocyte differentiation. J. Cell Biol. 182: 103-116.

Tai, G, Christodoulou, I, Bishop, AE, and Polak, JM. 2005. Use of green fluorescent fusion protein to track activation of the transcription factor osterix during early osteoblast differentiation. Biochem Biophys. Res. Commun. 333: 1116-1122.

Takada, I, Suzawa, M, Matsumoto, K, and Kato, S. 2007. Suppression of PPAR transactivation switches cell fate of bone marrow stem cells from adipocytes into osteoblasts. Ann. N. Y. Acad. Sci. 1116: 182-195.

Tjoelker, LW, Seyfried, CE, Eddy, RL, Jr., Byers, MG, Shows, TB, Calderon, J, Schreiber, RB, and Gray, PW. 1994. Human, mouse, and rat calnexin cDNA cloning: identification of potential calcium binding motifs and gene localization to human chromosome 5. Biochemistry 33: 3229-3236.

Toyooka, Y, Shimosato, D, Murakami, K, Takahashi, K, and Niwa, H. 2008. Identification and characterization of subpopulations in undifferentiated ES cell culture. Development 135: 909-918.

Trevillyan, JM, Chiou, XG, Chen, YW, Ballaron, SJ, Sheets, MP, Smith, ML, Wiedeman, PE, Warrior, U, Wilkins, J, Gubbins, EJ, Gagne, GD, Fagerland, J, Carter, GW, Luly, JR, Mollison, KW, and Djuric, SW. 2001. Potent inhibition of NFAT activation and T cell cytokine production by novel low molecular weight pyrazole compounds. J. Biol. Chem. 276: 48118-48126.

Vaillant, F, Blyth, K, Andrew, L, Neil, JC, and Cameron, ER. 2002. Enforced expression of Runx2 perturbs T cell development at a stage coincident with beta-selection. J. Immunol. 169: 2866-2874.

Vassilakos, A, Michalak, M, Lehrman, MA, and Williams, DB. 1998. Oligosaccharide binding characteristics of the molecular chaperones calnexin and calreticulin. Biochemistry 37: 3480-3490.

Wan, DC, Shi, YY, Nacamuli, RP, Quarto, N, Lyons, KM, and Longaker, MT. 2006. Osteogenic differentiation of mouse adipose-derived adult stromal cells requires retinoic acid and bone morphogenetic protein receptor type IB signaling. Proc. Natl. Acad. Sci. U. S. A. 103: 12335-12340.

90

Wang, A, Ding, X, Sheng, S, and Yao, Z. 2008. Retinoic acid inhibits osteogenic differentiation of rat bone marrow stromal cells. Biochem. Biophys. Res. Commun. 375: 435-439.

Wang, Y, Wan, C, Deng, L, Liu, X, Cao, X, Gilbert, SR, Bouxsein, ML, Faugere, MC, Guldberg, RE, Gerstenfeld, LC, Haase, VH, Johnson, RS, Schipani, E, and Clemens, TL. 2007. The hypoxia-inducible factor alpha pathway couples angiogenesis to osteogenesis during skeletal development. J. Clin. Invest 117: 1616-1626.

Wartenberg, M, Gunther, J, Hescheler, J, and Sauer, H. 1998. The embryoid body as a novel in vitro assay system for antiangiogenic agents. Lab. Invest. 78: 1301-1314.

Weston, AD, Rosen, V, Chahndraratna, RAS, and Underhill, TM. 2000. Regulation of skeletal progenitor differentiation by the BMP and retinoid signaling pathways. J. Cell Biol. 148286: 679-690.

Wheeler, DG, Horsford, J, Michalak, M, White, JH, and Hendy, GN. 1995. Calreticulin inhibits vitamin D3 signal transduction. Nucleic Acids Res. 23: 3268-3274.

Wiles, MV and Keller, G. 1991. Multiple hematopoietic lineages develop from embryonic stem (ES) cells in culture. Development 111: 259-267.

Winslow, MM, Pan, M, Starbuck, M, Gallo, EM, Deng, L, Karsenty, G, and Crabtree, GR. 2006. Calcineurin/NFAT signaling in osteoblasts regulates bone mass. Dev. Cell 10: 771-782.

Wobus, AM, Wallukat, G, and Hescheler, J. 1991. Pluripotent mouse embryonic stem cells are able to differentiate into cardiomyocytes expressing chronotropic responses to adrenergic and cholinergic agents and Ca2+ channel blockers. Differentiation 48: 173-182.

Woll, NL and Bronson, SK. 2006. Analysis of embryonic stem cell-derived osteogenic cultures. Methods Mol. Biol. 330: 149-159.

Xiao, G, Cui, Y, Ducy, P, Karsenty, G, and Franceschi, RT. 1997. Ascorbic acid-dependent activation of the osteocalcin promoter in MC3T3-E1 preosteoblasts: requirement for collagen matrix synthesis and the presence of an intact OSE2 sequence. Mol. Endocrinol. 11: 1103-1113.

Xiao, L, Naganawa, T, Obugunde, E, Gronowicz, G, Ornitz, DM, Coffin, JD, and Hurley, MM. 2004. Stat1 controls postnatal bone formation by regulating fibroblast growth factor signaling in osteoblasts. J. Biol. Chem. 279: 27743-27752.

Yamashita, A, Takada, T, Narita, J, Yamamoto, G, and Torii, R. 2005. Osteoblastic differentiation of monkey embryonic stem cells in vitro. Cloning Stem Cells 7: 232-237.

Yamashita, A, Takada, T, Nemoto, K, Yamamoto, G, and Torii, R. 2006. Transient suppression of PPARgamma directed ES cells into an osteoblastic lineage. FEBS Lett. 580: 4121-4125.

Zayzafoon, M, Abdulkadir, SA, and McDonald, JM. 2004. Notch signaling and ERK activation are important for the osteomimetic properties of prostate cancer bone metastatic cell lines. J. Biol. Chem. 279: 3662-3670.

91

Zhang, W, Deng, ZL, Chen, L, Zuo, GW, Luo, Q, Shi, Q, Zhang, BQ, Wagner, ER, Rastegar, F, Kim, SH, Jiang, W, Shen, J, Huang, E, Gao, Y, Gao, JL, Zhou, JZ, Luo, J, Huang, J, Luo, X, Bi, Y, Su, Y, Yang, K, Liu, H, Luu, HH, Haydon, RC, He, TC, and He, BC. 2010. Retinoic acids potentiate BMP9-induced osteogenic differentiation of mesenchymal progenitor cells. PLoS. ONE 5: e11917.

Zhou, X, Zhang, Z, Feng, JQ, Dusevich, VM, Sinha, K, Zhang, H, Darnay, BG, and de Crombrugghe, B. 2010. Multiple functions of Osterix are required for bone growth and homeostasis in postnatal mice. Proc. Natl. Acad. Sci. U. S. A 107: 12919-12924.

Zhu, GH, Huang, J, Bi, Y, Su, Y, Tang, Y, He, BC, He, Y, Luo, J, Wang, Y, Chen, L, Zuo, GW, Jiang, W, Luo, Q, Shen, J, Liu, B, Zhang, WL, Shi, Q, Zhang, BQ, Kang, Q, Zhu, J, Tian, J, Luu, HH, Haydon, RC, Chen, Y, and He, TC. 2009. Activation of RXR and RAR signaling promotes myogenic differentiation of myoblastic C2C12 cells. Differentiation 78: 195-204.

Zur Nieden, NI, Kempka, G, and Ahr, HJ. 2003. In vitro differentiation of embryonic stem cells into mineralized osteoblasts. Differentiation 71: 18-27.

92

APPENDIX Below are the cycle numbers (Ct values) for one representative experiment and the corresponding amplification curve, melting curve for each figure. Figure 5

L32 Runx2 OCN OSX

250 cells 24.07 25.01 25.07 

   

16.43 16.06 16.02 

24.38 24.22 23.61 

17.32 17.27 17.81 

500 cells 24.39 24.31 24.23 

20.28 20.36 20.73 

27.43 27.21 27.28 

20.41 20.86 20.18 

1000 cells 24.3 24.27 24.18 

19.3 19.29 19.01 

26.32 26.31 26.09 

19.52 19.65 19.61 

95

Figure 6

L32 Runx2 OCN OSX Control 24.09 

24.04 24.94 

24.46 

24.49 

24.5 

23.09 

22.91 

22.97 

29.18 

29.62 

29.64  Dex E 24.35 

23.36 24.13 

20.48 

20.77 

20.89 

21.43 

20.87 

20.88 

27.77 

27.38 

28.36  Dex L 25.3 

24.27 24.19 

20.24 

20.11 

20.16 

21.43 

21.6 

21.02 

26.31 

25.96 

25.75  GW 24.1 

24.04 24.08 

21.76 

21.89 

21.76 

20.28 

20.36 

20.03 

26.65 

26.68 

26.65  RA 24.07 

24.06 24.95 

20.04 

20.54 

20.21 

20.3 

20.29 

20.01 

24.12 

24.51 

24.74  RA Dex E 24.97 

24.87 24.9 

19.09 

18.41 

18.29 

19.83 

19.96 

19.44 

24.13 

24.15 

24.95  RA Dex L 24.41 

25.38 25.18 

18.43 

18.87 

18.88 

18.3 

18.09 

18.6 

     24.69 

23.89 

24.66 

98

Figure 10

L32 calreticulin Pluripotent 28.18 

27.62 

27.34 

N/A 

49.43 

56.9  Day 7 27.64 

27.77 

27.38 

27.05 

26.92 

27.3  Day 14 28.36 

27.31 

27.16 

25.33 

25.41 

25.5 

R1

Day 21 27.75 

28.65 

28.68 

26.07 

26.49 

25.94  Pluripotent 27.65 

28.12 

28.01 

N/A 

N/A 

N/A  Day 7 27.74 

27.13 

27.15 

N/A 

N/A 

N/A  Day 14 27.95 

27.69 

27.59 

N/A 

N/A 

N/A 

KO

Day 21 27.66 

28.32 

27.27 

N/A 

N/A 

N/A 

101

Figure 11

L32 Runx2 OSX BSP Pluripotent 28.18 

27.62 

27.34 

24.38 

24.62 

24.09 

24.23 

24.3 

24.27 

28.40 

28.78 

28.66  Day 7 27.64 

27.77 

27.38 

18.64 

18.07 

18.38 

23.18 

22.53 

23.6 

26.66 

26.86 

26.54  Day 14 28.36 

27.31 

27.16 

20.78 

20.62 

20.89 

18.43 

18.29 

18.08 

25.68 

25.27 

25.37 

R1

Day 21 27.75 

28.65 

28.68 

24.56 

24.41 

24.6 

24.18 

24.62 

24.64 

22.33 

22.64 

22.10  Pluripotent 27.65 

28.12 

28.01 

24.02 

24.3 

24.09 

26.77 

26.38 

26.36 

29.57 

29.90 

29.90  Day 7 27.74 

27.13 

27.15 

20.74 

21.13 

21.15 

25.31 

25.16 

25.75 

26.85 

26.56 

26.93  Day 14 27.95 

27.69 

27.59 

24.95 

24.69 

24.59 

25.05 

26.45 

26.65 

28.99 

28.29 

28.4 

KO

Day 21 27.66 

28.32 

27.27 

29.66 

29.32 

29.27 

28.15 

28.85 

28.22 

24.3 

24.49 

24.01 

104

Figure 12

L32 Col 2 Aggrecan R1 27.75 

28.65 

28.68 

22.56 

22.87 

22.01 

22.64 

22.98 

23.48  KO 27.66 

28.32 

27.27 

20.81 

20.46 

19.83 

18.69 

19.01 

18.95 

107

Figure 14

L32 Runx2 OSX BSP R1 20.40 

20.78 

19.66 

17.32

16.98

17.38

20.36 

20.65 

20.98 

19.98 

19.67 

20.06  R1 BAPTA-

AM 20.66 

19.86 

20.54 

18.09

18.27

19.05

23.67 

22.09 

22.87 

23.33 

24.64 

23.78  KO       19.68 

      19.27 

      20.37 

19.17

19.39

19.70

23.36 

23.65 

23.98 

26.89 

27.06 

26.57  KO ionomycin 20.33 

20.64 

20.10 

16.29

16.23

16.39

20.05 

21.61 

20.95 

21.96 

21.05 

21.36 

110

Figure 15

L32 Runx2 OSX BSP R1 20.40

20.78 19.66

20.89 20.53 20.41

24.19 24.36 24.89

22.77 22.89 23.19

R1 CsA 20.85 20.92 20.83

23.01 23.02 23.04

26.54 26.95 26.68

24.3 24.36 24.39

R1 A-285222 19.48 19.88 19.93

23.51 23.32 23.44

27.89 27.01 27.65

26.13 26.29 26.52

CN 19.98 20.45 20.06

19.96 20.02 20.08

24.39 23.43 23.42

24.19 24.36 24.89

CN CsA 20.11 20.37 20.97

22.69 22.69 22.43

26.58 26.19 26.43

27.92 27.37 27.80

CN A-285222 19.57 19.90 19.75

20.95 21.22 21.14

27.01 27.68 27.73

29.11 28.61 29.37