THE ROLE OF INTERACTIONS BETWEEN 4N1K AND CD47 ON INTEGRIN ACTIVATION AND INTEGRIN-MEDIATED CELL ADHESION by Pascal Leclair B.Sc., The University of British Columbia, 2011 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate and Postdoctoral Studies (Cell and Developmental Biology) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) June, 2014 © Pascal Leclair, 2014 ii Abstract Integrins are heterodimeric, cell-surface receptors that play a role in adhesion and chemoresistance. Integrins can be found in one of two states: an inactivated state where integrin-mediated adhesion is not supported, or in an activated state where integrin-mediated adhesion is possible. CD47 is a cell surface receptor that is said to regulate integrin functions by interacting with integrins and regulating their activation state. A thrombospondin C-terminal motif, RFYVVMWK, has been found to be responsible for cell adhesion and was subsequently shown to be specific for CD47. The 10-mer peptide derived from this sequence, 4N1K, has since been used as the prototypic ligand for CD47 and this interaction has been shown to decrease integrin-mediated adhesion and induce cell aggregation and apoptosis. However, the role of the 4N1K/CD47/integrin axis in chemoresistance has not been investigated. My thesis investigated the consequences of CD47 ligation by 4N1K on integrin-mediated functions. I found that the 4N1K peptide induces binding of a variety of antibodies, including non-specific control antibodies, to the surface of cells. In addition, cells that were deficient in CD47 expression were able to bind substrate-immobilized 4N1K as efficiently as their CD47-expressing parental cells. 4N1K was also found to block cell adhesion and induce cell aggregation in a manner that was independent on CD47 expression. These results suggest that 4N1K may produce artifacts in assays that use antibodies as reporters of integrin activation due to its hyper-adhesive nature, resulting in 4N1K binding to a variety of Ig-containing proteins, such as antibodies and cell-surface proteins. In iii addition, non-specific binding of 4N1K on cell surfaces and homotypic 4N1K interactions appear to be responsible for many of the observed phenotypes on cell adhesion and aggregation, and may explain the reported CD47-independent effects of 4N1K. As such, I propose that the 4N1K peptide not be used as a ligand to assess the role of thrombospondin/ CD47 interactions on cell functions, since the non-specific adherent nature of 4N1K could lead to erroneous interpretations of experimental data. iv Preface Dr. Chinten James Lim and I designed all the work contained in this manuscript. I performed all experiments and wrote the manuscript. Parts of this manuscript have been published: Leclair P, Lim CJ (2014) CD47-Independent Effects Mediated by the TSP-Derived 4N1K Peptide. PLoS ONE 9(5): e98358. doi:10.1371/journal.pone.0098358 v Table of Contents Abstract ........................................................................................................................................... ii Preface ............................................................................................................................................ iv Table of Contents ............................................................................................................................ v List of Tables.................................................................................................................................. vi Acknowledgements ....................................................................................................................... vii Chapter 1: Introduction ................................................................................................................... 1 1.1 The Bone Marrow Microenvironment and Cell Adhesion-Mediated Drug Resistance ........ 2 1.2 Integrins................................................................................................................................. 3 1.3 Regulation of Integrin Activation ......................................................................................... 6 1.4 Role of Integrins in Inflammation and Migration ............................................................... 13 1.5 Integrins and CAMDR ........................................................................................................ 17 1.6 CD47 ................................................................................................................................... 18 1.7 CD47 Ligands ..................................................................................................................... 21 1.8 CD47-Independent Effects of 4N1K ................................................................................... 25 1.9 CD47 Signaling ................................................................................................................... 25 1.10 Interactions Between Integrins and CD47 ........................................................................ 27 1.11 Hypothesis and Aims ........................................................................................................ 28 Chapter 2: Materials and Methods ................................................................................................ 30 Chapter 3: Results ......................................................................................................................... 37 3.1 Establishment of an Assay to Report β1-Integrin Activation (The “Activation Assay”) .... 37 3.2 Conditions Required for 4N1K Activity ............................................................................. 38 3.3 Consequences of CD47 Ligation by 4N1K on Integrin Activation .................................... 42 3.4 Characterization of the Non-specific Adherent Nature of 4N1K ........................................ 47 3.5 Consequences of CD47 ligation by 4N1K on Integrin-mediated Cell Adhesion ................ 51 3.6 Consequences of Cell Incubation with 4N1K on Jurkat Cell Aggregation ......................... 54 3.7 Generation of CD47-deficient Jurkat CRISPR Knock-outs. ............................................... 56 3.8 Discussion and Conclusion ................................................................................................. 66 References ..................................................................................................................................... 72 Appendix ....................................................................................................................................... 81 vi List of Figures Figure 1: Structure of Integrins ....................................................................................................... 5 Figure 2: Mechanism of Integrin Activation ................................................................................. 12 Figure 3: Cell Signaling in Polarized Cells ................................................................................... 16 Figure 4: CD47 .............................................................................................................................. 20 Figure 5: Structure of Thrombospondin and Receptor Binding Sites ........................................... 24 Figure 6: Integrin Activation Assay: Proof of Concept ................................................................ 37 Figure 7: 4N1K Must be Dissolved in bRPMI to be Active ......................................................... 39 Figure 8: Solubilization of 4N1K in Serum-free RPMI Results in its Aggregation ..................... 40 Figure 9: Titration of 4N1K in Activation Assays ........................................................................ 41 Figure 10: Selected Receptor Expression on Jurkat Derivative Cell Lines ................................... 44 Figure 11: 9EG7 Antibody Binding is Increased in 4N1K-treated Jurkat Cells ........................... 44 Figure 12: 4N1K Increases Binding of a Variety of Antibodies to Cell Surfaces ........................ 45 Figure 13: Integrin Activation Assay Using Various 4N1K and 4NGG Concentrations .............. 46 Figure 14: A Variety of Antibodies Bind Substrate-Immobilized 4N1K ..................................... 48 Figure 15: 4NGG and 4N1K Peptides Adequately Coat Wells .................................................... 49 Figure 16: Antibody Binding to Immobilized 4N1K Occurs in a Peptide Concentration-Dependent Manner ........................................................................................................................ 49 Figure 17: Cell-binding to 4N1K is Independent of CD47 Expression ........................................ 50 Figure 18: CD47 is not Required for 4N1K-mediated Decrease in Adhesion to Integrin Ligands.... ...................................................................................................................................... 52 Figure 19: 4N1K has Differential Effects on Cell Adhesion to Fibronectin Depending on its Concentration ................................................................................................................................ 53 Figure 20: 4N1K Induces Jurkat Cell Aggregation in a CD47-independent Manner ................... 55 Figure 21: CRISPR Knock-down of CD47. .................................................................................. 60 Figure 22: CD47 Expression on Clonal CD47 CRISPR Constructs ............................................. 61 Figure 23: Alignment of CD47 CRISPR Knock-down Clones with their Jurkat Parental ........... 63 Figure 24: Partial Sequencing Results of CRISPR Clone 1-5....................................................... 64 Figure 25: Jurkat CD47 CRISPR Predicted Amino Acid Sequences ........................................... 65 Figure 26: Receptor Profile of Polyclonal CD47 CRISPR Knock-down Cell Lines .................... 65 Figure 27: Proposed Mechanism for Non-specific 4N1K Effects on Cell Adhesion, Aggregation, and Antibody-based Assays .......................................................................................................... 71 Figure 28: 4N1K Mediates Non-specific Effects of Fibronectin and BSA Binding ..................... 81 Figure 29: 4N1K and MnCl2 Mediate Non-specific Binding of GST Peptides ............................ 83 Figure 31: 4N1K Induces AnnexinV Binding in a CD47-independent Manner ........................... 85 Figure 32: Anti-CD47 Antibody Clone CC2C6 Induces Apoptosis in a CD47-dependent Manner…. ..................................................................................................................................... 86 vii Acknowledgements I would like to thank my supervisor, Dr. Chinten James Lim, for his guidance throughout my training, as well as my supervisory committee, Dr. Cal Roskelley and Dr. Guy Tanentzapf, for their support and suggestions. I also appreciate all advice given by Dr. Chris Maxwell and Dr. Alex Beristain during weekly joint lab meetings. 1 Chapter 1: Introduction Childhood leukemia is a disease that occurs due to the transformation of hematopoietic cells, giving rise to cancerous cells that have unlimited potential for self-renewal, uncontrolled cell growth, and resistance to apoptosis. Acute Lymphoblastic Leukemia (ALL) is the most prevalent cancer for children between the ages of 2 and 5, accounting for one third of all childhood cancers [1, 2]. T-ALL, a form of ALL that affects thymocytes, is responsible for 15% of childhood leukemias and is said to be an aggressive and high-risk disease [2]. Current therapies for childhood ALL make use of highly toxic drugs administered during crucial years of development and lead to long-term and sometimes life-threatening complications, including cardiotoxicity, secondary malignancies, and neurocognitive deficits [3, 4]. Although the 5 year disease-free survival rate for childhood ALL is 80%, the remaining 20% will relapse due to minimal residual disease (MRD), the survival of a small population of leukemic cells, following chemotherapy treatment [5-7]. We now know that MRD occurs due to adherence of leukemic cells to components of the bone marrow niche via cell-surface receptors called integrins, conferring upon them the ability to resist chemotherapeutic drugs and evade apoptosis [6, 8, 9]. Following chemotherapy treatment, leukemic cells detach from the bone marrow niche, enter circulation, and start proliferating again giving rise to a subtype of leukemia that is resistant to drug therapies. The relapsed, chemoresistant form of ALL has a poor prognosis that often results in death [7]. 2 1.1 The Bone Marrow Microenvironment and Cell Adhesion-Mediated Drug Resistance The bone marrow (BM) is composed of two distinct niches: osteoblasts make up the osteoblastic niche where slow dividing, hematopoietic stem cells reside in a quiescent state, whereas endothelial cells are the main cell type of the vascular niche where hematopoietic stem cells proliferate and differentiate [6, 10, 11]. The main components of the BM extracellular matrix (ECM) are fibronectin, collagen, laminin, heparin sulfate, chondroitin, and hyaluronan [11]. Stromal-derived factor-1 (SDF-1, aka, CXCL-12) is a cytokine released by BM stromal cells that enables homing and retention of circulating cells, including leukemic cells, to the BM niche [12]. Once in the endosteal niche, leukemic cells bind to fibronectin in the extracellular matrix via cell-surface receptors of the integrin family where they gain resistance to chemotherapeutic drugs using a mechanism that is not well understood but likely responsible for this acquired chemoresistance [11, 13-15]; the mechanism by which cell adhesion leads to drug resistance is called cell adhesion-mediated drug resistance (CAMDR). As such, it has been shown that myeloma and acute myelogenous leukemia cell lines have a significant survival advantage to the chemotherapeutic drugs daunorubicin, doxorubicin, and AraC, when they are first allowed to adhere to α4- or α5-integrin substrates such as fibronectin [8, 15]. Finally, integrin-mediated drug resistance is not restricted to hematological malignancies since similar phenotypes have also been shown in fibrosarcomas [16], melanocytes [17], neuroblastomas [18], and Chinese hamster ovary cells [19]. 3 1.2 Integrins Integrins are transmembrane, heterodimeric proteins made up of one α- and one β-subunit, and are involved in adhesion, migration, and signal transduction, essential processes for extravasation and homing of immune cells [20, 21]. These cell-surface receptors were discovered due to their integral role in linking the extracellular matrix to the intracellular actin filament network [22]. Since then, 18 α- and 8 β-subunits have been discovered and these associate non-covalently to form up to 24 different heterodimers [21, 23]. Each integrin heterodimer has high affinity for specific ligands, either components of the extracellular matrix, such as fibronectin, laminin, thrombospondin, or fibrinogen, or counter-receptors expressed on cell surfaces, such as vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule-1 (ICAM-1), or mucosal vascular addressin adhesion molecule-1 (MadCAM-1) [23, 24]. Integrin dimers can be thought of as having three overall distinct domains: extracellular head domains that mediate binding to ligands, legs that include flexible knees and a transmembrane domain, and cytoplasmic tail domains that mediate intracellular signaling (Fig. 1 and 2) [25, 26]. The N-terminal of α-subunits consists of an extracellular β-propeller domain with 7 subdivisions, called blades, followed by a thigh and two membrane proximal calf domains; in addition, about half of the α-subunits also contain an inserted domain (I-domain) at the N-terminal; the C-terminal contains a single transmembrane (TM) domain and a short cytoplasmic tail (Fig. 1) [26, 27]. β-integrin subunits consist of an extracellular β-I domain, the ligand binding domain of non-αI-4 domain-containing integrin dimers, followed by a hybrid domain, a PSI domain, four EGF domains, and a membrane proximal β-tail domain (Fig. 1) [25, 26]. Metal cations, specifically, Ca2+, Mg2+, and Mn2+, have a significant role in integrin stability and integrity [28]. Integrin subunits have a total of 8 different cation binding sites, some specific for Ca2+ and others capable of binding any of the above (Fig. 1) [29]. Blades 4-7 of the α-subunit β-propeller each have a Ca2+ binding site that, together with the N-terminal of this domain, have been shown to be important in ligand recognition and affinity [30, 31]. In addition, mutations in these cation binding sites have been shown to prevent proper folding of the α-subunit and dimerization of the αβ molecules, leading to endoplasmic reticulum retention and subsequent degradation [32-34]. Another α-subunit Ca2+ binding site is found in its flexible knee, which was proposed to be important for stabilization of the extended conformation of integrins [27]. Finally, α-integrin subunits with an I-domain have an additional Ca2+ binding site called the metal ion-dependent adhesion site (MIDAS) which coordinates binding of ligands [26, 29]. β-integrin subunits contain three cation binding sites that are crucial for ligand binding (in non-αI-domain-containing integrins): a MIDAS site, a synergistic metal ion-binding site (SynMBS, also called ligand-induced metal binding site, LIMBS), and an adjacent to MIDAS (ADMIDAS) site (Fig. 1). Crystal structures determined that liganded integrins have all three sites occupied with metal ions, whereas ion occupancy of these sites in unliganded integrins seems to depend on the type of integrin heterodimer [35, 36]. Mutational studies of the MIDAS domain have shown that occupancy of this domain by Mg2+ is essential for ligand binding and this site coordinates with RGD-containing 5 ligands for binding [37, 38]. The LIMBS and ADMIDAS sites have two opposing functions. That is, the LIMBS positively regulates integrin activation by binding Ca2+ at physiological levels (low concentration), whereas the ADMIDAS site negatively regulates activation by high Ca2+ concentrations, which competes with the positive regulator Mn2+ for binding [38]. Finally, a synergy between the MIDAS and LIMBS sites exists such that low Ca2+ concentrations may help to stabilize Mg2+ binding to the MIDAS site [35]. Figure 1: Structure of Integrins. Integrins are formed from two heterodimers with short cytoplasmic tails, a transmembrane domain, and a long extracellular, ligand-binding domain. Each subunit also contains several Ca2+ binding sites that are important for integrin stability and function (shown as yellow circles). Red circles represent the MIDAS domains. Adapted from [26]. 6 1.3 Regulation of Integrin Activation Integrin activation is defined as a change in ligand-binding state of integrins, from low-to-high affinity [21]. Regulation of integrin affinity is crucial for the survival of organisms: among other things, leukocytes will not be able to arrest on endothelial cells if integrin activation is defective and immune cells will not able to reach sites of infection; conversely, cells with constitutively active integrins will be too “sticky” and impair efficient adhesion and migration responses. As such, integrins are found in one of two states to regulate a cell’s adhesiveness: a folded, inactive conformation, or an unfolded, active conformation (Fig. 2) [21]. Inactive integrins are characterized by having extracellular domains that are folded onto themselves, resulting in low-affinity of the receptor for extracellular ligands, and by having their cytoplasmic domains in close proximity to one another, preventing binding of downstream signaling proteins [21]. Stabilization of closed integrins has been shown to involve domains in the cytoplasmic, transmembrane, and extracellular domains: a highly conserved KxGFFKR motif in the α-subunit and a LLxxxHDR motif in the β-subunit cytoplasmic membrane-proximal domains, called the inner membrane clasp, was shown to be required for stabilization of integrins in the inactive state [39-41]; GxxxG motifs in both the α and β subunits are suggested to interact and form an outer membrane clasp that allows adequate packing of transmembrane helices [41-45]; a transmembrane dimer interface with a tilt angle of 25° was shown to maintain inner and outer membrane clasp integrity [41, 43, 46]; and finally, an extracellular clasp composed of a salt bridge between α and β subunits was found to stabilize the closed conformation [47]. To add to the complexity of integrin 7 activation, it was recently established that the number of residues present in the β1-subunit knee determines the ease with which these integrins can be activated [48]. “Inside-out” integrin activation can be initiated by a variety of extracellular chemokines [49, 50] or T-cell receptor agonists [50, 51], and terminates with the binding of the cytoskeletal protein talin to the β-integrin cytoplasmic tail [52, 53] (Fig. 2). Cytoplasmic signaling following extracellular stimulation begins with activation of Rap1-GTPase, which was found to form an “integrin activation complex” with Rap1-GTP-interacting adaptor molecule (RIAM) and talin [54, 55]. GTPases are proteins that act as molecular switches, being activated by guanine-nucleotide-exchange factor (GEFs), which stimulates the replacement of GDP by GTP, and inactivated by GTPase-activating proteins (GAPs), which hydrolyze GTP to GDP [56, 57]. Rap1-GTPase, the principle GTPase activated in response to chemokine stimulation leading to polarization and cell migration in lymphocytes, has a CAAX prenylation domain that restricts it to the plasma membrane [58]. Activated Rap1, therefore, results in the relocalization of RIAM and talin to the plasma membrane where talin can activate integrins [55]. RIAM is an adaptor protein that is part of the Mig10/RIAM/Lamellipodin (MRL) family of proteins that share PH, Ras-association, and proline-rich domains [59]. Talin is a cytosolic protein with a head domain consisting of 4 subdomains: F0, F1, F2, and F3, of which the F3 subdomain contains a phosphotyrosine-binding (PTB) domain responsible for integrin binding and activation, and a rod domain containing binding sites for vinculin and actin, providing a link between integrins and the cytoskeleton [25, 60]. Talin is normally in a conformation enabling interactions between its rod and F3 8 domains, which results in auto-inhibition and prevents premature activation of integrins [61] (Fig. 2). Release of talin auto-inhibition can be achieved in three ways: 1) by binding to RIAM [54, 55]; 2) by calpain-mediated proteolytic cleavage [62]; or 3) by binding to phosphatidylinositol 4,5-bisphosphate (PI(4,5)P2) [61]. With the release of auto-inhibition, talin binds to β-integrin tails via an NPxY domain [63], which results in the separation of the α- and β-integrin tail domains, thereby initiating integrin activation. Binding of the talin F3 domain to β-integrin subunits is proposed to destabilize the IMC and OMC, and change the tilt angle of the transmembrane subunits in a number of ways. Attraction between positive charges in the head domains and negatively charged phospholipids in the plasma membrane reorients talin, pulling the β-subunit tail along with it [64]. In addition, the charges on talin repel positive charges in the α-subunit cytoplasmic membrane proximal domain leading to the separation of the two integrin tails. Talin binding also results in a change in tilt angle between the two TM integrin domains which leads to the breaking of the inner and outer membrane clasps. These events cause a conformational change in integrins that leads to the unfolding of the extracellular domains in a switch-blade like motion, allowing high-affinity binding of ligands [41, 43, 46]. Outside-in integrin activation is also thought to occur as a result of extracellular ligands binding to inactive integrins, even though integrins in this state have low affinity for their ligands [65], or by binding of metal ions to binding sites in the extracellular domains [26]. In this case, ligand binding causes an entropic change in integrin conformation and 9 results in the unfolding of the extracellular domains and the separation of the cytoplasmic tails [21]. Finally, integrin clustering has been shown to result in changes in the degree to which integrins bind to ligands, although the mechanism for this phenomenon remains a controversial issue [52, 66]. Affinity of integrins describes the states of activation of integrins, either inactive (low affinity for ligands) or active (high affinity for ligands); in contrast, avidity is defined as the strength of cellular adhesion, which includes both the affinity state of integrins and the number of integrins adhered [66]. Integrin clustering has been shown to lead to the formation of a protein complex termed focal adhesion, which strengthens integrin-dependent adhesions by linking the actin cytoskeleton to the extracellular matrix via integrins [52, 67, 68]. One study showed that integrin clustering becomes possible only when four essential components are present: activated integrins, presence of PI(4,5)P2 and the adaptor protein talin, and binding to an immobilized ligand [67]. This evidence implies that although integrin activation leads to anchoring to the cytoskeleton, it also must precede clustering. But how does one resolve the fact that activated integrins anchored to the cytoskeleton can diffuse in the plasma membrane and lead to clustered formations? Cambi et al. attempted to shed light on this matter by investigating the organization of integrin clusters in monocytes and found that, contrary to previous studies, clusters can contain inactive integrins but that an individual cluster contains either activated or inactivated integrins, but not both [69]. In addition, they found that activated integrin clusters did not colocalize with talin or any other integrin tail-associated proteins, indicating that integrin 10 activation may not be a precursor to cluster formation. Another group found that although clusters containing activated integrins had multiple diffusion rates across the plasma membrane (stationary, slow, fast), their average mobility was significantly slower when compared to the mobility of all clusters [70]. In addition, they found an important role for extracellular Ca2+ in integrin cluster mobility in that there was a dramatic restriction of cluster diffusion at lower concentrations. As stated in a previous section, talin binding to α-integrin tail domains has been found to be the final step in integrin activation [52] (Fig.2). However, a recent study that developed a new assay to tease out integrin affinity changes from integrin clustering by labeling fixed cells with either a monovalent or a multivalent version of an activation-dependent antibody only found an increase in multivalent antibody binding to integrins following incubation with recombinant talin, indicating that talin increases integrin avidity and not integrin affinity [71]. This led the author to propose that previous assays were flawed in that they were reporting changes in integrin clustering rather changes in integrin affinity. Given the discrepancy in the evidence concerning the role of integrin clustering in integrin activation, or vice versa, it is obvious that further investigation will be required to elucidate the exact mechanism of action. A number of antibodies and small molecules have been developed to report integrin activation [72, 73]. Activation-dependent antibodies are divided into two classes: those that recognize epitopes that are exposed upon ligand binding (termed ligand-induced binding sites, LIBS) and those that bind activated integrins regardless of ligand occupancy [72]. HUTS-21 and 9EG7 are two antibodies of the latter class that bind 11 activated β1-integrin subunits and which have been used extensively in integrin research [72-78]. Specifically, the HUTS series of monoclonal antibodies (HUTS-4, HUTS-7, and HUTS-21) were determined to bind to β1-integrins when these were activated by an anti-β1-integrin antibody or by various cations, including Mn2+, where the degree of activation was most prominent when cells were incubated at 37°C [79]. An increase in immunoprecipitated β1-integrins was also detected following cell stimulation with the same reagents, confirming the specificity of these antibodies for activated β1-integrins. Similarly, Bazzoni et al. used the 9EG7 antibody to clarify the mechanism of integrin activation [80]. As such, this group determined that integrin activation preceded ligand binding since 9EG7 could bind unliganded integrins, and, in line with the known stimulatory effect of Mn2+ on integrin activation, 9EG7 binding was found to be induced by Mn2+. However, another study by the same group found that Ca2+ had opposing effects on different β1-heterodimers in that it was stimulatory for α4-integrins but inhibitory for α2-, α3-, α5-, and α6-integrins [81]. Furthermore, although 9EG7 was initially thought to be a LIBS antibody, Bazzoni et al. found that this was not universally the case since ligand occupancy only effectively increased 9EG7 binding in α4β1 and α5β1 integrins, whereas it minimally did so in other β1-integrins [81]. However, it was also suggested that an influential factor affecting antibody binding may be the extent to which conformational changes occur following integrin activation or ligand binding, such that α4β1 and α5β1 integrins undergo the greatest degree of change and α3β1 integrins undergoes the least. 12 Other strategies have also been developed to study integrin activation. An LDV peptide, the binding sequence included in some integrin-binding ligands, attached to a Fluorescein isothiocyanate (FITC) molecule, along with a non-fluorescently tagged counter-part, has been useful for real time analysis of integrin activation kinetics using competitive binding assays [73, 76, 82]. Förster Resonance Energy Transfer (FRET) assays using genetically altered α4-integrin subunits with a FITC molecule attached to the cytoplasmic tail, along with embedding of a plasma membrane FRET acceptor molecules, allowed the study of the various affinity states of integrins [83, 84]. Figure 2: Mechanism of Integrin Activation. Inactive integrins are found in a closed conformation that prevents downstream signaling (middle). Extracellular signals turn on pathways that terminate in the binding of talin to the cytoplasmic tail of β-integrin subunits, leading to the separation of the tails and extension of the extracellular domains. Activation allows integrins to adopt a high-affinity binding state for extracellular ligands, and allows cytoplasmic signaling proteins to bind to the tail domains. Figure taken from [21]. 13 1.4 Role of Integrins in Inflammation and Migration Integrins are especially important in homing of lymphocytes to areas of infection and inflammation in response to inflammatory chemoattractants [6, 20]. However, circulating leukocytes must first slow down and adhere to the vasculature before they can move from the blood to infected tissues. This is accomplished through a process of “rolling and arresting”, where loosely attached leukocytes tethering on vasculature endothelium roll along the endothelium through transient adhesions induced by shear stress on leukocytes and mediated by binding of selectins to sialoglycoproteins [20, 49]. Rolling cells are exposed to chemoattractants secreted by damaged tissue, resulting in the activation of integrins to a high affinity ligand binding state, which leads to the arrest of leukocytes on the surface of endothelia through tight adhesions provided by activated integrins [20]. Following firm attachment, leukocytes polarize in response to chemoattractant gradients and begin to migrate through the endothelium in the direction of increasing chemokine concentration. An integral part of this process, phosphatidylinositol(3,4,5)-triphosphate (PI(3,4,5)P3) is produced by phosphorylation of PI(4,5)P2 by phosphoinositide 3-kinase (PI3K), whereas the reverse process is achieved by phosphatase and tensin homologue deleted on chromosome ten (PTEN) [57] (Fig.3). Chemokine sensing by G-protein coupled receptors activates PI3K at the soon-to-be leading edge and results in the localized accumulation of PI(3,4,5)P3, whereas PTEN is excluded from the leading edge and restricted to the lateral and trailing edges of the cell, which further enriches PI(3,4,5)P3 to the leading edge [85]. Therefore, activation of cell-surface receptors by chemokines leads to the enrichment of plasma membrane-associated 14 PI(3,4,5)P3 at the leading edge of migrating cells due to the delicate balance between activated PI3K and exclusion of PTEN at the leading edge of the cells. Mediated by binding via a pleckstrin homology (PH) domains, accumulation of PI(3,4,5)P3 at the leading edge of migrating cells leads to the enrichment of several GEFs at the front of cells, which results in the activation of GTPases (importantly, Rac-GTPase and Rap1-GTPase) in this area [20, 57] (Fig.3). Cdc42, a Rho-subfamily GTPase that acts as the “master regulator of cell polarity” and also localized to the leading edge of migrating cells, regulates cell polarity in several ways: 1) It restricts where lamellipodia are formed by regulating the assembly of actin filaments. Srinivasan et al. showed that although Rac-GTPase was necessary and sufficient for polymerization of actin filaments, it was Cdc42 that regulated where these would ultimately form [86]. 2) It reorients the microtubule-organizing center (MTOC) and Golgi apparatus to the leading edge side of the nucleus during migration; in this way, protein cargo can quickly be moved to the front of the cell where it is most required [87, 88]. Finally, there is much evidence for a positive feedback loop between PI(3,4,5)P3 localization, Rac-GTPase and Cdc42 activation at the leading edge, and actin polymerization, that ensures that the leading edge will remain at the right location on migrating cells [89]. In addition to the heterogeneous distribution of cytosolic proteins, cell migration also involves the concerted activation of integrins at the leading edge of migrating cells and inactivation of integrins at the trailing edge (Fig. 3). Activation of integrins results in the separation of cytoplasmic domains that allow signaling proteins to bind to the tail domains of these receptors. One such signaling protein is paxillin, an adaptor protein that 15 restricts Rac1 activation by regulating GAP protein activity [20, 90]. Phosphorylation of α4-subunit tail domains by protein kinase A (PKA) prevents localized binding of paxillin and allows activation of Rac1 [91-93]. Furthermore, integrins regulate localization of PKA at the leading edge by serving as a type I cAMP-dependent protein kinase-anchoring protein (AKAP), to which PKA binds, resulting in preferential phosphorylation of α-subunit tail domains at the leading edge of cells [94-96]. Therefore, integrins induce cell polarity by acting as an AKAP which localizes PKA at the cell membrane, thereby facilitating the phosphorylation of α4-subunit tail domains, which prevents paxillin binding and allows Rac1 to remain active at the leading edge. 16 Figure 3: Cell Signaling in Polarized Cells. Ligation of G-protein coupled receptors by chemoattractants activates Rap-1, which in turns leads to activation of membrane-bound PI3K and enrichment of PI3K substrates at the leading edge of cells. PI3K catalyzes the formation of PI(4,5)P3 and results in localization of RAP1 at the cell membrane and generates positive feedback loop that increases activation of these proteins at the leading edge. Furthermore, exclusion of a phosphatase, PTEN, at the leading edge further enriches PI3K substrates. The leading edge is also formed by localized activation of Cdc42, resulting in actin polymerization and formation of protrusions, while disassembly of actin filaments at the rear results in retraction of the trailing edge. Adapted from [89] 17 1.5 Integrins and CAMDR α4β1 (VLA-4) and α5β1 (VLA-5) integrins bind to fibronectin, a protein highly expressed in the bone marrow environment [97]. Damiano et al. found that myeloma cells have a survival advantage when plated on fibronectin for 24 hours prior to incubation with doxorubicin, a chemotherapeutic drug commonly used in the treatment of leukemias, and this was shown to be mediated via 4β1-integrins [8]. Similarly, chronic myelogenous leukemia cells pre-treated with 5-integrin function-blocking antibodies reversed the CAMDR phenotype, thereby establishing a role for 5β1-integrins in this phenomenon [98]. Liu et al. also found a survival advantage for cells adhered to fibronectin via 4- and 5β1-integrins , but in addition, using a recombinant α4-integrin subunit with a truncated tail domain re-expressed in an otherwise α4-integrin-deficient Jurkat cell line, reported that the KxGFFKR motif in the α4-integrin tail domain regulates chemoresistance to doxorubicin in T-ALL cell lines [9]. Although one could argue for complete blockade of integrins as an effective therapy to prevent CAMDR, studies using α4-integrin-blocking antibodies have shown this to cause Progressive Multifocal Leukoencephalopathy, an opportunistic infection of the central nervous system that causes the demyelination of oligodendrocytes [99, 100]. Therefore, other avenues must be investigated to regulate integrin-mediated functions in hematological malignancies. One such avenue is to regulate β1-integrin heterodimer activity as it is one of the main players in CAMDR and a prominent integrin dimer expressed in leukemic cells [14, 100], in an effort to decrease adhesion of leukemic cells 18 to the bone marrow prior to chemotherapy. This would force these cells to remain in circulation during chemotherapy and prevent the survival of cells via CAMDR, thus minimizing the occurrence of minimal residual disease and improving the survivability of the patient’s battle with cancer. 1.6 CD47 CD47 is a 45kDa, five-pass membrane-spanning, cell-surface receptor that is highly glycosylated on its extracellular immunoglobulin variable (IgV) domain, the binding site for its ligands [101] (Fig. 4). Its cytosolic domain is alternatively spliced to produce four progressively longer isotypes, called isotypes 1 through 4, with isotype 4 being the longest and isotype 2 being the most expressed in hematopoietic cells [101, 102]. This receptor was initially discovered by three independent groups: one group identified it as a receptor that was absent in erythrocytes lacking the Rh antigen [103]; another group identified it as OA3, a receptor that was upregulated on ovarian carcinomas [104]; finally, one group identified it as “integrin-associated protein” (IAP) because they found it co-immunoprecipitated with integrin αvβ3 [105]. It was soon determined that all three of these proteins were one and the same [106]. CD47 localization to lipid rafts was found to be important in CD47-dependent cell spreading, actin polymerization, and synergy with T-cell receptor (TCR) in T-cell activation [107]. Recently, a disulfide bond between the Ig domain and one of the membrane spanning domains was found to be important for proper conformation of the receptor, for downstream signaling, and for localization to membrane rafts [107]. 19 CD47 is expressed in most tissue types studied so far and has been shown to be upregulated in a variety of cancer cells, including those of hematological origins like leukemia [101, 102, 108]. CD47 has also been found to be important in cell adhesion and motility [109, 110], cell aggregation [111], Ca2+ homeostasis and T-cell activation [107, 112], and apoptosis [113]. Importantly, it is a receptor that has been shown to be involved in recognition of self, such that cells that express high levels of this receptor will not be phagocytosed by macrophages and other professional phagocytes [114]. As such, CD47 is said to be a “don’t eat me signal” for a variety of cells [101]. The study of CD47 function was greatly simplified by the use of naturally CD47-deficient cell lines such as the human carcinoma OV-10 [104] and the intestinal epithelial cell CaCO2 [110], and by the genetic mutants 3656 and JinB8, fibroblasts from lung cell explants derived from a CD47-/- mouse [111, 115] and a Jurkat T-leukemic cell derivative (Jurkat IAP Negative) [109], respectively. Using these genetic models, in vitro assays have shown that cell adhesion was impaired in CD47-deficient cells when stimulated with CD47 ligands or antibodies, with integrin-activating antibodies, or by shear flow [109, 116, 117]. Similarly, Rebres et al. showed that although expression of CD47 in a monolayer of fibroblasts inhibited spontaneous migration of both Jurkat and JinB8 cells across the fibroblasts, it nevertheless improved migration efficiency during chemokine-stimulated migration [111]. In addition, another group showed that CD47 knock-down in HUVEC cells using siRNAs did not affect CD47-expressing T-cell adhesion to these cells, but rather decreased T-cell migration through them, indicating that CD47 is required on T-cells only for adhesion to endothelial cells but that efficient migration through them requires CD47 expression on both cells [118]. Interestingly, 20 migration of peripheral polymorphonuclear neutrophils (PMN) across a layer of epithelial cells was shown to be significantly delayed when CD47 on PMN was blocked using anti-CD47 antibodies, although the total number of migrating cells did not change [110]. It is clear from these results that the role of CD47 in cell migration is a complex one and requires further evaluation to fully elucidate the mechanism. Figure 4: CD47. CD47 consists of an extracellular IgV domain that mediates binding to ligands, a penta-spanning transmembrane domain, and an alternatively spliced cytoplasmic tail. IgV-domain glycosylation sites are shown as branches, and lines indicate disulfide bonds. Lines indicate known disulfide bridges between extracellular IgV domains and between the IgV and membrane spanning domain. Adapted from [119] 21 1.7 CD47 Ligands Thrombospondin (TSP), one of three ligands discovered for CD47, is a multifunctional protein composed of three monomers (in mammals) linked by disulfide bridges [120-122] (Fig. 5). It is composed of an N-terminal heparin-binding domain, a procollagen domain containing the cysteines responsible for dimerization, followed by type I, type II, and type III (which contains 33 Ca2+ binding sites) repeats, and finally a globular, cell-binding domain (CBD) at the C-terminus [120, 122, 123]. TSP exists in several isoforms, 1 to 5, with TSP-1 and TSP-2 being the best characterized forms. TSP-1, the prototypic form of TSP, is secreted by activated platelets and monocytes in response to inflammation where it is involved in clot formation and wound healing; TSP-2 is found in the nervous system and bone marrow stromal cells and also aids in wound healing [122]. TSP-1 has the ability to bind several different cell-surface receptors, including many integrins (Fig. 5). The N-terminus binds LDL-receptor related protein (LRP), heparin, and calreticulin, whereas the type I repeat has an affinity for CD36 [124]. Some β1-integrin dimers, including α4β1-integrins, can also bind TSP at its N-terminal domain via an LDV motif, whereas an RGD sequence found in the last C-terminal-proximal calcium domain mediates binding of α4β1, α5β1, and some β3-integrins [101, 116]. Finally, two Val-Val-Met (VVM) amino acid motifs at the C-terminus of TSP-1, named 4N1 (RFYVVMWK) and 7N3 (IRVVM), mediate cell binding to TSP [125]. CD47 was later identified as the receptor mediating cell binding to TSP, 4N1, and to a lesser extent, 7N3, via its IgV domain [126, 127]. 22 The 4N1 peptide has been used extensively since its discovery to study the consequences of CD47 ligation by TSP on cellular functions and signaling. When used as a peptide, one lysine residue is usually added to each end of 4N1 to increase its solubility and is therefore known as 4N1K [128]. As such, 4N1K and 7N3 were found to block cell adhesion to immobilized TSP or TSP fragments and to endothelium or epithelium layers [6, 48, 109, 129, 130]. However, some have also found an increase in adhesion to immobilized TSP following incubation with 4N1K or 7N3 [117, 129, 131], though the reason for this discrepancy remains elusive. Cell incubation with 4N1K was also found to increase platelet aggregation in several cell lines. Rebres et al. found that the mechanism for 4N1K-mediated cell aggregation was dependent on CD47 expression, although full length TSP could not induce the same phenotype [111]. Other studies confirmed these findings, but added that energy was not required in the induction of 4N1K-mediated aggregation since aggregation occurred in both energy-depleted cells and formaldehyde-fixed cells, indicating that this may be a passive reaction [132, 133]. 4N1K has also been used to elucidate the role of CD47 ligation by TSP on cell death. In this respect, immobilized 4N1K was found to increase breast cancer and T-ALL cell death as reported by an increase in annexinV binding [134, 135]. However, other studies found that 4N1K or 7N3 protected against the cytotoxic effects of chemotherapeutic drugs or radiotherapy [136, 137]. The seemingly contradictory effects of 4N1K on apoptosis (and possibly other CD47 functions) may be explained by the fact that CD47 on aged erythrocytes was found to have an increased ability to bind to ligands, leading to 23 increased phagocytosis of aged cells compared to younger cells upon incubation with 4N1K [138]. As mentioned above, an important role for CD47 is as a “don’t eat me” signal for professional phagocytes [139]. Signal-Regulatory Protein (SIRP) α (SHPS-1, CD172), another known ligand for CD47, is the co-receptor involved in this process [139, 140]. Of the three known SIRP isoforms, only SIRPα and SIRPγ have been shown to bind to CD47; SIRPγ has much lower binding affinity for CD47 than does SIRPα and its function in cell signaling remains poorly characterized [141-143]. All three isoforms were shown to have similar extracellular immunoglobulin domains, which, in the case of SIRPα and SIRPγ, mediate binding to the IgV domain of CD47 [144, 145]. Crystal structures of the SIRPα-CD47 complex were recently solved and have shown that a four-loop structure in the IgV domain of SIRPα and a two-loop structure in the IgV domain of CD47 are responsible for the interaction between these two receptors [144]. Cells that are apoptotic express cell-surface antigens, such as phosphatidylserine and calreticulin, that indicate to phagocytes that these cells need to be cleared [146]. However, these “eat me” signals also require down-regulation and redistribution of CD47 for efficient cell uptake. Indeed, Gardai et al. found that cell engulfment by macrophages did not occur if both calreticulin and CD47 were present on the cell surface of target cells, but did occur if CD47 was blocked from interaction with SIRPα [147]. In addition, there is a redistribution of CD47 on the cell surface of apoptotic cells, away from areas that are high in calreticulin and phosphatidylserine expression, so that detection of these antigens will not be nullified by activation of SIRPα on phagocytes by CD47 on 24 apoptotic cells. Finally, ligation of SIRPα in macrophages by platelet CD47 was shown to induce tyrosine phosphorylation of the SIRPα cytoplasmic tail by src homology-2 domain-containing protein tyrosine phosphatase (SHP)-1 and -2, whose downstream signaling inhibits presenting cell phagocytosis by the macrophage [148, 149]. This phenomenon was also shown to be true for leukocytes, as SIRPα phosphorylation was found to be significantly reduced in macrophages incubated with apoptotic Jurkat T-leukemic cells compared to viable non-apoptotic cells or cells incubated with a CD47-blocking antibody [93]. Figure 5: Structure of Thrombospondin and Receptor Binding Sites. The N-terminal heparin-binding domain (HBD) mediates adhesion to several receptors, including many β1-integrin heterodimers via an LDV motif. Other integrins also bind to the Type1 repeat domain and the last of the type 3 calcium-binding domains, which includes an RGD motif that mediates binding to αvβ3, αIIbβ3, and α5β1 integrins. The C-terminal domain includes two VVM motifs, including the 4N1 motif (shown) that mediate binding to CD47. Adapted from [120]. 25 1.8 CD47-Independent Effects of 4N1K Although the interaction between 4N1K and CD47 was initially found to be specific, a few reports have surfaced describing CD47-independent effects of 4N1K. The first report came from a study which found that 4N1K-induced platelet aggregation and protein phosphorylation could not be blocked by several different anti-CD47 antibodies [132]. A subsequent study only found a partial blockade of aggregation using one of these antibodies [133]. Using Jurkat and the CD47-decifient JinB8 cells, another study found that 4N1K- or 7N3-mediated increases in cell adhesion to TSP were independent of CD47 expression, as was the blockade of 7N3-mediated adhesion following incubation with pertussis toxin [117]. 1.9 CD47 Signaling A mechanism for signaling downstream of CD47 was first elucidated when it was shown that pertussis toxin, a specific inhibitor of Gαi, could block 4N1K-mediated cell spreading, thereby providing evidence that G proteins were involved in CD47 signaling [150, 151]. Co-immunoprecipitation studies provided additional support for this by showing that pertussis toxin could block associations between G proteins, CD47, and integrins [152], and other functional assays found G proteins to be important in 4N1K-mediated cell migration [153] and apoptosis [134]. Since G protein activation is usually as a result of heptaspanning protein stimulation, it has been proposed that the interaction of integrins and CD47 forms an ad hoc seven transmembrane receptor, leading to activation of G proteins [101, 152]. However, it should be noted that not all CD47-26 mediated functions are dependent on G-protein signaling as some could not be blocked by pertussis toxin [112, 154]. Ligation of CD47 by 4N1K or TSP leads to rapid tyrosine phosphorylation events, including focal adhesion kinase (FAK), SYK, and Lyn [129, 132, 155], and protein kinase C (PKC)-inhibitors were found to block cell spreading [150] and inhibit apoptosis [134, 135]. Although CD47 ligation by 7N3 was found to stimulate ERK1/2 phosphorylation in T leukemic cells, co-ligation of CD3 induced a much greater response, indicating that T-cell receptor activation synergizes with CD47 ligands to induce intracellular signaling [156]. However, others found Gαi-dependent inhibition of ERK phosphorylation and activity upon stimulation with 4N1K, the down-regulation of which was required for efficient cell migration [153]. Moreover, in agreement with CD47 signaling being Gαi-dependent, cell incubation with 4N1K was shown to decrease cyclic AMP (cAMP) levels in several models [152, 153], whereas PKA inhibitors were shown to rescue the apoptotic effects of anti-CD47 antibodies following inhibition by pertussis toxin [134]. One study using sickle cell reticulocytes also showed that CD47 ligation by 4N1K could induce protein kinase A (PKA)-dependent α4-integrin phosphorylation, which is important for regulation of cell migration [96, 131]. However, this study also found that 4N1K-mediated α4-integrin phosphorylation was independent of increases in cAMP; the reason for this has yet to be determined. 27 1.10 Interactions Between Integrins and CD47 As mentioned above, CD47 was initially known as “integrin-associated protein” and interaction between these two receptors was clear from the beginning. Co-immunoprecipitation studies have confirmed physical association between CD47 and β3- [105, 152, 155, 157] and β1-integrins [117, 158], and β2-integrins were recently found to interact with CD47 using fluorescence lifetime imaging microscopy [159]. One of the first roles discovered for CD47 was the regulation of intracellular Ca2+ following adhesion to fibronectin, which was found to be blocked by anti-CD47 antibodies [160]. In addition, binding of vitronectin-coated beads was found to be increased in CD47+/+ cells as compared to CD47-/- cells, which was inhibited following cell incubation with an anti-CD47 antibody [161, 162]. Another study subjected ICAM-1-adhered CD47+/+ or CD47-/- cells to increasing flow rates and found that, although more CD47+/+ cells adhered to this substrate under static conditions, the same percentage of each cell-type detached when cells were under shear flow, indicating that the presence of CD47 increases cell adhesion (integrin affinity) but does not affect the strengthening of β2-integrin adhesion (integrin avidity) [159]. Since TSP has been shown to bind to both CD47 and integrins, it has been proposed that this ligand mediates the interaction between the two receptors [116, 117, 122]. In many studies, soluble TSP, 4N1K, and 7N3 have been shown to decrease cell adhesion to various integrin ligands under static conditions [129, 150, 153, 163-165]. In contrast, 4N1K seems to have the opposite effect on cell adhesion under shear flow [129, 131]. It should be noted that some studies have found the opposite phenotypes stated above, i.e., 28 a decrease in adhesion under shear flow and an increase in adhesion in static assays [109, 117]. No explanation for these inconsistencies has been offered, although these could be due to inherent difference between the various cell lines used. LIBS antibodies have also been used to elucidate the consequences of CD47 ligation on integrin activation. Using the αIIbβ3-activation dependent antibody, PAC-1, Chung et al. determined that cell incubation with 4N1K, but not its control peptide 4NGG, led to significant increases in antibody binding, even beyond that seen using known integrin activators, such as epinephrine and thrombin, indicating that 4N1K was a potent activator of αIIbβ3-integrins [133, 155]. In a similar assay but using an anti-fibronectin antibody, it was determined that incubation with 4N1K increased binding of soluble fibronectin to sickle red blood cells [131]. 1.11 Hypothesis and Aims Relapse of childhood T-ALL, which occurs due to the expansion of a chemoresistant form of the disease, has been shown to be mediated by leukemic cells adhering to components of the bone marrow through adhesion receptors called integrins. In turn, CD47 has been shown to regulate the state of integrin adhesiveness through the extracellular protein thrombospondin, which binds to both receptors. A C-terminal domain motif of thrombospondin termed 4N1 and its peptide derivative, 4N1K, were shown to bind specifically to the extracellular IgV domain of CD47 to mediate its effects. Therefore, I hypothesize that adhesion of leukemic cells can be decreased by using 4N1K 29 to regulate integrin activation through CD47, thereby preventing cell adhesion mediated drug resistance. Aim 1: To investigate the consequences of CD47 ligation by 4N1K on β1-integrin activation. Aim 2: To investigate the role of CD47 ligation by 4N1K on integrin-mediated Jurkat cell adhesion. 30 Chapter 2: Materials and Methods Antibodies and Reagents TS2/16 (total β1) and BRIC-126 (CD47) were from Santa Cruz Biotechnology; 9EG7 and HUTS-21 (Activated β1) were from BD Pharmingen; 9F10 (α4-integrin), NKI-SAM-1 (α5-integrin), and FITC goat anti-rat were from Biolegend. Goat anti-mouse Dylight 488 was purchased from Thermo Scientific. All other reagents were purchased from Sigma-Aldrich unless otherwise specified. Cells Jurkat T-leukemic cells were purchased from American Type Culture Collection (ATCC); its α4-integrin-deficient derivative has been previously described [166]; the CD47-deficient Jurkat derivative, JinB8, was graciously provided by Dr. Roberts and previously characterized [167]. The Jurkat 6A and its β1-integrin-deficient A1 cell lines were graciously provided by Dr. Shimizu and previously characterized [168]. These cell lines were maintained in 5% CO2 at 37°C in complete RPMI (RPMI 1640 supplemented with 10% fetal bovine serum, non-essential amino acids (Invitrogen), and penicillin/streptomycin (Gibco)). All reagents were from Sigma-Aldrich unless otherwise specified. 31 Plasmids and Transfections The CD47isoform 4 plasmid was obtained from Dr. Ohdan [169]. The CD47isoform 2 was extracted from isoform 4 by polymerase chain reaction (PCR) using the following primers: Fwd: 5’-TGGAAGCTTATGTGGCCCCTGGTAGC-3’ Rev: 5’-TGCGGATCCTCAGTTATTCCTAGGAGGTTGTATAGTC-3’. The PCR product containing CD47isoform2 was subsequently cloned into pcDNA3.1(+) using the cloning sites included in each primer. JinB8 cells were transfected with this plasmid using Amaxa nucleofection (Lonza) and cells were serially sorted for expression using the FacsAria fluorescence-activated cell sorter (BD Biosciences), followed serial dilution to obtain clonal populations of CD47-expressing cell. CRISPR Constructs and CD47 Knock-down The following oligonucleotides were used to make the CD47-directed guide sequences: guide 1 Fwd primer: 5’-CACCGCAGCAACAGCGCCGCTACCA -3’ guide 1 Rev primer: 5’-AAACTGGTAGCGGCGCTGTTGCTGC-3’ guide 2 Fwd primer: 5’-CACCGCTGGTAGCGGCGCTGTTGCT-3’ guide 2 Rev primer: 5’-AAACAGCAACAGCGCCGCTACCAGC-3’ Guide sequences were inserted into the CRISPR plasmid pX330 (purchased from Addgene, plasmid 42230) according to manufacturer’s instructions. Transfection into Jurkat cells was performed using Amaxa nucleofection (Lonza) followed by one round of cell sorting using the FacsAria fluorescence-activated cell sorter (BD Biosciences) and serial dilution to obtain clonal populations of CD47-deficient Jurkat cells. 32 Sequencing of CRISPR-targeted CD47 Region in CRISPR CD47-knock-down Cells and Alignments Genomic DNA was extracted from Jurkat cells and all CRISPR knock-down cell lines using the Biobasic EZ-10 Spin column Genomic DNA Miniprep Kit. CD47 exon1 was amplified from the extracted genomic DNA by polymerase chain reaction (PCR) using the following primers: Fwd primer: 5’ GAC AGG AAC GGG TGC AAT GA 3’ Rev primer: 5’ TAA TTT TTG CGC GAG GTG CG 3’. PCR products were sequenced using the one or both of the following primers: Fwd primer: 5’-GAC AGG ACG TGA CCT GGA AG Rev primer: 5’ TAA TTT TTG CGC GAG GTG CG 3’. The exon 1 sequence from each CRISPR clone sequencing result was inserted in place of exon 1 nucleotides from a CD47iso2 cDNA sequence and the open reading frame for CD47 was translated. Each translation was then aligned with the protein product from a CD47iso2 cDNA. All analyses were performed using CLC sequence viewer6. Recombinant Proteins Full length fibronectin was purified from human plasma by affinity chromatography through gelatin-sepharose (GE). The thrombospondin C-terminal peptide 4N1K (KRFYVVMWKK) and its negative-control peptide 4NGG (KRFYGGMWKK) were purchased from GL Biochem (China). For all assays, 4N1K was solubilized in blank RPMI (NEAA, Pen/strep) at 525μM and incubated overnight at RT in the dark unless otherwise noted. 33 Flow Cytometry Cells were harvested, washed in Dulbecco’s Phosphate Buffered Saline, pH7.4 (PBS), resuspended at 4x106 cells/ml in 500ng primary antibodies diluted in 1%BSA/PBS, and incubated at 4°C for 20min. Cells were then washed in cold PBS, pH7.4 and resuspended in 200ng secondary antibodies diluted in 1%BSA/PBS and incubated at 4°C for 20min. After a final wash in cold PBS, pH7.4, data was acquired using a BD FACSCanto flow cytometer using an excitation wavelength of 488nm and a 530/30 emission filter while fluorescence-assisted cell sorting was conducted on a FACSAria (BD Biosciences). Post-acquisition analysis was conducted using FlowJo software (TreeStar). Integrin Activation Assays 6x105 cells per sample were harvested and washed in room temperature (RT) PBS, pH7.4 and incubated with 500ng HUTS-21 or 9EG7 antibody diluted in 1%BSA/serum-free RPMI (RPMI 1640, NEAA, Pen/strep) without or with treatment (50μM 4NGG, 50μM 4N1K, 1mM MnCl2) for 30min at 37°C. Cells were then washed in cold PBS and incubated with 200ng goat anti-rat (9EG7) or goat anti-mouse (HUTS-21) secondary antibodies diluted in 1%BSA/PBS for 20min at 4°C. After a final wash in cold PBS, cells were analyzed by flow cytometry using the BD FACSCanto flow cytometer at an excitation wavelength of 488nm and a 530/30 emission filter. The activation index was calculated as follows: Condition Geometric mean/ untreated Geometric mean. Microscope Imaging of 4N1K Aggregates 4N1K was dissolved in H2O, 1%BSA/PBS, bRPMI, or cRPMI to 0.525mM and incubated overnight at RT. 10μl of each 4N1K sample or dilution medium alone was 34 added to 100μl of 1% BSA/PBS and imaged with an Olympus IX81 microscope using differential interference contrast and a 20x LucPlanFL objective, CoolSnap HQ2 camera (Photometrics), H-117 linear-encoded stage (Prior Scientific), and controlled with MetaMorph (Molecular Devices). Protein Quantification Assay for Adsorbed Peptides Wells of a 96-well dish were coated with increasing concentrations of 4N1K and 4NGG at RT for 1hr and rinsed 2x with RT PBS to remove unbound peptides. A CBQCA protein quantification assay (Invitrogen) was then performed according to manufacturer’s instructions to quantify the amount of peptide adsorbed in each well. Peptide-Antibody Interaction Assays Wells of a 96-well dish were coated with 50μM 4NGG, 50μM 4N1K, 1%BSA/PBS, or 50μM poly-L-lysine (PLL) and incubated at RT for 1hr. Wells were then washed 3x in RT PBS, and 50μl of primary antibodies diluted to 2.5ng/μl in PBS were added to each well and incubated at 4°C for 20min. After 2x washes in cold PBS, 50μl of secondary antibodies diluted to 4ng/μl in PBS were added and incubated for 20min at 4°C. After a final wash in cold PBS, total fluorescence of each well was assessed using an Enspire spectrophotometer at 485nm excitation and 515nm emission filter. Cell Adhesion Assays Wells of a 96-well dish were coated with 50μl of 20μg/ml of Fn, 50μM 4NGG, 50μM 4N1K, or 50μM PLL and incubated at RT for 1hr. Cells were labeled with Celltracker™ Green CMFDA (Invitrogen) as follows: cells were washed 1x in a total of 50ml RT PBS 35 and resuspended at 1x106 cells/100μl in 2.5μM, pre-warmed Celltracker™ Green CMFDA solution (freshly dissolved in serum-free RPMI from a 10mM stock). Cells were incubated at 37°C for 30min before washing 1x in 10x volume of RT PBS and resuspending in pre-warmed serum-free RPMI at 1x106 cells/ml. After incubating at 37°C for 30min, cells were washed again in a total of 50ml RT PBS and resuspended in pre-warmed RPMI with 10% FBS at a concentration of 2x106 cells/ml. Substrate-coated wells were washed 3x in PBS, pH7.4 and 2x105 Celltracker™ Green CMFDA-labeled cells were added to each well and incubated at 37°C for 20mins for adhesion to peptides and for 45min for adhesion to fibronectin. Cell counts were assessed using an Enspire spectrophotometer at 485nm excitation and 515nm emission filter and background fluorescence of incubation medium was subtracted from each reading. % adhesion was calculated as follows: Fluorescence reading after washes/initial fluorescence reading*100 Aggregation Assays Cells were washed and resuspended in PBS at 0.25x106 cells/ml. 95μl of cells were then transferred to each well of a 96-well dish and cells were incubated untreated or treated with 50μM 4N1K for 90 min at 37°C, 5% CO2. Images were acquired following 90min incubation with an Olympus IX81 microscope using a 4x UPLFLNPHP objective and the 1.5x multiplier engaged, CoolSnap HQ2 camera (Photometrics), H-117 linear-encoded stage (Prior Scientific), and controlled with MetaMorph (Molecular Devices). Immunofluorescence Imaging Jurkat cells were washed and resuspended in PBS at 8x105 cells/ml, followed by incubation with 50μM 4NGG or 4N1K and anti-CD47 antibody (BRIC-126, diluted 36 1:200 in 1% BSA/PBS) for 30min at RT. After washing in PBS, cells were fixed in 3.7% formaldehyde for 20min at RT, washed in PBS, and stained with anti-mouse Alexafluor555 secondary antibody for 20min at RT. Cells were mounted on microscope slides with Prolong® gold antifade reagent with DAPI (Invitrogen) and visualized using an Olympus IX81 microscope using a 60x oil objective, CoolSnap HQ2 camera (Photometrics), H-117 linear-encoded stage (Prior Scientific), and controlled with MetaMorph (Molecular Devices). Statistical analysis The Student’s unpaired t test was used for all statistical analyses in this report 37 Chapter 3: Results 3.1 Establishment of an Assay to Report β1-Integrin Activation (The “Activation Assay”) I first established an assay to determine the state of activation of β1-integrins by using antibodies characterized as β1-integrin activation-dependent antibodies, HUTS-21 [79] and 9EG7 [80]. As proof of concept for this assay, a divalent cation that induces integrin activation, MnCl2 [79, 80], was used to verify that binding of these antibodies would increase when β1-integrins were activated. Indeed, incubation of Jurkat cells with MnCl2 resulted in an increase in 9EG7 antibody binding (Fig. 6). Therefore, this assay was appropriate to study changes in β1-integrin activation. Figure 6: Integrin Activation Assay: Proof of Concept. Cells were pre-incubated without or with 1mM MnCl2 and 9EG7 for 30min at 37°C. Cells were then washed and secondary antibody was added and incubated further for 20min at 4°C before a final wash and flow cytometric acquisition as stated in materials and methods. 38 3.2 Conditions Required for 4N1K Activity To investigate the role of CD47 ligation on integrin-mediated functions, I made use of a peptide derived from the C-terminal cell binding domain of TSP1 termed 4N1K, that was found to specifically mediate the binding of TSP to CD47 [126]. I initially used the above integrin activation assay to assess the activity of 4N1K. Although most lyophilized proteins can be solubilized in water or PBS and used as such, I found that 4N1K was only efficiently and consistently active when solubilized in serum-free RPMI media (bRPMI) (Fig.7). Along the same line, microscopic observations revealed the formation of aggregates when 4N1K was dissolved in serum-free RPMI (bRPMI, Fig. 8A), and are presumed to be important for 4N1K activity since dissolution of 4N1K in reagents that do not induce 4N1K aggregation (1%BSA/PBS, H2O, or cRPMI) yielded an inactive form of 4N1K (Fig 7, 8A, and not shown). 4NGG dissolved in bRPMI did not result in any observable aggregation of the peptide (Fig. 8B). In addition, I found that a 30min incubation period at 37°C using 50μM 4N1K yielded the most efficient increase in 9EG7 antibody binding (Fig. 9), conditions consistent with the reported use of 4N1K [109, 132, 165]. Therefore, these conditions have been used throughout this manuscript unless otherwise noted. 39 Figure 7: 4N1K Must be Dissolved in bRPMI to be Active. 4N1K was dissolved to 50μM in DMSO, H2O, cRPMI, or bRPMI and incubated with 9EG7 for 30min at 37°C before washing in PBS and incubating with secondary antibody for another 20min at 4°C. Secondary antibody only without treatment was used as negative control. Cells were analyzed using flow cytometry following a final wash in PBS. 40 A. B. Figure 8: Solubilization of 4N1K in Serum-free RPMI Results in its Aggregation. A. 4N1K was dissolved in 1%BSA/PBS, RPMI without serum, and incubated overnight at room temperature. The samples were diluted 1/10 in 1%BSA/PBS and imaged using DIC microscopy and a 20x LucPlanFL objective. B. 4N1K and 4NGG were dissolved in bRPMI and incubated overnight at room temperature. The peptides were added to Jurkat cells in suspension and imaged as in A, where the focus was on the supernatant to visualize 4N1K better. 41 Figure 9: Titration of 4N1K in Activation Assays. Jurkat cells were incubated with 4N1K at various concentrations for up to 2hrs and binding of 9EG7 antibody was used to report β1-integrin activation. Data was normalized to 4NGG as an activation index, which was calculated as stated in materials and methods. 42 3.3 Consequences of CD47 Ligation by 4N1K on Integrin Activation I began by characterizing the cell lines I was planning to use by assembling an expression profile of selected cell-surface receptors (Fig. 10). Jurkat-derivative cell lines, including the CD47-reconstituted JinB8 cell line, expressed CD47, α5-integrins, and β1- integrins to similar levels as the Jurkat 6A parental. A1 is a Jurkat derivative lacking β1-integrins [168], which accounts for the minimal α5-integrin expression due to loss of the paired β1-integrin subunit. I next used the integrin activation assay to assess changes in integrin activation mediated by 4N1K. As shown in Fig. 11, cell incubation with 4N1K significantly increased 9EG7 antibody binding compared to untreated samples. Surprisingly, I found that incubation with 4N1K still resulted in a significant increase in antibody binding when this assay was performed with the β1-integrin-deficient A1 cells (Fig. 11). To confirm these unexpected results, I repeated the assay using another known reporter of β1-integrin activation, HUTS-21 [79] (Fig. 12A). As expected, incubation with MnCl2 increased β1-integrin activation in Jurkat 6A but not A1 cells, as indicated by an increase in antibody binding in the former. Similar to 9EG7, cell incubation with 4N1K increased HUTS-21 antibody binding, whereas incubation with a similar concentration of 4NGG did not. Strikingly, the increase in 4N1K-mediated integrin activation was beyond that seen with MnCl2, suggesting that 4N1K is a potent activator of β1-integrin. To further evaluate these unexpected results, I repeated the assay using two important control antibodies: TS2/16, an activation-independent β1-integrin antibody (Fig. 12B) and non-specific IgG antibodies (Fig. 12C). A condition using only goat anti-mouse (GAM) 43 secondary antibodies (Fig. 12D) was also included to rule out primary antibody-specific effects. As expected, these antibodies had minimal binding in the A1 cell line even when pre-incubated with MnCl2, whereas TS2/16, but not the IgG control, bound significantly in untreated Jurkat cells only. TS2/16 binding was not altered by MnCl2 since this antibody is not dependent on the activation state of β1-integrins. Surprisingly, pre-incubation with 4N1K caused a significant increase in binding of both HUTS-21 and TS2/16 antibodies, as well as the GAM-antibody only condition, in both Jurkat and A1 cell lines (Fig 12). Finally, a peptide titration experiment indicated that this phenomenon occurred in a manner that was dependent on 4N1K concentration (Fig. 13). These results suggested that 4N1K was a peptide that mediated binding of a variety of IgG antibodies to cell surfaces, and therefore, no conclusions could be made regarding the effects of 4N1K on the state of β1-integrin activation from these assays. 44 Figure 10: Selected Receptor Expression on Jurkat Derivative Cell Lines. Cell lines were incubated with appropriate antibodies for 15min at 4°C and washed with PBS. Secondary antibodies were added and further incubated for 15min at 4°C. Receptor expression was assessed by flow cytometry following a final wash with PBS. Jurkat 6A: Jurkat parental cell line; A1: β1-integrin-deficient Jurkat derivative; JinB8: CD47-deficient Jurkat derivative; JinB8-CD47: CD47-reconstituted JinB8 cells. Figure 11: 9EG7 Antibody Binding is Increased in 4N1K-treated Jurkat Cells. 9EG7 antibody was incubated with or without 50μM 4N1K for 30min at 37°C, followed by a wash in PBS and incubation in secondary antibody for 20min at 4°C. Antibody binding was assessed by flow cytometry after a final wash in PBS as indicated in materials and method. 45 Figure 12: 4N1K Increases Binding of a Variety of Antibodies to Cell Surfaces. A-C. Antibodies were incubated without or with treatment (1mM MnCl2, 50μM 4N1K, or 50μM 4NGG) for 30min at 37°C, followed by a wash in PBS and incubation in secondary antibody for 20min at 4°C. Antibody binding was assessed by flow cytometry after a final wash in PBS as indicated in materials and method. D. Same as in A but only a goat anti-mouse (GAM) secondary antibody was used. 46 Figure 13: Integrin Activation Assay Using Various 4N1K and 4NGG Concentrations. An integrin activation assay was performed as indicated in materials and methods, but using only GAM secondary antibody and peptides were added at titrating concentrations as indicated. 47 3.4 Characterization of the Non-specific Adherent Nature of 4N1K To determine if 4N1K can bind directly to non-specific antibodies, I performed a cell-free, ELISA-type assay where I coated wells of a 96-well dish with BSA, 4NGG, 4N1K, or poly-L-lysine (PLL), a polypeptide used here as a positive control for protein binding, and determined to what extent various antibodies bound immobilized 4N1K. As shown in Fig. 14, both HUTS-21 and an IgG isotype control bound significantly to substrate-coated 4N1K and PLL but not to BSA or 4NGG. Adequate and similar peptide coating was confirmed by performing a CBQCA protein quantification assay for peptide adsorption (Fig. 15). In addition, peptide titration experiments using only goat anti-mouse secondary antibodies confirmed the non-specific interaction of IgG antibodies with 4N1K and showed that this occurred in a 4N1K concentration-dependent manner (Fig. 16). The secondary antibody only experiment was also repeated using a variety of different antibodies (goat anti-rat FITC, donkey anti-goat DyLight800, and goat anti-rabbit DyLight633) to exclude any isotype or fluorochrome contributions to this phenomenon and similar results were obtained (not shown). Together, these results support the interpretation that 4N1K is extremely sticky and adheres non-specifically to a variety of antibodies. I next determined if the non-specific, adherent nature of 4N1K is also applicable to cells in culture. As such, Jurkat, JinB8, and JinB8-CD47 cells were plated on 4NGG-, 4N1K-, or PLL-coated wells. Surprisingly, all cell lines used, including the CD47-deficient JinB8 cells, adhered to 4N1K in a concentration-dependent manner, significantly more than to 4NGG (Fig. 17). As expected, all cell lines adhered to PLL the most. These results 48 indicated that 4N1K may have properties that allow it to interact non-specifically with cell surfaces in addition to various antibodies. Figure 14: A Variety of Antibodies Bind Substrate-Immobilized 4N1K. Wells of a 96-well dish were coated with 1%BSA/PBS, 50μM 4NGG, 50μM 4N1K, or 50μM PLL. Antibodies were then added to each well and incubated for 20min at 4°C. After a final wash to remove non-adhered antibodies, total fluorescence of each well was determined using a spectrophotometer as indicated in materials and methods. 49 Figure 15: 4NGG and 4N1K Peptides Adequately Coat Wells. Wells of a 96-well dish were left uncoated or coated with various concentrations of 4NGG or 4N1K for 1hr at RT and then washed with PBS. The adsorbed proteins were then quantified using a CBQCA protein quantification assay as indicated in materials and Methods. Data is representative of three independently conducted experiments, where means ± standard error are shown. Figure 16: Antibody Binding to Immobilized 4N1K Occurs in a Peptide Concentration-Dependent Manner. Peptides were immobilized on plastic wells at various concentrations and dylight488-conjugated goat-anti mouse secondary antibodies were added and incubated for 20mins at 4°C. Wells were washed in cold PBS and fluorescence was assessed using a spectrofluorometer as indicated in materials and methods. 50 A. B. Figure 17: Cell-binding to 4N1K is Independent of CD47 Expression. A. Wells of a 96-well plate were coated with the indicated peptides at 50μM for 1hr at RT. Following 3x washes in RT PBS, Celltracker™ Green CMFDA-labeled cells were added and incubated for 20min at 37°C. Total fluorescence of each well was determined using a spectrofluorometer before and after 3x gentle washes with PBS; background fluorescence was subtracted from each reading. Values are the mean ± SEM of triplicate wells and data shown are representative results from at least 3 separate experiments. *, p ≤ 0.05; **, p≤0.002. B. Same as in A, but titrating concentrations of peptides were coated. Data are means of duplicate wells. 51 3.5 Consequences of CD47 ligation by 4N1K on Integrin-mediated Cell Adhesion Cell adhesion assays performed by others indicated that 4N1K could decrease cell adhesion to immobilized integrin ligands [125, 153, 165]. To determine if these results could be replicated in our model, I performed an adhesion assay where cells pre-incubated with or without 4N1K treatment were allowed to adhere to fibronectin (Fn)-coated wells, an extracellular matrix substrate that is a ligand for several integrins (Fig. 18 and 19). As expected, MnCl2 treatment significantly increased adhesion of both Jurkat and the β1-integrin-deficient A1 cells; the greater increase in Jurkat cell adhesion was not surprising since A1 cells do not express β1-integrin, which are important receptors in Fn-mediated adhesion (Fig. 18A). 4N1K treatment, but not 4NGG, resulted in a complete loss of adhesion in both Jurkat and A1 cells (Fig. 18A). Surprisingly, 4N1K also blocked adhesion to fibronectin in the CD47-deficient JinB8 cell line, suggesting that the inhibitory effects of 4N1K on cell adhesion to integrin ligands occurs independently of CD47 (Fig. 18B). I also performed peptide titration experiments to determine if 4N1K might have CD47-dependent effects at lower concentrations (Fig. 19). Surprisingly, 4N1K induced adhesion to Fn at concentrations of 12.5μM to 25μM, but again, this occurred in a CD47-independent manner. 52 A. B. Figure 18: CD47 is not Required for 4N1K-mediated Decrease in Adhesion to Integrin Ligands. A. Fluorescently-labelled Jurkat (6A) and β1-integrin-deficient A1 cells were plated on Fibronectin-coated wells and allowed to adhere for 45min at 37°C following a 20min pre-incubation period without or with treatment (50μM 4NGG, 50μM 4N1K, or 1mM MnCl2). Total fluorescence of each well was determined using a spectrofluorometer before and after 3x gentle washes with PBS; background fluorescence was subtracted from each reading. Values are the mean ± SEM of triplate wells and data shown are representative results from at least 3 separate experiments. *, p ≤ 0.001. B. Same as in A, but using Jurkat, CD47-deficient Jurkat cells (JinB8), and JinB8 cells reconstituted with CD47 (JinB8-CD47). Values are the mean ± SEM of triplicate wells and data shown are representative results from at least 3 separate experiments. *, p ≤ 0.05. 53 Figure 19: 4N1K has Differential Effects on Cell Adhesion to Fibronectin Depending on its Concentration. 20μg/ml of fibronectin was immobilized on plastic for 1hr at RT before washing with PBS. JinB8 and JinB8-CD47 cells were labeled with CellTracker Green™ as indicated in materials and methods and incubated with titrating concentrations of each peptide at 37°C for 20min. Cells were then added to each well and allowed to adhere for 30min at 37°C. Data are representative results from three separate experiments, where means ± SD are shown 54 3.6 Consequences of Cell Incubation with 4N1K on Jurkat Cell Aggregation Previous studies have shown that 4N1K could induce aggregation of platelets and fibroblasts, but the dependence for CD47 in this process has been put in question [111, 132] and the effects of 4N1K on Jurkat cell aggregation have not been studied. In addition, my results so far indicate that 4N1K mediates non-specific binding of antibodies and Jurkat cells. Therefore, I investigated the consequences of cell incubation with 4N1K on Jurkat cell aggregation and cell-surface immunofluorescence. Although 4NGG had minimal effects on cell aggregation, 4N1K caused a significant increase in aggregation not only in Jurkat cells, but also in the CD47-deficient JinB8 cell line, suggesting that this mechanism is independent of CD47 (Fig. 20). 55 Figure 20: 4N1K Induces Jurkat Cell Aggregation in a CD47-independent Manner. Cells were harvested, washed in PBS, and resuspended in 1x RT PBS. Cells were then transferred to each well of a 96-well dish and 4NGG or 4N1K was added to a final concentration of 50μM. Images were acquired using differential interference contrast imaging as indicated in methods and material. Images shown are after 90min incubation in treatment. 56 3.7 Generation of CD47-deficient Jurkat CRISPR Knock-outs. A surprising observation in my assays was that adhesion of untreated JinB8 cells was consistently higher than for untreated Jurkat cells, indicating that JinB8 cells are stickier than their parental Jurkat cells (Fig. 18B). In addition, the receptor profile assembled indicated that α5-integrin expression in JinB8 cells is similar but still higher than in Jurkat cells (Fig. 10). These differences make direct comparisons between these two cell lines difficult when assessing integrin functions such as activation and adhesion. Furthermore, since reconstituting JinB8 cells with CD47 increased cell adhesion to fibronectin (Fig 18B), it would be worthwhile to generate a CD47-null Jurkat cell line that more closely replicates Jurkat cell integrin receptor expression and cell adhesion phenotypes. Therefore, I used the CRISPR system (Clustered Regularly Interspaced Short Palindromic Repeats) to knock-out CD47 in the Jurkat cell line to obtain a better system to study the role of CD47 on integrin functions in T-leukemic cells. CRISPR plasmids were developed based on a bacterial adaptive immune system, where the Cas9 protein cleaves double-stranded DNA (dsDNA) fragments matching sequences present downstream of the Cas9 gene locus, in a region called protospacer [170, 171]. Sequences from this region are usually derived from foreign DNA and function as the target sequences for Cas9-mediated cleavage. Recognition of target sequences is achieved by transcription of the protospacer sequences into RNA molecules, which hybridizes with the target sequence and also binds the Cas9 protein. A final requirement for target dsDNA cleavage by Cas9 is the presence of a “protospacer adjacent motif” (PAM; 5’ NGG 3’, where N is any nucleotide) on the target sequence, where dsDNA 57 cleavage occurs at least 3bp away from the PAM sequence [172]. Studies later found that user-defined sequences, called guide sequences, could be inserted in the protospacer region to allow the knock-down of specific genes of interest and led to the design of plasmids optimized for cell transfection and gene knock-outs [173, 174]. To generate CD47-deficient Jurkat T-cells, we cloned two separate guide sequences matching a 20 nucleotide portion of CD47’s first exon into a CRISPR-Cas9 expression plasmid (see materials and methods). Jurkat cells were transfected and sorted three days post-transfection, at which point I found that CD47 had been partially knocked-down (Fig. 21A). Total knock-down of CD47 was observed one week post-sorting, as assessed by flow cytometry (Fig. 21B). After serial dilution to obtain clonal populations of CD47-deficient Jurkat cells (Fig. 22), genomic DNA was isolated and CD47- locus specific sequencing was performed to characterize the mutation that led to the loss of CD47 expression. Figure 23 shows the sequence alignment of each CRISPR-generated CD47-null clone compared to the wild-type parental Jurkat cells; note that the C1 constructs were sequenced twice due to high background level in the sequencing results starting around the ATG start codon (Fig. 24). However, the background in this case may have been as a result of different mutations being introduced on each genomic copy of CD47 following cleavage by Cas9. In this case, the PCR product of each construct might favor one gene copy rather than the other and lead to high background levels due to the presence of two DNA sequences in the PCR product, although one sequence would stand out as was the case with the C1 constructs. It should also be noted that each sequencing result was 58 obtained using PCR products from two separate PCR reactions and that the increase in background in each clone started in the region where the mutation was expected, supporting the fact that the C1 constructs may be heterozygous for their mutations. All C2 constructs had low background throughout and were only sequenced once; these are presumed to be homozygous for the indicated mutations. Analysis of CD47 CRISPR clone sequencing protein products revealed that, whereas C1-5 #1 had an insertion yielding a 23 amino acid protein and C1-20 #1 had a deletion yielding a 2 amino acid protein, both had the same 10 nucleotide deletion yielding a 25 amino acid protein when sequenced a second time (Fig. 25). The C1-7 construct had the same deletion as C1-5#2 and C1-20#2 both times it was sequenced. All C2 constructs (except C2-11) had a one nucleotide insertion which led to a premature stop codon, yielding a 23 amino acid protein. C2-11 had a 9-nucleotide deletion, which resulted in the whole protein being translated because the deletion was in-frame with CD47’s stop codon. Since the deleted segment was in the extracellular, antibody binding domain of CD47, it is possible that flow cytometry assessment of this clone showed a CD47-deficient clone because the antibody recognition epitope was lost despite surface expression of a mutant CD47 receptor. Therefore, this clone was set aside and was no longer used. The CRISPR system can result in off-target DNA cleavage; therefore we used flow cytometry on the CD47 CRISPR clones to verify that expression of various integrins of interest was not affected. As seen in Figure 22, only C2-4 expressed all integrins assessed to a similar extent as the Jurkat parental cell line. The next closest, C2-9, had lower α4-59 integrin expression and slightly lower β1-integrin expression. All C1 constructs had significant differences in α4-, α6-, and β1-integrin expression. Therefore, the C2-4 construct will be used for assessment of changes in integrin functions in CD47-deficient Jurkat cells. Finally, the polyclonal CD47 CRISPR#1 (JC47p1) and CRISPR#2 (JC47p2) knock-down cell lines were serially sorted for expression of selected integrins so that the expression pattern of these receptors are similar to the Jurkat parental, allowing ease of comparison of integrin-related functions (Fig. 26). Finally, since no gene knock-out experiment is complete without a proper rescue of the gene in question, I plan to reconstitute the C2-4, JC47p1, and JC47p2 cell lines with CD47. Work with these cell lines is ongoing and incomplete at the time of writing this thesis. 60 Figure 21: CRISPR Knock-down of CD47. A. Jurkat cells were transfected with CD47 CRISPR#1 or CRISPR#2 plasmids and sorted 3 days after transfection. The shaded populations were sorted and expanded. B. CD47 expression of the expanded cells in A. 10 days after the first sort. The shaded areas are populations that were sorted and expanded. 61 Figure 22: CD47 Expression on Clonal CD47 CRISPR Constructs. Sorted CD47 CRISPR knock-out cells were serially diluted and expanded at 1cell/2-3wells of a 96-well plate to obtain clonal populations. C1 indicates clones obtained using CD47 CRISPR guide sequence#1, and C2 are those using CD47 CRISPR guide sequence #2. Expression of selected receptors was then assessed using flow cytometry as indicated in materials and methods. 62 63 Figure 23: Alignment of CD47 CRISPR Knock-down Clones with their Jurkat Parental. A PCR-amplified region surrounding the expected Cas9-mediated cleavage at the beginning of the first CD47 exon was sequenced and aligned with the Jurkat parental sequencing of the same region. C1 represents clones obtained with CD47 CRISPR guide sequence#1, and C2 are those using CD47 CRISPR guide sequence#2. Yellow lines indicate the ATG start codon in exon 1. CLC Sequencing Viewer6 was used for analysis. 64 Figure 24: Partial Sequencing Results of CRISPR Clone 1-5. A PCR-extracted region of CD47 genomic DNA in CD47 CRISPR clone 1-5 was sequenced as indicated in materials and methods to assess the mutation induced by Cas9. The ATG start codon is at nucleotide 222. 65 Figure 25: Jurkat CD47 CRISPR Predicted Amino Acid Sequences. Amino acid sequence of protein products from CRISPR clones. C1 indicates clones obtained using CD47 CRISPR guide sequence#1, and C2 are those using CD47 CRISPR guide sequence #2. The exon 1 sequence from each CRISPR clone was inserted in place of the exon 1 nucleotides from a CD47 cDNA sequence and the open reading frame for CD47 was translated. Numbers on the right of all but C2-11 indicate the number of amino acids in the protein product. Note that the complete protein sequence for C2-11 is not shown, but the entire CD47 protein is translated except for the deleted amino acids; only the first 41 amino acids are shown. CLC Sequence Viewer6 was used for analysis. Figure 26: Receptor Profile of Polyclonal CD47 CRISPR Knock-down Cell Lines. Polyclonal Jurkat CD47 CRISPR knock-down cell lines were serially sorted for integrin expresion to be similar to their Jurkat parental. Flow cytometry and analysis was performed as indicated in materials and methods. 66 3.8 Discussion and Conclusion T-cell acute lymphoblastic leukemia (T-ALL) is a disease whose treatment makes use of highly toxic drugs that result in a high degree of morbidity and potentially life-threatening side-effects in children afflicted by this disease [3, 4]. One reason for the administration of these drugs is an attempt at killing cancer cells that hide in the bone marrow and acquire cell-adhesion mediated drug resistance, CAMDR, via a mechanism that remains poorly understood but involves proteins that mediate cell adhesion, such as integrins [8, 15]. 4N1K is a peptide derived from the C-terminal of TSP that is said to be specific for CD47, a pentaspanning cell-surface protein that has been shown to regulate integrin functions [101]. Here, I explored the consequences of 4N1K treatment on cellular integrin functions. The integrin activation assay I developed allowed me to assess changes in β1-integrin affinity-state following cell stimulation with 4N1K. The initial results of this assay agreed with the conclusions of others who performed similar assays using a different activation-dependent antibody [133, 155], in that 4N1K was found to be a highly potent activator of integrins (Fig. 11 and 12). Chung et al. found that cell incubation with 4N1K increased the binding of the αIIbβ3 LIBS antibody, PAC-1, even beyond that seen using epinephrine- and thrombin-receptor stimulators, both known integrin activators [155]. Another group used the same antibody to support these findings and also found that, at least in platelets, 4N1K-mediated integrin activation did not require energy, which indicated that such a mechanism was passive and independent of cell signaling [133]. A third group came to the same conclusion about the effects of 4N1K on integrin activation 67 by using an anti-fibronectin antibody to assess the degree of activated, fibronectin-bound integrins [131]. However, doubts about the interpretation of my integrin activation assay was raised upon finding that 4N1K also induced 9EG7 and HUTS21 binding to the β1-deficient A1 cells (Fig. 11 and 12). In addition, I repeated the assay using two control primary antibodies, TS2/16, an activation-independent β1-integrin antibody, and a non-specific IgG antibody, as well as a secondary antibody-only control condition. Surprisingly, binding of all these antibodies increased significantly when cells were stimulated with 4N1K, regardless of whether cells expressed β1-integrins or not (Fig. 12). These assays indicated that 4N1K may have a particular affinity for proteins that contain Ig domains, such as antibodies. This hypothesis was supported in my cell-free assay where a number of different antibodies were able to bind substrate-immobilized 4N1K to a significant degree (Fig. 14 and 16). In addition, I showed that the adhesive nature of 4N1K also mediates binding to cell surfaces, presumably via transmembrane proteins that contain extracellular Ig domains, in a manner that is independent of CD47 expression since both Jurkat and JinB8 cells were able to bind to immobilized 4N1K, but not 4NGG (Fig. 17). The adhesion assays I performed showed that 4N1K at 50μM concentrations blocked cell adhesion to integrin substrates such as fibronectin (Fig. 18), in agreement with results obtained in assays performed by others using similar 4N1K concentrations [125, 153, 165]. However, my peptide titration experiments revealed that 4N1K in fact induced adhesion when used at lower concentrations (Fig. 19). This prompted me to re-evaluate the literature concerning 4N1K-mediated changes in cell adhesion to integrin ligands. I 68 found that most studies which used 4N1K at 50μM or higher had concluded that 4N1K blocked adhesion to various substrates [109, 125, 130, 153, 165]. However, one study using 4N1K or 4N1 (4N1K without the added lysine residues) at 5-20μM did find that this peptide induced cell adhesion at lower concentrations [117]. Similarly, aggregation assays I performed agreed with previous results [117, 132, 133] in that I found that 4N1K induced cell aggregation and extended these findings to include T-leukemic Jurkat cells (Fig. 20). However, one of these assays revealed the same pattern as the adhesion assays, i.e. that 25μM 4N1 induced a higher degree of aggregation than 100μM 4N1 [132], indicating that the opposing effects of different 4N1K concentrations is not restricted to cell adhesion, but is also applicable to cell-cell interactions. As such, in light of the results obtained in my activation and cell-free assays, we can no longer ascribe the phenotypes of the aforementioned adhesion and aggregation assays on the specific binding of 4N1K to CD47, but must conclude that these effects are as a result of the non-specific, hyper-adhesive nature of 4N1K. I showed that 4N1K has the propensity to bind non-specifically to a variety of antibodies, which contain Ig motifs (Fig. 14). Many cell-surface proteins also have Ig domains with which 4N1K could interact and this may provide a mechanism for the results observed in the above functional assays (Fig. 27). Specifically, 4N1K may coat cells during incubation by binding to these cell-surface, Ig domain-containing proteins and hinder interaction of cell surface receptors with potential ligands. As such, 4N1K-mediated inhibition of cell adhesion to integrin ligands, as seen in Figure 18, may be as a result of 4N1K restricting access of cell surface integrins to fibronectin. Furthermore, I had 69 observed that 4N1K aggregates accumulate over time in solution, indicating that homotypic interaction between 4N1K molecules is possible (Fig. 8). Therefore, the adhesive nature of 4N1K may also allow 4N1K-coated cells to adhere to one another via 4N1K-4N1K interactions, resulting in cell aggregation. This would explain why 4N1K-mediated aggregation occurs in CD47-deficient cell lines (see next). The use of a CD47-deficient cell line, JinB8, in my assays revealed that both 4N1K-mediated adhesion and cell aggregation occurs independently of CD47 expression, indicating that 4N1K is not specific for CD47 (Fig. 18 and 20). This finding is supported by previous studies that found CD47-independent effects of 4N1K on cell adhesion and cell aggregation. In one study, 4N1K increased JinB8 cell adhesion to TSP as efficiently as it did Jurkat cells [117]. Another study found that 4N1K-mediated platelet aggregation was at best only partially dependent on CD47 and that this process did not require intracellular signaling since aggregation was not completely blocked in energy deprived cells or in cells fixed with formaldehyde [132, 133]. A recent study analyzing the crystal structure of TSP’s cell-binding domain (CBD) revealed that the 4N1 and 7N3 VVM motifs were buried within the globular domain of TSP, making them inaccessible for protein interactions [175]. Another group confirmed this finding but suggested that a break in ionic interactions between two residues in the TSP CBD, brought about by the proximity of TSP to CD47, would allow binding of CD47 to the VVM motifs by insertion of the IgV domain of CD47 into the open cleft of the TSP CBD [176]. However, this hypothesis remains to be validated in an experimental setting. Finally, a recent study using plasmon resonance successfully visualized 70 interactions between CD47 and SIRPα IgV domains, but failed to find any interaction between CD47 and the signature domain of TSP [177]. In conclusion, the TSP C-terminal-derived peptide, 4N1K, has been extensively used to study the consequences of CD47 ligation by TSP, despite the fact that contradicting evidence had been put forward concerning the specificity of 4N1K for CD47. I now show, using carefully controlled experiments, that 4N1K is a hyper-adhesive peptide that binds non-specifically to Ig-containing proteins, such as antibodies. Since cells express a variety of cell-surface proteins containing this domain, it is plausible that cell incubation with 4N1K leads to cells being coated by this adhesive peptide and may interfere with normal cell activity (Fig. 27). Consequently, many of the phenotypes ascribed to CD47 ligation by 4N1K can be explained by the non-specific, hyper-adhesive nature of 4N1K. For these reasons, I propose that 4N1K not be used to study the consequences of CD47 ligation by TSP. My adhesion studies using JinB8 and the reconstituted JinB8-CD47 showed that CD47 has a stimulatory effect on integrin-mediated adhesion, although Jurkat cells could not be used as a proper parental cell line due to the inherent differences between Jurkat and JinB8 cells (Fig. 18). Therefore, the CRISPR system was used to generate CD47 knock-outs in Jurkat cells and CD47 reconstitution of these cell lines is ongoing (Fig. 21 and 26). These CD47-null Jurkat cell lines will provide a superior tool to characterize the role of CD47 in integrin-mediated adhesion and CAMDR and will be the basis of future projects for the Lim laboratory. 71 Figure 27: Proposed Mechanism for Non-specific 4N1K Effects on Cell Adhesion, Aggregation, and Antibody-based Assays. 4N1K associates with Ig-containing proteins on the cell surface, which may interefere with receptors mediating adhesion to ligands. The propensity for homotypic interactions between cell surface-bound 4N1K would result in cell aggregation . Antibody-based assays would not report antibody binding to intended targets, but rather antibody binding to 4N1K. 72 References 1. Gaynon, P.S., et al., Survival after relapse in childhood acute lymphoblastic leukemia: impact of site and time to first relapse--the Children's Cancer Group Experience. Cancer, 1998. 82(7): p. 1387-95. 2. Harrison, C.J., Acute lymphoblastic leukemia. Clin Lab Med, 2011. 31(4): p. 631-47, ix. 3. Landier, W. and S. Bhatia, Cancer survivorship: a pediatric perspective. Oncologist, 2008. 13(11): p. 1181-92. 4. Fulbright, J.M., et al., Late effects of childhood leukemia therapy. Curr Hematol Malig Rep, 2011. 6(3): p. 195-205. 5. Pui, C.H., M.V. Relling, and J.R. Downing, Acute lymphoblastic leukemia. N Engl J Med, 2004. 350(15): p. 1535-48. 6. Meads, M.B., L.A. Hazlehurst, and W.S. Dalton, The bone marrow microenvironment as a tumor sanctuary and contributor to drug resistance. Clin Cancer Res, 2008. 14(9): p. 2519-26. 7. Hunger, S.P., et al., Improving outcomes for high-risk ALL: translating new discoveries into clinical care. Pediatr Blood Cancer, 2011. 56(6): p. 984-93. 8. Damiano, J.S., et al., Cell adhesion mediated drug resistance (CAM-DR): role of integrins and resistance to apoptosis in human myeloma cell lines. Blood, 1999. 93(5): p. 1658-67. 9. Liu, C.C., et al., The membrane-proximal KXGFFKR motif of alpha-integrin mediates chemoresistance. Mol Cell Biol, 2013. 33(21): p. 4334-45. 10. Konopleva, M.Y. and C.T. Jordan, Leukemia stem cells and microenvironment: biology and therapeutic targeting. J Clin Oncol, 2011. 29(5): p. 591-9. 11. Vincent, T. and N. Mechti, Extracellular matrix in bone marrow can mediate drug resistance in myeloma. Leuk Lymphoma, 2005. 46(6): p. 803-11. 12. Alon, R. and S.W. Feigelson, Chemokine-triggered leukocyte arrest: force-regulated bi-directional integrin activation in quantal adhesive contacts. Curr Opin Cell Biol, 2012. 24(5): p. 670-6. 13. Damiano, J.S., Integrins as novel drug targets for overcoming innate drug resistance. Curr Cancer Drug Targets, 2002. 2(1): p. 37-43. 14. Li, Z.W. and W.S. Dalton, Tumor microenvironment and drug resistance in hematologic malignancies. Blood Rev, 2006. 20(6): p. 333-42. 15. Matsunaga, T., et al., Combination therapy of an anticancer drug with the FNIII14 peptide of fibronectin effectively overcomes cell adhesion-mediated drug resistance of acute myelogenous leukemia. Leukemia, 2008. 22(2): p. 353-60. 16. Hoyt, D.G., et al., Integrin activation suppresses etoposide-induced DNA strand breakage in cultured murine tumor-derived endothelial cells. Cancer Res, 1996. 56(18): p. 4146-9. 17. Scott, G., L. Cassidy, and A. Busacco, Fibronectin suppresses apoptosis in normal human melanocytes through an integrin-dependent mechanism. J Invest Dermatol, 1997. 108(2): p. 147-53. 18. Rozzo, C., et al., Induction of apoptosis in human neuroblastoma cells by abrogation of integrin-mediated cell adhesion. Int J Cancer, 1997. 70(6): p. 688-98. 19. Zhang, Z., et al., The alpha 5 beta 1 integrin supports survival of cells on fibronectin and up-regulates Bcl-2 expression. Proc Natl Acad Sci U S A, 1995. 92(13): p. 6161-5. 20. Rose, D.M., R. Alon, and M.H. Ginsberg, Integrin modulation and signaling in leukocyte adhesion and migration. Immunol Rev, 2007. 218(Journal Article): p. 126-34. 73 21. Abram, C.L. and C.A. Lowell, The ins and outs of leukocyte integrin signaling. Annu Rev Immunol, 2009. 27(Journal Article): p. 339-62. 22. Tamkun, J.W., et al., Structure of Integrin, a Glycoprotein Involved in the Transmembrane Linkage between Fibronectin and Actin. Cell, 1986. 46(2): p. 271-282. 23. Barczyk, M., S. Carracedo, and D. Gullberg, Integrins. Cell Tissue Res, 2010. 339(1): p. 269-80. 24. Humphries, J.D., A. Byron, and M.J. Humphries, Integrin ligands at a glance. J Cell Sci, 2006. 119(Pt 19): p. 3901-3. 25. Moser, M., et al., The tail of integrins, talin, and kindlins. Science, 2009. 324(5929): p. 895-9. 26. Zhang, K. and J. Chen, The regulation of integrin function by divalent cations. Cell Adh Migr, 2012. 6(1): p. 20-9. 27. Xiong, J.P., et al., Crystal structure of the extracellular segment of integrin alpha Vbeta3. Science, 2001. 294(5541): p. 339-45. 28. Tiwari, S., et al., Divalent cations regulate the folding and activation status of integrins during their intracellular trafficking. J Cell Sci, 2011. 124(Pt 10): p. 1672-80. 29. Leitinger, B., et al., The regulation of integrin function by Ca(2+). Biochim Biophys Acta, 2000. 1498(2-3): p. 91-8. 30. Masumoto, A. and M.E. Hemler, Mutation of putative divalent cation sites in the alpha 4 subunit of the integrin VLA-4: distinct effects on adhesion to CS1/fibronectin, VCAM-1, and invasin. Journal of Cell Biology, 1993. 123(1): p. 245-53. 31. Loftus, J.C., et al., The amino-terminal one-third of alpha IIb defines the ligand recognition specificity of integrin alpha IIb beta 3. J Biol Chem, 1996. 271(4): p. 2033-9. 32. Wilcox, D.A., et al., A single amino acid substitution flanking the fourth calcium binding domain of alpha IIb prevents maturation of the alpha IIb beta 3 integrin complex. J Biol Chem, 1994. 269(6): p. 4450-7. 33. Basani, R.B., et al., Glanzmann thrombasthenia due to a two amino acid deletion in the fourth calcium-binding domain of alpha(IIb): Demonstration of the importance of calcium-binding domains in the conformation of alpha(IIb)beta(3). Blood, 1996. 88(1): p. 167-173. 34. Mitchell, W.B., et al., Two novel mutations in the alpha IIb calcium-binding domains identify hydrophobic regions essential for alpha IIbbeta 3 biogenesis. Blood, 2003. 101(6): p. 2268-76. 35. Zhu, J., et al., Structure of a complete integrin ectodomain in a physiologic resting state and activation and deactivation by applied forces. Mol Cell, 2008. 32(6): p. 849-61. 36. Campbell, I.D. and M.J. Humphries, Integrin structure, activation, and interactions. Cold Spring Harb Perspect Biol, 2011. 3(3). 37. Xiong, J.P., et al., Crystal structure of the extracellular segment of integrin alpha Vbeta3 in complex with an Arg-Gly-Asp ligand. Science, 2002. 296(5565): p. 151-5. 38. Chen, J., A. Salas, and T.A. Springer, Bistable regulation of integrin adhesiveness by a bipolar metal ion cluster. Nat Struct Biol, 2003. 10(12): p. 995-1001. 39. Hughes, P.E., et al., Breaking the integrin hinge. A defined structural constraint regulates integrin signaling. J Biol Chem, 1996. 271(12): p. 6571-4. 40. Lu, C., J. Takagi, and T.A. Springer, Association of the membrane proximal regions of the alpha and beta subunit cytoplasmic domains constrains an integrin in the inactive state. J Biol Chem, 2001. 276(18): p. 14642-8. 41. Lau, T.L., et al., The structure of the integrin alphaIIbbeta3 transmembrane complex explains integrin transmembrane signalling. EMBO J, 2009. 28(9): p. 1351-61. 42. Li, R., et al., Activation of integrin alphaIIbbeta3 by modulation of transmembrane helix associations. Science, 2003. 300(5620): p. 795-8. 74 43. Partridge, A.W., et al., Transmembrane domain helix packing stabilizes integrin alphaIIbbeta3 in the low affinity state. J Biol Chem, 2005. 280(8): p. 7294-300. 44. Li, W., et al., A push-pull mechanism for regulating integrin function. Proc Natl Acad Sci U S A, 2005. 102(5): p. 1424-9. 45. Luo, B.H., et al., Disrupting integrin transmembrane domain heterodimerization increases ligand binding affinity, not valency or clustering. Proc Natl Acad Sci U S A, 2005. 102(10): p. 3679-84. 46. Kim, C., et al., Interactions of platelet integrin alphaIIb and beta3 transmembrane domains in mammalian cell membranes and their role in integrin activation. Blood, 2009. 113(19): p. 4747-53. 47. Vomund, A.N., et al., A naturally occurring extracellular alpha-beta clasp contributes to stabilization of beta3 integrins in a bent, resting conformation. Biochemistry, 2008. 47(44): p. 11616-24. 48. Smagghe, B.J., et al., Modulation of integrin activation by an entropic spring in the {beta}-knee. J Biol Chem, 2010. 285(43): p. 32954-66. 49. Alon, R. and S. Feigelson, From rolling to arrest on blood vessels: leukocyte tap dancing on endothelial integrin ligands and chemokines at sub-second contacts. Semin Immunol, 2002. 14(2): p. 93-104. 50. Baker, R.G. and G.A. Koretzky, Regulation of T cell integrin function by adapter proteins. Immunol Res, 2008. 42(1-3): p. 132-44. 51. Sebzda, E., et al., Rap1A positively regulates T cells via integrin activation rather than inhibiting lymphocyte signaling. Nature Immunology, 2002. 3(3): p. 251-8. 52. Shattil, S.J., C. Kim, and M.H. Ginsberg, The final steps of integrin activation: the end game. Nat Rev Mol Cell Biol, 2010. 11(4): p. 288-300. 53. Anthis, N.J. and I.D. Campbell, The tail of integrin activation. Trends Biochem Sci, 2011. 36(4): p. 191-8. 54. Han, J., et al., Reconstructing and deconstructing agonist-induced activation of integrin alphaIIbbeta3. Curr Biol, 2006. 16(18): p. 1796-806. 55. Lee, H.S., et al., RIAM activates integrins by linking talin to ras GTPase membrane-targeting sequences. J Biol Chem, 2009. 284(8): p. 5119-27. 56. Ridley, A.J., et al., Cell migration: integrating signals from front to back. Science, 2003. 302(5651): p. 1704-9. 57. Charest, P.G. and R.A. Firtel, Big roles for small GTPases in the control of directed cell movement. Biochem J, 2007. 401(2): p. 377-90. 58. Shimonaka, M., et al., Rap1 translates chemokine signals to integrin activation, cell polarization, and motility across vascular endothelium under flow. Journal of Cell Biology, 2003. 161(2): p. 417-27. 59. Lafuente, E.M., et al., RIAM, an Ena/VASP and Profilin ligand, interacts with Rap1-GTP and mediates Rap1-induced adhesion. Dev Cell, 2004. 7(4): p. 585-95. 60. Kim, C., F. Ye, and M.H. Ginsberg, Regulation of integrin activation. Annu Rev Cell Dev Biol, 2011. 27(Journal Article): p. 321-45. 61. Goksoy, E., et al., Structural basis for the autoinhibition of talin in regulating integrin activation. Mol Cell, 2008. 31(1): p. 124-33. 62. Yan, B., et al., Calpain cleavage promotes talin binding to the beta 3 integrin cytoplasmic domain. J Biol Chem, 2001. 276(30): p. 28164-70. 63. Garcia-Alvarez, B., et al., Structural determinants of integrin recognition by talin. Mol Cell, 2003. 11(1): p. 49-58. 64. Wegener, K.L., et al., Structural basis of integrin activation by talin. Cell, 2007. 128(1): p. 171-82. 65. Xiao, T., et al., Structural basis for allostery in integrins and binding to fibrinogen-mimetic therapeutics. Nature, 2004. 432(7013): p. 59-67. 75 66. Carman, C.V. and T.A. Springer, Integrin avidity regulation: are changes in affinity and conformation underemphasized? Curr Opin Cell Biol, 2003. 15(5): p. 547-56. 67. Cluzel, C., et al., The mechanisms and dynamics of (alpha)v(beta)3 integrin clustering in living cells. J Cell Biol, 2005. 171(2): p. 383-92. 68. van Kooyk, Y., S.J. van Vliet, and C.G. Figdor, The actin cytoskeleton regulates LFA-1 ligand binding through avidity rather than affinity changes. J Biol Chem, 1999. 274(38): p. 26869-77. 69. Cambi, A., et al., Organization of the integrin LFA-1 in nanoclusters regulates its activity. Mol Biol Cell, 2006. 17(10): p. 4270-81. 70. Bakker, G.J., et al., Lateral mobility of individual integrin nanoclusters orchestrates the onset for leukocyte adhesion. Proc Natl Acad Sci U S A, 2012. 109(13): p. 4869-74. 71. Bunch, T.A., Integrin alphaIIbbeta3 activation in Chinese hamster ovary cells and platelets increases clustering rather than affinity. J Biol Chem, 2010. 285(3): p. 1841-9. 72. Byron, A., et al., Anti-integrin monoclonal antibodies. J Cell Sci, 2009. 122(Pt 22): p. 4009-11. 73. Njus, B.H., et al., Conformational mAb as a tool for integrin ligand discovery. Assay Drug Dev Technol, 2009. 7(5): p. 507-15. 74. Woods, M.L., C. Cabanas, and Y. Shimizu, Activation-dependent changes in soluble fibronectin binding and expression of beta1 integrin activation epitopes in T cells: relationship to T cell adhesion and migration. Eur J Immunol, 2000. 30(1): p. 38-49. 75. Zhao, Z., J. Gruszczynska-Biegala, and A. Zolkiewska, ADP-ribosylation of integrin alpha7 modulates the binding of integrin alpha7beta1 to laminin. Biochem J, 2005. 385(Pt 1): p. 309-17. 76. Chigaev, A., et al., Real-time analysis of conformation-sensitive antibody binding provides new insights into integrin conformational regulation. J Biol Chem, 2009. 284(21): p. 14337-46. 77. Green, J.A., et al., beta1 integrin cytoplasmic domain residues selectively modulate fibronectin matrix assembly and cell spreading through talin and Akt-1. J Biol Chem, 2009. 284(12): p. 8148-59. 78. McGinn, O.J., et al., Modulation of integrin alpha4beta1 by ADAM28 promotes lymphocyte adhesion and transendothelial migration. Cell Biol Int, 2011. 35(10): p. 1043-53. 79. Luque, A., et al., Activated conformations of very late activation integrins detected by a group of antibodies (HUTS) specific for a novel regulatory region (355-425) of the common beta 1 chain. Journal of Biological Chemistry, 1996. 271(19): p. 11067-11075. 80. Bazzoni, G., et al., Monoclonal antibody 9EG7 defines a novel beta 1 integrin epitope induced by soluble ligand and manganese, but inhibited by calcium. J Biol Chem, 1995. 270(43): p. 25570-7. 81. Bazzoni, G., et al., Divalent cations and ligands induce conformational changes that are highly divergent among beta1 integrins. J Biol Chem, 1998. 273(12): p. 6670-8. 82. Chigaev, A., et al., Real time analysis of the affinity regulation of alpha 4-integrin. The physiologically activated receptor is intermediate in affinity between resting and Mn(2+) or antibody activation. J Biol Chem, 2001. 276(52): p. 48670-8. 83. Chigaev, A., et al., FRET detection of cellular alpha4-integrin conformational activation. Biophys J, 2003. 85(6): p. 3951-62. 84. Chigaev, A., et al., Regulation of cell adhesion by affinity and conformational unbending of alpha4beta1 integrin. J Immunol, 2007. 178(11): p. 6828-39. 85. Janetopoulos, C., et al., Chemoattractant-induced phosphatidylinositol 3,4,5-trisphosphate accumulation is spatially amplified and adapts, independent of the actin cytoskeleton. Proc Natl Acad Sci U S A, 2004. 101(24): p. 8951-6. 76 86. Srinivasan, S., et al., Rac and Cdc42 play distinct roles in regulating PI(3,4,5)P3 and polarity during neutrophil chemotaxis. Journal of Cell Biology, 2003. 160(3): p. 375-85. 87. Lee, J.S., et al., Cdc42 mediates nucleus movement and MTOC polarization in Swiss 3T3 fibroblasts under mechanical shear stress. Mol Biol Cell, 2005. 16(2): p. 871-80. 88. Nobes, C.D. and A. Hall, Rho GTPases control polarity, protrusion, and adhesion during cell movement. Journal of Cell Biology, 1999. 144(6): p. 1235-44. 89. Charest, P.G. and R.A. Firtel, Feedback signaling controls leading-edge formation during chemotaxis. Curr Opin Genet Dev, 2006. 16(4): p. 339-47. 90. Nishiya, N., et al., An alpha4 integrin-paxillin-Arf-GAP complex restricts Rac activation to the leading edge of migrating cells. Nat Cell Biol, 2005. 7(4): p. 343-52. 91. Liu, S., et al., Binding of paxillin to alpha4 integrins modifies integrin-dependent biological responses. Nature, 1999. 402(6762): p. 676-81. 92. Han, J., et al., Phosphorylation of the integrin alpha 4 cytoplasmic domain regulates paxillin binding. J Biol Chem, 2001. 276(44): p. 40903-9. 93. Harburger, D.S. and D.A. Calderwood, Integrin signalling at a glance. J Cell Sci, 2009. 122(Pt 2): p. 159-63. 94. Goldfinger, L.E., et al., Spatial restriction of alpha4 integrin phosphorylation regulates lamellipodial stability and alpha4beta1-dependent cell migration. Journal of Cell Biology, 2003. 162(4): p. 731-41. 95. Lim, C.J., et al., Alpha4 integrins are type I cAMP-dependent protein kinase-anchoring proteins. Nat Cell Biol, 2007. 9(4): p. 415-21. 96. Lim, C.J., et al., Integrin-mediated protein kinase A activation at the leading edge of migrating cells. Mol Biol Cell, 2008. 19(11): p. 4930-41. 97. Bendall, L.J., K. Kortlepel, and D.J. Gottlieb, Human acute myeloid leukemia cells bind to bone marrow stroma via a combination of beta-1 and beta-2 integrin mechanisms. Blood, 1993. 82(10): p. 3125-32. 98. Damiano, J.S., L.A. Hazlehurst, and W.S. Dalton, Cell adhesion-mediated drug resistance (CAM-DR) protects the K562 chronic myelogenous leukemia cell line from apoptosis induced by BCR/ABL inhibition, cytotoxic drugs, and gamma-irradiation. Leukemia, 2001. 15(8): p. 1232-9. 99. Berger, J.R. and I.J. Koralnik, Progressive multifocal leukoencephalopathy and natalizumab--unforeseen consequences. N Engl J Med, 2005. 353(4): p. 414-6. 100. Meads, M.B., R.A. Gatenby, and W.S. Dalton, Environment-mediated drug resistance: a major contributor to minimal residual disease. Nature Reviews Cancer, 2009. 9(9): p. 665-74. 101. Brown, E.J. and W.A. Frazier, Integrin-associated protein (CD47) and its ligands. Trends Cell Biol, 2001. 11(3): p. 130-5. 102. Reinhold, M.I., et al., In vivo expression of alternatively spliced forms of integrin-associated protein (CD47). J Cell Sci, 1995. 108 ( Pt 11)(Pt 11): p. 3419-25. 103. Avent, N., et al., Monoclonal antibodies that recognize different membrane proteins that are deficient in Rhnull human erythrocytes. One group of antibodies reacts with a variety of cells and tissues whereas the other group is erythroid-specific. Biochem J, 1988. 251(2): p. 499-505. 104. Campbell, I.G., et al., An ovarian tumor marker with homology to vaccinia virus contains an IgV-like region and multiple transmembrane domains. Cancer Res, 1992. 52(19): p. 5416-20. 105. Brown, E., et al., Integrin-associated protein: a 50-kD plasma membrane antigen physically and functionally associated with integrins. Journal of Cell Biology, 1990. 111(6 Pt 1): p. 2785-94. 77 106. Mawby, W.J., et al., Isolation and characterization of CD47 glycoprotein: a multispanning membrane protein which is the same as integrin-associated protein (IAP) and the ovarian tumour marker OA3. Biochem J, 1994. 304 ( Pt 2): p. 525-30. 107. Rebres, R.A., et al., Membrane raft association of CD47 is necessary for actin polymerization and protein kinase C theta translocation in its synergistic activation of T cells. J Biol Chem, 2001. 276(10): p. 7672-80. 108. Lee, G., et al., Widespread expressions of immunoglobulin superfamily proteins in cancer cells. Cancer Immunol Immunother, 2012. 61(1): p. 89-99. 109. Ticchioni, M., et al., Integrin-associated protein (CD47/IAP) contributes to T cell arrest on inflammatory vascular endothelium under flow. FASEB J, 2001. 15(2): p. 341-50. 110. Liu, Y., et al., The role of CD47 in neutrophil transmigration. Increased rate of migration correlates with increased cell surface expression of CD47. J Biol Chem, 2001. 276(43): p. 40156-66. 111. Rebres, R.A., K. Kajihara, and E.J. Brown, Novel CD47-dependent intercellular adhesion modulates cell migration. J Cell Physiol, 2005. 205(2): p. 182-93. 112. Waclavicek, M., et al., T cell stimulation via CD47: agonistic and antagonistic effects of CD47 monoclonal antibody 1/1A4. J Immunol, 1997. 159(11): p. 5345-54. 113. Manna, P.P., et al., CD47 augments Fas/CD95-mediated apoptosis. J Biol Chem, 2005. 280(33): p. 29637-44. 114. Oldenborg, P.A., CD47: A Cell Surface Glycoprotein Which Regulates Multiple Functions of Hematopoietic Cells in Health and Disease. ISRN Hematol, 2013. 2013(Journal Article): p. 614619. 115. Lindberg, F.P., et al., Decreased resistance to bacterial infection and granulocyte defects in IAP-deficient mice. Science, 1996. 274(5288): p. 795-8. 116. Li, Z., et al., Interactions of thrombospondins with alpha4beta1 integrin and CD47 differentially modulate T cell behavior. Journal of Cell Biology, 2002. 157(3): p. 509-19. 117. Barazi, H.O., et al., Regulation of integrin function by CD47 ligands. Differential effects on alpha vbeta 3 and alpha 4beta1 integrin-mediated adhesion. J Biol Chem, 2002. 277(45): p. 42859-66. 118. Azcutia, V., et al., Endothelial CD47 promotes vascular endothelial-cadherin tyrosine phosphorylation and participates in T cell recruitment at sites of inflammation in vivo. J Immunol, 2012. 189(5): p. 2553-62. 119. Rebres, R.A., et al., Normal ligand binding and signaling by CD47 (integrin-associated protein) requires a long range disulfide bond between the extracellular and membrane-spanning domains. J Biol Chem, 2001. 276(37): p. 34607-16. 120. Adams, J.C., Thrombospondins: multifunctional regulators of cell interactions. Annu Rev Cell Dev Biol, 2001. 17(Journal Article): p. 25-51. 121. Adams, J.C. and J. Lawler, The thrombospondins. Int J Biochem Cell Biol, 2004. 36(6): p. 961-8. 122. Forslow, A., Z. Liu, and K.G. Sundqvist, Receptor communication within the lymphocyte plasma membrane: a role for the thrombospondin family of matricellular proteins. Cell Mol Life Sci, 2007. 64(1): p. 66-76. 123. Adams, J.C. and J. Lawler, The thrombospondins. Cold Spring Harb Perspect Biol, 2011. 3(10): p. a009712. 124. Bonnefoy, A., R. Moura, and M.F. Hoylaerts, The evolving role of thrombospondin-1 in hemostasis and vascular biology. Cell Mol Life Sci, 2008. 65(5): p. 713-27. 125. Kosfeld, M.D. and W.A. Frazier, Identification of a new cell adhesion motif in two homologous peptides from the COOH-terminal cell binding domain of human thrombospondin. J Biol Chem, 1993. 268(12): p. 8808-14. 126. Gao, A.G., et al., Integrin-associated protein is a receptor for the C-terminal domain of thrombospondin. Journal of Biological Chemistry, 1996. 271(1): p. 21-24. 78 127. Isenberg, J.S., et al., Differential interactions of thrombospondin-1, -2, and -4 with CD47 and effects on cGMP signaling and ischemic injury responses. J Biol Chem, 2009. 284(2): p. 1116-25. 128. Gao, A.G. and W.A. Frazier, Identification of a receptor candidate for the carboxyl-terminal cell binding domain of thrombospondins. J Biol Chem, 1994. 269(47): p. 29650-7. 129. Brittain, J.E., et al., Activation of sickle red blood cell adhesion via integrin-associated protein/CD47-induced signal transduction. J Clin Invest, 2001. 107(12): p. 1555-62. 130. Rock, M.J., et al., Cartilage oligomeric matrix protein promotes cell attachment via two independent mechanisms involving CD47 and alphaVbeta3 integrin. Mol Cell Biochem, 2010. 338(1-2): p. 215-24. 131. Brittain, J.E., et al., Mechanism of CD47-induced alpha4beta1 integrin activation and adhesion in sickle reticulocytes. J Biol Chem, 2004. 279(41): p. 42393-402. 132. Tulasne, D., et al., C-terminal peptide of thrombospondin-1 induces platelet aggregation through the Fc receptor gamma-chain-associated signaling pathway and by agglutination. Blood, 2001. 98(12): p. 3346-52. 133. Fujimoto, T.T., et al., Thrombospondin-bound integrin-associated protein (CD47) physically and functionally modifies integrin alphaIIbbeta3 by its extracellular domain. J Biol Chem, 2003. 278(29): p. 26655-65. 134. Manna, P.P. and W.A. Frazier, The mechanism of CD47-dependent killing of T cells: heterotrimeric Gi-dependent inhibition of protein kinase A. J Immunol, 2003. 170(7): p. 3544-53. 135. Manna, P.P. and W.A. Frazier, CD47 mediates killing of breast tumor cells via Gi-dependent inhibition of protein kinase A. Cancer Res, 2004. 64(3): p. 1026-36. 136. Rath, G.M., et al., The C-terminal CD47/IAP-binding domain of thrombospondin-1 prevents camptothecin- and doxorubicin-induced apoptosis in human thyroid carcinoma cells. Biochim Biophys Acta, 2006. 1763(10): p. 1125-34. 137. Maxhimer, J.B., et al., Radioprotection in normal tissue and delayed tumor growth by blockade of CD47 signaling. Sci Transl Med, 2009. 1(3): p. 3ra7. 138. Burger, P., et al., CD47 functions as a molecular switch for erythrocyte phagocytosis. Blood, 2012. 119(23): p. 5512-21. 139. Sick, E., et al., CD47 update: a multifaceted actor in the tumour microenvironment of potential therapeutic interest. Br J Pharmacol, 2012. 167(7): p. 1415-30. 140. Vernon-Wilson, E.F., et al., CD47 is a ligand for rat macrophage membrane signal regulatory protein SIRP (OX41) and human SIRPalpha 1. Eur J Immunol, 2000. 30(8): p. 2130-7. 141. Barclay, A.N. and M.H. Brown, The SIRP family of receptors and immune regulation. Nat Rev Immunol, 2006. 6(6): p. 457-64. 142. Jiang, P., C.F. Lagenaur, and V. Narayanan, Integrin-associated protein is a ligand for the P84 neural adhesion molecule. J Biol Chem, 1999. 274(2): p. 559-62. 143. Brooke, G., et al., Human lymphocytes interact directly with CD47 through a novel member of the signal regulatory protein (SIRP) family. J Immunol, 2004. 173(4): p. 2562-70. 144. Hatherley, D., et al., Paired receptor specificity explained by structures of signal regulatory proteins alone and complexed with CD47. Mol Cell, 2008. 31(2): p. 266-77. 145. Liu, Y., et al., Functional elements on SIRPalpha IgV domain mediate cell surface binding to CD47. J Mol Biol, 2007. 365(3): p. 680-93. 146. Gardai, S.J., et al., Recognition ligands on apoptotic cells: a perspective. J Leukoc Biol, 2006. 79(5): p. 896-903. 147. Gardai, S.J., et al., Cell-surface calreticulin initiates clearance of viable or apoptotic cells through trans-activation of LRP on the phagocyte. Cell, 2005. 123(2): p. 321-34. 79 148. Oldenborg, P.A., H.D. Gresham, and F.P. Lindberg, CD47-signal regulatory protein alpha (SIRPalpha) regulates Fcgamma and complement receptor-mediated phagocytosis. J Exp Med, 2001. 193(7): p. 855-62. 149. Ishikawa-Sekigami, T., et al., SHPS-1 promotes the survival of circulating erythrocytes through inhibition of phagocytosis by splenic macrophages. Blood, 2006. 107(1): p. 341-8. 150. Gao, A.G., et al., Thrombospondin modulates alpha(v)beta(3) function through integrin-associated protein. Journal of Cell Biology, 1996. 135(2): p. 533-544. 151. McDonald, J.F., J.M. Dimitry, and W.A. Frazier, An amyloid-like C-terminal domain of thrombospondin-1 displays CD47 agonist activity requiring both VVM motifs. Biochemistry, 2003. 42(33): p. 10001-11. 152. Frazier, W.A., et al., The thrombospondin receptor integrin-associated protein (CD47) functionally couples to heterotrimeric Gi. J Biol Chem, 1999. 274(13): p. 8554-60. 153. Wang, X.Q., F.P. Lindberg, and W.A. Frazier, Integrin-associated protein stimulates alpha2beta1-dependent chemotaxis via Gi-mediated inhibition of adenylate cyclase and extracellular-regulated kinases. Journal of Cell Biology, 1999. 147(2): p. 389-400. 154. Reinhold, M.I., et al., Costimulation of T cell activation by integrin-associated protein (CD47) is an adhesion-dependent, CD28-independent signaling pathway. J Exp Med, 1997. 185(1): p. 1-11. 155. Chung, J., A.G. Gao, and W.A. Frazier, Thrombspondin acts via integrin-associated protein to activate the platelet integrin alphaIIbbeta3. J Biol Chem, 1997. 272(23): p. 14740-6. 156. Wilson, K.E., et al., Beta 1 integrin- and proteoglycan-mediated stimulation of T lymphoma cell adhesion and mitogen-activated protein kinase signaling by thrombospondin-1 and thrombospondin-1 peptides. J Immunol, 1999. 163(7): p. 3621-8. 157. McDonald, J.F., A. Zheleznyak, and W.A. Frazier, Cholesterol-independent interactions with CD47 enhance alphavbeta3 avidity. J Biol Chem, 2004. 279(17): p. 17301-11. 158. Orazizadeh, M., et al., CD47 associates with alpha 5 integrin and regulates responses of human articular chondrocytes to mechanical stimulation in an in vitro model. Arthritis Res Ther, 2008. 10(1): p. R4. 159. Azcutia, V., et al., CD47 plays a critical role in T-cell recruitment by regulation of LFA-1 and VLA-4 integrin adhesive functions. Mol Biol Cell, 2013. 24(21): p. 3358-68. 160. Schwartz, M.A., E.J. Brown, and B. Fazeli, A 50-kDa integrin-associated protein is required for integrin-regulated calcium entry in endothelial cells. J Biol Chem, 1993. 268(27): p. 19931-4. 161. Lindberg, F.P., et al., Molecular cloning of integrin-associated protein: an immunoglobulin family member with multiple membrane-spanning domains implicated in alpha v beta 3-dependent ligand binding. Journal of Cell Biology, 1993. 123(2): p. 485-96. 162. Lindberg, F.P., et al., Integrin-associated protein immunoglobulin domain is necessary for efficient vitronectin bead binding. Journal of Cell Biology, 1996. 134(5): p. 1313-1322. 163. Kosfeld, M.D. and W.A. Frazier, Identification of active peptide sequences in the carboxyl-terminal cell binding domain of human thrombospondin-1. J Biol Chem, 1992. 267(23): p. 16230-6. 164. Anilkumar, N., et al., Trimeric assembly of the C-terminal region of Thrombospondin-1 or Thrombospondin-2 is necessary for cell spreading and fascin spike organisation. Journal of cell science, 2002. 115(11): p. 2357-2366. 165. Li, S.S., et al., Endogenous thrombospondin-1 is a cell-surface ligand for regulation of integrin-dependent T-lymphocyte adhesion. Blood, 2006. 108(9): p. 3112-20. 80 166. Rose, D.M., et al., Paxillin binding to the alpha 4 integrin subunit stimulates LFA-1 (integrin alpha L beta 2)-dependent T cell migration by augmenting the activation of focal adhesion kinase/proline-rich tyrosine kinase-2. J Immunol, 2003. 170(12): p. 5912-8. 167. Reinhold, M.I., et al., Cell spreading distinguishes the mechanism of augmentation of T cell activation by integrin-associated protein/CD47 and CD28. Int Immunol, 1999. 11(5): p. 707-18. 168. Romzek, N.C., et al., Use of a beta1 integrin-deficient human T cell to identify beta1 integrin cytoplasmic domain sequences critical for integrin function. Mol Biol Cell, 1998. 9(10): p. 2715-27. 169. Ide, K., et al., Role for CD47-SIRPalpha signaling in xenograft rejection by macrophages. Proc Natl Acad Sci U S A, 2007. 104(12): p. 5062-6. 170. Horvath, P. and R. Barrangou, CRISPR/Cas, the immune system of bacteria and archaea. Science, 2010. 327(5962): p. 167-70. 171. Jinek, M., et al., A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science, 2012. 337(6096): p. 816-21. 172. Mali, P., et al., RNA-guided human genome engineering via Cas9. Science, 2013. 339(6121): p. 823-6. 173. Cong, L., et al., Multiplex genome engineering using CRISPR/Cas systems. Science, 2013. 339(6121): p. 819-23. 174. Cho, S.W., et al., Targeted genome engineering in human cells with the Cas9 RNA-guided endonuclease. Nat Biotechnol, 2013. 31(3): p. 230-2. 175. Kvansakul, M., J.C. Adams, and E. Hohenester, Structure of a thrombospondin C-terminal fragment reveals a novel calcium core in the type 3 repeats. EMBO J, 2004. 23(6): p. 1223-33. 176. Floquet, N., et al., Human thrombospondin's (TSP-1) C-terminal domain opens to interact with the CD-47 receptor: a molecular modeling study. Arch Biochem Biophys, 2008. 478(1): p. 103-9. 177. Adams, J.C., et al., Extracellular matrix retention of thrombospondin 1 is controlled by its conserved C-terminal region. J Cell Sci, 2008. 121(Pt 6): p. 784-95. 81 Appendix Figure 28: 4N1K Mediates Non-specific Effects of Fibronectin and BSA Binding. Jurkat cells were treated as in Fig. 28 and FITC-conjugated fibronectin (Fn) or BSA binding was assessed using flow cytometry as indicated in materials and methods. CONCLUSION: In this assay, MnCl2 did not induce Fn binding (reasons for this is not known), whereas 4N1K induced significant Fn binding. Although there was minimal binding of BSA in untreated cells, 4N1K induced significant BSA binding. This indicates that 4N1K induces non-specific binding of various proteins contained in BSA. 82 A. B. 83 Figure 29: 4N1K and MnCl2 Mediate Non-specific Binding of GST Peptides. A. 2.5x105 Jurkat, JinB8, and JB4 cells were washed and resuspended in 90μl of bRPMI with 10μl of either bRPMI, 4NGG or 4N1K. Cells were incubated for 30min at 37°C. Solutions were also prepared with 90μl of 49.3µg\ml FITC-GST-CS1 or FITC-GST-Fn911 and 10μl of either bRPMI (untreated, 4NGG, 4N1K), 0.2mM EDTA (10mM final), or 0.02mM MnCl2. Cells were incubated for 30min at RT and spun down. Cells were resuspended in PBS and flow cytometry was performed to assess FITC peptide binding as indicated in materials and methods. B. As in A, but Jurkat and JB4 cells were either untreated or stimulated with MnCl2. GST-FITC or GST-CS1-FITC was used as reporters. CONCLUSION: As expected MnCl2 induces binding of FITC-peptides. However, it also induces binding of GST alone. Similarly, although 4N1K induces binding of FITC-peptides in Jurkat and JB4 cells, it also does so in the CD47-deficient JinB8 cells, indicating that 4N1K mediates non-specific effects on peptide binding. 84 Figure 30: 4N1K Induces Apoptosis in Jurkat Cells. 6x105 cells were washed and resuspended in 1%BSA/PBS with various 4N1K concentrations and incubated at 37°C for indicated times. Cells were spun down and incubated with propidium iodide and annexinV at 100:2:2 at RT for 15min in the dark. AnnexinV+/PI+ cells were assessed using flow cytometry. CONCLUSION: PI uptake increases with an increase in 4N1K concentration and incubation time; therefore, 4N1K induces cell apoptosis in a time- and concentration-dependent manner. 85 A. B. Figure 31: 4N1K Induces AnnexinV Binding in a CD47-independent Manner. 3 x 105 Jurkat or JinB8 cells were washed and resuspended in 0.8ml of blank RPMI. The following conditions were then added: 100μl of bRPMI and either: 100μl of bRPMI, 100μl 1mM 4NGG (100μM final), 100μl 0.5mM 4N1K (50μM final), 100μl 1mM 4N1K (100μM final). Cells were incubated for 2 days, washed, and incubated with Annexin V-Cy5 at 2:100 for 15min at RT in the dark. Flow cytometry was performed to assess AnnexinV binding. A. Raw data. B. Bar graph. CONCLUSION: 4N1K induces apoptosis in a CD47-independent manner. 86 Figure 32: Anti-CD47 Antibody Clone CC2C6 Induces Apoptosis in a CD47-dependent Manner. 4x105 Jurkat and JinB8 cells were incubated with IgG or anti-CD47 antibody clone B6H12 or CC2C6 (500ng in 100μl of 1%BSA/PBS each) for 30min. Cells were spun and incubated with 2:100 Annexin V-Cy5 for 15min at RT in the dark. Flow cytometry was performed to assess AnnexinV binding. CONCLUSION: Anti-CD47 antibody clone CC2C6, but not B6H12, induces apoptosis in Jurkat cells in a CD47-dependent manner.