Elucidating the molecular mechanisms of CALR and CD47 in ...

1st August 2021 of 2

Research Thesis Submission The following guidelines must be observed in all cases:

- The thesis should be saved as a PDF file and should be the same version as submitted to the College Research Degrees Board

- The file name should be the author’s name (surname first), subject area and month and year e.g. Smith, Jane – Psychology – November 2015

If you thesis contains confidential or sensitive material, or it is under embargo or other restrictions please give details below. Your Name (required) Your Student ID (required) Your Email Address (required) Your Thesis Title (required) Any details regarding confidential or sensitive material, restrictions or embargos

Kristian Boasman

06036783

[email protected]

Elucidating the molecular mechanisms of CALR and CD47 in myelodysplastic

syndromes, myeloproliferative neoplasms

1st August 2021 of 2

Please confirm the following statements that are applicable prior to submitting your thesis ☒ I am submitting the final version of my thesis as approved by the examining team. The content is the same as the fully bound version that will be submitted to the College Research Degrees Board ☒ My thesis does not contain any confidential or sensitive material that should be removed prior to publication ☐ My thesis does contain any confidential or sensitive material that should be removed prior to publication ☒ My thesis does not have any publication embargoes or restrictions ☐ My thesis does have any publication embargoes or restrictions Submit your thesis and this form to your College Research Mailbox College of Arts – [email protected] College of Science – [email protected] College of Social Science – [email protected] Lincoln International Business School – [email protected] * Submitted theses will be made available via the University of Lincoln library/repository and the British Library unless otherwise notified

mailto:[email protected]



Elucidating the molecular mechanisms of CALR and CD47 in myelodysplastic syndromes, myeloproliferative neoplasms

Kristian Boasman

A thesis submitted in partial fulfilment of the requirements of the University of Lincoln for the degree of Doctor of Philosophy

School of Life Science

March 2021

1

Acknowledgements First of all, I would like to thank the University of Lincoln for giving me the opportunity to conduct

my research and complete this thesis.

Like many PhD students I am thankful to my friends and family who helped me through the stress of writing a thesis. Specifically, my wife Kay Hancock Boasman and my daughter Grace for their

constant support. My parents for whom I would not be the person I am today without them. Also, my brother Adam Boasman and close friend Dr James Henshaw for always listening to my rants

when I need them to and for always supporting me.

Finally, I am indebted to my supervisors, Prof Ciro Rinaldi and Dr Matthew Simmonds. Their constant understanding, insight and encouragement enabled me to keep pushing myself in a drive to produce

the work I was capable of.

2

Abstract Myelodysplastic syndromes (MDS) and Myeloproliferative neoplasms (MPN) are myeloid

malignancies arising from clonal disorders of haematopoietic stem cells, with the tendency to

progress into acute myeloid leukaemia (AML). Whilst diagnostic criteria have been established,

treatment options remain relatively limited promoting the search for markers which could be used

to help predicting disease progression or be targeted therapeutically to provide new or enhanced

treatment options. In solid tumours, calreticulin (CALR) overexpression produces a pro-phagocytic

signal and is counteracted by concomitant expression of anti-phagocytic CD47, reflecting an

apoptosis vs survival mechanism. Increases of both CALR and CD47 on the cell membrane have been

observed in response to chemotherapy, however their role in myeloid malignancies is poorly

understood. The main aim of this work is to investigate the expression and cellular localisation of

CALR and CD47 in MDS and MPN, determine any changes in CALR /CD47 as disease progresses and

establish how cellular location of CALR and CD47 alters in response to treatment in MDS and MPN.

Initial investigation of cell line models of MDS and MPN indicate that when treated with disease

specific cyto-toxic drugs, total expression of CALR and CD47 increased. However, CALR was shown to

internalise away from the membrane upon treatment whereas CD47 was externalised, suggesting a

different mode of action for CALR and CD47 in MDS/MPN compared to solid tumours. Further work

investigating whether these changes in CALR and CD47 expression in patient samples found that

CALR and CD47 baseline protein expression was increased in both MDS and MPN patients compared

with controls. CD47 showed higher overall protein expression on MDS and MPN cell membranes

when compared with CALR. As MDS progressed, CD47 expression is seen to increase with the

severity of the disease as classified by IPSS subgroups. Interestingly, when investigating MPN

subtypes, significant increases in CD47 cell membrane expression, in particular, after treatment was

only seen in the MF and PV subtypes, suggesting an anti-phagocytic effect induced by cytotoxic

drugs. In the ET subtype, CD47 expression is reduced after cyto-reductive treatment, suggesting

3

instead a prophagocytic effect. To better understand the relationship between CALR and CD47 and

support any future work looking to use these in a more therapeutic setting, knockdown experiments

were undertaken. Knockdown experiments revealed a possible link between CALR and CD47. In fact,

it was observed that a mutual reduction in either CALR or CD47 when one or the other is repressed,

suggesting the influence on each other pathways. Reduction in CD47 expression resulted in higher

rate of cell death when cell were treated with cytotoxic drugs, however due to the concomitant

reduction of CALR the overall therapeutical effect is reduced. Leaving both CALR and CD47 intact on

the myeloid cancer cells but interfering with the CD47/SIRPα interaction instead, could actively

signal cell death messages in these myeloid neoplasms via monoclonal antibodies, and let the pro-

phagocytic message conducted by CALR to prevail. The overexpression of CD47 in MDS, PV and MF

could represent the novel target to block via monoclonal anti-CD47 antibody in monotherapy or in

combination to enhance treatment efficacy in myeloid cancers.

4

Contents Acknowledgements ................................................................................................................................. 1

Abstract ................................................................................................................................................... 2

Tables ...................................................................................................................................................... 8

Figures ..................................................................................................................................................... 8

Abbreviations list .................................................................................................................................. 11

Publications ........................................................................................................................................... 14

Conference posters ............................................................................................................................... 14

Chapter 1 – Introduction ................................................................................................................ 15

1.1: Myeloid Malignancies ............................................................................................................... 16

1.2: Characterisation of Different Myeloid Malignancies ................................................................ 16

1.3: Myelodysplastic syndromes (MDS) Prognostic Scoring ............................................................ 18

1.3.1: Initial Prognostic Scoring Generated by The French-American-British leukaemia and the World Health Organisation ................................................................................................................... 18

1.3.2: International Prognostic Scoring System and WHO Prognostic Scoring System .................. 19

1.4: WHO Classification (2001 – 2016)............................................................................................. 20

1.5: Myeloproliferative Neoplasms: WHO Classification (2001 – 2016) .......................................... 23

1.5.1: Essential Thrombocythaemia (ET) ........................................................................................ 24

1.5.2: Polycythaemia Vera (PV) ....................................................................................................... 25

1.5.3: PRE-MF and myelofibrosis (MF) ............................................................................................ 27

1.6: Targeted drugs for treatment ................................................................................................... 29

1.6.1: Azacytidine ............................................................................................................................ 29

1.6.2: Ruxolitinib ............................................................................................................................. 30

1.7: Immunosurveillance of cancer .................................................................................................. 31

1.8: Calreticulin ................................................................................................................................ 32

1.8.1: CALR in antigen presentation ............................................................................................... 36

1.8.2: CALR in cancer signalling ....................................................................................................... 37

1.8.3: Prognostic and predictive value of CALR .............................................................................. 38

1.9: CD47 .......................................................................................................................................... 39

1.9.1: Mechanism of function and intracellular signalling .............................................................. 41

1.9.2: CD47 pathways targeted for therapeutic intervention ........................................................ 43

1.9.3: Anti-CD47 approaches to treat cancer and AML .................................................................. 44

5

1.10: Purposed aims of research.................................................................................................... 47

Chapter 2: ............................................................................................................................................ 48

Methods ............................................................................................................................................... 48

2.1: Ethics ......................................................................................................................................... 49

2.2: Cell Line Models of MDS & MPN ............................................................................................... 49

2.3: Cell Culture ................................................................................................................................ 50

2.3.1: Testing cell lines for mycoplasma via qPCR ................................................................................ 51

2.4: Cell counting and seeding ......................................................................................................... 52

2.5: Growth curves ........................................................................................................................... 52

2.6: Drug titrations and drugging of cell lines .................................................................................. 53

2.7: Cell viability assay ..................................................................................................................... 54

2.8: Harvesting of cells to undertake RNA and Protein Analysis ..................................................... 55

2.9: Extraction of RNA from cell lines .............................................................................................. 55

2.10: cDNA synthesis from RNA ..................................................................................................... 56

2.11: Quantitative PCR (qPCR) ....................................................................................................... 57

2.12.1: Western blotting ................................................................................................................... 58

2.12.2: Buffer formulations for SDS-PAGE and Western blot experiments ...................................... 59

2.12.3: Protein isolation and quantification in preparation for SDS-PAGE and Western blotting ... 60

2.12.4: SDS-PAGE .............................................................................................................................. 61

2.12.5: Transfer to PVDF membrane ................................................................................................ 62

2.12.6: PVDF membrane blocking and antibody incubation ............................................................ 62

2.13: Fractionations of protein in cells .......................................................................................... 64

2.14: CALR and CD47 Knockdowns ................................................................................................ 65

2.15: Patient Samples Processing and Analysis ............................................................................. 66

2.15.1: Isolation of Peripheral Blood Mononuclear cells (PBMCs) from human donors .................. 66

2.15.2: Processing and collection of mononuclear cells from patient blood samples ..................... 67

2.15.3: Extraction of RNA using TRIzol® method from PBMCs ......................................................... 68

2.15.4: Clean-up of PBMC RNA obtained using the TRIzol® extraction method .............................. 69

2.16: Data Analysis and Statistics ................................................................................................... 70

2.16.1: Calculating relative gene expression .................................................................................... 70

2.16.2: Calculating relative protein expression ................................................................................ 71

Chapter 3: Establishing basal expression levels and location percentage of CALR and CD47 in MDS and MPN cell line models ..................................................................................... 72

3.1: Introduction .............................................................................................................................. 73

6

3.2: Aims........................................................................................................................................... 76

3.3: Methods .................................................................................................................................... 76

3.3.1: Cell culture and drug treatment ........................................................................................... 76

3.3.2: Protein analysis ..................................................................................................................... 77

3.3.3: Gene analysis ........................................................................................................................ 77

3.3.4: Fractionisations ..................................................................................................................... 77

3.4: Results ....................................................................................................................................... 77

3.4.1: CALR and CD47 changes and impact on survival in cell line models .................................... 77

3.4.2: Alterations in CALR and CD47 mRNA expression in untreated & treated cell lines ............. 78

3.4.3: CALR and CD47 protein expression of untreated and treated cell line models ................... 84

3.4.4: CALR and CD47 localization in MDS in response to treatment ............................................. 86

3.4.5: Examining changes in cellular locations of CALR and CD47 in intermediate MDS/MPN ...... 90

3.4.6: Examining changes in cellular locations of CALR and CD47 in MPN ..................................... 92

3.5: Discussion .................................................................................................................................. 95

Chapter 4: ......................................................................................................................................... 100

CALR & CD47 expression in patient with MDS and MPN .................................................... 100

4.1: Introduction ............................................................................................................................ 101

4.2: Aims ......................................................................................................................................... 102

4.3: Methods .................................................................................................................................. 102

4.3.1: Processing of patient samples ............................................................................................. 102

4.3.2: Protein analysis ................................................................................................................... 103

4.3.3: Gene analysis ....................................................................................................................... 103

4.3.4: Fractionisations ................................................................................................................... 103

4.4: Results ..................................................................................................................................... 103

4.4.1: Initial analysis of Total RNA & Protein Expression of CALR and CD47 in MDS and MPN patient samples 103

4.4.2: In-depth look at MDS disease severity sub-groups ............................................................. 105

4.4.4: Examining changes in cellular locations of CALR and CD47 in MDS subgroups ................... 107

4.4.5: In-depth look at MPN untreated Vs treated patient samples ............................................. 108

4.4.6: Examining the cellular locations of CALR and CD47 between different MPN subgroups .... 110

4.5: Discussion ................................................................................................................................ 111

Chapter 5: ......................................................................................................................................... 114

Investigating whether CALR and CD47 work together directly or independently in determining response to therapy within MDS and MPN cell models ............................... 114

5.1: Introduction ............................................................................................................................ 115

7

5.2: Aims ......................................................................................................................................... 116

5.3: Methods .................................................................................................................................. 117

5.3.1: Cell culture .......................................................................................................................... 117

5.3.2: Transfections ....................................................................................................................... 117

5.3.4: Gene Expression qPCR Analysis ........................................................................................... 117

5.3.5: Western Blotting Analysis .................................................................................................... 118

5.4: Results ..................................................................................................................................... 119

5.4.1: CALR and CD47 changes and impact on survival in transfected MDS & MPN cell line models . 119

5.4.2: CALR silenced transfection efficiency in cell line models .................................................... 120

5.4.5: CALR and CD47 silenced cells survival rate upon treatment with chemotherapeutics ....... 128

5.5: Discussion ................................................................................................................................ 129

Chapter 6: Conclusions ....................................................................................................... 132

6.1: Conclusions ............................................................................................................................. 133

6.2: Future studies ......................................................................................................................... 136

References .......................................................................................................................................... 137

Appendix ............................................................................................................................................. 152

8

Tables

Table 1: WHO classification of MDS

Table 1: 2016 WHO Diagnostic Criteria for ET. All four major criteria required for diagnosis or first

three major criteria and the minor criteria

Table 2: World Health Organisation’s 2016 PV diagnostic criteria. PV can be diagnosed either by the

presence of all three major criteria or with major criteria 1 and 2 with the minor criteria

Table 3: 2016 WHO Diagnostic Criteria for pre-MF. Diagnosis requires meeting all three major and

one minor criteria. Minor criteria to be confirmed twice on consecutive investigation

Table 5: 2016 WHO Diagnostic Criteria for overt MF. Diagnosis requires meeting all three major and

one minor criteria. Minor criteria to be confirmed twice on consecutive investigations

Table 6: Table of MDS cell line models (MOLM-13/SKM-1), intermediate cell line model (K562) and

MPN cell line models (HEL-92/GDM-1)

Table 7: Table of primer sequence use for GAPDH, CALR and CD47

Table 8: Buffer formulations for SDS-PAGE and Western blot experiments

Table 9: Antibodies and dilutions used in Western blot experiments

Figures

Figure 1.1: DNA methylation of azacytidine

Figure 1.2: The constitutive activity of the JAK-STAT signalling pathway, present within the

hematopoietic stem cells of patients with myelofibrosis

Figure 1.3: A potential model for CALR network signalling.

9

Figure 1.4: Linear diagram of human calreticulin indicating its amino-terminal (N-CALR), globular (P-

CALR), and carboxy-terminal (C-CALR) domains and the location of the sequences predicted by

computer modelling to determine an intrinsically disordered structure

Figure 1.5: The impact of the SIRPα–CD47 checkpoint and its blockade

Figure 2.1: Depiction of western blots antibodies on a membrane and the fluorescent image of

proteins

Figure 2.2: Depiction of a Bradford assay standard curve

Figure 3.1: Macrophages CD47/SIRPα interactions

Figure 3.2: Overall percentage survival rate of each cell line model treated with azacytidine and

ruxolitinib

Figure 3.3: Expression of untreated and treated MOLM-13 and SKM-1 cell line models

Figure 3.4: Expression of untreated and treated K562 cell line models

Figure 3.5: Expression of untreated and treated HEL-92 and GDM-1 cell line models

Figure 3.6: Representative example of western blot samples of CALR, CD47 and GAPDH proteins

Figure 3.7: Expression of CALR and CD47 protein in untreated Vs treated cell line models

Figure 3.8: Percentage of cellular protein location in untreated Vs treated MOLM-13 cell line models

Figure 3.9: Percentage of cellular protein location in untreated Vs treated SKM -1 cell line models

Figure 3.10: Percentage of cellular protein location in untreated Vs treated K562 cell line models

Figure 3.11: Percentage of cellular protein location in untreated Vs treated HEL-92 cell line models

Figure 3.12: Percentage of cellular protein location in untreated Vs treated GDM- 1 cell line models

10

Figure 4.1: Total RNA and protein fold expression of CALR and CD47 in healthy controls Vs MDS and

MPN patients

Figure 4.2: Total RNA and protein fold expression of CALR and CD47 in healthy controls Vs MDS

subgroups

Figure 4.3: Membrane expression of CALR and CD47 protein in MDS subgroups Vs control

Figure 4.4: Total protein fold expression of CALR and CD47 in untreated Vs treated and treated MPN

patients

Figure 4.5: Total protein fold expression of CALR and CD47 in untreated vs treated MPN subgroups

Figure 4.6: Membrane protein expression of CALR and CD47 in untreated vs treated MPN subgroups

Figure 5.1: Representative example of western blot samples of CALR, CD47 and GAPDH proteins

Figure 5.2: Survival rate percentage in transfected cell lines

Figure 5.3: Overall mRNA CALR expression in each cell line model treated with CALR silencing siRNAs

Figure 5.4: Overall mRNA CD47 expression in each cell line model treated with CALR silencing siRNAs

Figure 5.5: Overall mRNA CD47 expression in each cell line model treated with CD47 silencing siRNAs

Figure 5.6: Overall mRNA CALR expression in each cell line model treated with CD47 silencing siRNAs

Figure 5.7: Cell survival upon treatment with chemotherapeutics in MDS ad MPN cell lines

11

Abbreviations list

AML - Acute myeloid leukaemia

BAK1 - BCL2 antagonist/killer 1

BAX - Apoptosis regulator

Ca2+ - Calcium

CALR – Calreticulin

CANX - calnexin

CASP8 - Caspase 8

CBD - C-terminal binding

CMML - Chronic myelomonocytic leukaemia

CMPD - Chronic myeloproliferative disorders

eIF2α - Eukaryotic translation initiation factor 2 subunit alpha

ER - Endoplasmic reticulum

ET - Essential thrombocythemia

FAB - French-American-British

HSP70 - Heat shock protein family A

HSP90B1 - Heat shock protein 90 beta family A

Hu5F9-G4 - Humanised anti-CD47 antibody

Ig - Immunoglobulin

12

IPSS - International Prognostic Scoring System

IPSS-R - IPSS ranking system

ISR - Integrated stress response

ITAM - Immunoreceptor tyrosine activation motifs

ITIMs - Immunoreceptor tyrosine inhibitory motifs

JAK inhibitor - Janus kinase inhibitor

JAK/STAT - Janus kinase/signal transducer which is an activator of transcription

LRP - Lipoprotein receptor-related protein

LRP1 - LDL receptor-related protein 1

MBL - Mannose-binding lectin

MDS - Myelodysplastic syndromes

MHC - Major histocompatibility complex

MF - Myelofibrosis

MPL - Thrombopoietin receptor MCD - Mast cell disease

MPN - Myeloproliferative neoplasms

PD-L1 - Anti-programmed death-ligand 1

PDIA3 - Protein disulfide isomerase family A member 3

PERK - Eukaryotic translation initiation factor 2 alpha kinase 3

PLC - Peptide-loading complex

13

PTPases - Protein tyrosine phosphatases

PV - Polycythaemia vera

qPCR - Quantitative PCR

RA - Refractory anaemia

RAEB - Refractory anaemia with excess blasts

RAEB-T - Refractory anaemia with excess blasts in transformation

RARS - Refractory anaemia with ring sideroblasts

SIRPs - Signal regulatory proteins

SHPS-1 - Src homology 2 domain-containing protein tyrosine phosphatase substrate 1

SNAP25 - Synaptosome-associated protein 25.

SH2 - Src homology region 2

SHP - domain-containing phosphatase 1

THBS1 - Thrombospondin 1

TSPs – Thrombospondins

w/t - Wild type

WHO - World Health Organization

VAMP1 - Vesicle-associated membrane protein 1

14

Publications

1. Lally J, and Boasman K, et al. 2019. GATA-1 a potential novel biomarker for the differentiation of

essential thrombocythaemia and myelofibrosis. J Thromb Haemost. P14433

2. Boasman K, Simmonds MJ, Graham C, Saunthararajah Y, Rinaldi CR. 2018. Using PU.1 and Jun

dimerization protein 2 transcription factor expression in myelodysplastic syndromes to predict

treatment response and leukaemia transformation. Ann Hematol.

3. Boasman K, Simmonds MJ, Rinaldi CR, Rinaldi C. 2018.CALR and CD47: An Insight into their roles in

the disease pogression of MDS and MPN. J Blood Disord Transfus. (10).1.

Conference posters

American Society of Clinical Oncology Asia – 2019 (Bangkok): Expression of CD47 and CALR in

myeloproliferative neoplasms: Potential new therapeutic targets.

European Haematology Association – 2019 (Amsterdam): Expression of CD47 & CALR in

myeloproliferative neoplasms: potential new therapeutical targets.

British Society of Haematology – 2019 (Glasgow): Expression of CD47 and CALR in myeloproliferative

neoplasms: potential new therapeutical targets.

American Society of Clinical Oncology – 2018 (Chicago): PU.1 and JDP2 Expression in Myelodysplastic

Syndromes to Predict Treatment Response and Leukaemia Transformation.

European School of Haematology – 2018 (Dublin): Calreticulin and CD47 expression in a MPN cell

system K562: a model for outcome of therapeutic response.

British Society of Haematology – 2018 (Liverpool): Calreticulin & CD47 expression in a MDS/MPN cell

system K562: a model for outcome of therapeutic response.

British Society of Haematology – 2018 (Liverpool): PU.1 Expression Correlates with patient disease

status In Myelodysplastic Syndromes, A Potential New Prognostic Marker

European Haematology Association – 2017 (Madrid): Role of pro-phagocytic calreticulin and anti-

phagocytic CD47 in MDS and MPN models treated with azacytidine or ruxolitinib.

https://www.ncbi.nlm.nih.gov/pubmed/?term=Lally%20J%5BAuthor%5D&cauthor=true&cauthor_uid=30889303

https://www.ncbi.nlm.nih.gov/pubmed/?term=Boasman%20K%5BAuthor%5D&cauthor=true&cauthor_uid=30889303

https://www.ncbi.nlm.nih.gov/pubmed/30889303/?

15

Chapter 1 – Introduction

16

1.1: Myeloid Malignancies Myelodysplastic syndromes (MDS), Myeloproliferative neoplasms (MPN) and acute myeloid leukaemia

(AML) are Myeloid malignancies, which are a group of bone marrow diseases arising from clonal

disorders of haematopoietic stem cells. MDS and MPN generally effect the older populations. MDS

median age of diagnosis is 65-70, with patients under the age of 50 accounting for less than 10% of the

disease. Annually, 4 in 100,000 people are diagnosed with MDS, which increases to 40-50 people per

100,000 in the over 70s (Adès et al, 2014). MPNs have a similar age of diagnosis of between 59-65 years

with 1.6 in 100,000 diagnosed annually (Ayalew Tefferi, 2016; Titmarsh et al., 2014). MDS are

characterised by bone marrow failure and peripheral cytopenia’s in myeloid progenitors (Boasman,

2019), while MPN are characterised by differentiation of mature myeloid cell types (Lally et al., 2019).

MDS and MPN both show tendencies to develop into AML as the diseases progress (Murati et al., 2012).

In the UK AML is the second most prevalent leukaemia. AML is characterised by excessive myeloid blast

cell proliferation within the bone marrow, resulting in abnormal haematopoiesis (Yang et al,, 2018).

Whilst much has been achieved to classify these diseases over the years there are still challenges

associated with disease progression and treatment as laid out below.

1.2: Characterisation of Different Myeloid Malignancies

The formation of the components of blood from hematopoietic stem cells is called haematopoiesis,

with genetic changes and alterations in methylation status known to disrupt haematopoiesis. These

alterations in Haematopoiesis can lead to a variety of haematological neoplasms including

leukaemia and the chronic myeloid disorders (Orkin, 2000). In 2008, the World Health Organisation

re-classified Myelodysplastic syndromes (MDS), Myeloproliferative neoplasms (MPN) and acute

myeloid leukaemia (AML), to try and gain an International recognised criterion to aid diagnosis and

avoid overlap. Often myeloid neoplasms are characterised by dysmyelopoiesis (a hematologic

disorder characterised by the presence of cytopenia’s in the blood and dysplastic cells), abnormal

differentiation of hematopoietic cells, increased apoptosis with ineffective haematopoiesis, cytopenias

17

(lower than normal/expected blood cells) or leucocytosis and in particular MDS and MPN tend to

transform into AML (Korn & Méndez-Ferrer, 2017).

MDSs are classified based on the extent of cytopenia, the level of blast infiltration of the bone marrow,

and abnormal chromosomal structure (Gill et al., 2016). MPNs are characterized by overproduction of

mature blood cells (Park et al., 2015). Essential thrombocythemia (ET), polycythaemia vera (PV) and

myelofibrosis (MF) are the three most common forms of MPN (Vardiman et al., 2009), which often have

overlapping presentations but have been well defined from each other in the WHO revision of myeloid

classification in 2016 (Arber et al., 2016). In MPNs somatic mutations are common. In most cases the

patient harbour a Janus kinase inhibitor 2 (JAK2) mutation, which affects the JAK/STAT pathway

signalling (Campbell et al., 2006). Other mutations are also prevalent in MPN, such as the calreticulin

(CALR) mutations or cMPL mutations (Jonas & Greenberg, 2015).

The type and severity of MDS and MPN heavily effects the prognosis of the disease. Different

classifications have been used to indicate prognosis and clinical features, French-American-British [FAB]

classification and the World Health Organization [WHO] classification together with prognostic scoring

systems, such as the International Prognostic Scoring System (IPSS), have been used for over 25 years

(Jonas & Greenberg, 2015; Vardiman, 2012; Vardiman et al., 2009). Clinical indicators including: signs

and symptoms of anaemia, infection and bleeding complications predominate, with systemic symptoms

or features of autoimmunity being seen in some patients; perhaps indicative of disease pathogenesis

which result from the vast heterogeneity in these disorders (Adès., 2014).

Prognostic scoring systems used in MDS and MPN are mainly based on bone marrow/peripheral blood

features, cytogenetics and some clinical manifestations such as systemic symptoms or splenomegaly.

Although the presence of different mutations has been linked to these conditions, there are currently

no specific genetic markers to highlight specific subtypes or specific stages of these conditions that

predicts transformation and disease progression. As such, the standard IPSS revised system (IPSS-R) is

currently used in MDS, to rank risk factors for the disease into: very low, low, intermediate, high and

18

very high risk scores (Vardiman, 2012). A similar scoring system is used in MPN, in particular the

Dynamic International Prognostic Scoring System (DIPSS) prognostics on MF, describing potential

different stages in its progression (Gangat et al., 2011).

1.3: Myelodysplastic syndromes (MDS) Prognostic Scoring

1.3.1: Initial Prognostic Scoring Generated by The French-American-British leukaemia and the World Health Organisation

The French-American-British (FAB) leukaemia working group provided the first publication classifying

MDS (Bennett et al., 1982). With it taking nearly two decades later for a group of hematopathologists

and hematologists to write the World Health Organization (WHOs) first classification of hematologic

neoplasms including MDS, AML, MPNs and MDS/MPN overlap neoplasms which came out in 2001

(Vardiman et al., 2002).

The FAB group recognized 5 subtypes of MDS, refractory anaemia with excess blasts (RAEB), chronic

myelomonocytic leukaemia (CMML), refractory anaemia (RA), refractory anaemia with ring

sideroblasts (RARS) and refractory anaemia with excess blasts in transformation (RAEB-T) (Bennett et

al., 1982).

The FAB classification showed some issues with its classification system. Firstly, clarification on how

the RA and RARS subtypes were formed was necessary as to support how such patients should be

categorized. RA and RARS were characterized by dyserythropoiesis showing marrow blasts under 5%,

with “almost normal granulocytic and megakaryocytic morphology” (Bennett et al., 1982). However,

some MDS patients with marrow blasts under 5% also displayed multilineage dysplasia and reported

lower survival rates compared to patients with dyserythropoiesis only (Rosati et al., 1996). Secondly,

the FAB categorised that for RAEB the blast range was 5-20% and could be used as an important

prognostic factor. However, some authorities reasoned 5-20% was too broad of a range to define a

prognostically relevant subgroup (Greenberg et al., 1997). Third, RAEB-T survival rate and response to

19

treatment was similar to that of AML patients. It was suggested that RAEB-T should be encompassed

with AML and not MDS (Bernstein et al., 1996). Fourth, there was debate over whether CMML should

be reclassified or subdivided into two subgroups. CMML is clinically and morphologically

heterogeneous with some patients showing myeloproliferative features (marked leukocytosis and

splenomegaly but minimal hematopoietic dysplasia) whereas others have MDS features (modest

leukocytosis, no splenomegaly and marked dysplasia) (Michaux & Martiat, 1993). Fifth, a lack of

minimal diagnostic criteria was raised by some clinicians and pathologists who had patients with

suggested MDS but inconclusive morphologic features (Vardiman et al., 2002).

A number of prognostic scoring systems arose from studies trying to identify prognostic factors

beyond the FAB classification, including the most prominently used International Prognostic Scoring

System (IPSS) (Greenberg et al., 1997; Vardiman, 2012). Inclusion of the IPSS enhances the risk

stratification of patients with the various prognostic significance of the FAB subtypes, through the

scoring of the percentage of bone marrow blasts, number of cytopenia’s, and specific cytogenetic

abnormalities as independent prognostic variables (Vardiman, 2012).

1.3.2: International Prognostic Scoring System and WHO Prognostic Scoring System

For the diagnosis and treatment of MDS the IPSS has been the most widely used prognostic scoring

system. It was first published by FAB in 1997, intended for use with FAB classification system and has

been updated several times. Addition of the IPSS to the FAB classification system has been proven to

have reproducible and prognostic value. IPSS states that the patients risk factor is decided on points

scoring where it is scored on a scale from 0 to 2. Low risk - (0), Intermediate risk – 1 - (0.5-1),

Intermediate risk – 2 – (1.5-2) and High risk - (≥2). The points are scored by the percentage of blasts in

the bone marrow, - Karyotype (both scored from zero to 1) and cytopenia’s (scored as zero or 0.5). A

patient’s risk is classified according to the total score (Greenberg et al., 1997).

20

In 2012, a revised IPSS system (IPSS-R) was released now comprised of five risk groups (very low, low,

intermediate, high and very high). This is a revised tier system which gives greater specific detail into

cytogenetic prognostic categorization and cut-offs for haemoglobin levels, platelet counts and

neutrophil counts. A patient’s risk factor and prognostic scoring is now calculated by the severity of

cytopenia’s along with their bone marrow blast percentage (Greenberg et al., 2012). The IPSS and

IPSS-R are used only at the time of diagnosis, whereas the WHO Prognostic Scoring System (WPSS),

developed in 2005 based on the 2001 WHO classifications for MDS and revised in 2011 (Malcovati et

al., 2011), can constantly adjust allowing for patients to be reassigned categories as their disease

progresses. The WHO scoring system calculates a patients risk factor using a similar points system to

the IPSS-R; very low - (0) , low - (1) , intermediate - (2), high – (3-4) and very high - (≥5). The points are

scored by the WHO disorder classification (0-3), Karyotype (0-2) and presence or absence of severe

anaemia (0-1) (Malcovati et al., 2011).

1.4: WHO Classification (2001 – 2016)

The WHO classification was established to be used as a guide to combine all available information;

clinical, morphologic, cytochemical, immunophenotypic, genetic and other biologically significant

information. To aid in the specific diagnosis of each disease (Harris et al., 1999), the WHO modified

and expanded on the previously published data on the FAB subtypes, with their 2001 classification

(Vardiman et al., 2002). This was then further expanded on in the 2008 reclassification of

hematopoietic tumours, which included; morphology, immunophenotyping, genetics, morphologic

and histologic photomicrographs, and clinical information (Vardiman et al., 2009) (table 1). The WHO

published their latest classification in 2016 updating some key points (Arber et al., 2016).

21

Table 4: WHO classification of MDS (Vardiman, 2012).

Disease entity Blood findings Bone marrow findings

Refractory cytopenias with

unilineage dysplasia (RCUD):

refractory anemia (RA), refractory

neutropenia (RN), refractory

thrombocytopenia (RT)

Unicytopenia or

bicytopeniaa

No or rare

blasts (<1%)b

Unilineage dysplasia: ≥10% of the cells in

one myeloid lineage

<5% blasts

15% of erythroid precursors are ring

sideroblasts

Refractory anemia with ring

sideroblasts (RARS)

Anemia

No blasts

≥15% of the erythroid precursors are ring

sideroblasts

Dyserythropoiesis only

<5% blasts

Refractory anemia with multilineage

dysplasia (RCMD)

Cytopenia(s), no

or rare blastsb

No Auer rods

<1 × 109/L

monocytes

Dysplasia in >10% of cells in 2 or more

lineages

<5% blasts in marrow

No Auer rods

<1 × 109/L monocytes

Refractory anemia with excess

blasts-1 (RAEB-1)

Cytopenia(s)

<5% blasts

No Auer rods

<1 × 109/L

monocytes

Unilineage or multilineage dysplasia

5–9% blasts

No Auer rods

Refractory anemia with excess

blasts-2 (RAEB-2)

Cytopenia(s)

5–19% blasts

Auer rods ±c

Unilineage or multilineage dysplasia

0–19% blasts

Auer rods ±c

https://www.sciencedirect.com/science/article/pii/S0145212612003414?casa_token=nM7gcBNYBGQAAAAA:h0wIlY4igdZX0EIn8oNITwfZ0xjwrBhiJ9eCM9_Bee-7IOheXlNxgkkt0cVYRPrbbRoS0DTjdw#tblfn0005





22

<1 × 109/L

monocytes

MDS, unclassifiable (MDS-U) Cytopenias

<1% blastsb

Unequivocal dysplasia in less than 10% of

cells in one or more myeloid lines when

accompanied by a cytogenetic

abnormality considered as presumptive

evidence for a diagnosis of MDS

<5% blasts

MDS associated with isolated del(5q) Anemia

Usually normal

to elevated

platelets

No or rare

blasts

Normal to increased megakaryocytes with

hypolobated nuclei

<5% blasts del(5q) is the sole cytogenetic

abnormality

No Auer rods

a= Bicytopenia may be occasionally observed. Cases with pancytopenia should be classified as MDS-U

b = If marrow blast % is less than 5%, but there are 2–4% myeloblasts in blood, the diagnosis is RAEB-

1. Cases of RCUD and RCMD with 1% blasts in blood are classified as MDS-U

c =Cases with Auer rods and <5% myeloblasts in the blood and <10% in the marrow should be

classified as RAEB-2

In the latest 2016 classification WHO update (8 years after the last edition), several aspects of MDS

were updated. Flow cytometry may be used to identify atypical antigen expression on the cell surface,

when assessing whether pancytopenia occurs as a result of MDS. This is completed by checking that

there is no increase in the blast percentage resulting from immature proliferation of cells. At least


23

10% dysplastic erythroid, granulocytic, or megakaryocytic cells should be identified (Goasguen et al.,

2016; Kern et al., 2010).

1.5: Myeloproliferative Neoplasms: WHO Classification (2001 – 2016)

Myeloproliferative neoplasms weren’t classified until the 2008 revision of the 2001 WHO

classification. The WHO originally classified myeloproliferative disorders into four groups, 1- chronic

myeloproliferative disorders (CMPD), 2- myelodysplastic syndromes (MDS), 3- MDS/MPN crossover

diseases and 4- mast cell disease (MCD). In this classification system more recent subclassifications of

MPN, PV, ET & MF, were all classified under CMPD grouping along with chronic myeloid leukaemia

and chronic neutrophilic leukaemia, chronic eosinophilic leukaemia (and the hypereosinophilic

syndrome) and CMPD unclassified (Vardiman et al., 2002). A shared effective clonal

myeloproliferation was the major criteria for classification in the CMPD group, with the grouping

based purely on this feature, with other variability initially not take into account (Tefferi & Vardiman,

2008).

Research into molecular markers of CMPD including, the most prevalent JAK somatic mutation

(JAK2V617F) in 2005, led to a revised WHO classification in 2008. The main changes to the 2001

classification including myeloproliferative diseases being renamed to myeloproliferative neoplasms

(MPNs), and a separate group being created for eosinophilia and abnormalities of PDGFRA, PDGFRB

and FGFR1 for myeloid neoplasms. Due to its formation form clonal stem cells disorder, MCD was

grouped with the MPNs (Vardiman et al., 2009).

Along with the new revision of myeloid classification in 2008, MPNs were also given specific

diagnostic criteria. Mutations such as the JAK2V617F mutation provides additional information to

assist with diagnosis, although these mutations are not specific to any MPN they are used in

conjunction with other diagnostic criteria to reject other diagnoses. MPN is generally split into 3

subgroups: MF, ET and PV. There is debate on the specific determination of the individual MPNs

24

during diagnosis. Sometimes patients’ symptoms overlap making it hard to distinguish between

MPNs, leading to terms of “true ET” or “pre-fibrotic MF” (Campbell et al., 2009). In addition, after the

2008 revision of myeloid classification it was shown that there was a potential under-diagnosis of PV.

Patients with the disease transformation from MF to AML with consistent bone marrow morphology

stated in 2008 classification showed significantly increased risk of disease transformation when

compared to the disease transformation of PV to AML (Barbui et al., 2014). This led to the revised

2016 WHO MPN classification of myeloid neoplasms stating that the diagnostic criteria for PV now has

greater attention to bone marrow morphology and haemoglobin levels have been reduced (Arber et

al., 2016).

The WHO refined this issue in the new 2016 revision of myeloid classification to prevent misdiagnoses

of MF and ET, the updated criteria distinguished pre-MF from “true” ET (Arber et al., 2016). In the

new 2016 classification the diagnostic criteria was updated to recognise the CALR mutation found in

2013 (Klampfl et al., 2013; Nangalia et al., 2013a). Diagnostic criteria and guidance on myelofibrosis

grading was used to distinguish the different ET, pre-MF, and overt-MF subgroups of MPN (Arber et

al., 2016).

From the 2016 revision of myeloid neoplasms MPNs are now split into 4 sub-groups; with diagnosis of

each sub-group requiring elimination of the other myeloproliferative disorders (further details of each

subgroup are detailed below).

1.5.1: Essential Thrombocythaemia (ET)

This MPN subgroup is characterised by increases in platelet count circulating within blood (> 450 x 109

/L) combined with an increase of megakaryocytes - this is the criteria used to diagnose ET (Tefferi &

Vardiman, 2008). ET when compared to other MPNs has a longer overall median survival rate (19.8

years), whereas PV is (13.5 years). Although, ET has a high survival rate it is still much lower than the

age matched control population (Tefferi et al., 2014). Longer survival rates do not exempt ET from the

possibility of myeloid or leukemic disease transformation, with disease transformation seen within 0 -

25

10 years after diagnosis at 9.1% and 1.4%, respectively, increasing during 11-20 years (28.3% and 8.1%

chance of transformation, respectively) and in 21-30 years after disease onset the chance of disease

transformation after diagnosis significantly increases to 58.5% and 24.0%, respectively (Wolanskyj et

al., 2006). Criteria for ET diagnostics is shown in Table 2.

Table 5: 2016 WHO Diagnostic Criteria for ET. All four major criteria or the first three major criteria and the minor criteria

are required for diagnosis (Arber et al., 2016).

ESSENTIAL THROMBOCYTHAEMIA

Major Criteria

1 Platelet count ≥ 450 x 109 /L

2

BM biopsy showing proliferation mainly of the megakaryocyte lineage with

increased numbers of enlarged, mature megakaryocytes with hyperlobulated

nuclei. No significant increase or left shift in neutrophil granulopoiesis or

erythropoiesis and very rarely minor (grade 1) increase in reticulin fibres

3 Not meeting WHO criteria for CML, PV, MF, MDS or other myeloid neoplasm

4 Presence of JAK2, CALR, or MPL mutations

Minor Criteria 1

Presence of a clonal marker or absence of evidence for reactive

thrombocytosis

1.5.2: Polycythaemia Vera (PV)

PV is a subgroup within MPN, characterised by an elevated number of erythrocytes, greater than 25%

red cell mass above normal (Arber et al., 2016). Diagnostic criteria for PV is shown in Table 3. Vascular

and thrombotic complications can arise from the overproduction of red blood cells potentially having

serious implications for the patient’s quality of life (Stein et al., 2014). Each patient shows common

symptoms such as pruritus, fatigue, night sweats, bone pain, thrombosis, and bleeding

26

(Griesshammer et al., 2015). PV patients median age of diagnosis is between 55 and 60 years old

(Harrison & Lee, 2017), with an estimated 0.86 per 100,000 annual incidence rate (Titmarsh et al.,

2014). PV has the potential to transform into either myelofibrosis (MF) or AML. This transformation

can occur over 10-15 years and occurs in 10-15% and 5-10% of patients ,respectively (Harrison & Lee,

2017).

Table 6: World Health Organisation’s 2016 PV diagnostic criteria. PV can be diagnosed either by the presence of all three

major criteria or with major criteria 1 and 2 with the minor criteria (Arber et al., 2016).

POLYCYTHAEMIA VERA

Major Criteria

1

Haemoglobin > 165/160 g/L (men/women)

OR

Haemocrit > 49/48% (men/women)

OR

Elevated red cell mass > 25% above mean normal predicted value

2

BM biopsy showing hypercellularity for age with trilineage growth

(panmyelosis) including prominent erythroid, granulocytic, and

megakaryocytic proliferation with pleomorphic, mature megakaryocytes

(differences in size). This criterion may not be required in cases with sustained

absolute erythrocytosis, haemoglobin levels > 185/165 g/L and haematocrit

> 49.5%

3 Presence of JAK2V617F or similar mutation

Minor Criteria 1 Subnormal serum EPO level

27

1.5.3: PRE-MF and myelofibrosis (MF)

In the 2008 revision of the WHO myeloid classification PRE-MF and MF were classified together under

the integrated group MF. After the 2016 revision of myeloid classification PRE-MF and MF were

separated into different forms (Arber et al., 2016). Diagnostics criteria for PRE-MF is shown in Table 4.

Myelofibrosis can be presented as primary MF (if initial presentation), or secondary if MF is derived

from the progression of ET and PV (Curto-Garcia et al, 2018). Patients with PRE-MF can stay in a PRE-

MF state for many years and mimic the presentation of ET and PV. To distinguish, the presence of

leucoerythroblastosis cannot be present and the histopathological features have particular prominence

in discriminating pre-MF from ET or PV (Curto-Garcia et al., 2018).

Table 7: 2016 WHO Diagnostic Criteria for pre-MF. Diagnosis requires meeting all three major and one minor criteria. Minor

criteria to be confirmed twice on consecutive investigation (Arber et al., 2016).

PRE-MF

Major Criteria

1

Megakaryocytic proliferation and atypia, without reticulin fibrosis > grade 1,

accompanied by increased age-adjusted bone marrow cellularity, granulocytic

proliferation, and often decreased erythropoiesis

2 Not meeting WHO criteria for CML, ET, PV, MF, MDS, or other myeloid

neoplasm

3

Presence of JAK2, CALR, or MPL mutations or in the absence of these

mutations, presence of another clonal marker, or absence of reactive

myelofibrosis

Minor Criteria

1 Anaemia not attributed to a comorbid condition

2 Leucocytosis: ≥ 11 × 109/L

28

PRE-MF

3 LDH increased to above upper normal limit of institutional reference range

4 Palpable splenomegaly

MF is the least common of the three classical non-CML MPNs, as classified in the 2008 revision of

myeloid classification. MF accounts for around 0.47 new cases per 100,000 of diagnosed MPNs

(Titmarsh et al., 2014). Diagnostic criteria for MF are shown in Table 5. The median survival rate for MF

patients is 5.9 years, making MF the most serious MPN, when compared to PV and ET (Tefferi et al.,

2014). MF patients have a 25% potential of blast transformation and progression to AML, lowering the

prognostic outcome (Harrison & Lee, 2017). The physical symptoms of MF include sweating, pruritus,

fever and bone pain these result from a combination of bone marrow failure and/or splenomegaly,

leading to MF and causing a large negative impact on the quality of life for the patient (Harrison & Lee,

2017).

Table 8: 2016 WHO Diagnostic Criteria for overt MF. Diagnosis requires meeting all three major and one minor criteria.

Minor criteria to be confirmed twice on consecutive investigations (Arber et al., 2016).

MF

Major Criteria

1 Presence of megakaryocytic proliferation and atypia, accompanied by either

reticulin and/or collagen fibrosis grades 2 or 3

2 Not meeting WHO criteria for CML, ET, PV, MDS, or other myeloid neoplasm

3 Presence of JAK2, CALR, or MPL mutation or in the absence of these mutations,

presence of another clonal marker, or absence of reactive myelofibrosis

Minor Criteria 1 Anaemia not attributed to a comorbid condition

29

MF

2 Leucocytosis: ≥ 11 × 109/L

3 LDH increased to above upper normal limit of institutional reference range

4 Palpable splenomegaly

5 Leucoerythroblastosis

1.6: Targeted drugs for treatment

Different treatment options are used in the treatment of myeloid malignancies. Initially most

treatments start with chemotherapy/ targeted drugs before moving onto more extreme therapies if

the patients are able, e.g. stem cell transplants.

1.6.1: Azacytidine

A main drug used in the treatment of MDS is azacytidine. Azacytidines acts via multiple mechanisms

including DNA hypomethylation and cytotoxic effects on abnormal haematopoietic cells in the bone

marrow (Leone, Voso, Teofili, & Lübbert, 2003) (Figure 1.1). Azacytidine is a pyrimidine nucleoside

analogue of cytidine which is phosphorylated intracellularly to become its active form, azacytidine

triphosphate (Jones et al., 1983; Leone et al., 2003). Initial phosphorylation of azacytidine depends on

uridine cytidine kinase which contributes to a lack of resistance seen in certain cancer cell lines (Qin,

Jelinek, Si, Shu, & Issa, 2009). Upon administration of azacytidine it is incorporated into the RNA,

which affects the synthesis of nucleic acids and proteins resulting in cytotoxicity, it has also been seen

to be incorporated into DNA but to a lesser extent (Keating, 2009). Due to its incorporation into RNA

and DNA azacytidine mainly affects rapidly dividing cells like cancerous cells (Santini et al, 2001), while

slow/ non-proliferating cells are relatively unresponsive to the drug (Keating, 2009).

30

Figure 1.1: DNA methylation of azacytidine (Leone et al., 2003)

1.6.2: Ruxolitinib

Ruxolitinib is a janus kinase inhibitor (JAK inhibitor) with selectivity for subtypes JAK1 and JAK2

(Pardanani & Tefferi, 2011) (Figure 1.2). Approved for the treatment of patients with MPN, usually it is

used to treat disease-related splenomegaly or symptoms in adults with myelofibrosis, post PV and

post ET (Gotlib, 2013), which are types of myeloproliferative disorder that affects the bone marrow

(Mesa et al., 2012). PV patients previously treated with hydroxyurea as a first line of treatment may

be eligible for ruxolitinib when there has been an inadequate response to or intolerance of

hydroxyurea (Vannucchi et al., 2015). Patients with PV and ET have been shown to go into complete

molecular remission upon treatment with ruxolitinib after 5 years (Pieri et al, 2015). It dysregulates

activation of Janus kinase/signal transducer which is an activator of transcription in (JAK/STAT)

signalling, which controls the production of blood cells. Stopping the uncontrolled proliferation of

hematopoietic progenitors and cancer cells and stops the generation of a proinflammatory reaction

responsible for symptomatic burden (Cazzola & Kralovics, 2014).

31

Figure 1.2: The constitutive activity of the JAK-STAT signalling pathway, present within the hematopoietic stem cells of

patients with myelofibrosis (Mascarenhas & Hoffman, 2012)

1.7: Immunosurveillance of cancer

The hallmarks of cancer states that, cancers were thought to be initiated from genetic or epigenetic

defects within neoplastic precursor cells. These defects within the cell allow the cell to resist cell death,

over-proliferate and invade local as well as distant sites to generate metastases (Hanahan et al., 2011).

Work around the early 2000s indicated that cancer and its progression are also affected by other factors

such as the hosts immune system (Kaplan et al., 1998; Shankaran et al., 2001). Collected literature

amassed, stating that the immune system eliminates (pre)neoplastic cells as they form, while in some

cases (epi)genetic alterations can allow immunoevasion leading to growth of the (epi)genetic altered

cells (Dunn et al., 2004).

Oncogenesis and disease progression rely on intrinsic alterations and changes in expression of genes

and proteins in cancer cells, allowing a proliferative advantage and evasion of recognition from the

immune system that would signal them for death (Fucikova, Spisek, Kroemer, & Galluzzi, 2020). Current

models in the hallmarks of cancer state that oncogenesis and disease progression occur when increases

32

in genetic expression induces proliferative potential and amass a resistance to cell death. Also, the cells

react in response to disease progression and/or respond to the cancer therapy, resulting in the diseased

cell escaping the immunosurveillance of the hosts immune system (Hanahan et al., 2011; Galluzzi et al.,

2015).

Newly emerging targets of immune system escape include CALR and CD47, and their relationship

regarding one another. Overexpression of calreticulin (CALR) has been documented to produce a pro-

phagocytic signal in solid tumours and it is often counteracted by a concomitant expression of the anti-

phagocytic CD47, reflecting an apoptosis vs survival mechanism in response to chemotherapy which

will be discussed further(Boasman et al., 2019).

1.8: Calreticulin

Calreticulin (CALR) (Figure 1.3) is a calcium (Ca2+)-binding endoplasmic reticulum (ER) protein, which

is involved in many different functions including supporting the movement of secreted and inserted

proteins in the plasma membrane, maintaining homeostasis of proteins that accumulate in the ER,

aiding in the initial reaction to provide anticancer immunity relating to immunogenic cell death and

playing a role in regulated cell death by signalling an antigen-specific immune response to

damaged/irregular cells (Galluzzi et al., 2020; Matsuo et al., 1997; Ren et al., 2019; Saito et al., 1999;

Salvagno & Cubillos-Ruiz, 2019).

33

Figure 1.3: A potential model for CALR network signalling. (Left panel) CALR binds to calcium, actively involved in the

folding of newly synthesized glycoproteins, interacts with other molecular chaperones in the endoplasmic reticulum.

Promotes proliferation, cell cycle progression, migration, invasion and anoikic resistance by CALR regulating; STAT3, STAT5,

AKT, MAPK and FAK cell signalling. Under stress conditions, CALR translocates onto the plasma membrane by Golgi

apparatus-mediated exocytosis, which participates in immunogenic cell death. (Right panel) depicts a myeloproliferative

neoplasm cell with mutated-CALR (exon 9 indel mutations), which induces activation of MPL and JAK2/STAT signalling,

promoting cell proliferation and survival (Machado-Neto et al., 2016).

CALR is 46 kDa Ca2+-binding chaperone protein comprised of 417 amino acids that is principally

located in the lumen of the ER (Michalak, Groenendyk, Szabo, Gold, & Opas, 2009). CALR has a crystal

structure comprised of an extended proline-rich P-domain inserted between N- and C-terminal

domains (Boasman et al., 2019), which has 3 domains each with a specific function (Figure 1.4).

34

Figure 1.4: Linear diagram of human calreticulin indicating its amino-terminal (N-CALR), globular (P-CALR), and carboxy-

terminal (C-CALR) domains and the location of the sequences predicted by computer modelling to determine an

intrinsically disordered structure (indicated by the straight lines). (A) Normal human CALR and factors which may

potentially affect its tertiary structure. (B) Diagram of the structure of representative Type 1 (deletions) and Type 2

(insertions) CALR mutations found in Philadelphia-negative myeloproliferative disorders. The mutations found in these

maladies are all localized in exon 9 encoding the terminal C-region of the protein and encode a truncated (Type 1, top

diagram) or elongated (Type 2, bottom diagram) form of C-CALR (Varricchio et al., 2017).

The highly acidic C-domain found predominantly within the ER is responsible for calcium regulation

and binding, due to its low-affinity and high-capacity for Ca2+ binding. The globular N- terminal lectin-

like globular domain and extended P-domain are primarily responsible for the protein’s chaperone

function due to their high-affinity and low-capacity for Ca2+ binding (Chao & Gotlib, 2014; Michalak et

al., 2009). The C-domain contains the KDEL amino acid sequence which prevents CALR from being

secreted from the ER and ensures retrieval from the Golgi apparatus. Whilst the KDEL signal resides it

35

to the ER it can however be translocated by protein folding to other parts of the cell, including the

plasma membrane, when needed but normally CALR is found in the ER (Guglielmelli et al., 2014).

KDEL-containing proteins bind to KDEL receptors on the ER that affect its translocation. If the KDEL

receptor protein is missing or has a mutated KDEL region cellular mis-localisation can occur.

Localisation studies show that in myeloid cells CALR is upregulated, however when mutated (see

Figure 1.4), CALR stays located in the ER (Guglielmelli et al., 2014).

CALR is closely linked to another chaperone protein calnexin (CALR/CANX). Together these proteins

are a part of a multiple additional quality control system within cells that is responsible for

chaperoning newly synthesised proteins and glycoproteins within the ER. CALR/CANX supports

refolding and is responsible for preventing premature export of misfolded proteins from the ER. Other

chaperones include but are not limited to; heat shock protein family A (HSP70), heat shock protein 90

beta family A (HSP90B1) and protein disulfide isomerase family A member 3 (PDIA3) (Rutkevich &

Williams, 2011). Integrin and Ca2+ signalling in both excitable (neurones, skeletal muscle) and non-

excitable cells (red blood cells) are controlled by CALR (Coppolino et al., 1997). In addition to its ER

chaperone functions, CALR has also been found expressed at the surface of tumour and apoptotic

cells where it is believed to promote phagocytosis (Gardai et al., 2005; Obeid et al., 2007). Klampfl

(Klampfl et al., 2013) and Nangalia (Nangalia et al., 2013b), are the two groups who discovered the

mutation in the CALR gene, located in exon 9. Over 50 different mutations have been identified to

date, the most common mutations are type 1 mutation (52-bp deletion) and type 2 mutation (5-bp

insertion) (Cazzola & Kralovics, 2014). Ligand-independent signalling through JAK-STAT, PI3K and

MAPK pathways have been seen to be activated by CALR mutants via the thrombopoietin receptor

(MPL) (Chachoua et al., 2016). CALR mutations are suspected to have a contributory role in the

pathogenesis of MPNs (Luo & Yu, 2015).

36

Translocation of CALR from the ER lumen to the plasma membrane in cells acts in response

(survive/succumb to stress) to a specific change to the cells integrated stress response (ISR) (Pakos‐

Zebrucka et al., 2016). There are three functional models that trigger the translocation of CALR

throughout the cell. The first for CALR translation of the plasma membrane is driven by anthracyclines

and completed by inactivating the phosphorylation of the eukaryotic translation initiation factor 2

subunit alpha (eIF2α), through eukaryotic translocation initiation factor 2 alpha kinase 3 (PERK)

(Panaretakis et al., 2009). The second involves the pro-apoptotic Bcl-2 family and associated genes

involved in apoptotic signalling, including BCL2 associated X, apoptosis regulator (BAX) and BCL2

antagonist/killer 1 (BAK1), caspase 8 (CASP8) triggering CALR translocation (Panaretakis et al., 2009).

The third module is transport from ER to the Golgi apparatus, where CALR exposure driven by

anthracyclines is inhibited via downregulation of vesicle-associated membrane protein 1 (VAMP1) or

synaptosome-associated protein 25 (SNAP25) (Brandizzi & Barlowe, 2013; Panaretakis et al., 2008).

1.8.1: CALR in antigen presentation

Major histocompatibility complex (MHC) class I molecules are found on the cell surface of all

nucleated cells and present peptide fragments derived from intracellular proteins. The MHC class 1

pathway has an important role of alerting the immune system to damaged and infected cells (Hewitt,

2003). Loading of cellular antigens onto MHC Class I molecules is mediated by the peptide-loading

complex (PLC) of which CALR contributes to two main functions (Blees et al., 2017). Firstly, CALR

preserves steady-state levels of MHC Class I heavy chains within the ER, through binding of PDIA3 in a

glycan-dependent manner (Wearscha et al., 2011). Secondly, suboptimal assembled MHC Class I

molecules are retrieved by CALR from post ER compartments (ER-Golgi intermediate compartment

and cis-Golgi) (Howe et al., 2009). Cells with low expression or lacking CALR expression have a 50%–

80% reduction in properly loaded MHC class I on the cell surface (Gao et al., 2002). Increased

expression of CALR is seen in cancers with mutant and wild type (w/t) CALR, which avoid

immunodeletion due to upregulation of CD47 (Chao et al., 2012) and the inappropriate loading of the

37

MHC class I molecules with peptide (Galluzzi etal., 2018). Increased overall expression of CALR with

low surface expression has been seen in bladder and breast solid tumours, which result in a poor

disease prognosis (Cathro et al., 2010; Noblejas-López et al., 2019).

Specifically within MPNs, it has been suggested that expression of CALR creates a spontaneous

immune response in patients by creating shared tumour neoantigens (Holmström et al., 2018), under

the strict regulation of cytotoxic T lymphocyte-associated protein 4 and programmed cell death 1

signalling (Bozkus et al., 2019).

1.8.2: CALR in cancer signalling

CALR has been seen in different cancers to influence malignant transformation, disease progression

and response to therapy marking it as a prominent factor in multiple oncological settings. CALR plays

a major role in signalling cellular adjuvanticity, allowing the stressed or dying cell to deliver a

stimulatory signal rather than an inhibitory signal to immune cells. These stimulatory signals do not

originate from the internal CALR within the ER of the cell, they instead initiate a potent pro-

phagocytic signal from the CALR exposed on the surface of a stressed cell stimulatory (rather than co-

inhibitory) signals to immune cells (Obeid et al., 2007). This allows for the signalling cell to be targeted

by phagocytes (Gardai et al., 2005). LDL receptor-related protein 1 (LRP1) on the surface of

macrophages are commonly known to bind to extracellular CALR (Garg et al., 2012). However,

additional proteins including thrombospondin 1 (THBS1)(Goicoechea et al., 2002), C1q and mannose-

binding lectin (MBL) family (Stuartet al., 1997) have been shown to bind to extracellular CALR

resulting in CALR-dependent phagocytosis of apoptotic bodies (Ogden et al., 2001).

The pro-phagocytic properties of CALR are counteracted on the cell surface by two main mechanisms

and effected by others. Firstly, phosphatidylserine, a phospholipid found on the inner side of the

plasma membrane, is seen to be over produced on the cell surface (Suzuki et al., 2013). This binds to

arginine demethylase and lysine hydroxylase leading to the immunologically silent clearance of

38

resulting cell debris by macrophages (Fucikova et al., 2020). Therefore CALR must be overexpressed

on the cell surface to initiate phagocytosis prior to phosphatidylserine (Panaretakis et al., 2009).

Secondly phagocytosis of dying cells due to CALR can be inhibited by CD47, which binds the signal

regulatory protein α (SIRPα) on the surface of antigen presenting cells (Oldenborg et al., 2001).

Interestingly, upregulation of CALR on the surface of cancer cells is usually followed by a concomitant

upregulation in cell surface expression of the anti-phagocytic CD47, in a survival response to keep the

cell alive and evade immunosurveillance (as detailed later) (Feng et al., 2015). Cells being able to

evade immunosurveillance is a key hallmark for cancer (Hanahan and Weinberg, 2011).

1.8.3: Prognostic and predictive value of CALR

Preclinical models of CALR in prostate, colorectal cancers, urothelial carcinomas and AML suggest that

increased proliferation of CALR expression is lower in progressive/malignant stages and increases in

expression as the disease outcome improves (Alur et al., 2009; Cathro et al., 2010; Peng et al., 2010;

Schardt et al., 2009). In many cases upregulated levels of CALR correlate with the different markers of

cancer immunity activity. High levels of CALR on the surface of the cell in AML correspond with

increased levels of T cells specific for tumour-associated antigens (Fucikova et al., 2016) also with

limiting CD47 expression (Wemeau et al., 2010). These studies show AML blasts have the potential to

spontaneously expose CALR on their surface in a cellular response to changes in the

microenvironment of the bone marrow or in response to oncogenic stresses (Méndez-Ferrer et al.,

2020).

To apparently contradict the above, high expression of CALR has been seen in different cancers

including; breast (Erić et al., 2009), gastric (Chen et al., 2009), pancreatic (Sheng et al., 2017) and

neuroblastomas (Chao et al., 2010). Increased CALR expression in these cancers is associated with the

concomitant expression of the anti-phagocytic CD47 (Boasman et al., 2019), resulting in rapid tumour

progression and poor disease outcome. Indicating that along with being a key regulator of

39

Ca2+ homeostasis and integrin-dependent signalling in healthy cells it may also be involved in the

progression of cancers (Fucikova et al., 2020). Alternatively, some myeloid malignancies showed

overexpression of CD47 which could be independent of the CALR expression causing independent

negative prognostic impact in cancers (Chao et al., 2010).

Somatic frameshift mutations in CALR exon 9 have been identified in different cancers but are largely

associated with MPN. Over 50 different mutations have been identified to date, with type 1 mutation

(a 52-bp deletion) and type 2 (a 5-bp insertion) seen most frequently (Cazzola & Kralovics, 2014). A

protein secreted by the mutated CALR binds to the CALR receptors on the cell surface acting as a

decoy subverting CALR’s immunosurveillance activity in favour of tumour progression (P. Liu et al.,

2020). Patients with the CALR mutation tend to be younger than those with other non-mutated

variants and experience improved disease outcome. MF and ET patients bearing CALR mutations

display lower incidence of anaemia and leucocytosis, lower Dynamic International Prognostic Scoring

System (DIPSS) score and reduced frequency of spliceosome mutations, as well as higher platelet

counts than patients with wild-type CALR (Rotunno et al., 2014; Tefferi et al., 2014). CALR mutants

have been shown to activate the thrombopoietin receptor (MPL) resulting in ligand-independent

signalling through JAK-STAT, PI3K and MAPK pathways (Chachoua et al., 2016). Evidence suggests that

CALR mutation (as opposed to a JAK2 or MPL mutation) in ET patients have a lower risk of thrombosis

(Rotunno et al., 2014). Leukemic transformation is lower, in people with CALR mutations versus those

with JAK2 mutations (2.5% Vs 4.3%, respectively), with survival rates for CALR-mutated MF being

greater when compared to JAK2 mutated or triple-negative patients with a median survival of 17.7

years for CALR versus 9.2 years for JAK2 and 9.1 years for MPL (Rumi et al., 2014). This shows the

potential for changes in CALR to impact upon disease progression/prognosis.

1.9: CD47

CD47 is known as the “don’t eat me signal” and thus prevents phagocytosis by macrophages

(Boasman et al., 2019). It has a variety of key roles in immune regulation, atherogenesis, stress

40

resistance and tumour development (Eladl et al., 2020). CD47 is comprised of an extracellular

immunoglobulin (Ig) variable domain, a five times transmembrane-spanning domain, and a short

alternately spliced cytoplasmic tail (Oldenborg & Per-Arne, 2013). It is a 47 kDa protein that appears

70 kDa when resolved via SDS-PAGE due to its hyperglycosylated structure (Mushegian, 2002).

Different CD47 tissue expression patterns can be seen, which has been proposed to be due to

different isoforms of CD47 determined by alterations in length of the cytoplasmic tail from between

4-36 amino acid long (Jaiswal et al., 2009). The most abundantly expressed isoform is isoform 2, while

isoform 4 is the second most abundant one (Reinhold et al., 1995). Expression of CD47 is seen in

virtually all cells and affects cellular proliferation, apoptosis, adhesion, migration and immune system

homeostasis (Oldenborg & Per-Arne, 2013). Seen in different hematopoietic and non-hematopoietic

healthy cells CD47 plays a crucial role in the homeostatic system by being down regulated to allow for

phagocytosis (Oldenborg et al., 2000).

The most important role of CD47 is to relay the “do not eat me” signal to the immune system, it does

this via binding to the SIRPα on the macrophage surface membrane. The “do not eat me” signal is

seen in healthy cells and younger red blood cells thereby preventing cell phagocytosis. Older, stress

damaged cells which should be eliminated from the circulation, express CD47 in much lower amount

(or not at all for mature red blood cells) on their surface membranes, allowing for their phagocytosis

by macrophages (Oldenborg et al., 2000; Jaiswal., et al., 2009). CD47 has gained interest and is

considered as a promising target for the treatment of cancers. Studies into CD47 in cancer have

shown many cancers express high levels of CD47 on their surface, allowing cancer cells to evade

immunosurveillance by protecting them from phagocytosis by macrophages (Boasman et al., 2019).

CD47 is commonly seen and overexpressed in cancer cell, in the attempt to protect themselves from

phagocytosis (Eladl et al., 2020). Studies within MDS have specifically shown that blocking the CD47

signal leads to selective phagocytosis of the MDS cells by reducing the “don’t eat me” signal (Pang et

al., 2013).

41

1.9.1: Mechanism of function and intracellular signalling

CD47 plays an essential role in various cellular functions, these functions can occur through cell-cell as

well as cell-extracellular matrix interactions (Sick et al., 2012). Interactions of CD47 are mediated

through binding to extracellular ligands; signal regulatory proteins (SIRPs) and thrombospondins

(TSPs), membrane ligands and internal ligands (Eladl et al., 2020) (Figure 1.5).

SIRPα is also known as CD172a, Src homology 2 domain-containing protein tyrosine phosphatase

substrate-1 (SHPS-1) (Hatherley et al., 2008). It is expressed on haematopoietic stem cells and their

derived myeloid cells; granulocytes, monocytes, macrophages and dendritic cells (van Beek,

Cochrane, Barclay, & van den Berg, 2005). SIRPα is a transmembrane glycoprotein consisting of an

extracellular N-terminal domain composed of three immunoglobulin-like and a cytoplasmic domain,

two tyrosine phosphorylation sites and four immunoreceptor tyrosine inhibitory motifs (ITIMs)

(Hatherley et al., 2008).

42

Figure 1.5: The impact of the SIRPα–CD47 checkpoint and its blockade is depicted, the regulation of phagocytosis by SIRPα–

CD47 Immune Checkpoint. (A) On tumour cells surface SIRPα interacts with its ligand CD47, SIRPα undergoes tyrosine

phosphorylation and recruits the protein tyrosine phosphatases SHP-1 and SHP-2. These phosphatases inhibit the ability of

prophagocytic receptors to trigger phagocytosis when ligands are present on tumour cells. (B) prevention of the interaction

between CD47 and SIRPα , stops the prophagocytic receptors being able to induce productive activating signals, which lead

to phagocytosis (Veillette & Chen, 2018).

A phosphorylation reaction of the ITIM is induced by CD47 binding to the N-terminal of SIRPα on the

phagocytic cells, activating the protein tyrosine phosphatases (PTPases); Src homology region 2 (SH2)

domain-containing phosphatase1(SHP-1) and 2(SHP-2) (Tsai & Discher, 2008). Dephosphorylation of

immunoreceptor tyrosine activation motifs (ITAM) gives the “don't eat me” signal to the immune

system preventing the macrophage from engulfing the cancer cell, leading to the immunoevasion and

survival of the cancer cell (Murata et al., 2014).

43

CD47 also interacts with thrombospondins, extracellular matrix calcium-binding glycoproteins, that

are secreted by vascular inflammatory cells which regulate cellular functions including motility,

proliferation, and differentiation (Broom et al., 2009). The extracellular IgV section of CD47 binds to

the C-terminal binding domain (CBD) peptide 4N1K of TSPs. Disruption of CD47-TSP-1 interaction by

TSP-1-blocking antibodies has had positive outcomes in cancer therapies (Oaks et al., 2018), and has

shown a down-regulation of CD47 on tumour cells in multiple myeloma (Kukreja et al., 2009).

1.9.2: CD47 pathways targeted for therapeutic intervention

One of the overarching aims of CD47 targeted therapies is inhibition of the CD47-SIRPα axis. By

inhibiting CD47-SIRPα binding, anti-CD47 antibodies and anti-SIRPα antibodies stop the “do not eat

me” signalling in macrophages, allowing phagocytosis of the cancer cells by the macrophages.

Macrophage-mediated phagocytosis of cancer cells has been shown to be promoted when CD47-

SIRPα binding is targeted. As a result, APCs presents cancer cell antigens to T cells via MHC class I,

activating CD8+ T cells which can also target cancer cells, resulting in their destruction. Inhibition of

the CD47-SIRPα signalling also can increase cancer cell phagocytosis when dendritic cells migrate to T

cell zones in lymph nodes which present the cancer cell antigens to the T cells in turn activating the

CD8+ T cells (Hayat et al., 2020). Anti-CD47 treatments target the CD47-SIRPα signalling by; enhancing

the phagocytosis of cancer cells via macrophages, activation of apoptosis in cancer cells and inhibiting

the maturation of dendritic cells causing phagocytosis.

Mimicry peptides and gene silencing (siRNAs and miRNAs) have been used to inhibit CD47-SIRPα

binding to enhance the phagocytosis of cancer cell via macrophages, preventing the inhibitory signals

being received by the macrophages, resulting in increased phagocytosis (Willingham et al., 2012).

Activation of apoptosis in cancer cells can be activated by several anti-CD47 antibodies, initiating

independently of caspase-mediated cell death pathways (Manna & Frazier, 2004). The rate of the

44

induction of apoptosis of cancer cells differs with each anti-CD47 antibody, depending on the CD47-

targeting domain of the antibodies and the IgG isotype (Majeti et al., 2009).

In addition, anti-CD47 and SIRPα targeted therapies can also increase cancer cell phagocytosis by

promoting dendritic cell maturation and function. Binding of SIRPα on the surface of dendritic cells to

CD47 present on cancer cells, has been shown to be key in supressing dendritic cell maturation and

cytokine production, in turn decreasing phagocytosis of cancer cells (Matozaki et al., 2009). When the

interactions between CD47 and SIRPα are inhibited by anti-CD47, it leads to the dendritic cells being

able to mature, which in turn will force the production of anti-tumour cytokines produced by

dendritic cells, which in turn can promote cancer cell phagocytosis (Matozaki et al., 2009).

1.9.3: Anti-CD47 approaches to treat cancer and AML

In the immune system of cancer patients CD47 applies major effects. First the high increase in

expression of CD47 on the cellular membrane has been noted in many cancers, to protect the cancer

cells against phagocytosis by macrophages through CD47-SIRPα activation. Second, CD47 activates

and inactivates immune responses through reverse signalling pathways to protect the cancer cells. In

the treatment of cancer, CD47 therapies are being utilised in three main ways; SiRNA and miRNA-

based approaches, antibody-based approaches and peptide-based approaches. These anti-CD47

antibody therapies will usually be used in conjunction with cancer targeting chemotherapies, the aim

of this combination of therapies is to achieve higher anti-cancer efficacies (Chao et al., 2010).

Many studies in both cell lines, and mouse models have utilised anti-CD47 miRNAs and siRNAs to

reduce the expression of CD47, leading to the increased phagocytosis of cancer cells. Wang et al and

Lee et al studies used mouse models to show a suppression of the tumour growth in hepatocellular

carcinoma (Lee et al., 2014), and melanoma (Wang et al., 2013). CD47-targeting siRNAs studies have

also focused on oesophageal squamous cell carcinoma where a reduction of CD47 expression and a

suppressed tumour progression was shown and in colon cancer where tumour growth was inhibited

45

(Broom et al., 2009; Suzuki et al., 2012). Therapies using siRNAs and miRNAs to silence the CD47 gene

have exhibited low bioavailability and poor cellular uptake, resulting in poor delivery to intended site

leading to untargeted effect of the siRNAs and miRNAs causing safety concerns and limiting their

therapeutic application (Gao & Huang, 2013).

Antibody-based approaches use anti-CD47 antibodies to target specific cancer cells for macrophage-

mediated phagocytosis (Weiskopf et al., 2016). Unlike with siRNAs and miRNAs targeted therapies

anti-CD47 antibodies do not target non-specific cells, instead the antibodies tend to target cancer

cells more than normal cells due to the higher levels of CD47 on the cell surface (Gheibi Hayat et al.,

2019). Anti-CD47 antibodies are effective due to the specific “eat me” and “do not eat me” signals

that are expressed from dying cells and cancer cells, which encourage binding to their receptors on

the surface of macrophages resulting in phagocytosis (Tajbakhsh et al., 2019). Other pro-phagocytic

markers have been found to aid phagocytosis, via supporting CALR translocation through the cell from

the endoplasmic reticulum to the cytoplasmic membrane, where it binds the cancer cell to the

macrophages low-density lipoprotein receptor-related protein (LRP) initiating immune attack (Gardai

et al., 2005). Several different antibody-based compounds have shown promising signs. Preliminary

studies in non-human primates show the anti-CD47 humanized antibody Hu5F9-G4 prevents the

binding of CD47-SIRPα in haematological malignancies by binding to CD47, thus promoting

phagocytosis of cancer cells and giving the ability to control the growth of the malignancies in vivo

(Piccione et al., 2015). In vivo studies of the anti-CD47 antibody ALX148 is being used to treat solid

and hematologic malignancies. Studies have shown that ALX148 treatment increases phagocytosis by

macrophages, a shift in tumour-associated macrophages towards an inflammatory phenotype,

increased T cell effector function and increased proinflammatory cytokine production whilst reducing

the expression of suppressive cells in the tumour microenvironment (Kauder et al., 2018).

Peptide-based approaches use a mimicry peptide compound to bind to the target or use mimicry

peptides to bind to its receptor. Anti-CD47 peptide-based approaches will target either CD47 or its

46

receptor SIRPα rendering it inactive. 4N1K is a well-known anti-CD47 mimicry peptide, which binds to

CD47, mimicking part of the TSP-1 C-terminal domain (peptide sequence - KRFYVVMWKK) (Barazi et

al., 2002). In breast cancer cell lines 4NK1 activation of CD47 initiates caspase-independent apoptosis

(Manna & Frazier, 2004b). TTI-621 is a mimicry peptide that promotes phagocytosis of cancer cells

and is effective in controlling the growth of different tumour cells. It is a fully human recombinant

fusion protein that is used as a SIRPα Fc decoy, binding to the SIRPα N-terminal domain rendering

CD47 activation (Petrova et al., 2017).

Specially, AML leukemic stem cells and AML cells express CD47 highly, the high expression of CD47

can be considered as an independent prognostic factor for poor prognosis shown in two adult cohorts

of AML patients. Phagocytosis of AML leukemic stem cells in vitro has been initiated with the

treatment of monoclonal anti-CD47 antibodies (B6H12.2 and BRIC126), B6H12.2 and BRIC126 have

also shown to inhibit their growth in mice models (Jaiswal et al., 2009; Majeti et al., 2009). The anti-

CD47 antibody magrolimab has shown early promising results when combined with azacytidine in a

combination treatment of AML and MDS (Chao et al., 2020).

Earlier treatment with chemotherapeutic agents prior to the addition of anti-CD47 treatments have

been noted to stimulate the adaptive immune response against cancer-specific antigens resulting in a

better outcome (Liu et al., 2015). A different strategy of combining anti-CD47 antibodies with the

administration of anti-programmed death-ligand 1 (PD-L1) antibodies an immune checkpoint

inhibitor, has shown promising results in preclinical mice studies (Sockolosky et al., 2016). These early

preclinical results highlight the power and potential of chemotherapeutic drugs in combination with

anti-CD47 therapeutics in the future of myeloid malignancies.

47

1.10: Purposed aims of research

Whilst a role of CALR and CD47 has been established in cancer and to some extent investigated in

AML, less is known about MDS and MPN and as such this thesis will be looking to establishing the

implication of CALR and CD47 in treatment response and prognosis in MDS and MPN.

Chapter 3, will focus on establishing the differences in CALR and CD47 expression levels in untreated

vs treated MDS and MPN cell line models, which will provide insight on the relationship between the

expression rates of CALR and CD47 and how current therapeutic interventions alter CALR and CD47

localisation in untreated and treated cell line models in response to treatments.

Chapter 4 will analyse CALR/CD47 expression in MDS and MPN patient’s samples to confirm data in

cell lines.

Chapter 5 will assess the interactions, or lack thereof, between CALR and CD47 by determining what

effect knocking down one of them has on the other, and the impact on chemotherapy response in

MDS and MPN cell models.

48

Chapter 2: Methods

49

2.1: Ethics

This study gained ethical permission for this work to be undertaken via the University of Lincoln, UK

ethics board for the use of human derived cell line models and clinical samples. Clinical sample details

have been given later and further information on them and ethical approval is detailed in section -

Patient Samples Processing and Analysis.

2.2: Cell Line Models of MDS & MPN

Due to there not being perfect cell lines that completely follow the phenotype of these diseases, the

following 5 cell line models were used with 2 cell lines similar to MPN, 1 x Intermediate (which has

specific phenotypes of each disease) and 2 cells line similar to MDS (all have been extensively used

within the literature to explore these diseases). Cell line models used were purchased from DSMZ

(Deutsche Sammlung von Mikroorganismen, Website; www.dsmz.de) and ATCC (Website;

www.lgcstandards-atcc.org/Products/Cells_and_Microorganisms/Cell_Lines.aspx?geo_country=gb).

Details on the suspension cell lines, where they were derived from, recommended seeding volumes

and exact location acquired from as seen in Table 6.

Cell line Recommended

seeding volume

Type Acquired from

K-562 100,000 cell per

mL

established from the pleural effusion of a 53-

year-old woman with chronic myeloid leukemia

(CML) in blast crisis in 1970; cells carry the BCR-

ABL1 e14-a2 (b3-a2) fusion gene

ATCC

MOLM-

13

400,000 cells

per mL

established from the peripheral blood of a 20-

year-old man with acute myeloid leukemia AML

FAB M5a at relapse in 1995 after initial

DMSZ

http://www.dsmz.de/

http://www.lgcstandards-atcc.org/Products/Cells_and_Microorganisms/Cell_Lines.aspx?geo_country=gb

50

myelodysplastic syndromes (MDS, refractory

anemia with excess of blasts, RAEB)

SKM-1 400,000 cells

per mL

established from the peripheral blood of a 76-

year-old Japanese man with acute monoblastic

leukemia (AML M5) in 1989 following MDS

DMSZ

HEL-92 200,000 cells

per mL

Established from the bone marrow of a 30-year-

old Caucasian male with AML following MPN

ATCC

GDM-1 500,000 cells

per mL

established in 1979 from the peripheral blood of

a 66-year-old woman with a Philadelphia

chromosome negative myeloproliferative

disorder (resembling chronic myeloid leukaemia,

CML, in blast crisis) after transformation to

refractory acute myelomonocytic leukaemia

(AML)

ATCC

Table 9: Table of MDS cell line models (MOLM-13/SKM-1), intermediate cell line model (K562) and MPN cell line models

(HEL-92/GDM-1). Each cell line was acquired as stated from ATCC or DMSZ, were the descriptions were abbreviated from

DSMZ (Deutsche Sammlung von Mikroorganismen, Website; www.dsmz.de) and ATCC (Website; www.lgcstandards-

atcc.org/Products/Cells_and_Microorganisms/Cell_Lines.aspx?geo_country=gb).

2.3: Cell Culture

All cells were initially grown from frozen cell stocks kept in the liquid phase of liquid nitrogen, with

each vile containing 1x106 cells stored in a solution of 90% RMPI medium (1% penicillin/streptomycin

and 10% FBS) and 10% DMSO (a cell cryogenic preservative). Cells were cultured in class 2 cabinets

in sterile tissue culture rooms throughout using aseptic conditions as standard. Each cell line was

cultured in similar conditions. Any specific changes to conditions or supplementation will be

http://www.dsmz.de/

51

addressed in the methods section of each results chapter. All cell lines were grown in growth media

of RPMI-1640 (Invitrogen, Paisley, UK) supplemented with 1% penicillin/streptomycin (Invitrogen,

Paisley, UK) and 10% foetal bovine serum (Invitrogen, Paisley, UK) to prevent any infections and

support cell growth. Cells were incubated at 37oC with 5% CO2 throughout to support cell growth.

Upon removing cells from liquid nitrogen, they were thawed in a water bath at 37oC for 2-3 minutes,

then added to a falcon tube with 4 mL RMPI media (heated to 37oC to avoid shocking the cells) and

then centrifuged at 13,500 x RPM for 3.5 minutes at room temperature to pellet the cells (Micro Star

17R, VWR, Lutterworth, UK). The supernatant was then poured off to remove any DMSO contained

in the media, which may damage the cells and prevent future cell growth. The pellet was then

resuspended in 2 mL of fresh RPMI media initially to maintain density to support cell growth, added

to a T25 Flask (Sarstedt, Leicester, UK) and left to incubate at 37oC for 3 days to initially access cell

growth.

2.3.1: Testing cell lines for mycoplasma via qPCR

A premix solution was prepared for each cell line from the MycoSEQ™ Mycoplasma Real-Time PCR

Detection Kit (Thermo Scientific, Loughborough, UK). Each premix included 15 µL of Power SYBR™

Green PCR Master Mix, 2x and 3.0 µL of Mycoplasma Real‑Time PCR Primer Mix, 10x. Four microamp

tubes were set up for each cell line; Negative control - 18 µL of the premix solution with 12 µL of

Negative Control (water), Unknown sample reaction - 18 µL of the premix solution with 10 µL of

unknown sample and 2 µL of Negative Control (water), Inhibition-control reaction - 18 µL of the

premix solution with 10 µL of unknown sample and 2 µL of the Discriminatory Positive Control,

Positive control reaction - 18 µL of the premix solution with 10 µL of Negative Control (water) and 2

µL of the Discriminatory Positive Control. The samples were ran according to the 2.11 qPCR protocol.

The results would show different readings for the unknown samples compared to the controls in the

presence of mycoplasma.

52

2.4: Cell counting and seeding After 3 days incubation the cells were assessed to see if they had grown to confluency. To assess cell

growth, cell solutions were pipetted into a sterile 15 mL falcon and centrifuged at 13,500 x RPM for

3.5 minutes at room temperature to pellet the cells. Once pelleted, the supernatant was removed,

and cells were resuspended by gently pipetting up and down in 3 mL of pre-warmed RMPI media

and then counted. To perform the cell counting 20 µL of cell suspension was added to 20 µL 0.4%

trypan blue solution (Sigma-Aldrich, Poole, UK) and left to incubate at room temperature for 5

minutes to allow for the distinction between living and dead cells. Once incubated, 20 µL was

pipetted onto a haemocytometer chamber (Counting Chamber Improved Neubauer; Hawksley,

Sussex, UK), with a cover slip placed over the top to secure single layer of cell suspension. Cells were

counted in each of four corner quadrants (1 mm x 1 mm x 0.1 mm) of the 16-square grid. The sum of

the counted quadrants was divided by 4 (to get an average of the quadrants counted) and then

multiplied by the dilution factor of the trypan blue (2), this gave cell count per 1 x 10-4 ml. Final cell

count per mL was determined by multiplying by 10,000, to get number of cells per ml. Cell counts

were used to seed cells at required seeding density appropriate for specific cells (Table 6), and then

re-seeded in fresh T25 flasks, to continue growth/enable experiments to be undertaken.

2.5: Growth curves

Growth curves of untreated and treated cell lines were conducted using 12-well microtitre plates

(Sarstedt, Leicester, UK). Cells were seeded (1 x 105 cells per well) in 1 mL of RPMI 1640 growth

complete media. Cells to be drugged were seeded at the same density with the intended drug at

different dilutions ranging from control to high drug concentration. Drug treated samples were

gently mixed using a pipette. In 24 hour increments the cells were retreated with the specific drug

without changing the media, repeated until the time-period has ended. Running at the most for 96

hours. Both drugs half-life was around 8 hours. Therefore, retreatment was completed every 24

hours.

53

At each 24 hour time point cells were agitated to make sure all are in suspension and all of the cells

were removed from the wells. To assess cell growth at each time point, 20 µL of cell suspension was

added to an equal quantity of 20 µL 0.4% trypan blue solution (Sigma-Aldrich, Poole, UK) in a 500ul

Eppendorf and left to incubate at room temperature for 5 minutes. From this solution, 10 µL was

pipetted onto a haemocytometer chamber with cover slip attached. Dead cells with permeable

membranes took up the dye and stained blue; live cells did not stain and reflected light. Cells were

counted in each of four corner quadrants (1 mm x 1 mm x 0.1 mm) of the 16-square grid as

described earlier in the cell counting section with a modification that the number of both live and

dead cells were counted.

2.6: Drug titrations and drugging of cell lines

Cells lines were treated with azacytidine and ruxolitinib, drugs that are specific to MDS and MPN,

respectively, (as described in the Introduction) to investigate how therapy could impact upon

pathways within these cell line models. Azacytidine (Sigma-Aldrich, Poole, UK) and ruxolitinib

(Selleckchem, Newmarket, UK) (supplied at 5mg) were dissolved in 1 mL DMSO (Sigma-Aldrich,

Poole, UK) for drug stocks and applied to cell cultures diluted in media at a 1/100 dilution of drug to

final volume of culture media. Drug titrations were first performed to work out final concentration of

drug required. Drug titrations took place using 12-well microtitre plates (Sarstedt, Leicester, UK).

Determination of possible drug concentrations gathered from published data were used as a base

line then with the low/high concentrations adapted from them. Cells to be drugged were seeded at

5 x105 with the intended drug at different dilutions ranging from control to high drug concentration.

Cell solutions were then gently mixed to ensure equal distribution of drug through the cell media

using a pipette. In 24 hour increments, the cells were retreated with the specific drug every 24 hours

without changing the media to ensure a constant level of drugging, repeated up to a maximum of 96

hours.

54

At the desired time points, cells were agitated to make sure all are in suspension and all the cells

were removed from the wells using a 1 mL graduated pipette and placed into a sterile 1.5 mL

Eppendorf. To determine the impact of the drugging on the cell viability, as previously described 20

µL of cell suspension was added to 20 µL 0.4% trypan blue in separate 0.5ml Eppendorf and left to

incubate at room temperature for 5 minutes. After incubation, 10 µL was pipetted onto a

haemocytometer chamber with cover slip attached. Dead cells with permeable membranes took up

the dye and stained blue; live cells did not stain and reflected light. Live/dead cells were then

counted (as described below).

The titration optimisation experiments were measured for 96 hours with measurements taken every

24 hours. This revealed that an appropriate time point of 48 hours and 5µg/ml of drug every 24

hours, were chosen for cells to be drugged. This is within the ranges that were supported by the

literature.

2.7: Cell viability assay

After optimising drug dosing these were done to check toxicity. Trypan blue assays were conducted

to show cell viability and drug toxicity. Cells were grown over a period of 96 hours with live and dead

cells counted using a microscope after each 24-hour stage, using same method previously

mentioned in growth curves of cell line models. A percentage of live cells could then be calculated

from the total cells. Cell viability percentage was calculated by dividing live cells by total cells (dead

and live) and multiplying by 100. The non-treated cells were considered 100% viable with the

calculations of the viability of the drugged cells as a percentage of the control. Calculating the

percentage of the cell viability indicates the drug toxicity. If toxicity is to high all cells would die

including healthy cells. Therefore, the aim is to affect the cells enough that some cells survive and

are affected by the drug without destroying all of them.

55

2.8: Harvesting of cells to undertake RNA and Protein Analysis

At the desired time point (48 hours), cells were counted (in the presence or absence of drugging)

and 1x106 cells were pelleted to undertake the following RNA extractions and protein analysis.

2.9: Extraction of RNA from cell lines

RNA extraction on cell line models (1 x 106 cells) was performed using the Quick RNA Mini-Prep

(Zymo Research, USA) as per manufacturer’s instructions. The kit supplied all reagents unless

specified. Cell pellets were resuspended in 300 µL RNA lysis buffer and an equal amount of 100%

ethanol (Fisher Scientific, Loughborough, UK) was added to lyse the cells. The lysate ethanol mixture

was pipetted into a Zymo-Spin™ IIICG Column placed on top of a clean 2 mL collection tube and then

centrifuged (Micro Star 17R, VWR, Lutterworth, UK) at 15,000 x g for 30 seconds to filter out debris,

and the flow through was discarded. The spin column which retained the mRNA was prewashed with

400 µL of RNA wash buffer and centrifuge at 15,000 x g for 30 seconds. A DNase reaction mixture (75

µL DNase digestion buffer and 5 µL DNase I) was added to the spin column membrane then

incubated at room temperature for 15 minutes, to remove any DNA products. After incubation, 400

µL of RNA Prep Buffer was added to the spin column and centrifuged at 15,000 x g for 30 seconds to

wash the filter: with the flow through discarded. A further washing step was undertaken with 700

µL RNA wash buffer added to the spin column before being centrifuged at 15,000 x g for 30 seconds

and discarding the flow-through. As a final wash step another 400 µL RNA wash buffer was added to

the spin column and centrifuged at 15,000 x g for 2 minutes. The spin column was then transferred

to a clean RNase free 1.5mL Eppendorf and 50 µL of nuclease-free water pipetted directly on to the

membrane of the spin tube to help elute the mRNA from the column. The column was then

centrifuged at 15,000 x g for 30 seconds, the spin column was discarded and flow-through

containing the purified mRNA was collected.

RNA yield and purity were determined by absorbance readings at 230/260/280 nm on a Nanodrop

2000 spectrophotometer (Thermo Scientific, Loughborough, UK). Purity is assessed on the 230/260

56

absorbance for nucleic acids with protein contamination being registered at 280 nm (Olson &

Morrow, 2012). The nanodrop is used to determine the RNA quantity and purity through

absorbance. To use the nanodrop the nucleic acid option was chosen on the software and the

pedestal was cleaned with ultra-pure water. The nanodrop was blanked using 2µL of the same

nuclease-free water used to dilute the RNA, then 2µL of the sample obtained from the spin column

was added to the pedestal and read by the nanodrop in duplicate, this was repeated for each

individual sample.

2.10: cDNA synthesis from RNA

To obtain cDNA from the mRNA, required to enable quantitative qPCR to be undertaken, iScript

cDNA synthesis kit (Bio-Rad, Hercules, USA), was used to synthesise complementary DNA from total

RNA as per manufacturer’s instruction, with the kit suppling all reagents unless specified. Initially a

reaction mixture was prepared as follows (20 µL total).

4 µL iScript reaction mix

1 µL iScript Reverse Transcriptase

1000 ng Total RNA (as worked out from Nanodrop)

Nuclease-free H2O to make up to a 20 µL reaction

Controls were performed in each synthesis in the form of a no-reverse transcriptase control (NRT) in

which nuclease-free water was substituted for the reverse transcriptase control, to ensure reagents

had no contamination. No-template controls (NTC) were also used where nuclease-free water was

used in place of RNA sample, to ensure the nuclease free water was not contaminated. The 20 µL

reaction mixture was then run in a thermocycler (Model: T100, Manufacturer: Bio-Rad, Watford, UK)

per the following settings to support synthesis:

57

Step1: 25ºC 5 minutes

Step 2: 42ºC 30 minutes

Step 3: 85ºC 5 minutes

Step 4: 4ºC Hold

After the RNA extraction, the RNA product was immediately used in cDNA analysis or stored at -20oC

until needed.

2.11: Quantitative PCR (qPCR)

Using the cDNA produced above, qPCR reaction was completed on a StepOne Plus thermocycler

(Applied Biosystems). SYBR Green chemistry probe methods were used. The SYBR Green method is

commonly used in qPCR. Fluorescent dye binds to the cDNA samples that have been specifically

labelled with the forward and reverse primers. Once bound energy is released as fluorescence which

increases/decreases along with the concentration of cDNA in the sample (Bookout, Mangelsdorf, &

Hughes, 2003).

Samples were prepared in MicroAmp Fast 96-well Reaction Plate (Applied Biosystems, UK). SYBR

Green reaction mixture comprised of: 0.5 µL of 10 mM primer (Sigma-Aldrich, Dorset, UK) of each

forward and reverse primer. In addition, 5 µL iTaq SYBR Green (Bio-Rad, Hercules, USA) and 14 µL

1/50 diluted cDNA in nuclease-free water was also added (Sigma-Aldrich, Dorset, UK), this was

achieved by adding 4µL of cDNA generated in the previous step to 196µL of nuclease-free water.

Primers were selected from predesigned primers previously designed within labs or from published

data, then a series of quality checks were performed on the efficiency and melt curves to select the

used primers as follows. Internal validation of the primers was conducted by running several (3-5) n

numbers making sure that the cT values and melt curves were similar. Primers for CALR and CD47

along with GAPDH are shown in Table 7. GAPDH is a housekeeping gene that is readily found

58

throughout the cell, allowing for levels of CALR and CD47 to be standardised (Nygard, Jørgensen,

Cirera, & Fredholm, 2007).

Forward Reverse

GAPDH TGCACCACCAACTGCTTAGC GGCATGGACTGTGGTCATGAG

CALR TGTCAAAGATGGTGCCAGAC ACAACCCCGAGTATTCTCCC

CD47 TATCCTCGCTTGGTTGGAC TCTTTGAATGCATTAAGGGGTTCCT

Table 10: Table of primer sequence use for GAPDH, CALR and CD47

Duplicates of each sample were ran for n3 (each n was of the same cell line from different time

points in the cell lines growth), with NRT and NCT controls, as described earlier for the cDNA

amplification step. qPCR reactions were run on a predesigned SYBR Green fast protocol. Conditions

of the cycle times and temperatures for the qPCR are shown below as per recommended.

Step 1: 95ºC 30 seconds



Repeat Step 2 – 3 x 39

Step 4: 60ºC + 0.3ºC incremental to 95ºC 3 seconds

Analysis of the qPCR was then completed using the method detailed in the statistics and data

analysis section.

2.12.1: Western blotting

Western blotting is a technique used for protein analysis. Using a primary antibody, the target gene

is located and then visualised using a secondary antibody as shown in Figure 2.1. Individual proteins

are separated from a sample on an SDS gel. This is then transferred to an appropriate membrane

59

using electrical currents. Once on the membrane the primary antibodies from one host and

secondary antibodies from a different host can be used to visualise the proteins.

Figure 2.1: Depiction of western blots antibodies on a membrane and the fluorescent image of proteins (Fluorescent

Western Blotting | Sino Biological, 2020).

2.12.2: Buffer formulations for SDS-PAGE and Western blot experiments

A series of common buffers and solutions are used throughout. Buffers used for SDS-PAGE and

Western blot are given below:

All generic buffer were made up as detailed in table 8;

Modified RIPA buffer

Used for protein lysis 50 mM Tris pH 7.5

150 mM NaCl

1% NP-40

10% Glycerol

Fisher Scientific, Geel, Belgium

Fisher Scientific, Geel, Belgium Thermo Scientific, Rockford, USA

Fisher Scientific, Loughborough, UK

Fisher Scientific, Geel, Belgium Model: Purelab Flex,

60

5 mM EDTA

Ultrapure (18.2 MΩ) H2O (to make up to specific volume)

Manfacturer: Elga Lab Water, High Wycombe, UK

SDS Loading buffer

Used to load protein into the SDS gel

10% SDS

50% Glycerol

200 mM Tris pH 6.8

Ultrapure H2O

Bromophenol Blue

Fisher Scientific, Geel, Belgium

BDH Chemicals, Poole, UK

SDS Running Buffer

Used to run the SDS gel

25 mM Tris

192 mM Glycine

0.1% w/v SDS

Melford Biolaboratories, Chelsworth, UK

Transfer buffer

Used to transfer proteins from SDS gel to membrane

25 mM Tris

192 mM Glycine

20% Methanol

Fisher Scientific, Loughborough, UK

Phosphate-buffered saline (PBS)

Used to dilute block solution and to mix with tween

PBS tablets Sigma-Aldrich, Poole, UK

PBS-Tween (PBS-T)

Used to wash the membrane

0.1% Tween-20 Fisher Scientific, Loughborough, UK

Blocking buffer

Used to block membrane from unspecific binding and to dilute antibodies

PBS

Blocking solution

Sigma-Aldrich, Poole, UK

LI-COR Biosciences, Cambridge, UK

Table 11: Buffer formulations for SDS-PAGE and Western blot experiments

2.12.3: Protein isolation and quantification in preparation for SDS-PAGE and Western blotting

Cells were harvested at 1x106 as stated earlier. Once cells were harvested to undertake protein

analysis, they were then lysed within an 1.5ml Eppendorf using 100µL of modified RIPA buffer

61

(Sigma-Aldrich, Poole, UK) with protease inhibitors, to prevent degradation of the protein during

storage. Bradford assay was performed to determine the total protein concentration, using Bradford

reagent (Bio-Rad, Watford, UK) and reading the absorbance at 595nm. A serial dilution was

performed using BSA (bovine serum albumin), 5µL of each serial dilution and 5µL sample were

added to 250µL Bradford reagent and absorbance read at 595nm (infinite M20 PRO, Tecan). A

standard curve was composed from the serial dilution and protein concentrations were calculated

from this as depicted in Figure 2.2.

Figure 2.2: Depiction of a Bradford assay standard curve.

2.12.4: SDS-PAGE

To create the acrylamide gel within which to run the protein samples, a 10% acrylamide gel was

made; composed of 1.5 M Tris- pH 8.8 (Fisher Scientific, Geel, Belgium), 0.5 M Tris- pH 6.8, 30%

acrylamide/bisacrylamide (Bio-Rad, Watford, UK), N,N,N',N'-tetramethylethylene diamine (TEMED)

(Bio-Rad, Watford, UK), 10% SDS (Fisher Scientific, Geel, Belgium) and 10% ammonium persulfate

0

0.5

1

1.5

2

2.5

0 0.5 1 1.5 2 2.5 3 3.5 4 4.5

Abso

rban

ce 5

95 n

m

BSA standard mg/ml

Bradford assay - Absorbance 595

62

(Fisher Scientific, Geel, Belgium) which were mixed together in a universal tube. The 10% acrylamide

solution was then pipetted between 1mm SDS glass plates (Bio-Rad, Watford, UK), previously

secured in a gel casting station (Bio-Rad, Watford, UK) with a 15 well comb inserted to create the

wells. The gel was then left to set for 20-30 minutes.

Using the concentrations generated from the Bradford assay, 10mg of lysates were mixed with

loading buffer and loaded onto a 10% SDS-PAGE gel (together with a 0-120 bp ladder) within a mini

protean SDS-PAGE tank (Bio-Rad, Watford, UK) which was then filled with running buffer and ran at

120-150V (PowerPac™ Basic Power Supply, Bio-Rad, Watford, UK) until the proteins have separated

and the dye front had moved an adequate distance down the plate or reached the end of the plate.

2.12.5: Transfer to PVDF membrane

Proteins within the polyacrylamide gel were then transferred to a polyvinylidene fluoride (PVDF)

membrane using a wet-blot apparatus by passing a current across the gel to the membrane.

Tris/Glycine + 20% methanol transfer buffer (Fisher Scientific, Geel, Belgium) is then used during the

transfer step to aid in the disassociating of proteins from the SDS and to aid in the absorption of the

protein to the membrane. Proteins were transferred for 1 hour at +2 – 8ºC (to prevent overheating)

at 100V (Mini-PROTEAN tetra, Bio-Rad, Watford, UK).

2.12.6: PVDF membrane blocking and antibody incubation

The PVDF membrane was then incubated in blocking solution at room temperature for 1 hour, to

prevent any nonspecific binding. An antibody buffer was created using an appropriate amount of

primary antibody (see table 9), together with the required amount of fresh blocking buffer (antibody

dilutions are listed in table 8), and applied to the membrane and incubated overnight at 2 – 8ºC.

After incubation the primary antibody solution was removed, and the membrane washed 3 times for

5 minutes with PBS-T to remove any excess primary antibody left on the membrane.

63

Secondary fluorescent antibody dilutions were then prepared in 10ml of blocking buffer (see table 8)

and added to the membrane and left to be incubated in a light tight falcon tube for an hour at room

temperature, whilst gently rolling on a DigiRoll6 roller (SLS, Nottingham, UK) to ensure even

coverage. Use of a light tight falcon tube ensured that the florescent signal of the secondary

antibody would not be affected. After incubation the secondary antibodies were removed, and the

membrane washed 3 times for 5 minutes with fresh PBS-T each time to remove any excess

secondary antibody.

Visualisations of the fluorescent bands were completed using a digital developer (LI-COR

Biosciences, Cambridge, UK). Visualisation is achieved when the digital developer fires a targeted

light source onto the membrane. This excited the fluorescent dye (the secondary antibody) and

releases photons at its emission wavelength to return the molecule to its ground electronic state.

The light emitted from the protein is then captured at different wavelengths using specific filters by

the digital developer. The bands can then be further analysed, and densitometry undertaken as

detailed later in the statistics and data analysis section.

Table 12: Antibodies and dilutions used in Western blot experiments

Antibodies Primary or secondary antibody

Host Dilution in blocking buffer Manufacturer

CALR Primary Rabbit 1:2000 Abcam, Cambridge, UK

CD47 Primary Mouse 1:500 Abcam, Cambridge, UK

GAPDH Primary Rabbit 1:20000 Amsbio, Abingdon, UK

800 goat anti rabbit

Secondary Goat 1:50000 LI-COR Biosciences,

Cambridge, UK

700 donkey anti mouse

Secondary Donkey 1:50000 LI-COR Biosciences, Cambridge, UK

64

2.13: Fractionations of protein in cells

To investigate cellular localisation fractionation was undertaken, two kits were used to split samples

into the membrane, cytoplasm, cytosol and nucleus as no one kit for all for components is available.

Both kits work by using a series of buffers and centrifugations to break down and separate the

different components of the cell. The Nucper (Thermofisher scientific, Altringham, UK) kit was used

to, separate the sample into cytoplasm and nucleus samples and the Membranplus (Thermofisher

scientific, Altringham, UK), separating the sample into membrane and the cytosol.

In brief for the Nucper kits: Cell pellets (1x106) were resuspended in 100µL of nuc 1 solution then

incubated on ice for 10 minutes as per manufacturers guidelines. Then 5.5µL of nuc 2 solution was

added to the resuspended pellet solution and incubated for 1 min on ice. The sample was then spun

in a micro centrifuge at 16000g for 5 minutes. Once completed the supernatant of the sample

containing the cytoplasm is collected into a clean RNA free Eppendorf tube and placed on ice. The

remaining pellet was then resuspended in 50µl of nuc 3 solution. Samples were vortexed (VX100,

Labnet) for 15 seconds to fully suspend the content, then incubated for 40 mins at room

temperature, with a re-vortex every 10 minutes to ensure content doesn’t settle at the bottom of

the Eppendorf. The sample was spun in a micro centrifuge at 16000g for 10 minutes. Into a clean

RNA free Eppendorf tube the supernatant of the sample containing the nucleus, is collected and

placed on ice ready for protein extraction as detailed earlier.

For the Membrane plus kits: Cell pellets were resuspended in 100µL of mem 1 solution, then

incubated for 10 minutes at 4ºC whilst being gently agitated. The sample was spun in a micro

centrifuge at 16000g for 15 minutes. The supernatant containing the cytosol was collected and put

into a clean RNA free Eppendorf tube and placed on ice. The pellet was resuspended in 100µl of

mem 2 solution, then incubated for 30 minutes at 4ºC whilst being gently agitated. The supernatant

containing the membrane was collected and put into a clean RNA free Eppendorf tube and placed on

65

ice. All samples are either then used straight away for protein or frozen down at -80ºC. Protein was

extracted/analysed as described in earlier sections.

2.14: CALR and CD47 Knockdowns

CALR and CD47 knock downs using siRNA and lipofectamine LTX (Life technologies, California, US)

were performed on cell line models. siRNA transfections are used to supress a specific gene in cell

cultures. The technology degrades the mRNA after transcription preventing translation, therefore

the gene and the protein will not be expressed within the cell. As the exact relationship between

how CALR & CD47 work together is currently unknown in MDS and MPN, transfections of both genes

was undertaken to see how gene expression is altered without the other genes being present.

Using a 6 well plate (Sarstedt, Leicester, UK), for each cell line 2x105 cells per mL were seeded per

well. For each well seeded with cells, the following mixture was created in a clean 1.5ml Eppendorf -

1µg of specific siRNA and 1µL of plus reagent were added to 200µL of Opti-MEM media (Invitrogen,

Paisley, UK) and mixed gently and left to incubate for 5 minutes at room temperature. After

incubation, 2µL of lipofectamine LTX were added to each tube and incubated for 20 minutes at room

temperature to form DNA-lipofectamine complexes. Once incubation was completed, 200µL of the

DNA-lipofectamine complexes were added to each specific well and mixed by gently rocking back

and forth. The transfected cell was transferred into an incubator at 37ºC with 5% CO2 and left for 48

hours to enable knockdown to occur. After incubation, the cells were collected into a clean

Eppendorf tube and spun at 13500g for 5 mins to pellet them. The supernatant was discarded, and

the pellet frozen at -80ºC ready for future protein and RNA assays.

CALR siRNA sequence –

Sense -GGAGCAGUUUCUGGACGGAtt Anti-sense-UCCGUCCAGAAACUGCUCCtt

CD47 siRNA sequence –

66

Sense -GGUGAAACGAUCAUCGAGCtt Anti-sense-GCUCGAUGAUCAUUUCACCtt

2.15: Patient Samples Processing and Analysis

2.15.1: Isolation of Peripheral Blood Mononuclear cells (PBMCs) from human donors

All patients gave written informed consent before peripheral blood and bone marrow samples were

taken. All patients were greater than 18 years of age and had been diagnosed with an MPN or MDS

at the time of sampling via the doctors in charge of their treatment/care. Trained phlebotomists

were used to obtain the peripheral whole blood sample via venepuncture (vacutainer, vacutainer

tube EDTA and needles, Benton Dickinson, Winnersh, UK), and a surgeon took the bone marrow

samples. Each patient gave up to 12ml peripheral blood/6ml of bone marrow divided into 4mL

vacutainers containing EDTA (Ethylenediaminetetraacetic acid), to prevent the blood from clotting.

Three cohort groups were recruited for this research project;

Cohort 1: CALR and CD47 Expression levels in peripheral blood of patients with MPN

Patients recruited for this study were diagnosed according to the 2008 WHO criteria with essential

thrombocythaemia, polycytemia vera or myelofibrosis.

Exclusion criteria for this study were:

• Patients with any other myeloid neoplasms.

Cohort 2: CALR and CD47 Expression levels in peripheral blood of patients with MDS

Patients recruited for this study were diagnosed according to the 2008 WHO criteria with MDS.

Exclusion criteria for this study were:

• Patients with any other myeloid neoplasms.

67

Cohort 3: CALR and CD47 expression levels in peripheral blood of healthy aged matched controls.

Ethical approvals for the study are documented in the appendices.

Peripheral blood was obtained from MDS patients (n = 16) and MPN patients (n = 16), and healthy

control donors (n = 4). The healthy control samples were age matched and either taken from donors

at the Hospital or the University.

2.15.2: Processing and collection of mononuclear cells from patient blood samples

To separate out the blood samples the patient samples (whole blood) were diluted in a 15ml falcon

tube (Thermofisher, Oxoid, Basingstoke, UK) to a 1 in 3 dilution with the PBS/EDTA solution

(Thermofisher, Oxoid, Basingstoke, UK) and gently mixed by inversion. Diluted blood mixture was

then gently layered over 4ml (room temperature) of Ficoll-Paque Premium 1.077 g/ml (GE

Healthcare Life Sciences, Buckinghamshire, UK) in a 15 mL Falcon tube using an automated pipetboy

(Integra, Berkshire, UK) to provide a density gradient media to support fractionation of whole blood.

Layered blood samples were centrifuged at 400 x g for 40 minutes with the acceleration set to a

minimum in a swing bucket rotor centrifuge (Allegra X-15R, Manufacturer: Beckman-Coulter, High

Wycombe, UK) set to 18 ºC, which is the temperature for optimum yield for peripheral blood

mononuclear cell (PBMCs). Additionally, reduced acceleration and putting the brake to a minimum

reduced agitation to the contents of the falcon tube, keeping the separated layers intact. After

centrifugation, the blood sample and the Ficoll-Paque solution split into distinct layers. The interface

between the Ficoll-Paque and plasma layers is the PBMC layer, which was carefully extracted using a

Pasteur pipette (MERCK, Poole, UK), using a gentle circular motion around the falcon tube. The

collected PBMCs were then transferred to a clean 15 mL Falcon tube.

The PBMC layer was then washed in 10 mL of PBS/EDTA and centrifuged for 10 minutes at 300 x g

with brake switched on, then repeated at 200 x g for 10 minutes to support cell pelleting.

68

Supernatant was then discarded, and cells were gently resuspended in 5mL PBS. A cell count was

performed using 0.4% trypan blue (Sigma-Aldrich, Poole, UK) as described earlier. Using information

generated from the cell count, cells were split into 1x106/ml and aliquoted into 1.5mL Eppendorfs.

The sample was then centrifuged at full speed in a microcentrifuge (16000 rpm) to pellet remaining

PBMCs. The supernatant was aspirated completely, and the pellet stored at -80ºC freezer until use

ready for future RNA, protein or fractionation studies.

2.15.3: Extraction of RNA using TRIzol® method from PBMCs

Cell pellets containing 1 x 106 PBMC cells in a 1.5ml Eppendorfs were removed from -80oC and

solubilised from frozen in 500 µL TRIzol® reagent (Thermo-Fisher, Paisley, UK), to breakdown the

cells components whilst keeping the integrity of the RNA intact for 5 minutes. Then 100 µL of

chloroform (Sigma-Aldrich, Dorset, UK) was added and the tubes, shaken vigorously by hand for 15

seconds to mix the solution before being left to settle for 3 minutes. The TRIzol – chloroform mixture

was then centrifuged at 12,000 x g for 15 minutes at 4ºC to prevent RNA degradation (Micro Star

17R, VWR, Lutterworth, UK). After centrifugation, the mixture split into two layers, one containing

the RNA and the other containing DNA. The upper aqueous layer containing the RNA was removed

and placed in a clean RNA/DNA free Eppendorf where 0.25 µL of glycogen (20 µg/µL) (Life

Technologies, Paisley, UK) was also added to support co-precipitate of the RNA from the aqueous

phase, before being mixed gently by inversion. To the aqueous phase/glycogen mixture, 500 µL of

analytical grade isopropanol kept at room temperature (Fisher Scientific, Loughborough, UK) was

added and then left to incubate at room temperature for 10 minutes. After incubation, the sample

was centrifuged for 10 minutes at 12,000 x g at 4ºC in the micro centrifuge. Supernatant was

discarded and the pellet was washed with 75% ethanol (Fisher Scientific, Loughborough, UK) and

vortexed for 15 seconds. The pellet was then centrifuged for a further 5 minutes at 7500 x g at 4ºC.

Supernatant was discarded and the pellet was air-dried for 10 minutes at room temperature. After

air-drying, the pellet was dissolved in 50 µL of nuclease-free water, then incubated at 55-60ºC on a

69

block heater (Model: SBH130D/3, Manufacturer: Stuart, Stone, UK) for 10 minutes to support

resuspension of RNA before being taken forward ready for the clean-up phase.

2.15.4: Clean-up of PBMC RNA obtained using the TRIzol® extraction method

After phenol-chloroform extraction, a Qiagen RNeasy Mini kit (Qiagen, Hilden, Germany) was used

to clean-up the RNA as per manufactures details. Briefly, a 50 µL sample of previously prepared RNA

solution was placed in an Eppendorf and made up to 100ul with RNAse free water (Qiagen, Hilden,

Germany). To this solution, 350 µL of buffer RLT from the RNeasy kit was added to the sample and

mixed well, to which 250 µL of analytical grade ethanol (Fisher Scientific, Loughborough, UK) was

added and further mixed well by pipetting. The 700 µL sample was then transferred to the spin

column and centrifuged for 15 seconds at 10,000 x g (Micro Star 17R, Manufacturer: VWR,

Lutterworth, UK) at 4ºC using a RNeasy Mini kit spin column with a 2 mL collection tube; this allows

the spin column with a separate collecting tube below it to filter, clean and capture the RNA before

eluting. The flow-through was discarded.

DNase treatment was performed on the sample in the RNeasy Mini kit spin column using the RNase-

free DNase set (Qiagen, Hilden, Germany) to remove any potential remaining DNA contamination.

Then 350 µL of buffer RW1 was added directly to the column and centrifuged for 15 seconds at

10,000 x g, flow-through was discarded. An 80ul DNase solution (10 µL DNase I stock solution with

70 µL buffer RDD), was then added to the spin column membrane and incubated at room

temperature for 15 minutes. After incubation, 350 µL of buffer RW1 was added to the spin column

and centrifuged for 15 seconds at 10,000 x g, flow-through was discarded. Following centrifugation,

500 µL of buffer RPE was added to the spin column and centrifuged for 15 seconds at 10,000 x g,

flow-through was discarded and then this step repeated. After the wash steps, the spin column was

placed in a clean fresh collection tube and centrifuged at 17,000 x g for 1 minute to dry the

membrane. After spin drying 50 µL of nuclease-free water was added directly to the spin column

70

placed in a 1.5 mL sterile Eppendorf tube for collection, then centrifuged for 1 minute at 10,000 x g

and flow-through collected. RNA yield and purity were determined by absorbance readings at

230/260/280 nm on a Nanodrop 2000 spectrophotometer (Thermo Scientific, Loughborough, UK), as

previously described.

Protein was extracted from PBMCs using same method used for protein extraction in cell line models

(as previously described).

2.16: Data Analysis and Statistics

2.16.1: Calculating relative gene expression

Relative gene expression was calculated using the ΔΔCT method where the gene of interest was

normalised against the control gene GAPDH. Cell line and patient samples were processed into cDNA

and analysed on a qPCR machine which generated results in CT values. CT values were acquired from

the StepOne instrument software (v2.3).

ΔCT = CT (Gene of Interest) - CT(GAPDH)

ΔΔCT = ΔCT(Sample) – (Average ΔCT(Control Sample))

Relative quantification given by following transformation;

RQ = 2-(ΔΔCT)

Gene expression of patient data was collected then split into desired groups and the median gene

expression of the samples were compared against the healthy controls. Mann-Whitney U-tests were

utilised to determine the significance. All tests were carried out using Excel (Microsoft, New Mexico,

USA).

71

Gene expression of cell line data were calculated as described for patient sample data, with the

sample data being compared against the median gene expression. A t-test (α-= 0.05) (Excel) was

used to determine significance.

2.16.2: Calculating relative protein expression

Total relative protein expression was calculated by normalising the band intensity of the protein of

interest against the control protein GAPDH. Protein band intensity were acquired from the Licor

image studio software (digital developer FC, LI-COR Biosciences, Cambridge, UK), with relative

protein expression calculated by dividing the protein of Interest gene expression by the house

keeping gene GAPDH gene expression. Standard deviation error bars were calculated and a t-test (α-

= 0.05) (Excel, Microsoft, California, USA) was used to determine significance.

Samples that were fractionised were split into the four compartments then, using the total protein

expression of the sample’s, the percentage expression of each of the protein of interest were

calculated for each fraction. Standard deviation error bars were calculated as a percentage and a t-

test (α-= 0.05) (Excel) was used to determine significance. Significance stars are given as follows:

(*) p < 0.05; (**) p < 0.01; (***) p < 0.001; (****) p < 0.0001

A p value of less >0.05 was considered as significant. A T-test would be used on each sample Vs a

control to determine the significance, with all tests undertaken within Excel.

72

Chapter 3: Establishing basal expression levels and location percentage of CALR and CD47 in MDS and MPN cell line models

73

3.1: Introduction

Currently there are a lack of specific gene biomarkers that can differentiate between disease

subtypes or stage of the disease progression within myeloid malignancy. Evidence has started to

emerge within solid tumours that alternations in CALR and CD47 levels, two genes linked to a

possible apoptosis vs survival mechanism in malignant cells, could dictate successful response to

chemotherapy (Boasman et al., 2019). Whilst the role of CALR and CD47 in solid tumours has been

less well established; it has been proposed that their role within MDS and MPN could be similar to

their roles in solid tumours, where a concomitant upregulated expression is seen between CALR and

CD47 on the surface of the cell (Boasman et al., 2019), (Figure 3.1).

CALR and CD47 are both present and functional in cell homeostasis and are located throughout the

different components of the cell, from the nucleus through to the membrane (Eladl et al., 2020;

Fucikova et al., 2020). CALR acts as a pro-phagocytic signal and presence of CALR on the cell surface

acts as a marker for the clearance of aged or damaged cells by the immune system (Gardai et al.,

2005). CD47 is involved in many processes; cellular proliferation, apoptosis, adhesion, migration,

immune system homeostasis and signalling to inhibit phagocytosis; giving CD47 the title of the

“don’t eat me signal” (Oldenborg et al., 2016) (Figure 3.1).

74

Figure 3.1: (A) Macrophages ignore normal cells as a result of negative interactions in which the CD47–SIRP-α pair promote

the “don’t eat me” signal. (B) Cancer cells showing high levels of CD47 can avoid phagocytosis despite having a higher level

of calreticulin but (C) blocking CD47 with antibody favours their uptake (Unanue, 2013).

In different human cancers (i.e. AML, acute lymphocytic leukaemia, chronic myeloid leukaemia,

ovarian cancer, glioblastoma, and bladder cancer) (Chao et al., 2010) CALR plays a role in pathways

including activation of the unfolded protein response (UPR), which controls the upregulation of

molecular chaperones and degrading misfolded proteins, calcium signalling pathways and

macrophage mediated immune evasion. CALR upregulation is often counterbalanced by the

concomitant expression of the antiphagocytic CD47 in solid tumours (Chao & Gotlib, 2014). During

oncogenesis upregulation of CALR on the surface of the cell, targeted cancers cells can be detected

and bound by the LDL receptor related protein 1 (LRP1), which is present on the surface of a

macrophage and induce phagocytosis (Chao et al., 2010). When a total upregulation of CALR is

expressed in cancer cells, a concomitant upregulation of the anti-phagocytic CD47 protein on the cell

75

surface also occurs. This response is a key survival mechanism by cancer cells to evade

immunosurveillance and keep the cancer cells alive (Feng et al., 2015). Cells being able to evade

immunosurveillance is a key hallmark for cancer and is achieved by various different mechanisms

(Hanahan and Weinberg, 2011). Within cancer cells high CALR expression promotes signalling for

phagocytosis, while high expression of CD47 protects the cancer cell from an immune response. It is

this ratio of CALR to CD47 that directs the immune response, depending on which signal is

predominate in the cell’s fate (Feng et al., 2015).

Interestingly, it is overexpression of CALR with a concomitant CD47 expression (both mRNA and

protein cell surface levels) that has been correlated with poor prognosis in many different cancers

including mantle cell lymphoma, superficial and invasive bladder cancer, and neuroblastoma (Chao

et al., 2010). Chao et al. (Chao et al., 2010) showed a correlation between CALR overexpression,

together with counteracting CD47 overexpression, exhibiting lower survival rates. Patients across

different solid tumour cancers (lung, colon and ovarian) and in haematological cancers including

AML have exhibited higher expression rates of CD47 which correlate clinical outcomes (Majeti et al.,

2009; Willingham et al., 2012). When CD47 expression increases on the surface of cancer cells it

signals for survival by evading immune surveillance, acting as a marker of anti-phagocytosis (Tsai et

al, 2010; Galli et al., 2015).

Within the haematological disorders, including MDS and MPN, the role of CALR and CD47 has not

been as well characterized. However CD47 over expression on the membrane of cancer cells has

been measured in AML (Jaiswal et al., 2009; Majeti et al., 2009), suggesting CD47 as a potential

therapy. Patients with MDS, can progress on to AML throughout the course of their treatment and

these patients show a gradual increase of CD47 expression throughout the stages of their disease

progresses (Jiang et al., 2013). Jiang et al showed that MDS expressed the CALR/CD47 ratio

differently to solid tumours (Jiang et al., 2013). Initially over expression of CALR and lower levels of

CD47 were witnessed in low risk patients, with this ratio altering as disease progress with increasing

76

CD47 expression on the cell surface being seen with concurrent lower levels of CALR seen in higher

risk patients, suggesting a potential role of CALR/CD47 ratio in myeloid malignancies (Chao et al.,

2010). This suggests that the role of CALR/CD47 ratio in MDS and MPN could act differently to that

seen in solid tumours.

3.2: Aims

The overarching aim of the project is to establish the differences of CALR and CD47 expression levels

in untreated and treated MDS and MPN cell lines, which will provide insight on how current

therapeutic strategies could affect levels of CALR and CD47 in MDS and MPN patients. Specifically,

this chapter will:

• Examine the differences between CALR and CD47 basal levels mRNA and protein comparing

MDS and MPN cell line models

• Expose each cell lines to azacytidine and ruxolitinib to determine the CALR and CD47

expression modifications when cells are treated

• Identify changes in CALR and CD47 within cell compartments during and after treatment

3.3: Methods

Methods are described in brief here only where they differ from the main Methods in Chapter 2 (p

43).

3.3.1: Cell culture and drug treatment

Cell lines were cultured and treated as previously described, with cell survival rates calculated using

trypan blue assay. All cell lines were drugged and grown/harvested to produce n3 (The experiment

was completed three independent times). All treatment concentrations were derived through an

optimisation process. Initially, an appropriate concentration range would be selected informed

through data available across multiple literature sources. Concentrations ranges below and above

77

this given starting concentration were then tested on each cell line for each treatment and then a

suitable concentration for each treatment was picked for use across each cell line model. The final

drug dose used across each cell line was 5µg/ml.

3.3.2: Protein analysis Protein expression analysis was conducted using protein lysate extracted from untreated and

treated samples of all cell line models. Western Blotting was completed. Densitometry readings

were taken from each band for each cell line model on western blots and qualitative analysis was

generated from the densitometry readings, resulting in normalised signals of the target bands.

3.3.3: Gene analysis

Gene expression analysis of CALR and CD47 was conducted using qPCR generated from RNA

extracted from untreated and treated samples of all cell line models.

3.3.4: Fractionisations

All the cell line models were fractionated into the cellular components of membrane, cytoplasm,

cytosol, and nucleus using two fractionation kits.

3.4: Results

3.4.1: CALR and CD47 changes and impact on survival in cell line models

To determine the effect of standard-of-care drugs on cell survival, cell counts from 24h and 48h after

incubating with 5µg/mL of specific drug with re-drugging occurring at 24 hours were evaluated. All

cell lines were subjected to treatment with both azacytidine (MDS treatment) and ruxolitinib (MPN

treatment) as shown in Figure 3.2. Survival rate was calculated as fraction of alive cells from the

total cell pool treated and expressed as %.

78

When treated with azacytidine MOLM-13 and SKM-1 (MDS models) exhibited an overall survival rate

of 71% and 74% respectively. HEL-92 and GDM-1 (MPN models) exhibited an overall survival rate of

86% and 92% respectively, with K562 (MDS/MPN overlap model) exhibiting a survival rate of 77%.

When treated with ruxolitinib, the HEL-92 and GDM exhibited an overall survival rate of 71% and

76%, respectively. MOLM -13 and SKM-1 exhibited an overall survival rate of 97% and 93%,

respectively, K562 exhibiting a survival rate of 81%. The cell line models acted as expected showing

higher death rate when exposed to the appropriate drugs as per specificity.

Figure 3.2: Overall percentage survival rate of each cell line model treated with both azacytidine and ruxolitinib. All cell

lines and drugging were grown/harvested at different time points to the n3 (mean of three independent experiments).

Total cell counts were determined after 48 hours from Live and dead cells after treatment with 5µg/mL of specific drug

with re-drugging occurring at 24 hours. The error bars represent the standard deviation between the n3.

3.4.2: Alterations in CALR and CD47 mRNA expression in untreated & treated cell lines In both MDS (MOLM-13 and SKM-1) cell line models, low expression of CALR and CD47 was seen at

baseline (Figure 3.3). When treated with azacytidine, increased median fold expression of CALR was

shown in both MOLM – 13 and in SKM – 1 (p=0.017); (19-fold and 42-fold, respectively). Increases in

0

20

40

60

80

100

MOLM-13 SKM-1 k562 HEL - 92 GDM - 1Perc

enta

ge o

f liv

ing

cells

pos

t tre

atm

ent

Overall % survival rate of each cell line model treated with azacytidine and ruxolitinib

Azacytidine Ruxolitinib

79

CD47 median fold expression were also shown in both MOLM – 13 (p=0.023) and SKM – 1 (68-fold

and 5-fold respectively). When treated with ruxolitinib, a similar increase of CALR expression to 35-

fold was shown in MOLM-13, however, CD47 expression was unchanged compared to the untreated.

SKM-1 exhibited a significant increase in both CALR and CD47 (10-fold; p=0.045 and 15-fold;

p=0.044) when treated with ruxolitinib.

80

Figure 3.3: Mean expression of CALR and CD47 mRNA in the MDS cell line models - MOLM-13 (A) and SKM-1 (B). Each cell

line model was treated with azacytidine and ruxolitinib. Each mean was comprised from n3 for each cell line model.

Significant increases between the drug treated compared to the untreated were seen in: MOLM-13 showed no significance

in CALR azacytidine treated or in CALR and CD47 ruxolitinib treated. CD47 azacytidine showed significance (T Test; p =

0.023). SKM-1 showed no significance in CD47 levels when treated with azacytidine. Significance increases was shown in

0

10

20

30

40

50

60

70

80

90

Untreated Azacytidine Ruxolitinib

2^(-Δ

ΔCt)

fold

cha

nges

*

Expression of mRNA in untreated and treated MOLM-13 cell line models

CALR CD47

A

0

10

20

30

40

50

60

70

80

90


2^(-Δ

ΔCt)

fol

d ch

ange

s

* * *

Expression of mRNA inuntreated and treated SKM-1 cell line models

CALR CD47

B

81

CALR levels upon azacytidine treatment (p = 0.017) and in CALR upon ruxolitinib treatment (p = 0.045) and CD47 (p = 0.044).

The error bars represent the standard deviation of the n3 between the drug treated samples compared to the control.

When treated with azacytidine and ruxolitinib the intermediate K562 cell line model showed a

similar trend to the ones witnessed in the MDS cell line models with CALR having the greatest

increase when treated with azacytidine and ruxolitinib (70-fold and 35-fold respectively) with

azacytidine treatment showing significance (p=0.034) (Figure 3.4). Increases in CD47 were also seen,

with 11-fold (p=0.035) and 13-fold (p=0.012) increases seen upon azacytidine and ruxolitinib

treatment, respectively.

Figure 3.4: Mean expression of CALR and CD47 mRNA in K562 cell line models. Each cell line model was treated with

azacytidine and ruxolitinib. Each mean was comprised from n3 of the cell line model. Significant increases between the

drug treated compared to the untreated were seen: K562 showed no significance in CALR ruxolitinib treated. Significance

was shown in azacytidine treated CALR (p = 0.034) CD47 (p = 0.035) and in ruxolitinib treated CD47 (p = 0.012). The error

bars represent the standard deviation of the n3 between the drug treated compared to the control.

0

10

20

30

40

50

60

70

80

90

100


2^(-Δ

ΔCt)

fol

d ch

ange

s

* * *

Expression of mRNA in untreated and treated K562 cell line models

CALR CD47

82

When treated with azacytidine the MPN cell lines (HEL-92 and GDM-1) (Figure 3.5) exhibited a

similar trend, showing an increase in both CALR and CD47 expression. CALR in both MPN cell line

models showed greater increases than the CD47 genes, however, in HEL-92 the CALR expression

levels showed a larger fold increase than in GDM-1 (110-fold; p=0.044 and 25-fold, respectively).

CD47 expression levels showed a smaller fold increases than CALR in both HEL-92 and GDM-1 cell

lines (5-fold and 8-fold, respectively). This pattern of gene expression matches the K562 cell line

model but differs to the MDS cell line models.

When treated with ruxolitinib the MPN cell line models showed a different pattern of expression to

the MDS cell line models. Both HEL-92 and GDM-1 exhibited increasing expression of CALR and

CD47, with HEL-92 cells having a higher expression of CALR 49.9-fold (p=0.031) and CD47 68.2-fold

(p=0.041). GDM-1 has lower expression of CALR and CD47 (25-fold and 5-fold, respectively) while

still showing increases, GDM-1 followed a similar pattern to the K562 cell line model. Every cell line

showed an increase in CALR and CD47 expression, indicating that there is a gene response to

treatment in MPN and MDS cell lines.

83

Figure 3.5: Mean expression of CALR and CD47 mRNA in HEL-92 (A) and GDM-1 (B) cell line models. Each cell line model

was treated with azacytidine and ruxolitinib. Each mean was comprised from n3 of each cell line model. Significant

increases between the drug treated compared to the untreated were seen: HEL-92 showed no significance in azacytidine

treated. Ruxolitinib treated showed significance in CALR (p = 0.031) and CD47 (p = 0.041). GDM-1 showed no significance in

0

20

40

60

80

100

120

140

160


2^(-Δ

ΔCt)

fol

d ch

ange

s

* *

Expression of mRNA in untreated and treated HEL-92 cell line models

CALR CD47

A

0

5

10

15

20

25

30

35

40


2^(-Δ

ΔCt)

fol

d ch

ange

s

*

Expression of mRNA in untreated and treated GDM-1 cell line models

CALR CD47

B

84

CD47 azacytidine treated and ruxolitinib treated. Significance was shown in CALR azacytidine treated (p = 0.044). The error

bars represent the standard deviation of the n3 between the drug treated compared to the control.

3.4.3: CALR and CD47 protein expression of untreated and treated cell line models

To determine whether alternations seen in CALR and CD47 mRNA in cell lines translated into protein

levels changes, western blotting was performed. Sample Western Blot plots for CALR, CD47 and

GAPDH in untreated (A), azacytidine-treated (B) and ruxolitinib-treated (C) in all cell line can be seen

in Figure 3.6. A full Western blot example with molecular markers can also be found depicted within

Appendix 3.1.

CALR CD47 GAPDH

A B C A B C A B C MOLM - 13

SKM – 1 K562

HEL – 92 GDM - 1

Figure 3.6: Representative example of western blot samples of CALR, CD47 and GAPDH proteins. Each protein had an

untreated sample = A, azacytidine treated sample = B and ruxolitinib treated = C, for each cell line model. Each protein was

conducted to the n3 of each cell line model

CALR and CD47 protein expression data was generated from the densitometry readings that were

taken from each band for each cell line model on western blots shown in Figure 3.6 Qualitative

analysis was generated from the densitometry readings, resulting in normalised signals of the target

bands. Showing the proteins response to treatment, shown in Figure 3.7.

85

Figure 3.7: Mean protein expression of CALR (A) and CD47 (B) rate expression (normalised against GAPDH expression) in

untreated sample, azacytidine treated sample and ruxolitinib treated for each cell line model. Normalised ratio (NR)

expression rate was determined from normalised western blot analysis. Significant increases between the drug treated

compared to the untreated were seen: Significant changes were witnessed in azacytidine treated MOLM - 13 CALR (p =

0.033) and K562 CALR (p = 0.044), ruxolitinib treated MOLM -13 CALR (p = 0.035), K562 CALR (p = 0.029), HEL - 92 CALR (p =

0.035) and SKM -1 CD47 (p = 0.016). The error bars represent the standard deviation of the n3.

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

MOLM - 13 SKM - 1 K562 HEL - 92 GDM - 1

Nor

mal

ised

rate

exp

ress

ion

* * * * *

Expression of CALR protein in untreated Vs treated cell line models


A

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

MOLM - 13 SKM - 1 K562 HEL - 92 GDM - 1

Nor

mal

ised

rate

exp

ress

ion

*

Expression of CD47 protein in untreated Vs treated cell line models


B

86

Overall protein normalised ratio (NR) expression upon treatment with azacytidine and ruxolitinib,

showed increased expression of CALR in MOLM-13 cells (0.70-NR; p=0.033 and 0.72- NR; p=0.035 vs

control 0.59- NR), respectively (Figure 3.7). Increased CALR protein expression was also observed in

HEL-92 when treated with ruxolitinib (0.51- NR: p=0.035 vs control 0.50- NR). Interestingly CALR

protein showed a decrease expression in K562 cells upon treatment with azacytidine and ruxolitinib

(0.61- NR; p=0.044 and 0.62- NR; p=0.029 vs control 0.64- NR). CD47 protein exhibited a range in

expression over the cell lines, with only a significant reduction seen when treated with ruxolitinib in

SKM-1 cells (0.27- NR: p=0.016 vs control 0.30- NR)

3.4.4: CALR and CD47 localization in MDS in response to treatment

Whilst basal RNA and protein expression has shown alterations within CALR and CD47 expression

upon treatment with azacytidine and ruxolitinib this does not provide insights into how this could

impact the immune response and in particular in terms of cell surface expression. Cell fractionation

was employed to analysed the movement of CALR and CD47 within cell compartments (membrane,

cytoplasm, cytosol and nucleus) when cells were exposed to drugs, enabling the determination of

how treatment impacts upon cellular localisation/membrane expression to start to better link to

how the cells are recognised by macrophages.

In both MDS cell line models CALR protein expression rates on the membrane are seen to

significantly decrease when treated with azacytidine and ruxolitinib (MOLM-13 – 18% membrane

expression in untreated Vs 13% and 15% (p =0.0001), respectively, see Figure 3.8) & SKM-1 - 17%

membrane expression in untreated Vs 7% (p=0.004) and 3%(p=0.001), respectively, see Figure 3.9.

At the same point higher CD47 membrane expression was detected (MOLM-13 –13% upon

azacytidine treatment and 9% upon ruxolitinib treatment (p=0.03) vs 8% in untreated: SKM-1 –8%

upon azacytidine and 7% upon ruxolitinib vs 5% untreated), although only MOLM - 13 treated with

ruxolitinib showed significance. However, higher CD47 expression is witnessed in the azacytidine

87

treated cells which is the drug specific to their treatment. Localisation of the CD47 indicates that it is

being externalised as the cytoplasm and the cytosol as a combined average decrease, this is the

opposite pattern of expression compared to CALR localisation where upon treatment it is becoming

internalised.

88

B MOLM-13

UNTREATED AZA RUXO M CP CS N M CP CS N M CP CS N

CALR

CD47

Figure 3.8: Percentage expression of CALR and CD47 within fractionated cells are shown taken from densitometry of

western blots (A). Representative example of western blot samples of CALR and CD47 proteins in different fractions of MDS cell line model MOLM-13 (B). Each cell was split into cellular fractions (membrane (M), cytoplasm (CP), cytosol (CS) and

nucleus (N)), proteins for each fraction have been shown in the untreated sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. Mean protein expression of CALR and CD47 in fractionated MDS cell line

model MOLM-13 with untreated sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. These bands were achieved using the same concentration of cells using the same protocol and volumes in the fractionation kits. Each set of protein fractionations then had their average band expressions calculated across the 4 fractions giving a total % for each fraction. Azacytidine treated cells showed no significance. Significance was seen in ruxolitinib treated cells in the membrane in both CALR (p= 0.0001) and CD47 (p=0.03). The error bars represent the

standard deviation of the n3.

0

10

20

30

40

50

60

70

80

90

100

% w

ithin

the

cells

*** *

Percentage of cellular protein location in untreated Vs treated MOLM-13 cell line models

CALR CD47

A

89

B SKM1


CALR

CD47


western blots (A). Representative example of western blot samples of CALR and CD47 proteins in MDS cell line model SKM-1 (B). Each cell was split into cellular fractions (membrane (M), cytoplasm (CP), cytosol (CS) and nucleus (N)), proteins for

each fraction have been shown in the untreated sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. Mean protein expression of CALR and CD47 in fractionated MDS cell line model SKM-1 with untreated

sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. These bands were achieved using the same concentration of cells using the same protocol and volumes in the fractionation kits. Each set of protein

fractionations then had their average band expressions calculated across the 4 fractions giving a total % for each fraction. Significant increases between the drug treated compared to the untreated were seen: SKM1 showed significance in

azacytidine treated cells and ruxolitinib treated cells in the membrane in CALR (p= 0.004 and 0.001, respectively). The error bars represent the standard deviation of the n3.

0

10

20

30

40

50

60

70

80

90

100

% w

ithin

the

cells

** **

Percentage of cellular protein location in untreated Vs treated SKM -1 cell line models

CALR CD47

A

90

3.4.5: Examining changes in cellular locations of CALR and CD47 in intermediate MDS/MPN

As with MDS models the fractionated K562 intermediate protein samples were also analysed (Figure

3.10). CALR expression shows a relatively similar percentage expression on the membrane across the

untreated, azacytidine treated and ruxolitinib treated cells of around 2%. Interestingly, the CD47 is

seen to significantly decrease in expression when the untreated is compared to the azacytidine and

ruxolitinib treated cells (26% Vs 21% p=0.012 and 22% p=0.048, respectively). The intermediate

K562 cell line showed much higher level of CD47 both in untreated and after treatment compared to

the MDS lines (as seen in Figure 3.8 and 3.9).

91

B K562


CALR

CD47


western blots (A). Representative example of western blot samples of CALR and CD47 proteins in intermediate cell line model K562 (B). Each cell was split into cellular fractions (M = membrane, CP=cytoplasm, CS=cytosol and N= nucleus),

proteins for each fraction have been shown in the untreated sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. Mean protein expression of CALR and CD47 in fractionated intermediate cell line model K562 with untreated sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. These

bands were achieved using the same concentration of cells using the same protocol and volumes in the fractionation kits. Each set of protein fractionations then had their average band expressions calculated across the 4 fractions giving a total %

for each fraction. Significant increases between the drug treated compared to the untreated were seen: SKM1 showed significance in azacytidine treated cells and ruxolitinib treated cells in the membrane in CD47 (p= 0.012 and 0.048,

respectively). The error bars represent the standard deviation of the n3.

0

20

40

60

80

100

120

% e

xpre

ssio

n w

ithin

the

cells

* *

Percentage of cellular protein location in untreated Vs treated K562 cell line models

CALR CD47

B

92

3.4.6: Examining changes in cellular locations of CALR and CD47 in MPN

Fractionations of the cell line models were continued in the MPN cell line models HEL-92 (Figure 3.11) and GDM-1 (Figure 3.12).

93

B HEL-92


CALR HEL -92

CD47


western blots (A). Representative example of western blot samples of CALR and CD47 proteins in the MPN cell line model HEL-92 (B). Each cell was split into cellular fractions (membrane (M), cytoplasm (CP), cytosol (CS) and nucleus (N)), proteins for each fraction have been shown in the untreated sample, azacytidine treated sample and ruxolitinib treated sample for

each cell line model. Mean protein expression of CALR and CD47 in fractionated MPN cell line model HEL-92 with untreated sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. These bands were achieved using the same concentration of cells using the same protocol and volumes in the fractionation kits. Each set of protein fractionations then had their average band expressions calculated across the 4 fractions giving a total % for each

fraction. Azacytidine treated cells showed no significance. Significance was seen in ruxolitinib treated cells in the membrane in CD47 (p=0.01). The error bars represent the standard deviation of the n3.

0

10

20

30

40

50

60

70

80

90

100

% e

xpre

ssio

n w

ithin

cel

ls

*

Percentage of cellular protein location in untreated Vs treated HEL-92 cell line models

CALR CD47

A

94

B GDM -1


CALR

CD47


western blots (A). Representative example of western blot samples of CALR and CD47 proteins in the MPN cell line model GDM-1(B). Each cell was split into cellular fractions (membrane (M), cytoplasm (CP), cytosol (CS) and nucleus (N)), proteins for each fraction have been shown in the untreated sample, azacytidine treated sample and ruxolitinib treated sample for

each cell line model. Mean protein expression of CALR and CD47 in fractionated MPN cell line model GDM-1 with untreated sample, azacytidine treated sample and ruxolitinib treated sample for each cell line model. These bands were achieved using the same concentration of cells using the same protocol and volumes in the fractionation kits. Each set of protein fractionations then had their average band expressions calculated across the 4 fractions giving a total % for each

fraction. Significant increases between the drug treated compared to the untreated were seen: SKM1 showed significance in azacytidine treated cells and ruxolitinib treated cells in the membrane in CD47 (p= 0.04 and 0.008, respectively).

Ruxolitinib showed CALR significance in the membrane (P=0.03). The error bars represent the standard deviation of the n3.

0

10

20

30

40

50

60

70

80

90

100

% e

xpre

ssio

n w

ithin

cel

ls

* * **

Percentage of cellular protein location in untreated Vs treated GDM- 1 cell line models

CALR CD47

A

95

The MPN cell line models show that HEL-92 cells exhibit larger protein alterations in CALR/CD47

upon drugging. CALR is seen to decrease from 29% in the untreated cells to 24% in the azacytidine

treated and 13% in ruxolitinib treated. CD47 is seen to increase from 2% in the untreated cells to

13% in the azacytidine treated and a significant increase of 33% in ruxolitinib treated (p=0.01).

Greater differences in CALR and CD47 membrane expression were seen in the ruxolitinib treated

cells in the MPN cell line models, which is the intended treatment for MPN. Mimicking the HEL-92

cell line model trend the GDM-1 cell line model exhibited a small decrease in CALR and an increase

of CD47 membrane expression. CALR is seen to decrease from 16% in the untreated cells to 14% in

the azacytidine and a significant decrease of 11% in ruxolitinib treated (p=0.03). CD47 is seen to

significantly increase from 11% in the untreated cells to 12% in the azacytidine (p=0.04) and 17% in

ruxolitinib treated (p=0.008).

The protein expression percentage exhibits a trend that shows the internalisation of CALR from the

membrane Vs an externalisation of the CD47 gene after treatment with both azacytidine treated

sample and ruxolitinib compared to untreated cells. The membrane protein in the MPN cell lines

supports what is also seen in the MDS model SKM -1 and K562 cell lines with regards to changes in

CALR and CD47 upon treatment.

3.5: Discussion

Overexpression of CALR has been documented to produce a pro-phagocytic signal in solid tumours

and it is often counteracted by a concomitant over expression of the antiphagocytic CD47 (Daitoku

et al., 2016), however this function is unknown in MDS and MPN. Once treated with either

azacytidine or ruxolitinib the MDS and MPN cell line models elicited changes in CALR and CD47 in

respond to both treatments. Unsurprisingly, each disease had a greater response to the drug

intended for that disease (MDS – azacytidine and MPN – ruxolitinib). The MDS cell line models upon

treatment with azacytidine (used clinically to treat MDS) showed a reduced overall survival rate of

71% for MOLM-13 and 74% for SKM-1, compared to the overall survival rate when treated with

96

ruxolitinib of 97% and 93% respectively. Alternately, the MPN cell lines models showed an opposite

pattern when treated. Where upon treatment with azacytidine, HEL-92 and GDM-1 showed an

overall survival rate of 86% for and 92%, respectively, compared to MOLM-13 and SKM-1 overall

survival rate of 71% and 76, respectively. Azacytidine has been shown to dramatically reduce survival

rate in both MDS cell line models (Dubois et al., 2019); with the cell survival rates corresponding

with published data that MDS has a sensitivity to azacytidine whereas MPN has less sensitivity to

azacytidine. This sensitivity to drug treatment was also seen in MPN cell lines models where they

showed greater cell sensitivity to ruxolitinib (used to treat MPN) without inducing complete cellular

death (Ishida et al., 2018). Each cell line model behaved as expected to the treatment before looking

specifically at CALR and CD47 expression upon treatment.

Survival rates of the treated cell line models showed that a specific cell line models does not have a

sensitivity in response to the unintended drug, indicating that intended drug for the specific disease

induces cells into cell death. However, lower expression rates in the anti-phagocytotic signal of

CD47 (Brightwell et al., 2016) and the pro-phagocytosis signals of CALR (Araki & Komatsu, 2017) in

response to the untreated cells and cells treated with the intended drug treatment, indicating that

the unintended drug could have the potential for better treatment response compared to intended

drug treatment in certain settings.

CALR and CD47 exhibit a concomitant expression, reflecting a mechanism of apoptosis vs survival in

response to chemotherapy (Boasman et al, 2019). Increasing expressions of CALR and CD47 are also

witnessed in K562 and the MPN GDM-1 cell line models. However, they exhibit higher CALR

expressions compared to CD47 in the treated samples, following the same expression patterns as

seen in solid tumours (Chao et al., 2010). MDS cell line models (MOLM-13 and SKM-1) and the MPN

cell line model HEL-92, when treated with azacytidine, show an increase of both CALR and CD47,

however CD47 is shown to be higher than the CALR gene expression. Interestingly in response to

treatment with unintended drug for that disease the CALR and CD47 gene still increases in

97

expression, however, the mean expression is a lot lower than when treated with intended drug.

Indicating that a similar process was occurring but less specifically.

The protein translations do not fully support the mRNA data in response to treatment as different

cell lines exhibits increases in CALR and CD47 expression to treatment while others exhibit

decreases. Upon treatment with azacytidine decreases in CALR expression were seen in SKM-1,

K562, HEL-92 and GDM-1. While upon ruxolitinib treatment increases in CALR expression was seen in

MOLM-13, SKM-1, HEL-92 and GDM-1 cell lines, CD47 only showed increases in azacytidine treated,

K562, HEL-92 and GDM-1 cells and ruxolitinib treated K562 and GDM-1 cells. Total expressions of

CALR were larger than CD47 in all models: untreated, azacytidine and ruxolitinib treated cells in

both MDS and MPN cell line models. This differs to the published data where we would expect to

see an increase of protein in response to treatment where CALR is expressed greater than CD47

(Chao et al., 2010), instead it is shown that CD47 is expressed greater making it the dominate signal.

CALR and CD47 membrane expression is vital to determining cell fate. Interestingly, when

fractionated, the location of the proteins followed a similar pattern across the samples and differed

to what was previously witnessed in other cancers (Chao et al., 2010). CALR has different roles

within the cell, one of which is to signal for cell death in regular and cancer cells (Unanue, 2013).

CALR interacts with the protein LDL receptor-like protein on the macrophage, telling it that the cell is

damaged and needs to be destroyed (Fucikova et al., 2018); a mechanism that is widely witnessed in

cancer cell phagocytosis (Chao et al., 2010). Total CALR expression percentage was seen to be

internalised away from the membrane into the cytoplasm and cytosol, when comparing untreated to

treated samples. CD47 showed an opposite reaction to treatment, exhibiting greater expression on

the cell membrane. CD47 on the cell membrane interacts with SIRPα on the cell membrane of the

macrophage (Unanue, 2013). Interaction between the CD47 on the diseased cell and the SIRPα on

the macrophage cell signals a recognition of “self”, preventing the macrophage from phagocytosing

the diseased cell (Oldenborg & Per-Arne, 2013). Many types of cancers exhibit large amounts of

98

CD47 leading them to avoid phagocytosis by evading recognition of macrophages (Chao et al, 2012).

The percentage expression of CD47 increases upon treatment with azacytidine and ruxolitinib

indicating how the cell is attempting to protect itself, hiding from the immune system.

Lots of data was generated in this chapter. The significant results can be found in the mRNA data

azacytidine treated showed CALR expression in SKM-1, K562, and GDM-1 cells were shown to

increase significantly with ruxolitinib treated SKM-1 and HEL-92 cells shown to also significantly

increase CALR expression. CD47 was also seen to be significantly increase in MOLM-13 and K562

were shown to increase significantly with ruxolitinib treated SKM-1 and HEL-92 cells shown to also

significantly increase CD47 expression. Protein expression in MOLM-13 azacytidine and ruxolitinib

cells showed significant increases in CALR as did ruxolitinib treated HEL-92 cells. Interestingly, K562

cells exhibited a significant decrease in expression. CD47 only showed a decrease in expression in

ruxolitinib treated SKM-1 cells. The expression of proteins on the membrane showed significant

decreases in CALR expression in azacytidine treated SKM-1 cells and ruxolitinib treated MOLM-13,

SKM-1, GDM-1 cells. Significant increases in CD47 protein expression were seen in azacytidine

treated GDM-1 cells and ruxolitinib treated MOLM-13, HEL-92 and GDM-1 cell. Interestingly, CD47

membrane expression was seen to significantly decrease in both azacytidine and ruxolitinib treated

K562 cells.

This work has shown that in the untreated vs treated samples there is a concomitant increase of

total gene expression of CALR and CD47 and some of the cell lines protein expressions. Solid

tumours have been witnessed across different cancers to show increases of CALR and CD47 in mRNA

and protein expression (Chao et al., 2010). Some MDS and MPN cell lines showed no increases with

CALR, whilst still being expressed greater compared to CD47 upon treatment. This indicates that in

MDS and MPN cell lines models CALR and CD47 do not correlate to the published data showing no

concomitant expression as an increase in CALR is not always shown by an increase in CD47. In some

cell lines CD47 is shown to increase independently of CALR. Indicating that CALR and CD47 do not

99

work together and CD47 does not react to increases in CALR in response to treatment. The location

of CALR and CD47 were then looked at from the total expression. It was also demonstrated that the

percentage of the protein that had migrated to the cell membrane was a lot higher for CD47 which

externalised its expression compared to the CALR, which internalised its expression upon treatment.

This differs to the published data on solid tumours, concluding that myeloid malignancies MDS and

MPN could potentially behave differently to solid cancers.

100

Chapter 4: CALR & CD47 expression in patient with MDS and MPN

101

4.1: Introduction

Previously in chapter 3, work suggested that CALR and CD47 act differently in MDS and MPN cell

lines compared to solid cancers. In 2007 Obeid et al observed increased levels of CALR on the cell

surface when patients developed cancer, as an attempt to signal abnormal cells to the immune

system (Klampfl et al., 2013). (Obeid et al., 2007). Many different solid tumours (Chao et al., 2010),

including: Gastric cancers (Lee et al., 2019), breast cancer (Wu., et al 2019), pancreatic (Sheng et al.,

2014) and neuroblastoma (Hsu et al., 2005) have shown a degree of CALR overexpression.

Whilst not currently extensively studied, some investigation of CALR in MPN patients has been

undertaken. Interestingly, high levels of CALR have been observed in MPN, in both CALR wild-type

and mutated forms of the disease (Klampfl et al., 2013), despite CALR mutations only being found in

only 4.4% of MPN patients (Mikic et al., 2019). Interestingly, patients that have mutated CALR has

been seen to have a greater survival rate upon treatment compared to patients that over express

wild-type CALR (Heuser et al., 2014).

CD47 expression strongly correlates with worsening clinical outcomes in many solid tumours (Jaiswal

et al., 2009; Willingham et al., 2012) and it is found elevated together with CALR in lung cancers

(Arrieta et al., 2020), colorectal cancers (Fujiwara-Tani et al., 2019) breast cancer and glioblastomas

(M. Liu et al., 2019). As previously described, inhibition of phagocytosis occurs when CD47 binds to

the signal regulatory protein alpha (SIRPα) on the surface of the macrophage, tumours then use this

mechanism to evade immune surveillance (Zhao et al., 2018). Taken together with the cell line work

in MDS and MPN undertaken in Chapter 3, the data suggests that further investigating of CALR and

CD47 within patient samples is now warranted

102

4.2: Aims

In this chapter, the level of expression of CALR and CD47 was measured in patient peripheral blood

samples collected from MDS and MPN patients to determine any correlation with disease subtype

and stage of progression. The following steps were performed:

• Initial screening of MDS and MPN patient samples baseline CALR and CD47 expression was

undertaken to better understand potential differences between MDS and MPN in

comparison to healthy age matched controls.

• Determine if any changes in CALR and CD47 total RNA and protein expression within MDS

and MPN also correlated with alterations in the subcellular localisation.

• Correlate the data generated on CALR and CD47 with different stages of disease progression

and/or subgroupings within the MDS and MPN.

4.3: Methods

Methods are described in brief here; for full methodology refer to Chapter 2 (p 48).

4.3.1: Processing of patient samples

Patient samples peripheral blood were collected from Boston and Lincoln hospitals, UK. In this study

mutation status was not looked at, but samples were collected from 16 MDS patients – 4 low risk, 4

intermediate-1 risk, 4 intermediate-2 risk and 4 high risk patients. All MDS patients were receiving

treatment. MPN patients consisted of 8 ET patient samples (4 untreated, 4 treated), 4 PV patients (2

untreated, 2 treated) and 4 MF patients (2 untreated, 2 treated). Sample collection was performed

as previously described in Chapter 2 using ficol-plaque. Optimisations on technique and ficol-plaque

used were conducted on healthy volunteer patient samples. Once processed optimisations on the

number of cells used per technique was conducted, it was determined that 1x106 /ml of cells gave

the most consistent results across the different techniques and was therefore employed across all

experiments.

103

4.3.2: Protein analysis

Protein expression analysis was conducted using protein lysate extracted from untreated and

treated samples of patient samples. Western Blotting was completed. Densitometry readings were

taken from each band for each cell line model on western blots and qualitative analysis was

generated from the densitometry readings, resulting in normalised signals of the target bands.

4.3.3: Gene analysis

Gene expression analysis of CALR and CD47 was conducted using qPCR generated from RNA

extracted from untreated and treated samples of all cell line models.

4.3.4: Fractionisations

All the cell line models were fractionated into the cellular components of membrane, cytoplasm,

cytosol, and nucleus using two fractionation kits.

4.4: Results

4.4.1: Initial analysis of Total RNA & Protein Expression of CALR and CD47 in MDS and MPN patient samples

Total RNA and protein expression of both CALR and CD47 was measured in a cohort of patients with

MDS (16 samples) and MPN (16 samples) and compared against a control group (4 controls) (Figure

4.1), with increases in both CALR and CD47 observed in both MDS and MPN patients compared to

controls.

MDS patients show a greater increase in CALR mRNA levels (11.07-fold) (P=0.041) and a smaller

increase in CD47 mRNA (2.02-fold) when compared to the healthy controls. These changes were

mirrored in the protein expression showing a large increase in CALR normalised ratio (NR) expression

(14.2-NR) rather than a smaller increase in the CD47 expression (2.37- NR).

104

MPN showed a different expression pattern with no increase in CALR when compared to the control

(1.50-fold) but a significant increase in CD47 (4.83-fold). Similarly, the mRNA results, the protein

expression of CALR and CD47 in MPN patients shows a small decrease in the CALR (4.36- NR), with a

larger increase in the CD47 (11.3- NR). The presence of large error bars was seen in MDS, indicating a

potential further analysis required into the different MDS sub-groups.

C Whole protein CALR CD47 GAPDH

MDS MPN

Figure 4.1: RNA (A) and protein (B) expressions were gathered from patient samples. Mean expression of cDNA from RNA and protein of CALR and CD47 in control, MDS and MPN are depicted. Representative example of western blot samples of

CALR and CD47 proteins (C). Each group was comprised of patient across the disease and treatment progression. Each

0

5

10

15

20

25

CONTROL MDS MPN

Chan

ges i

n RN

A fo

ld e

xpre

ssio

n vs

hou

seke

epin

g ge

ne

*

Total RNA fold expression of CALR and CD47 in healthy controls Vs MDS and MPN patients

CALR CD47

A

0

5

10

15

20

25

30

CONTROL MDS MPN

Nor

mal

ised

rate

exp

ress

ion

Total protein fold expression of CALR and CD47 in healthy controls Vs MDS and MPN patients

CALR CD47

B

105

mean was comprised from a minimum of 4 patient per group. Significant increases between the patient samples compared to the healthy controls were seen: MDS patients showed significance in CALR RNA expression (p = 0.041) against controls.

Representative example of western blot samples of CALR, CD47 and GAPDH proteins in the MDS and MPN patient samples. Each cell was lysed with RIPA buffer then loaded to a gel at 10mg/well. Each sample was run in duplicate on the same blot

then replicated. The error bars represent the standard deviation across the group of patient samples, resulting in high error bars.

4.4.2: In-depth look at MDS disease severity sub-groups

When MDS patients were stratified according to the IPSS, low risk, intermediate-1, intermediate-2

and high risk (Figure 4.2) the levels of CALR and CD47 mRNA increased with disease progression

overall, however the pattern is not consistent, with intermediate-1 patients exhibiting an increase

that is still greater than the control group, albeit at a lower level than the other stages (low risk =

6.28-fold, intermediate-1 = 2.02-fold, intermediate-2 = 18.76-fold and high risk = 17.93-fold). CD47

initially exhibits a decrease in expression in low- risk patients compared to the control (0.04-fold vs

1.01-fold, respectively), however as disease progresses a progressive increase in CD47 expression is

observed, from low to high risk (intermediate-1 0.36-fold, intermediate-2 1.47-fold and high risk

7.34-fold). Interestingly, CD47 throughout the disease progression shows a positive association with

an increase at each stage, with low risk patient’s expressing lower levels of CD47 and high-risk

patients showing higher expression levels. Intermediate-2 exhibited some variability between

samples in mRNA levels as demonstrated by the high error bars.

When looking at corresponding protein levels, CALR showed an initial slight decrease in expression in

low risk patients compared to controls (from 5.1 NR in controls to 3.7- NR in low risk), but then

increased expression in the following MDS subgroups (intermediate-1 = 6.1-fold, intermediate-2 =

31.83- NR p = 0.037 and high risk = 10.39- NR), similar to the mRNA. CD47 initially exhibited a

significant decrease in expression in low- risk patients compared to the control (0.41- NR (p=0.0033)

vs 2.2- NR, respectively). A progressive increase in expression is then observed as disease

progressed in CD47, from low to high risk (intermediate-1 = 0.47- NR, intermediate-2 = 3.9- NR

(p=0.0012) and high risk = 4.87- NR). Interestingly, CD47 throughout the disease progression

increases at each stage, with low risk patient’s expressing lower levels of CD47 and high-risk patients

106

showing higher expression levels. This suggests the protein backs up the mRNA expression as they

show similar patterns.

Figure 4.2: Mean expression of cDNA from RNA (A) and protein (B) of CALR and CD47 in different MDS risk groups (low, intermediate – 1, intermediate – 2, High) Vs a control group. Each group was comprised of patient across disease

progression. Significant increases between the MDS disease progression groups compared to the healthy controls were seen: MDS low risk patients showed significance in RNA expression in CD47 expression (p = 0.027). MDS patients showed

significance in protein expression in Int-2 CALR expression (p = 0.037), Low risk CD47 expression (p=0.0033) and Int-1 CD47

0

5

10

15

20

25

30

35

40

CONTROL LOW INT-1 INT-2 HIGH

Chan

ges i

n RN

A fo

ld e

xpre

ssio

n vs

ho

usek

eepi

ng g

ene

Total RNA fold expression of CALR and CD47 in healthy controls Vs MDS subgroups

CALR CD47

A

0

5

10

15

20

25

30

35

40

45

CONTROL low int-1 int-2 high

Nor

mal

ised

rate

exp

ress

ion ** * **

Total protein fold expression of CALR and CD47 in healthy controls Vs MDS subgroups

CALR CD47

B

107

(p=0.0012). The error bars represent the standard deviation across the group of patient samples, resulting in high error bars.

4.4.4: Examining changes in cellular locations of CALR and CD47 in MDS subgroups

The work in this chapter was informed by the results from chapter 3. Therefore, although the other

compartments were examined, membrane expression was therefore focused on. Upon examining

cellular localisation of CALR and CD47 protein within the MDS subgroups (Figure 4.3), when

comparing the protein localisation to the controls, the CALR membrane expression showed large

increases in membrane protein expression (Low risk MDS – 26.73% CALR membrane expression Vs

control – 12.98%), however this was more variable amongst the proceeding subgroups with Int-1

group showing a large decrease compared to the low risk group (5.05%) and both int-2 and high risk

showing similar levels of protein to that seen in the low group (28.07% and 24.22%, respectively)

compared to the control, suggesting the CALR is being internalised.

CD47 membrane protein shows an initial decrease in the low risk subgroups vs the controls and then

a progressive increase in expression was seen at every corresponding subgroup. Significant

decreases were seen in the low risk and int-groups when compared to the controls with a further

decrease seen in Int-2, followed by a large increase witnessed in the high-risk group (Low – 1.41%

(p=0.00014), Int-1 – 3.33% (p=0.034),Int-2 - 14.66%, high-risk - 47.30% Vs Control 21.91%). The

expression rates indicate a trend where CALR is internalised away from the membrane as the

disease progresses, while CD47 is externalised.

108

B Fractionated protein CALR CD47 M CY CS N M CY CS N

MDS

MPN

Figure 4.3: Protein (A) expressions were gathered from patient samples. Mean expression of CALR and CD47 protein on the membrane of the cells in different stages of (low, intermediate – 1, intermediate – 2, High) Vs a control group. Each group was comprised of patient across treatment progression. Representative example of western blot samples of CALR and CD47 proteins (C). Each cell was split into cellular fractions (membrane (M), cytoplasm (CY), cytosol (CS) and nucleus

(N)). Significant increases between the MDS disease progression groups compared to the healthy controls were seen: MDS patient’s protein showed significance in CD47 low risk and int-1 expressions (p=0.00014 and p=0.034, respectively) vs

controls. Representative example of western blot samples of CALR and CD47 proteins in MDS and MPN patients. The error bars represent the standard deviation across the group of patient samples, resulting in high error bars.

4.4.5: In-depth look at MPN untreated Vs treated patient samples

To understand the impact of treatment in CALR and CD47 expression, the level of expression in

treated MPN patient’s vs untreated was measured (Figure 4.4) and then stratified according to the

MPN subtypes (Figure 4.5). Different lines of therapy (hydroxyurea, interferon alpha or ruxolitinib)

were used, but due to the small size of each treatment sub-group, we could not further stratify

according to different cytoreductive drug. As described in the introduction use of different

0

10

20

30

40

50

60

70

80

90

CONTROL LOW INT -1 INT -2 HIGH

% e

xpre

ssio

n w

ithin

cel

ls

*** **

Membrane expression of CALR and CD47 protein in MDS subgroups Vs control

CALR CD47

A

109

treatment options could account for the variability seen in the overarching subgroups above and

variability seen by the error bars.

CALR protein shows a down regulation in MPN cells upon treatment (untreated - 16.45- NR vs

treated – 3.44- NR). CD47 protein expression shows a large increase (untreated – 8.38- NR vs

control – 10.75- NR). These changes indicate that as the disease progress, there is a recognition of an

unhealthy cell and upregulation of CALR occurs to signal for its removal. An upregulation of CD47 in

treated patients comparing with untreated suggesting a strong “do not eat me” signal expressed by

cells when attacked by a cytotoxic agent. Interestingly, despite a reduction in CALR being observed

after treatment, the level of protein reduced suggesting how the “do not eat” message prevails over

the pro-phagocytic signal.

Figure 4.4: Mean expression of CALR and CD47 protein in untreated (8 samples) and treated MPN (8 samples) patients. Each group was comprised of patient across subgroups treatment progression. The error bars represent the standard

deviation across the group of patient samples, resulting in high error bars.

When MPN samples are split into MF, PV and ET subgroups, they still maintain similar changes in

CALR and CD47 protein expression in response to treatment (Figure 4.5). MF patients showed CALR

and CD47 protein responded to treatment with a decrease in CALR expression (untreated - 9.37- NR

0

5

10

15

20

25

30

35

40

45

Untreated Treated

Nor

mal

ised

rate

exp

ress

ion

Total protein fold expression of CALR and CD47 in untreated Vs treated and treated MPN patients

CALR CD47

110

vs treated 4.20- NR) andCD47 protein exhibited a down regulation upon treatment (untreated –

33.5- NR vs treated 5.30- NR).

PV patients showed CALR and CD47 protein responded to treatment with a decrease in CALR

expression (untreated – 1.30- NR vs treated 0.20- NR), while CD47 protein exhibited a large

upregulation upon treatment (untreated – 6.34- NR vs treated 15.96- NR). ET patients exhibited a

decrease in CALR expression (untreated – 36.7- NR vs treated 1.19- NR), while CD47 protein

exhibited an upregulation upon treatment (untreated – 0.84- NR vs treated 1.99- NR).

Figure 4.5: Mean expression of protein of CALR and CD47 in untreated and treated MPN subgroups patients (ET, PV, MF). Each group was comprised of patient across subgroups treatment progression. MPN patients showed no significance in expression. The error bars represent the standard deviation across the group of patient samples, resulting in high error bars.

4.4.6: Examining the cellular locations of CALR and CD47 between different MPN subgroups

To understand whether the changes seen in mRNA and protein levels between untreated and

treated MPN subgroups reflect in modifications in cell membrane levels of CALR and/or CD47

proteins cell fractions membrane were analysed (Figure 4.6).

0

10

20

30

40

50

60

70

80

MF Untreated MF Treated PV Untreated PV Treated ET Untreated ET Treated

Nor

mal

ised

rate

exp

ress

ion

Total protein expression of CALR and CD47 in untreated vs treated MPN subgroups

CALR CD47

111

The membrane expression is shown as a percentage of the overall expression. Interestingly MF and

PV membrane proteins responded differently upon treatment compared to ET. MF patients had

significant reduction in CALR levels in membrane levels after treatment (untreated – 43.3% vs

treated 2%) while CD47 protein exhibited a large upregulation on membrane expression (untreated

– 11.8% vs treated 34.3%). PV patients also showed a decrease in CALR membrane expression

(untreated – 36% vs treated 2.2%) and increase in CD47 membrane protein expression upon

treatment (untreated – 11.4% vs treated 35.9%). Interestingly, ET patients showed a decrease in

both CALR and CD47 membrane expression upon treatment: CALR (untreated – 21% vs treated

1.19%), CD47 (untreated – 22% vs treated 16.6%), suggesting perhaps a separate mechanism of

immune-response within this MPN subgroup.

Figure 4.6: Mean expression of CALR and CD47 protein of membrane expression in untreated and treated MPN subgroups patients (ET, PV, MF). Each group was comprised of patient across subgroups treatment progression. The error bars

represent the standard deviation across the group of patient samples, resulting in high error bars.

4.5: Discussion

When compared to healthy aged-matched controls, MDS and MPN patients exhibited an increase in

CALR and CD47 expression. Overall, level of CALR and CD47 increased as seen in solid cancers

0

10

20

30

40

50

60

70

MF Untreated MF Treated PV Utreated PV Treated ET Untreated ET treated

% e

xpre

ssio

n w

ithin

cel

ls

Membrane protein expression of CALR and CD47 in untreated vs treated MPN subgroups

CALR CD47

112

although the ratios of these seen in MDS and MPN vary (Chao et al., 2010). MDS showed larger

expression of CALR whereas MPN exhibited larger expression of CD47.

Upon closer analysis of MDS, when split according to the IPSS, inconsistent patterns of CALR were

seen with an initial decrease in the low risk group before increases were then seen in intermediate-

1, intermediate-2 and high-risk groups while CD47 showed a more consistent increase as the disease

progressed. Fractionation data confirmed a large upregulation of CALR on the membrane of the cells

initially and then a decrease in expression during evolution into high-risk MDS. CD47 externalised

throughout disease progression with lower percentages of CD47 in low risk and progressive increase

moving into high risk. The Initial upregulation of CALR indicates that the cell wants to signal for cell

phagocytosis (Schcolnik-Cabrera et al., 2019) but is quickly counteracted by CD47 upregulation with

the anti-phagocytic message to prevail (Liu et al., 2017).

Total protein expression in MPNs upon treatment showed decreases of CALR in all subgroups (ET,

PV, MF), interestingly, CD47 expression significantly increases during therapy in MF and PV but not

in ET, suggesting a potential different mechanism of interaction with the immune system in ET

patients comparing with MF and PV. Fractionations of the MPN subgroups showed patients of all

subgroups responded similarly with decreases in membrane CALR protein percentage expression

upon treatment but increases in CD47 externalization in MF and PV (Jaiswal et al., 2009). Overall

prognosis in ET patients is favourable, and the reduced expression of CD47 in response to treatment

may correlate with the positive outcome because the ET cells can be more easily recognised and

destroyed by the immune system.

Further data are required to understand if ET progression into MF or AML causes changes to the

CD47/CALR ratio during transformation. Protein expression of CALR have shown little consistencies

in expressions patterns in MDS and MPN. CD47 however, follows a more consistent pattern in MDS

progression as well as after therapy in PV and MF and could represent a new therapeutic target. For

113

example, the inhibition of CD47 interaction with macrophages could suppress the do not eat

message and help favour the immune response against MDS and MPN cells.

114

Chapter 5: Investigating whether CALR and CD47 work together directly or independently in determining response to therapy within MDS and MPN cell models

115

5.1: Introduction

The phagocytic message expressed via CALR in cancer cells is thought to counteract the strong “do

not eat me” immune escape mechanism deployed via CD47 overexpression (Chao et al., 2010).

Interestingly, previous work has looked at therapeutically targeting CALR and CD47 in solid cancers

separately but do not expand on the relationship between the two.

Targeting CALR as an anti-cancer strategy, has been mainly focused on targeting molecules able to

affect pathways indirectly implicated in enhancing CALR expression, and as a result should have

stimulated phagocytosis of the cancer cell (Michalak et al., 2009). An increased rate of apoptosis has

been show in prostate and lung cancer cells when treated with CALR targeting adenovirus (Ad-

CALR/MAGE-A3) which upregulates total surface CALR, inhibiting cell proliferation (Liu et al., 2015).

Treatment with anthracycline, a chemotherapeutic drug used across a variety of different cancers

(Fucikova et al., 2016), stimulates the activation of a tumour-targeting immune response through

increasing CALR expression on the cell surface, with levels of CALR membrane expression correlating

with the overall removal of cancer cells and survival of patients with AML, neuroblastoma, non-small

cell lung carcinoma, ovarian cancer, and colorectal carcinoma (Fucikova et al., 2016).

Targeting CD47 interaction with the key SIRPα dependent phagocytosis pathway (as described in

detailed in section in the introduction) using specific anti-CD47 monoclonal antibodies has been

tested in several clinical trials in solid tumours with a certain degree of success, suggesting a

potential new anti-cancer strategy (Majeti et al., 2009). Application of CD47 monoclonal antibodies

blocking CD47 activity, lead to increases in the SIRPα dependent phagocytic signals resulting

macrophage activation, promoting them to phagocytose cancer cells, inducing cancer cell death

(Feng et al., 2015) . In leukemic, lymphoma, pancreatic, and ovarian carcinomas cell lines the

combination of anti-CD47 and anti-CD49 antibodies (CD49 mediates general cell surface signalling),

showed an increase in cell death, through the same mechanism of SIRPα activation (Dheilly et al.,

2017). In addition, use of anti-CD47 antibody Hu5F9-G4 in brain tumours demonstrated inhibition of

116

CD47, leading to increased presence of activated macrophages (Gholamin et al., 2017). Blocking

CD47 using CD47 siRNA could also represent a therapeutic strategy, with CD47 siRNA administered

intravenously in melanoma and lung metastases resulting in a reduction of tumour size (Wang et al.,

2013).

Whilst chemotherapy drugs are considered the first line treatment for cancer, cytotoxic drugs are

not always effective so work has also been done looking at whether therapeutically targeting CALR

and CD47 molecules could be used in combination with chemotherapy to enhance treatment

efficacy (Xiang & Liu, 2017). In AML, humanised anti-CD47 antibody (Hu5F9-G4) when administered

in combination with ARA-C or Azacytidine, resulted in an initial reduction of all blasts in vivo in a pre-

clinical assessment (Liu et al., 2015). With initial trials showing how altering expression in CALR and

CD47 in AML could lead to improvement in disease, this suggest there is potential to target CALR or

CD47 with a therapeutic effect in MDS and MPN. As earlier work within this thesis has shown a role

for alteration in CALR and CD47 cell surface expression in MDS and MPN disease

progression/response to therapy different to that seen in other cancer settings, it will be important

to understand the relationship between the separate CALR and CD47 pathways and explore

potential cross interaction between the two specifically in the MDS and MPN setting.

5.2: Aims

The overarching aim of the chapter is to assess the interactions, or lack thereof, between CALR or

CD47 within an MDS/MPN specific setting by determining what effect knocking down one of them

has on the other, and the potential effect on chemo-response in vitro. This will be achieved by the

following steps

• Transfect cell line models with either CALR and CD47 siRNAs to generate knockdowns, and

evaluate the impact upon mRNA and protein expression on the other molecule

117

• Treat the transfected cell line models with azacytidine and ruxolitinib, to enlighten how

knocking down either CALR or CD47, impacts upon both molecules response to treatment

5.3: Methods

Methods are described in brief here only where they differ from the main Methods in Chapter 2 (p

48).

5.3.1: Cell culture

Cell culture and drug treatment was performed as previously described in Chapter 2 (p 52) using

azacytidine (MDS specific) and ruxolitinib (MPN specific), each drug was used to treat both MDS and

MPN cell lines as previously used in Chapter 3 (p72).

5.3.2: Transfections

Transfections using siRNAs were performed using the optimised protocol as described in Chapter 2,

section 2.14: CALR and CD47 Knockdowns (p 65). Cell lines were incubated for 48 hours with siRNAs

(Sigma, Dorset, UK) and lipofectamine LTX Reagent (Thermo Fisher, Paisley, UK) to achieve this. To

optimise the transfections different time points and concentrations of lipofectamine were assessed

across all cell lines. Due to different cell line models reacting to knockdowns differently, total

knockdowns were not achieved. A reasonable concentration was chosen that suited all the cell line

models whilst showing no effect on the lipofectamine control or the scatter siRNA (1µg/2x105 cells).

Having no effect on the control and the scatter siRNA indicates that the knockdown is specific to the

siRNAs used. Graph to show efficiency is included in the Appendix 4.1.

5.3.4: Gene Expression qPCR Analysis

Gene expression analysis of CALR and CD47 was conducted using qPCR (as described in Chapter 2:

Quantitative PCR (qPCR) (p 57)).

118

5.3.5: Western Blotting Analysis

Protein of the transfected samples were collected to assess via western blotting (as described in

Chapter 2: 2.12.1 Western blotting (p 59-64)). Densitometry readings were taken from western blots

for each cell line model and qualitative analysis was gathered from the densitometry resulting in

normalised signals of the target bands against the GADPH housekeeping gene. Figure 5.1 is an

example of n1 for CALR, CD47 and GAPDH in all cell line models. With each sample there was

untreated, azacytidine treated and ruxolitinib treated.

Anti-CALR CALR CD47 GAPDH A B C A B C A B C

MOLM - 13 SKM – 1

K562 HEL – 92

GDM - 1 Anti-CD47

A B C A B C A B C MOLM - 13

SKM – 1

K562

HEL – 92 GDM - 1

Figure 5.1: Representative example of western blot samples of CALR, CD47 and GAPDH proteins. Each protein had an

untreated sample = A, azacytidine treated sample = B and ruxolitinib treated = C, for each transfected cell line model. Full

representative blot in Appendix 4.1.

119

5.4: Results

5.4.1: CALR and CD47 changes and impact on survival in transfected MDS & MPN cell line models

To determine the effect of transfections on cell survival, cell counts were evaluated 48h after

incubating with either CALR or CD47 silencing RNA. All cell lines were subjected to treatment with

both CALR silencing and CD47 silencing siRNAs as shown (Figure 5.2). Survival rates were calculated

as fraction of alive cells from the total cell pool treated and expressed as a percentage. When

treated with CALR silencing siRNAs, MOLM-13 and SKM-1 (MDS models) exhibited an overall survival

rate of 68% - p=0.029 and 67% - p=0.013, respectively. HEL-92 and GDM-1 (MPN models) exhibited

an overall survival rate of 64% p= 0.0016 and 69% - p=0.017 respectively, with K562 (MDS/MPN

overlap) exhibiting a survival rate of 66% - p=0.0016. When treated with CD47 silencing siRNAs

however, the MOLM -13 and SKM-1 exhibited an overall survival rate of 58% - p=0.01 and 58% -

p=0.01, respectively. HEL-92 and GDM exhibited an overall survival rate of 64% and 59% - p=0.0049,

respectively, and K562 exhibiting a survival rate of 60% - p=0.0056, suggesting that when the anti-

phagocytic signal CD47 is reduced, cells died more independently of the type of cell line. As

expected, cells transfected with either CALR or CD47 silencing siRNAs exhibited an overall decrease

in survival across all cell lines.

120

Figure 5.2: Overall mean percentage survival rate of each cell line model treated with CALR and CD47 silencing siRNAs. All cell lines and transfected at different time points to the n3 (representing 3 independent experiments). Total cell counts

were determined after 48 hours from live and dead cells after transfections. Statistics were performed against non-transfected cells. The error bars represent the standard deviation of the n3 between the treated compared to the control.

5.4.2: CALR silenced transfection efficiency in cell line models

CALR knockdowns were confirmed in all cell lines. Each cell line was transfected with CALR silencing

siRNAs which knocked down the expression of CALR (Figure 5.3) and CD47 (Figure 5.4) within the

cells. After transfection, cell lines showed suppression of CALR rather than a complete knockdown,

in fact complete knockdown was shown to result in all cells dying (data not shown). Along with CALR

silencing siRNAs a scatter siRNA was also used to treat cells. Within the scatter treated cells, no

change in expression was observed (data not shown).

MDS cell line models (MOLM-13 and SKM-1), showed low expression of CALR mRNA in transfected

cells compared to non-transfected (0.42-fold and 0.22-fold Vs 1.32-fold and 1.09-fold, respectively).

The MPN cell line models (HEL-92 and GDM-1) showed decreased CALR mRNA expression in

transfected cells compared to non-transfected (0.38-fold and 0.82-fold Vs 1.05-fold and 1.59-fold,

respectively). The K562 cell line showed a similar change in CALR mRNA expression patterns to those

0

10

20

30

40

50

60

70

80

90

MOLM-13 SKM-1 K562 HEL-92 GDM-1

Perc

enta

ge o

f liv

ing

cells

pos

t tre

atm

ent

* * * * ** ** ** * **

Overall percentage survival rate of each transfected cell line models after 48 hours of transfection

Anti-CALR Anti-CD47

121

seen in the MDS cell lines. These same patterns of alterations in CALR expression were also seen

when investigating protein expression.

The impact of CALR suppression on CD47 expression was also determined. Interestingly, in MDS cell

lines (MOLM-13 and SKM-1) transfected with CALR silencing, CD47 mRNA in responded similarly to

the CALR mRNA showing large decreases in expression (0.35-fold and 0.31-fold Vs 1.32-fold and

1.02-fold in non-transfected cells, respectively). Within the MPN cell line GDM-1 CD47 mRNA

expression showed a small decrease in (0.98-fold Vs 1.05-fold in non-transfected cells). In contrast to

other cell lines, the HEL-92 MPN cell line showed an increase in expression of CD47 upon CALR

silencing (1.39-fold Vs 1.13-fold in non-transfected cells) in and the MDS/MPN intermediate K562

CD47 mRNA was also seen to decrease upon CALR silencing transfection (0.40-fold Vs 1.21-fold).

CD47 protein expression was also affected similarly to the mRNA patterns as reported in Figure 5.4.

122

Figure 5.3: Overall mean mRNA (A) and protein (B) levels of CALR in each cell line model when treated with CALR silencing siRNAs. All cell lines and drugging were grown/harvested at different time points to the n3. Showing how the siRNAs are

affecting the cell line models upon transfection. The error bars represent the standard deviation of the n3.

0

0.5

1

1.5

2

2.5

3


2^(-Δ

ΔCt)

fold

exp

ress

ion

RNA expression of CALR in anti-CALR siRNA transfected cells line models

Non-transfected transfected

A

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


norm

alise

d ra

te c

hang

es

Protein expression of CALR in anti-CALR siRNA transfected cells line models

Non transfected Transfected

B

123

Figure 5.4: Overall mean mRNA (A) and protein (B) transfection rate of CD47 in each cell line model when treated with CALR silencing siRNAs. All cell lines and drugging were grown/harvested at different time points to the n3. Total cell counts

were determined after 48 hours. The error bars represent the standard deviation of the n3.

5.4.4: CD47 silenced transfection efficiency in cell lines

The impact of CD47 suppression was also analysed using CD47 silencing siRNAs. As previously

described, when total knockdown was obtained, cells did not survive (data not shown), so the assay

was optimised to produce a significant, but not complete, gene suppression. Upon transfection with

0

0.5

1

1.5

2

2.5


2^(-Δ

ΔCt)

fold

exp

ress

ion

RNA expression of CD47 in anti-CALR siRNA transfected cells line models


A

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


norm

alise

d ra

te c

hang

es

Protein expression of CD47 in anti-CALR siRNA transfected cells line models


B

124

CD47 silencing siRNA all cell line models showed decreases in CD47 mRNA and protein (Figure 5.5),

when compared to non-transfected cells. Along with CD47 silencing siRNAs a scatter siRNA was also

used to treat cells. Within the scatter treated cells no change in expression was observed.

Interestingly, also in this case, the suppression of CD47 resulted in reduction of CALR expression in

most of the cell lines (Figure 5.6).

MDS cell line models (MOLM-13 and SKM-1), showed low expression of CALR mRNA in CD47 silenced

transfected cells compared to non-transfected (0.63-fold and 0.53-fold Vs 1.32-fold and 1.09-fold,

respectively). MPN cell line HEL-92 showed a decrease in CALR mRNA expression (0.54-fold Vs 1.05-

fold in non-transfected cells), whereas the GDM-1 MPN cell line showed only a small decrease in

CALR expression upon CD47 silencing (1.55-fold Vs 1.59-fold in non-transfected cells). Upon

analysing protein expression in all cell line models upon which CD47 silencing transfections had been

undertaken, MDS cell line models (MOLM-13 and SKM-1) showed low expression of CALR protein in

CD47 silenced transfected cells compared to non-transfected (0.02- NR and 0.26- NR Vs 0.59- NR and

0.34- NR, respectively). The MPN cell line models (HEL-92 and GDM-1), also showed decreased CALR

protein expression in transfected cells compared to non-transfected (0.29- NR and 0.026- NR Vs

0.50- NR and 0.59- NR, respectively).

Upon CD47 silencing transfections, in MDS cell lines (MOLM-13 and SKM-1) CD47 mRNA responded

similarly to the CALR mRNA with large decreases in expression seen Vs non-transfected cells (0.75-

fold and 0.50-fold Vs 1.32-fold and 1.02-fold, respectively). The MPN cell line models (HEL-92 and

GDM-1), also showed decreases in CD47 mRNA expression in transfected cells compared to non-

transfected (0.44-fold and 0.64-fold Vs 1.13-fold and 1.05-fold, respectively). Protein expression

supported the mRNA in both MDs and MPN cell line. CD47 protein responded with decreases in the

MDS (MOLM-13 and SKM-1) (0.02- NR and 0.24- NR Vs 0.48- NR and 0.30- NR in non-transfected

cells, respectively), upon CD47 silencing transfections. While HEL-92 and GDM-1 cell lines also

125

showed CD47 expression to decreased upon CD47 silencing (0.15- NR and 0.02- NR Vs 0.26- NR and

0.48- NR in non-transfected cells, respectively).

The K562 cell line showed a similar protein expression pattern to the MDS cell lines with CALR mRNA

was shown to decrease upon transfection (0.55-fold Vs 1.29-fold), while CD47 mRNA was also seen

to decrease upon CD47 silencing transfection (0.49-fold Vs 1.21-fold). The K562 cell line showed a

similar expression pattern to the MDS and MPN cell lines. CALR protein was shown to decrease upon

transfection (0.08- NR Vs 0.64- NR), while CD47 protein was also seen to decrease upon CD47

silencing transfection (0.04- NR Vs 0.28- NR).

126

Figure 5.5: Overall mean mRNA (A) and protein (B) transfection rate of CD47 in each cell line model treated with CD47 silencing siRNAs. All cell lines and drugging were grown/harvested at different time points to the n3. Total cell counts were

determined after 48 hours. The error bars represent the standard deviation of the n3.

0

0.5

1

1.5

2

2.5


2^(-Δ

ΔCt)

fold

exp

ress

ion

RNA expression of CD47 in anti-CD47 siRNA transfected cells line models


A

0

0.1

0.2

0.3

0.4

0.5

0.6


Nor

mal

ised

ratio

cha

nges

Protein expression of CD47 in anti-CD47 siRNA transfected cells line models


B

127

Figure 5.6: Overall mean mRNA (A) and protein (B) transfection rate of CALR in each cell line model treated with CD47

silencing siRNAs. All cell lines and drugging were grown/harvested at different time points to the n3. Total cell counts were

determined after 48 hours. The error bars represent the standard deviation of the n3.

0

0.5

1

1.5

2

2.5

3


2^(-Δ

ΔCt)

fold

exp

ress

ion

RNA expression of CALR in anti-CD47 siRNA transfected cells line models


A

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8


Nor

mal

ised

ratio

cha

nges

Protein expression of CALR in anti-CD47 siRNA transfected cells line models


B

128

5.4.5: CALR and CD47 silenced cells survival rate upon treatment

with chemotherapeutics

To model the effects of chemotherapeutic agents upon cell survival within MDS or MPN cells where

either CALR or CD47 has been knocked down, the MOLM-13 cell line was used to represent MDS

and the HEL-92 cell line was used to represent MPN. Each MDS and MPN cell line model showed

decreases in the percentage of cell survival in response to treatment with chemotheraputic agents

compared to the untreated transfected cells (Figure 5.7). Both cells exhibited a decreased cell

survivall in response to both intended treatment options (azacytidine - MDS and ruxolitinib – MPN)

and non-intended treatment option in both CALR silencing and CD47 silencing. When the pro-

phagocytic CALR is silenced MOLM-13 showed reductions in cell survival upon treatment with

azacytidine and ruxolitinib (43% and 54%, respectively) compared to normal cell survival when

previously treated in Chapter 3 (p 78). However, as expected when the anti-phagocytic CD47 is

silenced, cells showed a deeper drop in survival upon treatment with azacytidine and ruxolitinib

(38% and 42%, respectively). MPN cell line model HEL-92 responded with a similar cell survival

pattern, showing a greater percentage of cell death when CD47 is suppressed, after exposure to

azacytidine and ruxolitinib (survival rate 42% and 39%, respectively) when compared to CALR

suppressed cells (51% and 47%, respectively).

129

Figure 5.7: Cell survival upon treatment with azacytidine and ruxolitinib in representative MDS (MOLM-13) (A) and MPN

(HEL-92) (B) cell lines. The error bars represent the standard deviation of the n3 between the drug treated compared to the control.

5.5: Discussion

In this chapter the effect of CALR and CD47 on cell survival and the impact of each gene on the other

using knockdown models in MDS and MPN cell lines was studied. Transfection efficiency has been

challenging throughout this work, as only partial knockdowns could be undertaken due to the

triggering of complete cell death upon higher doses of the silencing siRNAs, however a significant

0

10

20

30

40

50

60

70

80

90

100

Anti-CALR Anti-CD47

Perc

enta

ge o

f liv

ing

cells

pos

t tre

atm

ent

Overall percentage survival rate of transfected MOLM-13 cells after treatment with chemotherapeutic agents

Azazytidine Ruxolitinib

A

0

10

20

30

40

50

60

70

80

90

100

Anti-CALR Anti-CD47Perc

enta

ge o

f liv

ing

cells

pos

t tre

atm

ent

Overall percentage survival rate of transfected HEL-92 cells after treatment with chemotherapeutic agents

Azazytidine Ruxolitinib

B

130

reduction in both genes and protein expression was successfully obtained. Upon knockdown with

CALR or CD47 silencing siRNAs, a reduction of CALR and CD47 expression of mRNA and protein was

seen in all cell lines. In all cells it is interesting to note that when CALR was knocked down using CALR

silencing siRNAs, CD47 expression was also seen to decrease. Similarly, when CD47 was knocked

down using CD47 silencing siRNAs, CALR expression was also seen to decrease suggesting a link

between CALR and CD47 expression in MDS and MPN cell lines. It was also shown that when either

of them is silenced, cells survival rate decreased. This is expected as both CALR and CD47 are

involved in many homeostatic functions in the cell and therefore, when reduced, the cell may not

function as needed (Adachi et al., 2018).

When cells were treated, survival rates of MDS and MPN cell lines are seen to further decrease, as

expected but the cell lines where the pro-phagocytic CALR was silenced had a lower reduction in cell

survival, in comparison with the anti-phagocytic CD47 slienced cells which showned a larger

reduction in cell survival. This indicates that when CALR is silenced, CD47 is still being expressed to

keep the cell alive and when CD47 is silenced the expression of CALR allows for more cells to induce

phagocytosis (Feng et al., 2018).

Whilst this work suggests that modulation of CALR or CD47 expression could play an important role

in MDS and MPN cell survival, it also suggests that a simple reduction in levels and altered ratio

between the two might not be enough to direct the cell toward either survival or death. In solid

tumours, strategies used to reduce CD47 expression levels demonstrated some efficacy in reducing

tumour size and increasing macrophage activity (Wang et al., 2013; Gholamin et al., 2017). Also,

increased CALR expression signals for cell death by interacting with the LDL receptor-like protein on

the macrophage, telling the cell is damaged and needs to be destroyed (Fucikova et al., 2018). Lui et

al. 2019, previously showed that manipulating linked signalling systems via siRNA knock downs was

not successful in inducing in vitro phagocytosis of tumour cells or in vivo anti-tumour responses (Liu

et al., 2019), which findings from Chapter 5 further support. Due to the cross interaction between

131

the two genes expression, knocking out one of the genes will potentially affect the other and the

overall signal might not be therapeutically effective. Leaving the gene and protein intact but directly

interfering with the CD47 anti phagocytic signal with a compound/antibody able to disrupt the

interaction with macrophages, could potentially allow the pro phagocytic message transduced by

CALR to prevail (Lian., 2019) and provide a more appropriately way of therapeutically targeting CALR

and CD47 expression in MDS and MPN in the future.

132

Chapter 6: Conclusions

133

6.1: Conclusions The knowledge and understanding of the molecular mechanisms underpinning the role of CALR and

CD47 in cancers have progressed in recent years but are still not fully understood. Established

research in to CALR and CD47 have given insights into solid cancer and some of the myeloid

malignancies, with CALR and CD47 shown to increase in their expression. Little research had been

done in CALR and CD47 expression in MDS and MPN, therefore, this project aimed to investigate and

establish CALR and CD47 role in MDS and MPN and their implication in therapy response and disease

prognosis.

Within MDS and MPN cell line models CALR and CD47 expression was shown to change when cell

lines were exposed to chemotherapy drugs such as azacytidine and ruxolitinib, and often the

changes in total expression or membrane exposure reflected the treatment specificity as described

in Chapter 3 (p 72-97). Not surprisingly, localisation studies in both MDS and MPN models, found

that when cells are exposed to cytotoxic drugs, CD47 is externalised more than CALR indicating a

strong ‘don’t eat me’ signal released by the cell in danger in the attempt to stay alive. However, this

reflects a substantial difference with previous studies in solid cancer, where CD47 and CALR are

concomitantly expressed on the membrane during chemotherapy exposure and might represent a

very important feature in myeloid neoplasms to target for future therapeutical strategies.

Chapter 4 (p 98-111) focused on the expression levels of CALR and CD47 in patients with MDS and

MPN, to determine whether the changes observed in cell lines could be replicated in patient

samples. When compared to healthy aged-matched controls, overall levels of CALR and CD47 have

been shown to increase in solid cancers although the ratios of these in MDS and MPN vary (Chao et

al., 2010). MDS showed larger expression of CALR whereas MPN exhibited larger expression of CD47.

When the MDS cohort was split into subgroups in accordance to IPSS, changes in total CALR and

CD47 seemed to follow a pattern related to disease progression, with CALR expression seen to be

higher in intermediate-2 and high-risk patients as previously described by Schcolnik‑Cabrera et al.

134

2019 (Schcolnik-Cabrera et al., 2019), suggesting the attempt by the leukemic cell to still try to

identify itself as abnormal particularly during the late evolution into AML. CD47 expression showed a

consistent pattern during progression, demonstrating a progressive increase from low risk into

intermediate-2 and high-risk, indicating the cells are counteracting CALR signalling and trying to

prevent phagocytosis (Jaiswal et al., 2009). Not surprisingly, most of the relevant changes in CALR

and CD47 happened on the membrane of the cells as confirmed by fractionation study. The MDS cell

exposes more CALR and CD47 on the membrane during AML progression, but the CALR/CD47 ratio

indicate a prevalent “do not eat” message, indicating that this may be an immune-escape

mechanism needed by the leukaemia cell throughout MDS evolution.

Fractionation data confirmed a large upregulation of CALR on the membrane of the cells initially and

then a decrease in expression during evolution into high-risk MDS. CD47 externalised throughout

disease progression with lower percentages of CD47 in low risk and progressive increase moving into

high risk. The Initial upregulation of CALR indicates that the cell wants to signal for cell phagocytosis

but is quickly counteracted by CD47 upregulation with the anti-phagocytic message to prevail.

When separately analysing MPN patients split by PV, ET and MF classification, they showed different

patterns of expression. Total protein expression in MPN groups upon treatment showed decreases

in CALR in all subgroups, while CD47 protein expression was shown to increase in PV and MF but not

in ET. Total protein expression in MPNs upon treatment also showed decreases of CALR in all

subgroups (ET, PV, MF), interestingly, CD47 expression significantly increases during therapy in PV

and MF but not in ET, supporting the mRNA data, and suggests a potential different mechanism of

interaction with the immune system in ET patients comparing with MF and P, and with it potentially

differences in prognosis.

Fractionations of the MPN subgroups showed patients of all subgroups responded similarly with

decreases in membrane CALR protein expression upon treatment but increases in CD47

externalization in MF and PV. Overall prognosis in ET patients is favourable, and the reduced

135

expression of CD47 in response to treatment may correlate with the positive outcome because the

ET cells can be more easily recognised and destroyed by the immune system. Further data are

however required to understand if ET progression into MF or AML changes the CD47/CALR ratio

during transformation.

To start investigating the therapeutic potential of manipulating CALR and CD47 in treating MDs and

MPN, transfection assays were undertaken to alternatively supress CALR or CD47 to evaluate the

impact of on to the other. In all cell line models, when CALR or CD47 were knocked down, both

decreased, suggesting a potential direct interaction of CALR and CD47 intra-cellular pathways. This

hypothesis however needs further corroborating by additional studies.

When transfected cells were treated azacytidine and ruxolitinib, survival rates of MDS and MPN cell

lines were seen to further decrease, as expected, but the pro-phagocytic CALR silenced cells reduced

their survival less, in comparison with the anti-phagocytic CD47 slienced cells. This ndicates that

when CALR is silenced, CD47 is still being expressed to keep the cell alive and when CD47 is silenced

the expression of CALR allows for more cells to induce phagocytosis. Chapter 5 (p 112- 128) supports

the results found Lui et al. 2019 studies, where manipulating linked signalling systems siRNA knock

downs were not successful on their own at inducing in vitro phagocytosis of tumour cells or in vivo

anti-tumour responses (Liu et al., 2019). Due to the cross interaction between the two genes

expression, knocking out the genes will potentially affect the other and the overall signal might not

be therapeutically effective. Leaving the gene and protein intact but directly interfering with the

CD47 anti phagocytic signal with a compound/antibody able to disrupt the interaction with

macrophages, could potentially allow the pro-phagocytic message transduced by CALR to prevail.

In summary, this work describes the role of CALR and CD47 in MDS and MPN. Changes in CALR and

CD47 expression not only follow different diseases but also disease progression. The expression

relative ratio between the pro-phagocytic signal CALR and the do not eat me message CD47 seem to

play a key role in the immune response of both MDS and MPN cells in both cell lines and patient’s

136

samples. CALR and CD47 expression pathways, even if transducing different messages via very

different intra-cellular pathways, might be interdependent but further studies are required to

confirm this. Studies within MDS have specifically shown that blocking the CD47 signal leads to

selective phagocytosis of the MDS cells by reducing the “don’t eat me” signal (Pang et al., 2013). The

availability a of monoclonal antibody targeting CD47 could represent a great opportunity for

therapeutic intervention round these newly elucidated CALR and CD47 pathways in myeloid

malignancy. Phase 1 trials using Magrolimab in AML have demonstrated efficacy in monotherapy

and in combination (Chao et al., 2020). This study suggests the possible impact of such therapy in

MDS and MPN as well. Targeting the “do not eat signal” CD47 in MDS and MPN patients could slow

down progression and increase efficacy of the standard chemotherapy.

6.2: Future studies Future research following on from this project could have a few different directions. The first is to

continue with the CD47 story. This would progress by looking at using a drug such as Magrolimab in

both MDS and MPN (Chao et al., 2020). By looking in both cell lines and eventually patient samples

the effects of targeting the “do not eat signal” CD47, either in combination with current therapies or

by itself, to determine any potential benefits for the treatment of MDS and MPN patients.

Alternately looking at other genes such as BCL-2, calnexin, GATA-1 and PU-1 that have been

proposed to play a role in MDS/MPN onset and seeing their reaction in relation to CALR and CD47.

Techniques that could be utilised to aid in this work would include flow cytometry, immune

precipitations and florescent microscopy to determine interactions between these molecules and

the impact on the immune response or response to treatment. This would elaborate on any possible

connection within MDS and MPN patients.

137

References Adachi, K., Kano, Y., Nagai, T., Okuyama, N., Sakoda, Y., & Tamada, K. (2018), IL-7 and CCL19 expression in CAR-T cells improves immune cell infiltration and CAR-T cell survival in the tumor. Nature Biotechnology, 36(4). p. 346–351.

Adès, L., Itzykson, R., & Fenaux, P. (2014), Myelodysplastic syndromes. Lancet (Vol. 383). p. 2239–2252.

Alur, M., Nguyen, M. M., Eggener, S. E., Jiang, F., Dadras, S. S., Stern, J., Kimm, S., Roehl, K., Kozlowski, J., Pins, M., Michalak, M., Dhir, R., Wang, Z. (2009), Suppressive roles of calreticulin in prostate cancer growth and metastasis. American Journal of Pathology, 175(2). p. 882–890.

Araki, M., & Komatsu, N. (2017), Novel molecular mechanism of cellular transformation by a mutant molecular chaperone in myeloproliferative neoplasms. Cancer Science, 108(10). p. 1907–1912.

Arber, D. A., Orazi, A., Hasserjian, R., Thiele, J., Borowitz, M. J., Le Beau, M. M., Bloomfield, C. D., Cazzola, M., Vardiman, J. W. (2016), The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood, 127(20). p. 2391–2405.

Arrieta, O., Aviles-Salas, A., Orozco-Morales, M., Hernández-Pedro, N., Cardona, A. F., Cabrera-Miranda, L., Barrios-Bernal, P., Soca-Chafre, G., Cruz-Rico, G., Peña-Torres, M. L., Moncada-Claudio, G., Ramirez-Tirado, L., Ramirez-Tirado, L. (2020), Association between CD47 expression, clinical characteristics and prognosis in patients with advanced non-small cell lung cancer. Cancer Medicine, 9(7). p. 2390–2402.

Barazi, H. O., Li, Z., Cashel, J. A., Krutzsch, H. C., Annis, D. S., Mosher, D. F., & Roberts, D. D. (2002), Regulation of integrin function by CD47 ligands. Differential effects on alpha vbeta 3 and alpha 4beta1 integrin-mediated adhesion. J Biol Chem. p. 42859-42866.

Barbui, T., Thiele, J., Vannucchi, A. M., & Tefferi, A. (2014), Rethinking the diagnostic criteria of polycythemia vera. Leukemia. p. 1191–1195.

Belcic Mikic, T., Pajic, T., & Sever, M. (2019), CALR mutations in a cohort of JAK2 V617F negative patients with suspected myeloproliferative neoplasms. Scientific Reports, 9(1). p. 1–9.

Bennett, J. M., Catovsky, D., Daniel, M. T., Flandrin, G., Galton, D. A. G., Gralnick, H. R., & Sultan, C. (1982), Proposals for the classification of the myelodysplastic syndromes. British Journal of Haematology, 51(2). p. 189–199.

Bernstein, S. H., Brunetto, V. L., Davey, F. R., Wurster-Hill, D., Mayer, R. J., Stone, R. M., Bloomfield, C. D. (1996), Acute myeloid leukemia-type chemotherapy for newly diagnosed patients without antecedent cytopenias having myelodysplastic syndrome as defined by French-American-British criteria: A cancer and leukemia group B study. Journal of Clinical Oncology, 14(9) p. 2486–2494.

Blees, A., Januliene, D., Hofmann, T., Koller, N., Schmidt, C., Trowitzsch, S., Tampé, R. (2017), Structure of the human MHC-I peptide-loading complex. Nature, 551(7681). p. 525–528.

138

Boasman, K., Simmonds, M. J., Graham, C., Saunthararajah, Y., & Rinaldi, C. R. (2019). Using PU.1 and Jun dimerization protein 2 transcription factor expression in myelodysplastic syndromes to predict treatment response and leukaemia transformation. Annals of Hematology.

Boasman, K., Simmonds, M. J., & Rinaldi, C. R. (2019), p. 1-5. CALR and CD47: An Insight into Their Roles in the Disease Progression of MDS and MPN. Journal of Blood Disorders & Transfusion, 10(1).

Bookout, A. L., Mangelsdorf, D. J., & Hughes, H. (2003), Nuclear Receptor Signalling, 1, 12. p. 801-810.

Bozkus, C. C., Roudko, V., Finnigan, J. P., Mascarenhas, J., Hoffman, R., Iancu-Rubin, C., & Bhardwaj, N. (2019), Immune checkpoint blockade enhances shared neoantigen-induced T-cell immunity directed against mutated calreticulin in myeloproliferative neoplasms. Cancer Discovery, 9(9). p. 1192–1207.

Brandizzi, F., & Barlowe, C. (2013), Organization of the ER-Golgi interface for membrane traffic control. Nature Reviews Molecular Cell Biology. p. 382-392.

Brightwell, R. M., Grzankowski, K. S., Lele, S., Eng, K., Arshad, M., Chen, H., & Odunsi, K. (2016), The CD47 “don’t eat me signal” is highly expressed in human ovarian cancer. Gynecologic Oncology, 143(2). p. 393–397.

Broom, O. J., Zhang, Y., Oldenborg, P.-A., Massoumi, R., & Sjölander, A. (2009), CD47 Regulates Collagen I-Induced Cyclooxygenase-2 Expression and Intestinal Epithelial Cell Migration. PLoS ONE, 4(7). e6371.

Campbell, P. J., Bareford, D., Erber, W. N., Wilkins, B. S., Wright, P., Buck, G., Wheatley, K., Harrison, C. N., Green, A. R. (2009), Reticulin accumulation in essential thrombocythemia: Prognostic significance and relationship to therapy. Journal of Clinical Oncology. p. 2991-2999.

Campbell, P. J., Baxter, E. J., Beer, P. A., Scott, L. M., Bench, A. J., Huntly, B. J. P., Green, A. R. (2006), Mutation of JAK2 in the myeloproliferative disorders: Timing, clonality studies, cytogenetic associations, and role in leukemic transformation. Blood, 108(10). p. 3548–3555.

Cathro, H. P., Smolkin, M. E., Theodorescu, D., Jo, V. Y., Ferrone, S., & Frierson, H. F. (2010), Relationship between HLA class i antigen processing machinery component expression and the clinicopathologic characteristics of bladder carcinomas. Cancer Immunology, Immunotherapy, 59(3). p. 465–472.

Cazzola, M., & Kralovics, R. (2014), From Janus kinase 2 to calreticulin: The clinically relevant genomic landscape of myeloproliferative neoplasms. Blood, 123(24). p. 3714–3719.

Chachoua, I., Pecquet, C., El-Khoury, M., Nivarthi, H., Albu, R. I., Marty, C., Constantinescu, S. N. (2016), Thrombopoietin receptor activation by myeloproliferative neoplasm associated calreticulin mutants. Blood, 127(10). p. 1325–1335.

Chao, M. P., & Gotlib, J. (2014), Two faces of ET: CALR and JAK2. Blood, 123(10). p. 1438–1440.

139

Chao, M. P., Jaiswal, S., Weissman-Tsukamoto, R., Alizadeh, A. A., Gentles, A. J., Volkmer, J., Weiskopf, K., Willingham, S. B., Raveh, T., Park, C. Y., Majeti, R., Weissman, I. L. (2010), Calreticulin is the dominant pro-phagocytic signal on multiple human cancers and is counterbalanced by CD47. Science Translational Medicine. p. 63-94.

Chao, M. P., Majeti, R., & Weissman, I. L. (2012), Programmed cell removal: a new obstacle in the road to developing cancer. Nature Reviews Cancer, 12(1). p. 58–67.

Chao, M. P., Takimoto, C. H., Feng, D. D., McKenna, K., Gip, P., Liu, J., Volkmer, J. P., Weissman, I. L., Majeti, R. (2020). Therapeutic Targeting of the Macrophage Immune Checkpoint CD47 in Myeloid Malignancies. Frontiers in Oncology.

Chao, M. P., Weissman, I. L., & Majeti, R. (2012), The CD47-SIRPα pathway in cancer immune evasion and potential therapeutic implications. Current Opinion in Immunology. p. 225-232.

Chen, C. N., Chang, C. C., Su, T. E., Hsu, W. M., Jeng, Y. M., Ho, M. C., Hsieh, F. J., Lee, P. H., Kuo, M. L., Lee, H., Chang, K. J. (2009), Identification of calreticulin as a prognosis marker and angiogenic regulator in human gastric cancer. Annals of Surgical Oncology, 16(2). p. 524–533.

Coppolino, M. G., Woodside, M. J., Demaurex, N., Grinstein, S., St-Arnaud, R., & Dedhar, S. (1997), Calreticulin is essential for integrin-mediated calcium signalling and cell adhesion. Nature, 386(6627). p. 843–847.

Curto-Garcia, N., Ianotto, J.-C., & Harrison, C. N. (2018), What is pre-fibrotic myelofibrosis and how should it be managed in 2018? British Journal of Haematology, 183(1). p. 23–34.

Daitoku, S., Takenaka, K., Yamauchi, T., Yurino, A., Jinnouchi, F., Nunomura, T., Akashi, K. (2016), Calreticulin mutation does not contribute to disease progression in essential thrombocythemia by inhibiting phagocytosis. Experimental Hematology, 44(9). p. 817-825.

Dheilly, E., Moine, V., Broyer, L., Salgado-Pires, S., Johnson, Z., Papaioannou, A., Cons, L., Calloud, S., Majocchi, S., Nelson, R., Rousseau, F., Ferlin, W., Kosco-Vilbois, M., Fischer, N., Masternak, K. (2017), Selective Blockade of the Ubiquitous Checkpoint Receptor CD47 Is Enabled by Dual-Targeting Bispecific Antibodies. Molecular Therapy : The Journal of the American Society of Gene Therapy, 25(2). p. 523–533.

Dubois, A., Furstoss, N., Calleja, A., Zerhouni, M., Cluzeau, T., Savy, C., Robert, G. (2019), LAMP2 expression dictates azacytidine response and prognosis in MDS/AML. Leukemia, 33(6). p. 1501–1513.

Dunn, G. P., Old, L. J., & Schreiber, R. D. (2004), The Three Es of Cancer Immunoediting. Annual Review of Immunology, 22(1). p. 329–360.

Eladl, E., Tremblay-Lemay, R., Rastgoo, N., Musani, R., Chen, W., Liu, A., & Chang, H. (2020), Role of CD47 in Hematological Malignancies. Journal of Hematology and Oncology. BioMed Central. p. 96-101.

140

Erić, A., Juranić, Z., Milovanović, Z., Marković, I., Inić, M., Stanojević-Bakić, N., & Vojinović-Golubović, V. (2009), Effects of humoral immunity and calreticulin overexpression on postoperative course in breast cancer. Pathology and Oncology Research, 15(1). p. 89–90.

Feng, M., Chen, J. Y., Weissman-Tsukamoto, R., Volkmer, J. P., Ho, P. Y., McKenna, K. M., Weissman, I. L. (2015), Macrophages eat cancer cells using their own calreticulin as a guide: Roles of TLR and Btk. Proceedings of the National Academy of Sciences of the United States of America, 112(7). p. 2145–2150.

Feng, M., Marjon, K. D., Zhu, F., Weissman-Tsukamoto, R., Levett, A., Sullivan, K., Weissman, I. L. (2018), Programmed cell removal by calreticulin in tissue homeostasis and cancer. Nature Communications, 9(1). e3194.

Fucikova, J., Kasikova, L., Truxova, I., Laco, J., Skapa, P., Ryska, A., & Spisek, R. (2018), Relevance of the chaperone-like protein calreticulin for the biological behavior and clinical outcome of cancer. Immunology Letters. p. 25-34.

Fucikova, J., Spisek, R., Kroemer, G., & Galluzzi, L. (2020), Calreticulin and cancer. Cell Research. Springer Nature. p. 5-16.

Fucikova, J., Truxova, I., Hensler, M., Becht, E., Kasikova, L., Moserova, I., Vosahlikova, S., Klouckova, J., Church, S., Cremer, I., Kepp, O., Kroemer, G., Galluzzi, L., Salek, C., Spisek, R. (2016), Calreticulin exposure by malignant blasts correlates with robust anticancer immunity and improved clinical outcome in AML patients. Blood, 128(26). p. 3113–3124.

Fujiwara-Tani, R., Sasaki, T., Ohmori, H., Luo, Y., Goto, K., Nishiguchi, Y., Kuniyasu, H. (2019), Concurrent Expression of CD47 and CD44 in Colorectal Cancer Promotes Malignancy. Pathobiology, 86(4). p. 182–189.

Galli, S., Zlobec, I., Schürch, C., Perren, A., Ochsenbein, A. F., & Banz, Y. (2015), CD47 protein expression in acute myeloid leukemia: A tissue microarray-based analysis. Leukemia Research, 39(7). p. 749–756.

Galluzzi, L., Buqué, A., Kepp, O., Zitvogel, L., & Kroemer, G. (2015,), Immunological Effects of Conventional Chemotherapy and Targeted Anticancer Agents. Cancer Cell. p. 690-714.

Galluzzi, L., Chan, T. A., Kroemer, G., Wolchok, J. D., & López-Soto, A. (2018), The hallmarks of successful anticancer immunotherapy. Science Translational Medicine. American Association for the Advancement of Science. eaat7807.

Galluzzi, L., Vitale, I., Warren, S., Adjemian, S., Agostinis, P., Martinez, A. B., Chan, Timothy A., Coukos, G., Demaria, S., Deutsch, E., Draganov, D., Edelson, R. L., Formenti, S. C., Fucikova, J., Gabriele, L., Gaipl, U. S., Gameiro, S. R., Garg, A. D., Golden, E., Han, J., Harrington, K., Hemminki, A., Hodge, J., Hossain, Dewan Md S., Illidge, T., Karin, M., Kaufman, H. L., Kepp, O., Kroemer, G., Lasarte, Juan J., Loi, S., Lotze, M. T., Manic, Gwenola., Merghoub, T., Melcher, A., Mossman, K., Prosper, F., Rekdal, Ø., Rescigno, M., Riganti, C., Sistigu, A., Smyth, M., Spisek, R., Stagg, J., Strauss, B., Tang, D., Tatsuno, K., Van Gool, S., Vandenabeele, P., Yamazaki, T., Zamarin, D., Zitvogel, L., Cesano, A.,

141

Marincola, F. M. (2020), Consensus guidelines for the definition, detection and interpretation of immunogenic cell death. Journal for ImmunoTherapy of Cancer. e000337.

Gangat, N., Caramazza, D., Vaidya, R., George, G., Begna, K., Schwager, S., Van Dyke, D., Hanson, C., Wu, W., Pardanani, A., Cervantes, F., Passamonti, F., Tefferi, A. (2011), DIPSS plus: a refined Dynamic International Prognostic Scoring System for primary myelofibrosis that incorporates prognostic information from karyotype, platelet count, and transfusion status. Journal of Clinical Oncology : Official Journal of the American Society of Clinical Oncology, 29(4). p. 392–397.

Gao, B., Adhikari, R., Howarth, M., Nakamura, K., Gold, M. C., Hill, A. B., Elliott, T. (2002), Assembly and antigen-presenting function of MHC class I molecules in cells lacking the ER chaperone calreticulin. Immunity, 16(1). p.99–109.

Gao, K., & Huang, L. (2013), Achieving efficient RNAi therapy: progress and challenges. Acta Pharmaceutica Sinica B, 3(4). p. 213–225.

Gardai, S. J., McPhillips, K. A., Frasch, S. C., Janssen, W. J., Starefeldt, A., Murphy-Ullrich, J. E., Henson, P. M. (2005), Cell-surface calreticulin initiates clearance of viable or apoptotic cells through trans-activation of LRP on the phagocyte. Cell, 123(2). p. 321–334.

Garg, A. D., Krysko, D. V., Verfaillie, T., Kaczmarek, A., Ferreira, G. B., Marysael, T., Rubio, N., Firczuk, M., Mathieu, C., Roebroek, A. J.M., Annaert, W., Golab, J., De Witte, P., Vandenabeele, P., Agostinis, P. (2012), A novel pathway combining calreticulin exposure and ATP secretion in immunogenic cancer cell death. EMBO Journal, 31(5). p. 1062–1079.

Gheibi Hayat, S. M., Bianconi, V., Pirro, M., & Sahebkar, A. (2019), Efferocytosis: molecular mechanisms and pathophysiological perspectives. Immunology and Cell Biology. p. 124-133.

Gholamin, S., Mitra, S. S., Feroze, A. H., Liu, J., Kahn, S. A., Zhang, M., Cheshier, S. H. (2017), Disrupting the CD47-SIRPα anti-phagocytic axis by a humanized anti-CD47 antibody is an efficacious treatment for malignant pediatric brain tumors. Science Translational Medicine, 9(381). e2968.

Gill, H., Leung, A., & Kwong, Y.-L. (2016), Molecular and Cellular Mechanisms of Myelodysplastic Syndrome: Implications on Targeted Therapy. International Journal of Molecular Sciences, 17(4). p. 440.

Goasguen, J. E., Bennett, J. M., Bain, B. J., Brunning, R. D., Vallespí, M. T., Tomonaga, M., Renault, A. (2016), Quality control initiative on the evaluation of the dysmegakaryopoiesis in myeloid neoplasms: Difficulties in the assessment of dysplasia. Leukemia Research, 45. p. 75–81.

Goicoechea, S., Pallero, M. A., Eggleton, P., Michalak, M., & Murphy-Ullrich, J. E. (2002), The anti-adhesive activity of thrombospondin is mediated by the N-terminal domain of cell surface calreticulin. Journal of Biological Chemistry, 277(40). p. 37219–37228.

Gotlib, J. (2013) JAK inhibition in the myeloproliferative neoplasms: lessons learned from the bench and bedside. Hematology / the Education Program of the American Society of Hematology. American Society of Hematology. Education Program. Hematology Am Soc Hematol Educ Program. p. 529-537.

142

Greenberg, P., Cox, C., LeBeau, M. M., Fenaux, P., Morel, P., Sanz, G., Sanz, M., Vallespi, T., Hamblin, T., Oscier, D., Ohyashiki, K., Toyama, K., Aul, C., Mufti, G., Bennett, J. (1997), International scoring system for evaluating prognosis in myelodysplastic syndromes. Blood, 89(6). p. 2079–2088.

Greenberg, P., Tuechler, H., Schanz, J., Sanz, G., Garcia-Manero, G., Solé, F., Haase, D. (2012), Revised international prognostic scoring system for myelodysplastic syndromes. Blood, 120(12). p. 2454–2465.

Griesshammer, M., Gisslinger, H., & Mesa, R. (2015), Current and future treatment options for polycythemia vera. Annals of Hematology. Springer Verlag. p. 901-910.

Guglielmelli, P., Nangalia, J., Green, A. R., & Vannucchi, A. M. (2014), CALR mutations in myeloproliferative neoplasms: Hidden behind the reticulum. American Journal of Hematology, 89(5). p. 453–456.

Hanahan, D., & Weinberg, R. A. (2011), Hallmarks of Cancer: The Next Generation. Cell, 144(5). p. 646–674.

Harris, N. L., Jaffe, E. S., Diebold, J., Flandrin, G., Muller-Hermelink, H. K., Vardiman, J., Bloomfield, C. D. (1999), World health organization classification of neoplastic diseases of the hematopoietic and lymphoid tissues: Report of the clinical advisory committee meeting - Airlie house, Virginia, November 1997. Journal of Clinical Oncology. American Society of Clinical Oncology. p.3835-3849.

Harrison, C. N., & Lee, J. S. (2017), Myeloproliferative neoplasms. Medicine. p. 275-279.

Hatherley, D., Graham, S. C., Turner, J., Harlos, K., Stuart, D. I., & Barclay, A. N. (2008), Paired Receptor Specificity Explained by Structures of Signal Regulatory Proteins Alone and Complexed with CD47. Molecular Cell, 31(2). p. 266–277.

Hayat, S. M. G., Bianconi, V., Pirro, M., Jaafari, M. R., Hatamipour, M., & Sahebkar, A. (2020), CD47: role in the immune system and application to cancer therapy. Cellular Oncology. p. 19-30.

Heuser, M., Panagiota, V., Koenecke, C., Fehse, B., Alchalby, H., Badbaran, A., Thol, F. (2014,), Low frequency of calreticulin mutations in MDS patients. Leukemia. p. 1933–1934.

Hewitt, E. W. (2003), October. The MHC class I antigen presentation pathway: Strategies for viral immune evasion. Immunology.

Holmström, M. O., Martinenaite, E., Ahmad, S. M., Met, Friese, C., Kjær, L., Riley, C. H., Thor Straten, P., Svane, I. M., Hasselbalch, H. C., Andersen, M. H. (2018), The calreticulin (CALR) exon 9 mutations are promising targets for cancer immune therapy. Leukemia, 32(2). p. 429–437.

Howe, C., Garstka, M., Al-Balushi, M., Ghanem, E., Antoniou, A. N., Fritzsche, S., Springer, S. (2009), Calreticulin-dependent recycling in the early secretory pathway mediates optimal peptide loading of MHC class I molecules. The EMBO Journal, 28(23). p. 3730–3744.

Hsu, W. M., Hsieh, F. J., Jeng, Y. M., Kuo, M. L., Chen, C. N., Lai, D. M., Chen, W. J. (2005) Calreticulin expression in neuroblastoma - A novel independent prognostic factor. Annals of Oncology. P. 314-321.

143

Ishida, S., Akiyama, H., Umezawa, Y., Okada, K., Nogami, A., Oshikawa, G., Miura, O. (2018), Mechanisms for mTORC1 activation and synergistic induction of apoptosis by ruxolitinib and BH3 mimetics or autophagy inhibitors in JAK2-V617F-expressing leukemic cells including newly established PVTL-2. Oncotarget, 9(42). p. 26834–26851.

Jaiswal, S., Jamieson, C. H. M., Pang, W. W., Park, C. Y., Chao, M. P., Majeti, R., Traver, D., van Rooijen, N., Weissman, I. L. (2009), CD47 is upregulated on circulating hematopoietic stem cells and leukemia cells to avoid phagocytosis. Cell, 138(2). p. 271–285.

Jiang, H., Fu, R., Wang, H., Li, L., Liu, H., & Shao, Z. (2013), CD47 is expressed abnormally on hematopoietic cells in myelodysplastic syndrome. Leukemia Research, 37(8). p. 907–910.

Jonas, B. A., & Greenberg, P. L. (2015), MDS prognostic scoring systems – past, present, and future. Best Practice & Research. Clinical Haematology, 28(1). p. 3–13.

Jones, P. A., Taylor, S. M., & Wilson, V. (1983), DNA modification, differentiation, and transformation. Journal of Experimental Zoology, 228(2). p. 287–295.

Kaplan, D. H., Shankaran, V., Dighe, A. S., Stockert, E., Aguet, M., Old, L. J., & Schreiber, R. D. (1998). Demonstration of an interferon-dependent tumor surveillance system in immunocompetent mice. Immunology (Vol. 95).

Kauder, S. E., Kuo, T. C., Harrabi, O., Chen, A., Sangalang, E., Doyle, L., Rocha, S. S., Bollini, S., Han, B., Sim, J., Pons, J., Wan, H. I. (2018), ALX148 blocks CD47 and enhances innate and adaptive antitumor immunity with a favorable safety profile. PLoS ONE. 13(8). e0201832.

Keating, G. M. (2009), September. Azacitidine: A review of its use in higher-risk myelodysplastic syndromes acute myeloid leukaemia. Drugs.

Kern, W., Haferlach, C., Schnittger, S., & Haferlach, T. (2010), Clinical utility of multiparameter flow cytometry in the diagnosis of 1013 patients with suspected myelodysplastic syndrome: Correlation to cytomorphology, cytogenetics, and clinical data. Cancer, 116(19). p. 4549–4563.

Klampfl, T., Gisslinger, H., Harutyunyan, A. S., Nivarthi, H., Rumi, E., Milosevic, J. D., Kralovics, R. (2013), p. 2379–2390. Somatic Mutations of Calreticulin in Myeloproliferative Neoplasms. New England Journal of Medicine, 369(25), p. 2379–2390.

Korn, C., & Méndez-Ferrer, S. (2017), Myeloid malignancies and the microenvironment. Blood, 129(7). p. 811–822.

Kukreja, A., Radfar, S., Sun, B. H., Insogna, K., & Dhodapkar, M. V. (2009), Dominant role of CD47-thrombospondin-1 interactions in myeloma-induced fusion of human dendritic cells: Implications for bone disease. Blood, 114(16). p. 3413–3421.

Lally, J., Boasman, K., Brown, L., Martinelli, V., Cappuccio, I., Sovani, V., Marinaccio, C., Crispino, J. D., Graham, C., Rinaldi, C. (2019), GATA -1 a potential novel biomarker for the differentiation of essential thrombocythaemia and myelofibrosis. Journal of Thrombosis and Haemostasis. p. 14433.

144

Lee, P.-C., Chiang, J.-C., Chen, C.-Y., Chien, Y.-C., Chen, W.-M., Huang, C.-W., Lee, H. (2019), Calreticulin regulates vascular endothelial growth factor-A mRNA stability in gastric cancer cells. PLoS ONE 14(11). e0225107.

Lee, T. K. W., Cheung, V. C. H., Lu, P., Lau, E. Y. T., Ma, S., Tang, K. H., Ng, I. O. L. (2014), Blockade of CD47-mediated cathepsin S/protease-activated receptor 2 signalling provides a therapeutic target for hepatocellular carcinoma. Hepatology, 60(1). p. 179–191.

Leone, G., Voso, M. T., Teofili, L., & Lübbert, M. (2003), Inhibitors of DNA methylation in the treatment of hematological malignancies and MDS. Clinical Immunology, 109(1). p. 89–102.

Lian, S., Xie, X., Lu, Y., & Jia, L. (2019). Checkpoint CD47 function on tumor metastasis and immune therapy. OncoTargets and Therapy. Dove Medical Press Ltd.

Liu, J., Wang, L., Zhao, F., Tseng, S., Narayanan, C., Shura, L., Willingham, S., Howard, M., Prohaska, S., Volkmer, J., Chao, M., Weissman, I. L., Majeti, R. (2015), Pre-Clinical Development of a Humanized Anti-CD47 Antibody with Anti-Cancer Therapeutic Potential. PLOS ONE, 10(9). e0137345.

Liu, M., O’Connor, R. S., Trefely, S., Graham, K., Snyder, N. W., & Beatty, G. L. (2019), Metabolic rewiring of macrophages by CpG potentiates clearance of cancer cells and overcomes tumor-expressed CD47−mediated ‘don’t-eat-me’ signal. Nature Immunology, 20(3). p. 265–275.

Liu, P., Zhao, L., Loos, F., Marty, C., Xie, W., Martins, I., Lachkar, S., Qu, B., Waeckel-Énée, E., Plo, I., Vainchenker, W., Perez, F.,Rodriguez, D., López-Otin, C., van Endert, P., Zitvogel, L., Kepp, O., Kroemer, G. (2020), Immunosuppression by Mutated Calreticulin Released from Malignant Cells. Molecular Cell, 77(4). p. 748-760.

Liu, X., Kwon, H., Li, Z., & Fu, Y. xin. (2017), January. Is CD47 an innate immune checkpoint for tumor evasion? Journal of Hematology and Oncology.

Liu, X, Pu, Y., Cron, K., Deng, L., Kline, J., Frazier, W. A., Xu, M. M. (2015). CD47 blockade triggers T cell-mediated destruction of immunogenic tumors. Nature Medicine.

Liu, X., Sun, N., Dong, Y., Li, J., Liu, Y., Ren, Y., Yang, C., Zhang, L., Zhou, Y., Tong, Z., Ding, J., Zhao, X., Liu, Y., He, R., Xu, K., Wu, Y., Liu, Y. (2015), Anticancer effects of adenovirus-mediated calreticulin and melanoma-associated antigen 3 expression on non-small cell lung cancer cells. International Immunopharmacology, 25(2). p. 416–424.

Luo, W., & Yu, Z. (2015), Calreticulin (CALR) mutation in myeloproliferative neoplasms (MPNs). Stem Cell Investigation, 2(8). p. 16.

M.I. Reinhold, F.P. Lindberg, D. Plas, S. Reynolds, M.G. Peters, E. J. B. (1995), In vivo expression of alternatively spliced forms of integrin-associated protein (CD47). Journal of Cell Science, 108. p .3419–3425.

Majeti, R., Chao, M. P., Alizadeh, A. A., Pang, W. W., Jaiswal, S., Gibbs, K. D., Weissman, I. L. (2009), CD47 is an adverse prognostic factor and therapeutic antibody target on human acute myeloid leukemia stem cells. Cell, 138(2). p. 286–299.

145

Malcovati, L., Porta, M. G. D., Strupp, C., Ambaglio, I., Kuendgen, A., Nachtkamp, K., Cazzola, M. (2011), Impact of the degree of anemia on the outcome of patients with myelodysplastic syndrome and its integration into the WHO classification-based prognostic scoring system (WPSS). Haematological, 96(10). p. 1433–1440.

Manna, P. P., & Frazier, W. A. (2004), CD47 Mediates Killing of Breast Tumor Cells via Gi-Dependent Inhibition of Protein Kinase A. Cancer Research, 64(3). p. 1026–1036.

Mascarenhas, J., & Hoffman, R. (2012), Ruxolitinib: The first FDA approved therapy for the treatment of myelofibrosis. Clinical Cancer Research, 18(11). p. 3008–3014.

Matozaki, T., Murata, Y., Okazawa, H., & Ohnishi, H. (2009), February. Functions and molecular mechanisms of the CD47-SIRPα signalling pathway. Trends in Cell Biology. Trends Cell Biol.

Matsuo, Y., MacLeod, R. a, Uphoff, C. C., Drexler, H. G., Nishizaki, C., Katayama, Y., Kimura, G., Fujii, N., Omoto, E., Harada, M., Orita, K. (1997), Two acute monocytic leukemia (AML-M5a) cell lines (MOLM-13 and MOLM-14) with interclonal phenotypic heterogeneity showing MLL-AF9 fusion resulting from an occult chromosome insertion, ins(11;9)(q23;p22p23). Leukemia, 11(9). p. 1469–1477.

Méndez-Ferrer, S., Bonnet, D., Steensma, D. P., Hasserjian, R. P., Ghobrial, I. M., Gribben, J. G., Krause, D. S. (2020), May. Bone marrow niches in haematological malignancies. Nature Reviews Cancer. Nature Research.

Mesa, R. A., Yasothan, U., & Kirkpatrick, P. (2012), Ruxolitinib. Nature Reviews Drug Discovery, 11(2). p 103–104.

Michalak, M., Groenendyk, J., Szabo, E., Gold, L. I., & Opas, M. (2009), Calreticulin, a multi-process calcium-buffering chaperone of the endoplasmic reticulum. The Biochemical Journal, 417(3). p 651–666.

Michaux, J. L., & Martiat, P. (1993), Point of view: Chronic myelomonocytic leukaemia (CMML)-a myelodysplastic or myeloproliferative syndrome? Leukemia and Lymphoma, 9(1–2). p. 35–41.

Murata, Y., Kotani, T., Ohnishi, H., & Matozaki, T. (2014). The CD47-SIRPα signalling system: Its physiological roles and therapeutic application. Journal of Biochemistry.

Murati, A., Brecqueville, M., Devillier, R., Mozziconacci, M.-J., Gelsi-Boyer, V., & Birnbaum, D. (2012), Myeloid malignancies: mutations, models and management. BMC Cancer, 12(1). p. 304.

Mushegian, A. (2002), Refining structural and functional predictions for secretasome components by comparative sequence analysis. Proteins: Structure, Function, and Genetics, 47(1). p. 69–74.

Nangalia, J., Massie, C. E., Baxter, E. J., Nice, F. L., Gundem, G., Wedge, D. C., Avezov, E., Li, J., Kollmann, K., Kent, D.G., Aziz, A., Godfrey, A.L., Hinton, J., Martincorena, I.Van Loo, P., Jones, A.V., Guglielmelli, P., Tarpey, P., Harding, H.P., Fitzpatrick, J.D., Goudie, C.T., Ortmann, C.A., Loughran, S.J., Raine, K., Jones, D.R., Butler, A.P., Teague, J.W., O'Meara, S., McLaren, S., Bianchi, M., Silber, Y., Dimitropoulou, D., Bloxham, D., Mudie, L., Maddison, M., Robinson, B., Keohane, C., Maclean, C., Hill, K., Orchard, K., Tauro, S., Du, M-Q., Greaves, M., Bowen, D., Huntly, B.J.P., Harrison, C.N., Cross,

146

N.C.P., Ron, D., Vannucchi, A. M. M., Papaemmanuil, E., Campbell, P.J. J., Green, A. R. (2013a), Somatic CALR mutations in myeloproliferative neoplasms with nonmutated JAK2. The New England Journal of Medicine, 369(25). p. 2391–2405.

Noblejas-López, M. del M., Nieto-Jiménez, C., Morcillo García, S., Pérez-Peña, J., Nuncia-Cantarero, M., Andrés-Pretel, F., Ocana, A. (2019). Expression of MHC class I, HLA-A and HLA-B identifies immune-activated breast tumors with favorable outcome. OncoImmunology, 8(10).

Nygard, A. B., Jørgensen, C. B., Cirera, S., & Fredholm, M. (2007), Selection of reference genes for gene expression studies in pig tissues using SYBR green qPCR. BMC Molecular Biology, 8(1). p. 1–6.

Oaks, J., Wang, M., & Zou, H. (2018), Abstract 5623: Development of a CD47-blocking antibody as a cancer therapy. In Cancer Research (Vol. 78). American Association for Cancer Research (AACR). p. 5623–5623.

Obeid, M., Tesniere, A., Ghiringhelli, F., Fimia, G. M., Apetoh, L., Perfettini, J.-L., Castedo, M., Mignot, G., Panaretakis, T., Casares, N., Métivier, D., Larochette, N., van Endert, P., Ciccosanti, F., Piacentini, M., Zitvogel, L., Kroemer, G. (2007), Calreticulin exposure dictates the immunogenicity of cancer cell death. Nature Medicine, 13(1). p. 54–61.

Ogden, C. A., DeCathelineau, A., Hoffmann, P. R., Bratton, D., Fadok, B., Ghebrehiwet, V. A., & Henson, P. M. (2001), C1q and mannose binding lectin engagement of cell surface calreticulin and CD91 initiates macropinocytosis and uptake of apoptotic cells. Journal of Experimental Medicine, 194(6). p. 781–795.

Oldenborg, P.-A., & Per-Arne. (2013), CD47: A Cell Surface Glycoprotein Which Regulates Multiple Functions of Hematopoietic Cells in Health and Disease. ISRN Hematology. p. 614-619.

Oldenborg, P. A., Gresham, H. D., & Lindberg, F. P. (2001), CD47-signal regulatory protein α (SIRPα) regulates Fcγ and complement receptor-mediated phagocytosis. Journal of Experimental Medicine, 193(7). p. 855–861.

Oldenborg, P. A., Zheleznyak, A., Fang, Y. F., Lagenaur, C. F., Gresham, H. D., & Lindberg, F. P. (2000), Role of CD47 as a marker of self on red blood cells. Science, 288(5473). p. 2051–2054.

Olson, N. D., & Morrow, J. B. (2012), DNA extract characterization process for microbial detection methods development and validation. BMC Research Notes, 5. p. 668.

Orkin, S. H. (2000). Diversification of haematopoietic stem cells to specific lineages. Nature Reviews Genetics. European Association for Cardio-Thoracic Surgery.

Pakos-Zebrucka, K., Koryga, I., Mnich, K., Ljujic, M., Samali, A., & Gorman, A. M. (2016), The integrated stress response. EMBO Reports, 17(10). p. 1374–1395.

Panaretakis, T., Joza, N., Modjtahedi, N., Tesniere, A., Vitale, I., Durchschlag, M., Kroemer, G. (2008), The co-translocation of ERp57 and calreticulin determines the immunogenicity of cell death. Cell Death and Differentiation, 15(9). p. 1499–1509.

147

Panaretakis, Theocharis, Kepp, O., Brockmeier, U., Tesniere, A., Bjorklund, A.-C., Chapman, D. C., Kroemer, G. (2009), Mechanisms of pre-apoptotic calreticulin exposure in immunogenic cell death. The EMBO Journal, 28(5). p. 578–590.

Pang, W. W., Pluvinage, J. V, Price, E. A., Sridhar, K., Arber, D. A., Greenberg, P. L., Schrier, S. L., Park, C. Y., Weissman, I. L. (2013), Hematopoietic stem cell and progenitor cell mechanisms in myelodysplastic syndromes. Proceedings of the National Academy of Sciences of the United States of America, 110(8). p. 3011–3016.

Pardanani, A., & Tefferi, A. (2011), Targeting myeloproliferative neoplasms with JAK inhibitors. Current Opinion in Hematology, 18(2). p. 105–110.

Park, J. H., Sevin, M., Ramla, S., Truffot, A., Verrier, T., Bouchot, D., Girodon, F. (2015). Calreticulin mutations in myeloproliferative neoplasms: Comparison of three diagnostic methods. PLoS ONE, 10(10).

Peng, R. Q., Chen, Y. B., Ding, Y., Zhang, R., Zhang, X., Yu, X. J., Zhou, Z. W., Zeng, Y. X., Zhang, X. S. (2010), p. 2428–2434. Expression of calreticulin is associated with infiltration of T-cells in stage III B colon cancer. World Journal of Gastroenterology, 16(19).

Petrova, P. S., Viller, N. N., Wong, M., Pang, X., Lin, G. H. Y., Dodge, K., Uger, R. A. (2017). TTI-621 (SIRPαFc): A CD47-blocking innate immune checkpoint inhibitor with broad antitumor activity and minimal erythrocyte binding. Clinical Cancer Research.

Piccione, E. C., Juarez, S., Liu, J., Tseng, S., Ryan, C. E., Narayanan, C., Wang, L., Weiskopf, K., Majeti, R. (2015). A bispecific antibody targeting CD47 and CD20 selectively binds and eliminates dual antigen expressing lymphoma cells. MAbs.

Pieri, L., Guglielmelli, P., & Vannucchi, A. M. (2015), Ruxolitinib-induced reversal of alopecia universalis in a patient with essential thrombocythemia. American Journal of Hematology, 90(1). p. 82–83.

Qin, T., Jelinek, J., Si, J., Shu, J., & Issa, J. P. J. (2009), Mechanisms of resistance to 5-aza-2’-deoxycytidine in human cancer cell lines. Blood, 113(3). p. 659–667.

Ren, L., Leisegang, M., Deng, B., Matsuda, T., Kiyotani, K., Kato, T., Harada, M., Park, J. H., Saloura, V., Seiwert, T., Vokes, E., Agrawal, N., Nakamura, Y. (2019), Identification of neoantigen-specific T cells and their targets: implications for immunotherapy of head and neck squamous cell carcinoma. OncoImmunology, 8(4). e1568813.

Rosati, S., Mick, R., Xu, F., Stonys, E., Le Beau, M. M., Larson, R., & Vardiman, J. W. (1996), Refractory cytopenia with multilineage dysplasia: Further characterization of an “Unclassifiable” myelodysplastic syndrome. Leukemia, 10(1). p. 20–26.

Rotunno, G., Mannarelli, C., Guglielmelli, P., Pacilli, A., Pancrazzi, A., Pieri, L., Fanelli, T., Bosi, A., Vannucchi, A. M. (2014), Impact of calreticulin mutations on clinical and hematological phenotype and outcome in essential thrombocythemia. Blood, 123(10). p. 1552–1555.

148

Rumi, E., Pietra, D., Ferretti, V., Klampfl, T., Harutyunyan, A. S., Milosevic, J. D., Them, Nicole C.C. Berg, T., Elena, C Casetti, I. C., Milanesi, C., Sant'Antonio, E., Bellini, M., Fugazza, E., Renna, M. C., Boveri, E., Astori, C., Pascutto, C., Kralovics, R., Cazzola, M. (2014), JAK2 or CALR mutation status defines subtypes of essential thrombocythemia with substantially different clinical course and outcomes. Blood, 123(10). p. 1544–1551.

Rutkevich, L. A., & Williams, D. B. (2011), April. Participation of lectin chaperones and thiol oxidoreductases in protein folding within the endoplasmic reticulum. Current Opinion in Cell Biology. Elsevier Current Trends.

Saito, Y., Ihara, Y., Leach, M. R., Cohen-Doyle, M. F., & Williams, D. B. (1999). Calreticulin functions in vitro as a molecular chaperone for both glycosylated and non-glycosylated proteins. The EMBO Journal (18).

Salvagno, C., & Cubillos-Ruiz, J. R. (2019), The impact of endoplasmic reticulum stress responses in dendritic cell immunobiology. In International Review of Cell and Molecular Biology (349). p. 153–176.

Santini, V., Kantarjian, H. M., & Issa, J. P. (2001) April. Changes in DNA methylation in neoplasia: Pathophysiology and therapeutic implications. Annals of Internal Medicine. American College of Physicians.

Schardt, J. A., Weber, D., Eyholzer, M., Mueller, B. U., & Pabst, T. (2009), Activation of the unfolded protein response is associated with favorable prognosis in acute myeloid leukemia. Clinical Cancer Research, 15(11). p 3834–3841.

Schcolnik-Cabrera, A., Oldak, B., Juárez, M., Cruz-Rivera, M., Flisser, A., & Mendlovic, F. (2019). Calreticulin in phagocytosis and cancer: opposite roles in immune response outcomes. Apoptosis. Springer New York LLC.

Shankaran, V., Ikeda, H., Bruce, A. T., White, J. M., Swanson, P. E., Old, L. J., & Schreiber, R. D. (2001). IFNγ, and lymphocytes prevent primary tumour development and shape tumour immunogenicity. Nature.

Sheng, W., Chen, C., Dong, M., Wang, G., Zhou, J., Song, H., Li, Y., Zhang, Ji., Ding, S. (2017), Calreticulin promotes EGF-induced EMT in pancreatic cancer cells via Integrin/EGFR-ERK/MAPK signaling pathway. Cell Death & Disease, 8(10). e3147.

Sheng, W., Chen, C., Dong, M., Zhou, J., Liu, Q., Dong, Q., & Li, F. (2014), Overexpression of Calreticulin Contributes to the Development and Progression of Pancreatic Cancer. Journal of Cellular Physiology, 229(7). p 887–897.

Sick, E., Jeanne, A., Schneider, C., Dedieu, S., Takeda, K., & Martiny, L. (2012). CD47 update: A multifaceted actor in the tumour microenvironment of potential therapeutic interest. British Journal of Pharmacology. Br J Pharmacol.

Sockolosky, J. T., Dougan, M., Ingram, J. R., Ho, C. C. M., Kauke, M. J., Almo, S. C., Garciaa, K. C. (2016). Durable antitumor responses to CD47 blockade require adaptive immune stimulation. Proceedings of the National Academy of Sciences of the United States of America.

149

Stein, B. L., Moliterno, A. R., & Tiu, R. V. (2014). Polycythemia vera disease burden: contributing factors, impact on quality of life, and emerging treatment options. Annals of Hematology.

Stuart, G. R., Lynch, N. J., Day, A. J., Schwaeble, W. J., & Sim, R. B. (1997), The C1q and collectin binding site within C1q receptor (cell surface calreticulin). Immunopharmacology, 38(1–2). p 73–80.

Suzuki, J., Denning, D. P., Imanishi, E., Horvitz, H. R., & Nagata, S. (2013), Xk-related protein 8 and CED-8 promote phosphatidylserine exposure in apoptotic cells. Science, 341(6144). p 403–406.

Suzuki, S., Yokobori, T., Tanaka, N., Sakai, M., Sano, A., Inose, T., Sohda, M., Nakajima, M., Miyazaki, T., Kato, H., Kuwano, H. (2012), CD47 expression regulated by the miR-133a tumour suppressor is a novel prognostic marker in esophageal squamous cell carcinoma. Oncology Reports, 28(2). p 465–472.

Tajbakhsh, A., Gheibi Hayat, S. M., Butler, A. E., & Sahebkar, A. (2019). Effect of soluble cleavage products of important receptors/ligands on efferocytosis: Their role in inflammatory, autoimmune and cardiovascular disease. Ageing Research Reviews.

Tefferi, A., & Vardiman, J. W. (2008). Classification and diagnosis of myeloproliferative neoplasms: The 2008 World Health Organization criteria and point-of-care diagnostic algorithms. Leukemia.

Tefferi, Ayalew. (2016), Myeloproliferative neoplasms: A decade of discoveries and treatment advances. American Journal of Hematology, 91(1). p 50–58.

Tefferi, Ayalew, Guglielmelli, P., Larson, D. R., Finke, C., Wassie, E. A., Pieri, L., Gangat, N., Fjerza, R., Belachew, A. A., Lasho, T., Ketterling, R. P., Hanson, C. A., Rambaldi, A., Finazzi, G., Thiele, J., Vannucchi, A. M. (2014), Long-term survival and blast transformation in molecularly annotated essential thrombocythemia, polycythemia vera, and myelofibrosis. Blood, 124(16). p 2507–2513.

Titmarsh, G. J., Duncombe, A. S., Mcmullin, M. F., O’Rorke, M., Mesa, R., De Vocht, F., Horan, S., Fritschi, L., Clarke, M., Anderson, L. A. (2014), How common are myeloproliferative neoplasms? A systematic review and meta-analysis. American Journal of Hematology, 89(6). p 581–587.

Tsai, R. K., & Discher, D. E. (2008), Inhibition of “self” engulfment through deactivation of myosin-II at the phagocytic synapse between human cells. Journal of Cell Biology, 180(5). p 989–1003.

Tsai, R. K., Rodriguez, P. L., & Discher, D. E. (2010), Self inhibition of phagocytosis: the affinity of “marker of self” CD47 for SIRPalpha dictates potency of inhibition but only at low expression levels. Blood Cells, Molecules & Diseases, 45(1). p 67–74.

Unanue, E. R. (2013, July 2). Perspectives on anti-CD47 antibody treatment for experimental cancer. Proceedings of the National Academy of Sciences of the United States of America. National Academy of Sciences.

van Beek, E. M., Cochrane, F., Barclay, A. N., & van den Berg, T. K. (2005), Signal Regulatory Proteins in the Immune System. The Journal of Immunology, 175(12). p7781–7787.

150

Vannucchi, A. M., Kiladjian, J. J., Griesshammer, M., Masszi, T., Durrant, S., Passamonti, F., … Verstovsek, S. (2015), Ruxolitinib versus Standard Therapy for the Treatment of Polycythemia Vera. New England Journal of Medicine, 372(5). p 426–435.

Vardiman, J. (2012). The classification of MDS: From FAB to WHO and beyond. Leukemia Research.

Vardiman, J. W., Harris, N. L., & Brunning, R. D. (2002). The World Health Organization (WHO) classification of the myeloid neoplasms. Blood.

Vardiman, J. W., Thiele, J., Arber, D. A., Brunning, R. D., Borowitz, M. J., Porwit, A., Bloomfield, C. D. (2009). The 2008 revision of the World Health Organization (WHO) classification of myeloid neoplasms and acute leukemia: Rationale and important changes. Blood.

Variable, T., & Imager, M. (2020). Fluorescent Western Blotting. Work, 1–4.

Varricchio, L., Falchi, M., Dall’Ora, M., De Benedittis, C., Ruggeri, A., Uversky, V. N., & Migliaccio, A. R. (2017), Calreticulin: Challenges posed by the intrinsically disordered nature of calreticulin to the study of its function. Frontiers in Cell and Developmental Biology, 5. P 96.

Veillette, A., & Chen, J. (2018). SIRPα–CD47 Immune Checkpoint Blockade in Anticancer Therapy. Trends in Immunology. Elsevier Ltd.

Wang, Y., Xu, Z., Guo, S., Zhang, L., Sharma, A., Robertson, G. P., & Huang, L. (2013), Intravenous Delivery of siRNA Targeting CD47 Effectively Inhibits Melanoma Tumor Growth and Lung Metastasis. Molecular Therapy, 21(10). p 1919–1929.

Wearscha, P. A., Peapera, D. R., & Cresswell, P. (2011), Essential glycan-dependent interactions optimize MHC class I peptide loading. Proceedings of the National Academy of Sciences of the United States of America, 108(12). p 4950–4955.

Weiskopf, K., Jahchan, N. S., Schnorr, P. J., Cristea, S., Ring, A. M., Maute, R. L., Volkmer, A K., Volkmer, J-P., Liu, J., Lim, J S., Yang, D., Seitz, G., Nguyen, T., Wu, D., Jude, K., Guerston, H., Barkal, A., Trapani, F., George, J., Poirier, J T., Gardner, E E., Miles, L A., Stanchina, E., Lofgren, S M., Vogel, H., Winslow, M M., Dive, C., Thomas, R K., Rudin, C M., Van de Rijn, M., Majeti, R., Garcia, K., Weissman, C., Irving L.., Sage, J. (2016), CD47-blocking immunotherapies stimulate macrophage-mediated destruction of small-cell lung cancer. The Journal of Clinical Investigation, 126(7). p 2610–2620.

Wemeau, M., Kepp, O., Tesnière, A., Panaretakis, T., Flament, C., De Botton, S., Zitvogel, L., Kroemer, G., Chaput, N. (2010). Calreticulin exposure on malignant blasts predicts a cellular anticancer immune response in patients with acute myeloid leukemia. Cell Death and Disease, 1(12).

Willingham, S. B., Volkmer, J.-P., Gentles, A. J., Sahoo, D., Dalerba, P., Mitra, S. S., Weissman, I. L. (2012), The CD47-signal regulatory protein alpha (SIRPa) interaction is a therapeutic target for human solid tumors. Proceedings of the National Academy of Sciences of the United States of America, 109(17). p 6662–6667.

Wolanskyj, A. P., Schwager, S. M., McClure, R. F., Larson, D. R., & Tefferi, A. (2006), Essential thrombocythemia beyond the first decade: Life expectancy, long-term complication rates, and prognostic factors. Mayo Clinic Proceedings, 81(2). p 159–166

151

Wu, L., Wang, T., He, D., Li, X., & Jiang, Y. (2019), EVI-1 acts as an oncogene and positively regulates calreticulin in breast cancer. Molecular Medicine Reports, 19(3). p 1645–1653.

Xiang, Y.-R., & Liu, L. (2017). “Eating” Cancer cells by blocking CD47 signaling: Cancer therapy by targeting the innate immune checkpoint. Cancer Translational Medicine, 3(6). p200.

Yang, X., Yao, R., & Wang, H. (2018).. Update of ALDH as a Potential Biomarker and Therapeutic Target for AML. BioMed Research International, 2018. p 1–5.

Zhao, C. L., Yu, S., Wang, S. H., Li, S. G., Wang, Z. J., & Han, S. N. (2018). Characterization of cluster of differentiation 47 expression and its potential as a therapeutic target in esophageal squamous cell cancer. Oncology Letters, 15(2). P 2017–2023.

152

Appendix

EA1 [doc version 09.02]

Ethical Approval Form: Library/Desk/Lab/Studio-based Research Projects

This form must be completed for each piece of research activity whether conducted by academic staff, research staff, graduate students or undergraduates. Applications by students must be endorsed by an academic member of staff acting as Principal Investigator/supervisor. The completed form must be sent to the designated Ethics Committee within the College.

Please complete all sections. If a section is not applicable, write N/A.

1 Name of Applicant

Kristian Boasman

2 School

LIFE SCIENCE

3 Position in the University

PhD Student

4 Role in relation to this research

Postgraduate student

5 Name(s) of collaborators/co-workers and their relationship to the project (e.g. supervisor, assistant etc.)

Name, and role in project: 1. Rachael Simpson (postgraduate student) 2. 3. 4. 5.

6 Brief statement of main Research Question or Project Title

Elucidating the molecular mechanisms of calreticulin and CD47 in myelodysplastic syndrome (MDS) and myeloproliferative neoplasms (MPN) – using model cell systems.

7 Ethical checklist Does the research involve living human participants, or human tissue? Yes No If you answered “yes”, submit form EA2 for Ethical Approval. Does the research involve living animals, or animal tissue? Yes No If you answered “yes”, submit form EA3 for Ethical Approval. Does the research involve confidential data, or data not in the public domain? Yes No Does the project potentially put you or your collaborators at physical or psychological risk? Yes No Could the topic or results of this research be seen as illegal, or attract legal action against the University from an outside agency? Yes No Could the topic or results of this research attract unwelcome media attention, or affect the reputation or standing of the University? Yes No Could the topic, results or conduct of this research be regarded as offensive, immoral or destructive by some reasonable people? Yes No Does this research need to be undertaken under a relevant professional code of conduct? Yes No Are there any potential conflicts of interest in conducting this research, including financial gain for the researchers, or for individuals or external organizations affiliated with the researchers? Yes No Are there any factors inhibiting the application of the University’s ethical guidelines, including those on proper treatment of data, research design and publication of results? Yes No Does the research require the approval of any external body? Yes No

153

Does the research involve living human participants, or human tissue? Yes No If you answered “yes”, submit form EA2 for Ethical Approval. Does the research involve living animals, or animal tissue? Yes No If you answered “yes”, submit form EA3 for Ethical Approval. Does the research involve confidential data, or data not in the public domain? Yes No Does the project potentially put you or your collaborators at physical or psychological risk? Yes No Could the topic or results of this research be seen as illegal, or attract legal action against the University from an outside agency? Yes No Could the topic or results of this research attract unwelcome media attention, or affect the reputation or standing of the University? Yes No Could the topic, results or conduct of this research be regarded as offensive, immoral or destructive by some reasonable people? Yes No Does this research need to be undertaken under a relevant professional code of conduct? Yes No Are there any potential conflicts of interest in conducting this research, including financial gain for the researchers, or for individuals or external organizations affiliated with the researchers? Yes No Are there any factors inhibiting the application of the University’s ethical guidelines, including those on proper treatment of data, research design and publication of results? Yes No Does the research require the approval of any external body? Yes No If the answer to all questions above is “No”, you may complete section 8 to certify that there are no ethical issues, submit this form to the relevant Ethics Committee, and proceed with the research immediately. You accept professional responsibility for this decision, and if unsure should instead submit to the Committee.

If the answer to any of the above questions is “Yes”, complete the rest of the form, submit to the relevant Ethics Committee, and await approval before proceeding with the research. Answering “Yes” does not necessarily imply that the research is problematic, only the Ethics Committee needs to consider the research to ensure that it can proceed, and that the research design conforms to best practice. Having reviewed the ethical implications of this research, I certify that there are no issues requiring Ethical Approval. I certify that the research will be carried out in compliance with the University’s ethical guidelines for library/desk/laboratory/studio-based research, with Health and Safety regulations, and with all other relevant University policies and procedures. If there are any changes to the research requiring ethical clearance, I shall apply for such clearance before continuing with the research. Signed: Principal Investigator Note. This section must be endorsed by the member of academic staff responsible for the project. In the case of research by students, the supervising member of academic staff must sign. The signed form should then be submitted to the relevant Ethics Committee within the College, and the research may proceed. Yes No If “No”, provide an ethical justification for your project and explain why you wish to continue with the research in breach of normal ethica

154

13 If there are any other ethical issues, to which the attention of the approving committee should be drawn, please state them in this section, and explain how you have taken the issues into account, so that the research should be approved. Please consult the University’s ethical guidelines for advice.

Please also include here, or attach separately, a brief description of the research, to allow the approving committee to reach judgement.

Introduction - The myelodysplastic syndromes (MDS), and myeloprolife disorders that have the potential to progress to acute leukaemia (cancer MDS/MPN to acute myeloid Leukaemia (AML) occurs not just in blood b Overall survival rates can range from low risk (4-18 months) to high risk CALR is a multi-process calcium buffering chaperone of the endoplasmic used as a biomarker of cancer. In recent studies it has been show that cancer patient’s show increased (CALR) (2), and anti-phagocytic signal CD47 (4). The high levels presen needed to protect the cells from CALR mediated phagocytosis. CD47 an marrow cells. CALR has been shown to be upregulated in certain AML s contrast with previous studies in solid tumours, there is a down-regulatio with MDS (1).Whilst the full mechanisms of CALR and CD47 in relation t understanding of these molecules could potentially be exploited for use a and bone marrow samples. Aim- The overarching aim of the project is to dissect the molecular mech haematopoietic transcription factors such as: CALR and CD47, and their will access the levels of CALR and CD47 mRNA in a larger cohort of MD will explore the role of CALR in MPN/MDS to elucidate the signalling me Elucidating these signalling mechanisms will indicate future therapeutic t MDS and MPN. Methods -The purpose of this study is to dissect the molecular pathwa and partner transcription factors in a series of commercially available ce and further investigate any changes in these pathways. Total RNA and pro RT-qPCR for changes in gene transcription. Proteomic investigations of employed to look for changes in protein expression in commercially avai Ethical Issues - For the methodological work-up of these investigations and procedures, we will use commercially available cell lines, with no im of disease’. However, due to the very nature of MDS disease there is available. We will also use commercially available mouse ES cells to ge cell model with hematopoietic potential as an additional model of disease techniques as already stated. For further information on these cells lines please see associated EA3 fo References

(1) Panga, 2016. PNAS. 3011–3016 (2) Park, 2015. Plos ONE, 10. 1-8. (3) YI, 2016. Oncology Letters,1861-1867 (4) Chao, 2010. Sci Transl Med 2: 63-94. (5) Kourie, 2016 Myeloma Leuk , 264-268 (6) Park, 2015. Cancer Genet 630-635

155

EA2 Ethical Approval Form: Human Research Projects

Please word-process this form, handwritten applications will not be accepted

This form must be completed for each piece of research activity whether conducted by academic staff, research staff, graduate students or undergraduates. The completed form must be approved by the designated authority within the College.

Please complete all sections. If a section is not applicable, write N/A.

1 Name of Applicant

Kristian Boasman School: Life Sciences

College: Science

2 Position in the University

PhD Student

3 Role in relation to this research

Postgraduate researcher

4 Brief statement of main Research Question

Elucidating the role of Calreticulin and CD47 on myelodysplastic syndrome (MDS) and myeloproliferative neoplasms (MNP). (Human samples)

5 Brief Description of Project

Introduction - The myelodysplastic syndromes (MDS), and myeloproliferative neoplasms (MPN), are a group of blood disorders that have the potential to progress to acute leukaemia (cancer of the white blood cells). The progression of MDS/MPN to acute myeloid Leukaemia (AML) occurs not just in blood but initiates in the bone marrow once it is affected. Overall survival rates can range from low risk (4-18 months) to high risk (4-6 months). CALR is a multi-process calcium buffering chaperone of the endoplasmic reticulum. The over expression of CALR is used as a biomarker of cancer. In recent studies it has been show that cancer patient’s show increased levels of the pro-phagocytic signals calreticulin (CALR), and anti-phagocytic signal CD47. The high levels present in cancer cells potentially show that CD47 is needed to protect the cells from CALR mediated phagocytosis. CD47 and CALR are present in regular blood and bone marrow cells. CALR has been shown to be upregulated in certain AML studies. However recent data suggest, in contrast with previous studies in solid tumours, there is a down-regulation of CALR mRNA in the bone marrow of patients with MDS. Whilst the full mechanisms of CALR and CD47 in relation to MPN and MDS are not fully understood, greater understanding of these molecules could potentially be exploited for use as a biomarker, especially if seen in both blood and bone marrow samples. Aim: The purpose of this study is to dissect the molecular pathways involved in the expression of Calreticulin, CD47, and partner transcription factors. We will use bone marrow and peripheral blood samples from 80 patients affected by MDS/MPN (20 low risk; 20 Int-1; 20 Int-2 and 20 high risk patients). Samples will be collected by a trained phlebotomist. Method: Mononuclear cells will be separated using Ficoll method and total RNA and proteins will be extracted, rendering samples acellular. RNA and protein will be stored only and not functional cells. All processes are exempt from the Human Tissue Act. We will measure, evaluate, and further investigate any changes in these pathways. Total RNA and protein will be extracted and assessed by quantitative RT-qPCR for changes in gene transcription. Proteomic investigations of western blot and mass spectrometry will also be employed to look for changes in protein expression. FACS analysis will also be carried out to show cell surface abundance.

156

9 Statement of the ethical issues involved and how they are to be addressed –including a risk assessment of the project based on the vulnerability of participants, the extent to which it is likely to be harmful and whether there will be

Obtaining Informed Consent and Exclusion Criteria to Prevent any Vulnerable Participants Taking Part: The following measures will be employed

- An easy to read yet comprehensive leaflet detailing the full study will be provided and all and any questions will be answered in full before any consent is given (Please see the participation information sheet).

- No volunteer below the age of 18yrs will be permitted to take part.

- A signed informed consent form will be required before any samples are taken and a cool off period will be allowed.

5 Brief Description of Project

Introduction - The myelodysplastic syndromes (MDS), and myeloproliferative neoplasms (MPN), are a group of blood disorders that have the potential to progress to acute leukaemia (cancer of the white blood cells). The progression of MDS/MPN to acute myeloid Leukaemia (AML) occurs not just in blood but initiates in the bone marrow once it is affected. Overall survival rates can range from low risk (4-18 months) to high risk (4-6 months). CALR is a multi-process calcium buffering chaperone of the endoplasmic reticulum. The over expression of CALR is used as a biomarker of cancer. In recent studies it has been show that cancer patient’s show increased levels of the pro-phagocytic signals calreticulin (CALR), and anti-phagocytic signal CD47. The high levels present in cancer cells potentially show that CD47 is needed to protect the cells from CALR mediated phagocytosis. CD47 and CALR are present in regular blood and bone marrow cells. CALR has been shown to be upregulated in certain AML studies. However recent data suggest, in contrast with previous studies in solid tumours, there is a down-regulation of CALR mRNA in the bone marrow of patients with MDS. Whilst the full mechanisms of CALR and CD47 in relation to MPN and MDS are not fully understood, greater understanding of these molecules could potentially be exploited for use as a biomarker, especially if seen in both blood and bone marrow samples. Aim: The purpose of this study is to dissect the molecular pathways involved in the expression of Calreticulin, CD47, and partner transcription factors. We will use bone marrow and peripheral blood samples from 80 patients affected by MDS/MPN (20 low risk; 20 Int-1; 20 Int-2 and 20 high risk patients). Samples will be collected by a trained phlebotomist. Method: Mononuclear cells will be separated using Ficoll method and total RNA and proteins will be extracted, rendering samples acellular. RNA and protein will be stored only and not functional cells. All processes are exempt from the Human Tissue Act. We will measure, evaluate, and further investigate any changes in these pathways. Total RNA and protein will be extracted and assessed by quantitative RT-qPCR for changes in gene transcription. Proteomic investigations of western blot and mass spectrometry will also be employed to look for changes in protein expression. FACS analysis will also be carried out to show cell surface abundance. Approximate Start Date: Oct 2016

Approximate End Date: Oct 2019

6 Name of Principal Investigator or Supervisor

Dr Ciro Rinaldi Dr Matt Simmonds Email address: [email protected] [email protected]

Telephone: 01205 446311 01522 886896

7 Names of other researchers or student investigators involved

Rachael Simpson

8 Location(s) at which project is to be carried out

Molecular Research Lab, Joseph Banks Laboratories

157

9 Statement of the ethical issues involved and how they are to be addressed –including a risk assessment of the project based on the vulnerability of participants, the extent to which it is likely to be harmful and whether there will be significant discomfort. (This will normally cover such issues as whether the risks/adverse effects associated with the project have been dealt with and whether the benefits of research outweigh the risks)

Obtaining Informed Consent and Exclusion Criteria to Prevent any Vulnerable Participants Taking Part: The following measures will be employed

- An easy to read yet comprehensive leaflet detailing the full study will be provided and all and any questions will be answered in full before any consent is given (Please see the participation information sheet).

- No volunteer below the age of 18yrs will be permitted to take part.

- A signed informed consent form will be required before any samples are taken and a cool off period will be allowed.

Obtaining Samples and Sample Anonymization:

- All consent forms and patient information will be kept at Boston hospital in a secured area by Professor Ciro Rinaldi and the samples will be labelled with a specific code to ensure anonymity to the patient from their given sample.

- Peripheral blood samples from healthy volunteers will be drawn by a registered haematologist ensuring no risk to the volunteer.

- No information regarding the identity of the donor volunteer will be held at the University of Lincoln, UK and only anonymised sample ID and matched anonymised data will ever be given.

Sample Processing at the University of Lincoln:

- All samples will be rendered acellular immediately, by extracting the Mononuclear cells using Ficoll method. The total RNA and proteins will be extracted and kept for a maximum of 5 years.

Data Protection:

- All data protection and sample preparation will be processed and held in conjunction with the NHS and university guidelines.

- Data on these samples will be kept for six years after it is published in line with journal guidelines on how long raw data should be kept.

8 Location(s) at which project is to be carried out

Molecular Research Lab, Joseph Banks Laboratories

158

s research require

of an external body

Yes No If “Yes”, please state which body:- NHS IRAS form

cal approval

n from that body ?

N/A Yes -Please append documentary evidence to this form. No If “No”, please state why not:- An IRAS form has been sent and is currently awaiting a decision. Previously approval has been given for a project using MNP samples and there is currently approval for MDS samples to be used. IRAS has been sent to confirm the approval of both for this specific project. Please note that any such approvals must be obtained and documented before the project begins.

159

CONSENT FORM – Healthy Volunteer Consent

Elucidating the role of Calreticulin and CD47 on MDS/MPN Name of Researcher: Dr Ciro Rinaldi – Consultant Haematologist, Dr Matt Simmonds (Senior Lecturer) & Dr Ciaren Graham (Senior Lecturer)

Please initial box each box

1. I confirm that I have read and understand the information sheet dated 07/01/2015 (Final Version 1.0) for the above study. I have had the opportunity to consider the information, ask questions and have had these answered satisfactorily. 2. I understand that my participation is voluntary and that I am free to withdraw at any time 3. I understand that data collected during the study, may be looked at by authorised individuals from the University of Lincoln, the NHS Trust and from regulatory authorities, where it is relevant to my taking part in this research. I give permission for these individuals to have access to my records. 4. I agree to take part in the above study.

Name Signature Date ----------------------------------------- ----------------------------------- ------------- Name of Person taking consent Signature Date ----------------------------------------- ----------------------------------- -------------

160

Appendix 3.1: Example of full western blots ran to the n3 depicting CALR & GAPDH, CD47 & GAPDH

40kDa CALR 46kDa 36kDa GAPDH 36kDa

40kDa CD47 47-50kDa 36kDa GAPDH 46kDa

161

Figure 4.1. Example of knockdown optimisation blot and graph

1 2 3 4 5 6 7 8 9 40 kDa CALR 47kDa 30kDa GAPDH 36kDa

Figure 4.1 representative western blot of transfected k562 cells (A). This cell line was transfected with anti CALR siRNA (1µg/2x105cells). 1- ladder, 2- 3- anti CALR treated siRNA, 4-5 -untreated cells, 6-7 - scatter siRNA treated cells, 8-9 -

lipofectamine only treated cells. A graph representing the normalised bands of the western blot (B).

00.10.20.30.40.50.60.70.80.9

CALR siRNA Untreated Scatter siRNA Lipofectamine

Nor

mal

ised

ratio

exp

ress

iom

Expression of CALR and CD47 in anti CALR siRNA treated K562 cells

CALR CD47

B

Elucidating the molecular mechanisms of CALR and CD47 in ...

Documents

Transcript of Elucidating the molecular mechanisms of CALR and CD47 in ...