Sunday, 18 December 2011

MSc Dissertation

"Extracting Circulating Endothelial Cells from Whole Blood"

The endothelium lines the blood and lymphatic vessels of the body and is made up of a monolayer of endothelial cells. It plays a crucial role in several physiological activities including the regulation of vasotone, leukocyte trafficking, coagulation and the maintenance of vessel wall permeability. Endothelial cells exhibit both structural and functional heterogeneity making it a potential therapeutic target. The ability to determine the phenotype of the various endothelial cells lies in the ability to culture them in vitro.  Most studies rely on Human Umbilical Vein Endothelial Cells (HUVECs) in culture to establish the various endothelial cell characteristics. HUVECs are however of foetal origin and exhibit different phenotypical characteristics to that of the adult endothelium. An alternative source of endothelial cells which is more representative of the adult endothelium such as from whole blood will be most welcome. 
Circulating endothelial cells (CECs) are mature endothelial cells that have come detached from the endothelial cell wall. They have recently been identified as a key marker in assessing vascular injury brought about by damage to the endothelium and has been identified as a potential therapeutic target for assessing vascular injury.
 In health, they are present in low levels of about 1 – 20 CECs/ml of whole blood and elevated levels have been identified in disease. To date, circulating endothelial cells are yet to be phenotyped as culturing them have proven challenging due to the cells apoptosing before reaching culture stage or soon after initiating culture.
Immunomagnetic bead isolation is one of the techniques currently employed to extract and enumerate CECs from whole blood. 
Isolating endothelial cells from human umbilical vein endothelial cells in culture followed by whole blood by use of Immunomagnetic bead separation was carried out. This was based on the principle of streptavidin – biotin conjugation with the magnetic beads coated in streptavidin and the cells binding to Anti – CD146 antibody labelled with biotin. An extra step was incorporated to detach the cells from the beads by incubating the bead bound cells in a release buffer to break the streptavidin – biotin bond hence releasing the cells. The previous protocol for isolating CECs from whole blood using this method had beads attached to the cells at the end of the isolation process. Bead free cells allow for further feasibility studies such as cell / Tissue culture to be carried out on positive isolated bead free cells.
When HUVECs were isolated from culture using this method, there was a percentage cell loss of over 50% by the end of the isolation process. A significant drop in cell numbers was observed between bead bound cells and bead free cells inferring lack of efficiency in the release buffer or in the technique of detaching the cells.
Enumerating CECs from whole blood after fluorescent staining with Ulex europeus lectin yielded a count of 64 CECs/ ml of whole blood in one volunteer. Incubating the cells with release buffer to detach the beads from the cells again saw a significant drop in the cell numbers.
Culturing Bead bound and bead free HUVECs post isolation saw a higher proliferation rate with the bead free cells as compared to that of the bead bounds cells. The growth rate of the bead bound cells increased with each media change as the beads were washed off.
Post isolated CECs in culture failed to grow. 

Keywords: Circulating endothelial cells, Human Umbilical Vein Endothelial Cells,    Immunomagnetic isolation, Whole blood.                                                                                                   


My most sincere gratitude goes out to Dr. Francesca Arrigoni for giving me the opportunity to work on this project and for her guidance and direction throughout the course of the project.

Special thanks to Ammara Abdullah for her hints and tips, support and technical assistance and to Blerina Ahmetayj for her assistance in initiating the Human umbilical vein endothelial cell (HUVEC) culture.

List of Figures………………………………………………………………………………….....IX
List of Tables…………………………………………………………………………………........X

        1.3.3 Functional Heterogeneity. 10
   1.4 Endothelial Cell Function and Dysfunction 12
   1.4.1 Vasoregulation. 12
   1.4.2 Coagulation. 15
   1.7 Cell Density, Signalling and Proliferation. 32
2.0 Aim.. 34
    2.1 Objectives. 34
    2.2 Materials and Methodos. 35
       2.2.1 Subjects. 35
       2.2.2 Obtaining Whole Blood. 35

Appendix A
Appendix B
Appendix C
Appendix D
Appendix E
Appendix F
Appendix G
Appendix H

Figure 1: Structural heterogeneity of the endothelium showing the continuous, discontinuous and          fenestrated endothelium.
Figure 2: Mechanism of membrane transport of albumin by caveoli across the endothelium.
 Figure 3: The normal pulmonary endothelium in its quiescent state and when activated due to a malfunction.
Figure 4: Development of endothelial cells from the early angioblast.
Figure 5: Tight and adherens junctions of the endothelial cell necessary for controlling and maintaining the permeability of the endothelium.
Figure 6:  Regulation of vasodilation by Nitric Oxide. 
Figure 7: The coagulation Cascade showing the events that occur to the endothelium once activated changing from its antithrombic state to the active prothrombic state.
Figure 8: Mechanisms of the passage of molecules across the endothelium.
Figure 9: The four stage multistep process of the extravasation cascade.  Rolling, stopping, activation and transmigration.
Figure 10: The process of atherosclerosis.
Figure 11: Mechanism of circulating endothelial cell detachment from the vessel wall.
Figure 12: Principle of Immunomagnetic isolation using streptavidin coated beads
Figure 13: The cell cycle.
Figure 14: The proliferative stages of Human Umbilical Vein Endothelial Cells (HUVECs) in culture from initiating culture from cryopreserved cells till they are harvested for use.
Figure 15:  Cell counts on a haemocytomer at each stage of the isolation process.
Figure 16: Cell count from all 40 lines on a nageotte chamber before and after detachment of beads.
Figure 17: A Q-Q Plot showing the distribution of the cell numbers
Figure 18: Average cell counts before and after detachment of beads on a nageotte chamber.
Figure 19: Appearance of bead bound cells and bead free cells before culture and during culture.
Figure 20: An average frequency distribution chart showing the number of Dynabeads attached per Human Umbilical Vein Endothelial Cells (HUVEC) counted over a 100 bead bound cells. B) Beads attached onto the cell surface at varying frequencies. C) x40 magnified image showing degree of binding on a cell.
Figure 21: Monocytes, Leukocytes and Platelets adhered to a tissue culture plate after incubating blood for 2 hours at 370C.
Figure 22 Ulex europeus staining of circulating endothelial cell.

Table 1:  Phenotypic profiling of endothelial cells.
Table 2: Heterogeneity of endothelial cells in different adult tissues and organs.
Table 3:  Examples of Functional Heterogeneity of the Endothelium in Normal Adult Vasculature.
Table 4: Vasoregulatory substances synthesised by the endothelium and their effects.
Table 5: Substances involved in the regulation of haemostasis and thrombosis by the endothelium.
Table 6: Reported numbers of Circulating endothelial cells in health and disease using Immunomagnetic bead isolation.
Table 7:  Reported phenotypic differences between Circulating endothelial cells in different disorders.
Table 8: Endothelial cell markers present on other non-endothelial cell surfaces.
Table 9: Comparison of Immunomagnetic bead capture method and flow cytometry for isolating circulating endothelial cells.
Table 10: Comparison of two Immunomagnetic bead separation methods.
Table 11: Cell counts on a haemocytomer at each stage of the isolation process.
Table 12: SPSS output of a Paired sample t-Test comparin the mean cell counts before and after detachment of beads
Angiotensin-converting enzyme
Cell adhesion molecule
Cluster of Differentiation
Circulating endothelial cell
Cyclic guanosine triphosphate
Central Nervous System
Cardiovascular Disease
Dimethyl sulfoxide
E -
Endothelial Cell
Endothelium-derived growth factor
Ethylenediaminetetraacetic acid
Endothelial Growth media
Endothelial Progenitor cell
Foetal bovine serum
Fibroblast growth factor
Guanylate Cyclase
Granulocyte- colony stimulating factor
Granulocyte-macrophage colony-stimulating factor
Guanosine triphosphate
Human Leukocyte antigen
Human umbilical vein endothelial cell
Inter cellular adhesion molecule
Insulin – like growth factor
Junctional adhesion molecule
Low density lipoprotein
Leukotriene B -4
Magnetic Cell Sorting
Major Histocompatibility Complex
Sodium Acetate
Nitric Oxide
Nitric Oxide Synthase
Platelet activating factor
Plasminogen activator inhibitor
Platelet-derived growth factor
Platelet/endothelial cell adhesion molecule
Tissue Growth Factor
Tissue plasminogen activator
Vascular Cell adhesion molecule
VE -
Vascular endothelial
Vascular endothelial growth factor
Vesiculo-vacuolar organelles
Von willebrand Factor

1.1 Function of the Endothelium
The endothelium refers to the inner cell lining of the blood and lymphatic vessels and is made up of a monolayer of endothelial cells (1, 2). It forms a structural barrier between the vascular space and the tissues (3) with the primary essential function of maintaining permeability of the vessel wall (4). The cell surface in an adult has about 1 – 6 x 1013 cells, an approximate weight of 1kg and covers a surface area of about 1 – 7cm2(2 - 5). Each endothelial cell is between 25 – 50μm in length, 10–15μm in width and up to 5μm in depth (3). They contain a central nucleus and are aligned in the direction of blood flow in straight segments of the arteries but not at branch points. (1)
Endothelial cells line the entire vascular system and are found in almost all body parts including the cardiac muscle, skeletal muscle, subcutaneous tissue, pancreas, intestine, pituitary glands, lungs and cerebral capillaries (6). The endothelium exhibits structural and functional heterogeneity and the endothelial cells found in each of these body tissues/ organs vary in their characteristics. The location of the endothelial cell determines its structure which in turn defines its function. They could be continuous, others discontinuous, some fenestrated, vary in thickness, permeability and uniformity (1, 7) (Figure 1)1.  
Endothelial cells are highly metabolically active and play a crucial role in physiological activities related to metabolism, coagulation, transport, vasoconstriction and vasodilation.  These include the regulation of transendothelial permeability, haemostatic balance, proliferation, blood flow, and inflammation (1, 8 -9).
The endothelium is involved in almost all disease states either as a primary determinant of pathophysiology or as a victim of collateral damage (1). The activated endothelium produces cytokines and other growth factors such as platelet-derived growth factor (PDGF), basic fibroblast growth factor (bFGF) and insulin-like growth factor (IGF) which trigger the excess proliferation of smooth muscle cells. (3).
The characteristics of the endothelium make it an attractive therapeutic target serving as a channel of communication to the various body organs. This remains a key area of scientific interest as its full therapeutic potential is barely tapped. A suitable diagnostic technique to phenotype endothelial cells is yet to be uncovered mainly due to its heterogeneity and diverse functions. 

Figure 1: Continuous, discontinuous and fenestrated endothelium. The structure differs based on the location. They have pores/ gaps in the structure to aid with their primary permeability role. The continuous endothelium also found in the CNS, lymph nodes and muscle plays a role in exchange and transport as well as maintaining the blood brain barrier and leukocyte trafficking. The continuous fenestrated endothelium aids with secretion, absorption and filtration. The discontinuous fenestrated also found in the bone marrow and spleen plays a role in the exchange of particles, blood cell processing and hemopoisis. The clathrin coated pits is also necessary for receptor mediated endocytosis (1).
1.1.1 Structure of the lungs in relation to function: A case study
The heterogeneity of the endothelium with regards to structure in relation to function is typical of the lungs. The adult lung weighs about 1 kg of which 40% - 50% is blood (10). The pulmonary endothelium forms a continuous layer on the luminal surface of the lungs vasculature which is metabolically active (11). The structure of these endothelial cells varies in the various pulmonary vasculatures (12, 13).
The lungs serves as the gateway between the non-systemic and systemic circulation by filtering the entire blood before it enters the circulatory system as well as maintaining and regulating the entry of hormones, enzymes and other active substances to their target organs. (11).  
The main function of the lungs is the exchange of gases. Other functions include filtration of blood flowing to the systemic circulation and of microthrombi arriving from systemic veins, excretion or deactivation of serotonin, bradykinin, norepinephrine, acetylcholine and drugs such as propranolol and chlorpromazine. They also play a role in the metabolism of metabolically active peptides in the circulation such as the conversion of angiotensin I to angiotensin II (11, 14-15).
The vast majority of the wall of the airways is composed of alveoli which create the surface area for the exchange of substances between lungs and its surrounding environment (Figure 2) (9, 11). Caveoli are vesicle carriers invaginated in the cell membrane which play a role in transcellular transport of molecules across the endothelium.

Figure 2: Mechanism of membrane transport of albumin by caveoli across the endothelium. A) Albumin binds to its receptor protein, glycoprotein gp60 which associates with the scaffolding protein caveolin-1. B) This activates the enzyme Src Kinase which causes the phosphorylation of the proteins caveolin-1 and dynamin causing the alveolae to engulf the albumin and transport it across the endothelium (c) (9).

The barrier separating the pulmonary capillaries from the alveolar air is composed of endothelial cells, an attenuated interstitial space and pulmonary epithelial cells (11). The pulmonary endothelium is uniquely positioned to execute its functions. It produces a variety of vasoactive mediators in response to changes in blood Oxygen and Carbon dioxide tensions, pressure, and flow such as nitric oxide which is a key player in the regulation and maintenance of vasomotor tone (12). It also plays a role in homeostasis, leukocyte trafficking, transduction of luminal signals to abluminal vascular tissues, secretion of growth factors, activation of cell signals with autocrine and paracrine effects as well as its barrier function (13).
A dysfunction in the endothelium of the pulmonary vasculature leads to diseases such as pulmonary hypertension (13) (Figure 3).

Figure 3: The normal pulmonary endothelium in its quiescent state plays key roles in homeostasis, leukocyte trafficking, transduction of luminal signals to abluminal vascular tissues, secretion of growth factors, activation of cell signals with autocrine and paracrine effects and barrier functions. When the endothelium is activated due to a malfunction, chemokines, cell markers and growth factors are activated which disturbs the normal homeostatic balance triggering a sequence of events which leads to the disease state. Squares indicate endothelial cells; ovals, smooth muscle cells; and closed circles, platelets. (12)

1.2 Origin of endothelial cells
Endothelial cell development begins with hematopoiesis after implantation of an embryo when blood islands begin to form within the primitive yolk sac (4, 16).  The cardiovascular system is the first organ system to develop. Subsequent developments involve vasculogenesis and then angiogenesis. Vasculogenesis refers to the de novo organisation of endothelial cells into vessels in the absence of any pre-existing vascular system and only occurs in the early embryo (4). Angiogenesis on the other hand occurs in the avascular regions of the embryo and involves the continual expansion of the vascular tree from pre-existing vessels. (3-4). This is thought to be one of the key processes involved in the progress of pathological conditions such as neovasculization of tumour cells, the proliferative phase of diabetic retinopathy and inflammatory diseases such as rheumatoid arthritis (3).
The exact mechanism of differentiation of endothelial cells from the early mesoderm is however unclear. Knowledge of this mechanism is of interest as the inhibition of angiogenesis could be a potential therapeutic strategy in the regulation of angiogenesis (3)
Endothelial cells are postulated to originate from the same precursor cell as hematopoietic stem cells, angioblast / hemangioblast as studies observed a close developmental association between the haematopoietic and endothelial lineages (16-17) (Figure 2). The angioblast form the outer layer of the endothelial cells encasing the blood island whiles the hematopoietic stem cells form the inner cluster from which the first embryonic blood cells develop (4).
The Precursor cell development is thought to arise from the ventral floor of the dorsal aorta within the aorta-gonad-mesonephros region. Splanchno- pleuricmesoderm transforms into mesenchymal cells that differentiate into the hemangioblast. The hemangioblast transforms into an intermediate pre-endothelial cell that then differentiates into either a committed hematopoietic cell line or an endothelial cell” (2).
Certain growth factors have been identified from experiments carried out in mouse embryos as playing a role in the regulation of endothelial cell differentiation and the subsequent development of the vessels. Fibroblast growth factor (FGF) and vascular endothelial cell growth factor (VEGF) are two of such growth factors (18).

Figure 4: Development of endothelial cells from the angioblast. Endothelial cells differentiate from mesodermal precursors. Vasculogensis is the development process from the early embryonic yolk sac and angiogenesis is the continual development and expansion of the vascular tree.  The process is induced by growth factors (2).

1.3 Endothelial Cell Heterogeneity
1. 3.1 Surface markers of endothelial cells
A unique characteristic amongst endothelial cells is the presence of Weibel – Palade bodies.
Weibel – Palade bodies were first identified by Ewald R. Weibel as rod shaped bodies associated with vascular/blood physiology whose function remained uncertain at the time (19).  They have since been defined as membrane-bound, elongated vesicles of 0.1 x 2-3µm in size that contain regularly spaced tubular structures aligned parallel to the longitudinal axis and serve as a storage and/or processing vesicles for von Willebrand protein (vWF)(20). vWF protein is a large multimeric glycoprotein synthesized by megakaryocytes that mediates adhesion of platelets to the sub endothelium after vascular damage (20).
Based on the location of endothelial cells, phenotypic variations are observed even within the same individual such that cells from different locations express different markers and can generate different responses to the same stimulus (2). (See table 1 for phenotypic profiling of endothelial cells). The Heterogeneity of the endothelium has made attempts to phenotype the cells an elusive task. Having a specific phenotype for all endothelial cells would allow for their easy detection and distinction in health and disease.
Some endothelial cell markers have been identified and serve as a marker for the clinical identification of certain diseases. The earliest markers used in the identification of endothelial cells were Angiotensin converting enzyme (ACE), von Willebrand factor (vWF), Vascular endothelial growth factor receptor - 1 (VEGFR 1) and Vascular endothelial growth factor receptor - 2
(VEGFR -2) (17). The current more accepted and widely used markers and adhesion molecules include Cluster of Differentiation 31(CD31) also known as platelet adhesion cell molecule – 1 (PECAM – 1), Cluster of Differentiation 34 (CD34), P- Selectin, Vascular endothelial – Cadherin (VE – Cadherin), von Willebrand Factor (vWF), Endothelial Selectin (E-Selectin), Vascular cell adhesion -1 (VCAM-1) and Inter cellular adhesion molecule (ICAM-1). Most of these markers are not exclusive to endothelial cells. ACE, vWF, PECAM-1 P-Selectin and CD34 are also present in megakaryocytes, platelets and other hemopoietic cell types whereas VE-Cadherin and the VEGF receptors are more endothelial cell specific (17).
Table 1: Phenotypic profiling of endothelial cells. (2, 7)
Protein Marker
Thrombomodulin, Protein C, Protein S, PGI2 Synthesis
PGI2, Thrombomodulin,
heparin sulphate
2, 7
Pro coagulant
Extrinsic pathway initiation
VIII activation, platelet adhesion
vWF, Factor V, TXA2 Thromboplastin, PAF,
 PAI – 1, PAI – 2.
2, 7
Leukocyte adhesion
Adherens junctions
Tight junction
VE-cadherin, E- selectin, P- selectin, ICAM-1, ICAM-2, VCAM-1, PECAM -1, CD34, CD44
2, 7
Angiogenesis and Proliferation
2, 7
Lipid Metabolism
Lipid metabolic pathway
LDL- receptor, lipoprotein lipase
2, 7
Renin / angiotensin system
NO system
Endothelin -1,Prostacyclin
2, 7
Inflammatory and Immune response
H- antigen
HLA system
IL-1, IL- 6, IL-8.
MCP-1, MCP-2
2, 7

1.3.2 Structural heterogeneity
Aside the different markers expressed on the surface of the various endothelial cells, they also exhibit structural differences depending on the location in the vascular tree which in turn defines its function (17) (See Table 2). Human umbilical vein endothelial cell (HUVEC) appears flat and elongated in culture but other endothelial cells vary in shape based on the location. For example in high endothelial venules, they are plump and cuboidal;  they are broad, short and rectangular in shape in the pulmonary artery; large and round in the pulmonary vein; long narrow and rectangular in the inferior vena cava(1). Thorin et al reported distinct differences in the growth rate, protein synthesis, endotholin – 1 release and responsiveness to endogenous factors in human endothelial cells in culture (21). They also vary in size from 0.1μm in capillaries and veins to 1μm in the aorta (1). Even with the primary permeability function of the endothelium, the different vascular beds exhibit different characteristics.
The microvascular endothelium in an adult can be grouped into three phenotypes on the basis of morphology, namely, continuous, discontinuous and fenestrated. With regards to structural relation to function, the microvasculature in the brain and retina is continuous with tight junctions and plays a key role in maintaining the blood brain barrier. That in the liver is discontinuous and necessary for the efficient clearing of molecular entities in the body. Likewise, endothelial cells lining the bone marrow and spleen sinusoids are discontinuous and allow cellular trafficking between intercellular gaps.  The glomerular endothelium in the kidney is lined by fenestrated endothelial cells and acts as a semi-permeable membrane for the filtration of blood-borne components, whiles that in the descending and ascending vasa recta or peritubular capillaries engage in the re-absorption and excretion of components respectively from the blood circulation. Endothelial cells in the intestinal villi and endocrine glands are also fenestrated and facilitate the selective permeability required for efficient absorption, secretion and filtering (1, 4, 17, 22).
The mechanism underlining endothelial cell heterogeneity has been attributed to genetic factors and micro environmental influences from soluble mediators, mechanical forces, the extra cellular matrix and cell – cell interactions (24). The physical and chemical environment between organs and even along a single blood vessel within the same organ is different thereby response to the same stimuli will be different. William Aird’s theory of nurture versus nature proposes that the phenotypic variation of the endothelium is as a result of evolution from the ability to sense and respond to the extracellular environment and site specific epigenic modification (25)

Table 2: Heterogeneity of endothelial cells in different adult tissues and organs (17)
Tissue/ Organ
Central nervous system
Low number of vesicles, complex tight junctions
Blood-brain barrier
Lymph nodes
High endothelial venules
Lymphocyte homing
High number or vesicles
Endocrine glands
Gastrointestinal tract
Choroid plexus
Kidney glomeruli
Large gaps
Exchange of particles
Splenic sinus of red pulp
Blood cell processing

1.3.3 Functional Heterogeneity
The structure of the endothelial cells in the different locations is suited to its functions (Table 3).
The primary function of the endothelium is the maintenance of vessel wall permeability. In support of this, endothelial cells possess clathrin coated pits, clathrin coated vesicles, multivesiclar bodies and lysosomes, which represent the structural components of the endocytotic pathway (1). The endocytotic pathway is the route via which macromolecules are transported to lysosomes for degradation. This could be dependent or independent of receptors. Endothelial cells also possess caveolae and vesiculo-vacuolar organelles (VVOs) which represent the structural components of the transcytotic pathway (1). The transcytotic pathway refers to the transcellular transfer of molecules across the endothelium. It involves the uptake of molecules via endocytosis at one end, transport across the cell via vesicular carriers and exocytosis at the opposite end (26). Caveolae are 70nm membrane – bound, flask shaped vesicles with a smooth inner surface which open to the luminal or abluinal side but are occasionally free in the cytoplasm (26).
In aid of endothelial permeability function, they also possess tight and adherens intercellular junctions (Figure 3). Tight junctions which are usually found at the apical region of the intercellular cleft form a barrier to transport between endothelial cells and help maintain cell polarity between the luminal and abluminal side of the endothelial cells (1). Adherens junctions are cellular membrane contacts formed by cadherins as transmembrane glycoproteins that mediate the physical attachment between cell membrane and an intracellular undercoat network of cytoplasmic proteins and actin microfilaments (23).
Other functions of the endothelium include, maintenance of haemostasis, leukocyte trafficking, regulation of the vasomotor tone, proliferation and angiogenesis. These functions are performed by specific subsets of blood vessel types or vascular beds (1). A few of the endothelial cell function and what happens to the dysfunctional endothelium will be reviewed in the next section.

Figure 5: Tight and adherens junctions of the endothelial cell necessary for controlling and maintaining the permeability of the endothelium. The junctions are made up of transmembrane proteins and their corresponding intercellular molecules. Cell adhesion at these junctions is mediated by adhesion molecules and proteins (23).

Table 3: Examples of Functional Heterogeneity of Endothelium in Normal Adult Vasculature (1).

1.4 Endothelial cell function and dysfunction
1.4.1 Vasoregulation
One of the key functions of the endothelium is the regulation of vascular flow and basal vasomotor tone. This is mediated by vasoactive substances activated in response to humoral and mechanical stimuli (See Table 4). The mechanism of regulation occurs both dependently and independent of the endothelium. With endothelial-dependent regulation, endothelial cells produce vasoconstrictors such as endothelin 1, thromboxane A2, platelet-activating factor, and vasodilators such as nitric oxide and prostacyclin (Figure 4) (9). Of these substances nitric oxide and endothelins are the main regulators of the basal vascular tone and the others come into play only when vascular function / hemodynamics is disturbed (3). The regulatory mechanism independent of the endothelium includes vasoconstriction via the sympathetic nervous system and vasodilation via gene regulated peptides such as calcitonin (19).

Table 4: Vasoregulatory substances synthesised by the endothelium

Nitric Oxide (NO) is a heterodiatomic free radical produced by the oxidation of L-arginine to L-citrulline by the enzyme Nitric oxide Synthases (NOS). NO induces vasodilatation by stimulating soluble guanylate cyclase to produce Cyclic GMP (4, 27). An isoform of NOS known as endothelial NOS (eNOS) is constitutively active in endothelial cells. eNOS is further stimulated by receptor-dependent agonist such as thrombin, adenosine 5 – diphosphate, bradykinin, muscarinic and substance P. Stimulated eNOS increases the intracellular calcium and perturb the plasma membrane phospholipid asymmetry. eNOS is the most potent endogenous vasodilator known (3, 4, 27). Shear stress and cyclin strain also stimulate eNOS activity as a result of a shear response consensus sequence, GAGACC in the promoter of the Nos3 gene. An example of this shear stress which stimulates eNOS activity is during exercise where the activated eNOS contributes to the phenomenon of flow mediated vasodilatation. This is the mechanism by which blood flow increases in response to exercise.   Another isoform of nitric oxide believed to be derived from endothelial cells after stimulation by cytokines is inducible nitric oxide (iNOS) also known as the Nos2 gene (27).
Nitric oxide is not only involved in vasodilatation. It also has antiplatelet and antimitogenic properties (27) and plays a vital role in the inhibition of thrombosis by inhibiting platelet adhesion, activation, secretion and aggregation. It also promotes platelet desegregation and inhibits leukocyte/ endothelial cell adhesion as well as the proliferation and migration of smooth muscle cells (3).

Figure 6: Maintenance of vasodilation by NO.  The endothelium contains type II and III NOS which catalyses the production of NO from L-arginine. NOS is activated via receptor mediated changes or in response to changes in shear force brought about by changes in intracellular calcium. NO   subsequently activates   guanylate cyclase (GC) in smooth muscle cells. This is a soluble enzyme which converts GTP to cGMP. cGMP in turn activates a protein kinase which brings about changes in the levels of calcium in the smooth muscle cell leading to changes with the end result of vasodilation (9).  

The other main regulator of vascular flow and basal vasomotor is the vasoconstrictor endothelin – 1 (ET-1). Endothelin -1 is one of three 21 amino acid peptides produced by different cell types (3). Endothelial cells only produce endothelin – 1 which is also secreted by the vascular smooth muscle cell. The other two types, endothelin -2 and endothelin -3 are involved in the regulation of cellular proliferation and hormone production (3). The secretion of Endothelin – 1 begins with transcription of the inactive precursor of endothelin – 1 after stimulation by hypoxia, shear stress and ischemia. The endothelin -1 produced then binds to endothelin receptor A coupled to G- protein. The type A receptor is expressed on the vascular smooth muscle cell and cardiac myocytes (3, 28). Binding of endothelin -1 to the abundant G-protein - coupled ET-A receptor is thought to be the main mechanism behind ET-1 induced vasoconstriction. This bond causes an increase in intracellular calcium concentration which in turn increases vascular smooth muscle cell tone (28).  The vasoconstriction effect is maintained even after ET-1 dissociates from the ET-A receptor through long – lived effects on intracellular calcium. Nitric oxide and prostaglandin counteracts this effect and acts to restore the intra cellular calcium to basal levels (28).

1.4.2 Coagulation
The normal vasoregulatory properties of the endothelium provide an antithrombic surface that inhibits platelet adhesion and clotting. When activated either by physical or chemical factors, the cells undergo a series of chemical changes converting the antithrombic surface to an active prothrombotic surface. The injured endothelium returns to its antithrombic state once the prothrombotic cascade ceases. The activated endothelium recruits platelets, monocytes and neutrophils which play a role in initiating / amplifying the coagulation process. A key player in initiating this coagulation cascade is tissue factor enzyme which catalyses a series of events leading to the activation of prothrombin. Prothrombin is in turn converted to thrombin which further converts fibrinogen to fibrin. Thrombin also stimulates the procoagulant pathway on the endothelial cells and is responsible for the activation of platelets and other coagulation enzymes and cofactors.
Platelets are crucial to the formation of clots on the damaged endothelial cell layer. This requires the interaction of platelets with vWF (a well-known marker of endothelial cells) bound to collagen on the sub endothelial layer (3, 29, 30). (See table 5 for regulators of haemostasis and thrombosis by the endothelium).
Table 5: Regulation of haemostasis and thrombosis by the endothelium (4).

Figure 7: Cascade of event that occurs once the endothelium is activated changing from its antithrombic state to the active prothrombic state. A) The normal endothelium in its anti thrombic state. The endothelial cell surface express TM and EPCR which reacts with thrombin to produce activated PC (APC). The cell surface also secretes tissue-type plasminogen activator (tPA), which work together with TF pathway inhibitor (TFPI) and antithrombin (AT) on to the cell surface to promote fibrinolysis. 
B) The damaged endothelium triggers a series of events leading to coagulation. The role of the anti-coagulant proteins TFPI, AT, EPCR, and TM expressed on the cell surface is impaired and APC and AT deactivated. Fibrinolysis is disrupted and platelets, vWF and other prothrombotic factors are activated promoting coagulation (30).

There is a whole series of anti-coagulant and procoagulant events which take place during this process (Figure 7). Due to the nature of this report it will not be discussed in depth.


1.4.3 Permeability
The endothelium is a semipermeable barrier that controls the passage of fluids and solutes across the different vascular beds. This semi permeable property distinguishes the endothelium from the epithelium and is necessary for maintaining the fluid homeostatic balance. The passage of water insoluble substances from the blood into the interstitium depends on endothelial permeability and often on specific carrier proteins (31).
In the capillaries there is a continuous exchange of fluids and solids between blood and the underlying interstitium. In the arteries and blood-brain barrier, permeability is regulated by the tight and adherens junctions. The number and complexity of these junctions is inversely proportional to the permeability (31). The continuous fenestrated endothelium has high permeability of fluids and small solutes but not macromolecules.  The transport of macro molecules across the endothelial barrier is   dependent on their size as well as the barrier properties of the particular endothelium. This size-selective nature of the barrier to plasma proteins is a vital in creating protein gradients needed for maintains fluid balance of tissues. In addition, plasma proteins, such as albumin, act as circulating chaperones for hydrophobic substances, fatty acids, and hormones, molecules whose transport is crucial for cell functions vital to the organism (32).
Loss of the endothelium permeability function results in tissue inflammation (32).

Figure 8: Passage of molecules across the endothelium. Mechanism of passage is dependent on the size of the molecules and the vascular bed. Large molecules such as albumin require the help of receptors and are transported across the transcellular route. Water molecules pass freely across aquaporins and smaller molecules such as urea and glucose are transported via the paracellular route (31)

1.4.4 Leukocyte trafficking
The normal functional endothelium constantly adapts to its local extracellular environment such that when there is a slight sense of foreign invasion or tissue damage, the endothelium is activated and endothelial cells are induced locally to release inflammatory mediators. This in turn recruits leukocytes from the intravascular space to extravascular sites of infection (3). Injury to the endothelial cell wall also activates the coagulation pathway and recruits platelets and other molecules to the injury site as previously described.
Leukocyte trafficking describes this process of inflammatory response whereby lymphocytes and leukocytes are recruited to the site of infection to combat the inflammation. Also known as the extravasation cascade, it involves a four stage multistep process of rolling, stopping, activation and transmigration (33). (Figure 4). 

Figure 9: Stages of the extravasation cascade. 1. Leukocytes roll along the endothelium due to interactions between
selectin on the endothelium and siayl-lewis on the leukocyte.  2. Cells of the activated endothelium express selectins on their surface. 3. Interaction between integrins on the Leukocytes and adhesion molecules on the endothelium enables tight binding of the Leukocytes onto the endothelium. 4. Chemotactic gradient is created due to the secretion of chemokines and integrins causing the leukocytes to migrate towards the centre of infection along the chemotactic gradient (34).

 Lymphocytes interact with the endothelial cells through the L-selectin receptor and also express integrins that interact with endothelial cell adhesion molecules, intramolecular cell adhesion molecules and vascular cell adhesion molecules.  These adhesion molecules include selectins, integrins, adhesion molecules of the junction adhesion molecule (JAM) family (JAM 1 – 3), those of the immunoglobulin super family, endothelial cell adhesion molecules (VCAM-1 and CD31), CD39, VE-Cadherin Occludin, Claudins and other variants of the CD44 family (35).
Selectins are transmembrane glycoproteins that recognise cell – surface carbohydrate ligands found on leukocytes. There are three identified molecules (L- selectin, P-selectin E-selectin) all involved in leukocyte trafficking. They each have an amino – terminal ca2+ -dependent lectin domain, an endothelial growth factor (EGF) domain, a transmembrane domain and a cytoplasmic tail (9).
During inflammation, vasodilation occurs and the activated endothelium increases in permeability induced by TNFα, lipopolysaccharides or interleukins to allow for the influx of leukoctytes to the inflammation site (28). The process begins with the tethering and rolling of leukocytes on the endothelial cell surface. This is mediated by E- selectin and P selectin. Rolling is followed by activation and firm adhesion on the cell surface brought about by interactions between leukocyte integrins {(leukocyte function associated antigen -1 (LFA-1) or very late antigen – 4 (VLA-4)} and adhesion molecules I-CAM-1 and VCAM -1. The leukocytes then flatten and migration follows via the process of diapedesis. The exact molecular mechanism underpinning the process of migration is not fully understood but is known to be mediated by additional adhesion molecules (1, 3).
One proposed mechanism is believed to be as a result of the process of vasodilation which occurs during inflammation and causes an influx of intra cellular calcium within the endothelial cells. This influx is restricted to endothelial cells adjacent a transmigrating leukocyte. The enzyme Myosin kinase light chain is activated which causes phosphorylation of myosin II and it undergoes conformational changes. These changes causes the actin – myosin bundles to contract which in turn puts a strain on the actin filaments near the cell border causing retraction of the endothelial cell at this border and allowing the passage of leukocytes (28).
Extravasation occurs by the migration of the leukocytes (towards the chemotactic gradient created by the secretion of cytokines and other inflammatory mediators) through the endothelial cell junctions towards the site of inflammation (36)
During the leukocyte trafficking process, endothelial cells may undergo necrosis or apoptosis as tissue is reabsorbed and repaired. Dysfunction in this case occurs where there is net liability to the host as seen in the case of severe sepsis where there is an excessive, sustained, and generalized activation of the endothelium (37). The continuous inflamed endothelium also leads to the events which mark the beginning of atherogenesis.

1.5 Endothelial cell dysfunction and Atherosclerosis
Atherogenesis marks the beginning of Atherosclerosis which is a chronic inflammatory process affecting arteries (38). For as long as there is damage to the endothelium leading to inflammation, the endothelium remains activated and leukocyte trafficking continues.  As leukocytes and monocytes are continually recruited to the affected sites they begin to accumulate and with time transform into foam cells (macrophages) and eventually form a plaque. The plaque formed is also as a result of inflammatory mediators enhancing the uptake of lipoprotein particles and formation of lipid – filled macrophages. Smooth muscle cells are also attracted to the site and a fatty streak is formed. As the process continues the smooth muscles together with collagen form a fibrous cap on the outside of the plaque while the initial macrophages and monocytes become necrotic (39) (Figure 10).
This plaque is vulnerable and is subject to rupturing via the action of macrophages which can weaken the connective tissue framework of the fibrous cap of the plaques by secreting extracellular matrix degrading enzymes. The ruptured plaque induces acute clinical syndromes such as thrombosis and acute coronary syndrome (39 -41).

Figure 10: The process of atherosclerosis. Damage to the endothelium triggers leukocyte trafficking. The continuous influx of leukocytes and monocytes leads to the formation of foam cells and eventually the fibrous cap on the outside with a necrotic core. Rupturing of the plaque releases the necrotic cells into the circulation causing diseases such as stroke (43, 44).

The lesions formed on the endothelial cell surface impede the normal blood flow which begins the atherosclerotic process. Atherosclerosis can develop into other diseases such ischemia and stroke (41). Lack of the normal laminar shear stress from disruption of blood flow may reduce the local production of endothelium derived Nitric Oxide which has anti- inflammatory roles aside that of vasodilation and may cause a reduced expression of VCAM - 1. Disruption of the normal flow may also increase the production of certain adhesion molecules such as ICAM-1 thereby aggravating the inflammatory response. Thirdly it may enhance the proliferation of smooth muscle cells of proteoglycans which can clutch onto lipoprotein particles and promote their oxidative modification thereby stimulating an inflammatory response (42). This final mechanism is the more accepted hypothesis as the genesis of the atherosclerotic process. It is postulated that the increased permeability of the vascular endothelium once activated triggers the penetration of low density lipoproteins into the intima of the arterial wall where they are oxidised. The oxidised low density lipoprotein (LDL) is recognised by the immune system as foreign triggering an immune response. In this response monocytes are also recruited to the affected site where they are converted to macrophages which engulf the oxidised LDL via phagocytosis. This however further transforms the macrophages into lipid – laden / foam cells which the immune system recognises as foreign and so the process of leukocyte trafficking including recruitment of monocytes continues (43).
A.    B. C.  D.

1.6 Circulating endothelial cells (CECs)
CECs are mature endothelial cells that have become detached from the vessel wall (45). They are a part of the non-hematopoietic cells in the blood but the exact anatomic origin is unclear although there have been suggestions of it arising from the microvascular or the macrovascular portion of the vascular tree (46). Their exact origin and the mechanism of their detachment from the vessel wall are also unclear. These mature detached cells could be apoptotic or necrotic depending on the situation causing their detachment.
The proposed mechanisms of detachment include;
1.                      Defective adhesion properties of the endothelial cell
2.                      Action of proteases and or cytokines
3.                      Mechanical injury to the vascular wall (46)

Mechanism of detachment
Figure 11: Mechanism of CEC detachment from the vessel wall. Injury to the endothelial cell disrupts the adhesive properties of the endothelium releasing the cells into circulation. Cadherins, fibronectin, vitronectin and integrins are thought to protect against detachment. Their mechanism of clearance from circulation once released is unknown (45, 47 - 49).

Circulating Endothelial Cell have recently become of interest since being identified as a key marker in the assessment of endothelial injury and repair (45 -48). Other markers of endothelial injury include thrombomodulin, von Willebrand factor, E-selectin and endothelial microparticles. Several articles have reported elevated levels in vascular associated disease and very low levels in health. (See table 6).
The average number of CECs/ ml of blood in health are about 5 cells / ml as they are usually in the senescent state (63). The elevated levels also correspond with plasma and physiological markers of endothelial damage / dysfunction such as flow mediated dilatation, vWF and soluble E-selectin (64). Endothelial progenitor cells (EPCs) and endothelial microparticles (EMPs) are other endothelial cells types of interest as marker of vascular disease. These differ from CEC in maturity, growth characteristics and cell surface expression amongst other things. Whereas CECs are mature, EPCs are immature and differ from the former by expressing Cluster of differentiation (CD) 133 and CD34. They also differ in that EPCs have the ability to proliferate readily in culture post isolation from polymononuclear cells (PBMCs) (16). Endothelial microparticles (EMP) are vesicles formed by the EC membrane after injury or activation, harboring cell surface proteins and cytoplasmic elements and expressing endothelial-specific surface markers reflective of parent cell status (65).
This report will be on CECs. There have been reported phenotypic differences between CECs in different diseases. Sheets of relatively intact cells have been isolated in patients with acute coronary syndrome and vasculitis whereas severely damaged necrotic cells, membrane fragments and smaller particles were detected in inflammatory disorders (63)

Table 6: Reported numbers of CECs in health and disease using immunomagnetic beads isolation.
Diseased state
Mean CEC numbers in health
Mean CEC numbers in disease
ANCA- associated small – vessel vasculitis
Septic shock
Acute myocardial infarction
Coronary angioplasty
No data
Congestive heart failure
Kawasaki Disease
Type II Diabetes
Sickle cell anaemia
13.2 – 22.8
Renal Transplantation
Bone marrow transplantation
Thrombotic Thrombocytopenia
6 – 220
Peripheral Vascular Disease
1.1 – 3.5

Table 7:  Reported phenotypic differences between of CECs in different disorders.
Acute coronary syndrome
Sheets of intact cells
Rickettsial infection and vasculitis
Severely damaged and necrotic cells
Cytomegalovirus infection
Giant cells

Elevated numbers are also indicative of the extent of the endothelial lesion (48). Diseases most associated with these elevated levels of CECs have endothelial injury being central to the pathogenesis e.g., sickle cell anaemia, Cardiovascular disease (CVD), Pulmonary vascular disease (PVD) and systemic vasculitis. The levels are also representative of the anatomical stage i.e. low levels in health, medium levels in localised damage as seen in patients undergoing coronary angioplasty and high levels in diseases with widespread vascular damage such as vasculitis (49).
CECs are currently defined phenotypically as expressing CD146, vWF and VE-cadherin in the absence of CD45 and CD133 (69). Of these markers, CD146 is the key marker for their detection although others markers are necessary in the case of cancer as CD146 is reported to be present on malignant tissues, trophoblast, mesenchymal stem cells and periodontal tissue (69).

Table 8: Endothelial cell markers present on other non-endothelial cell surfaces (48).
Expression by non- endothelial cells
Platelets, monocytes, neutrophils, T-cells
Activated endothelial cells
Endothelial cells, activated B and T lymphocytes, monocytes
Endothelial cells, activated monocytes, Tissue macrophages, erythroid marrow precursors
Activated endothelial cells, stromal cells
Endothelial cells, keratinocytes, platelets, monocytes, neutrophils
P1H12, S-endo-1
Endothelial cells, activated T-lymphocytes, melanoma cells, trophoblast
Tissue Factor
Endothelial cells, monocytes/macrophages

 1.6.1 Isolation and Enumeration of Circulating Endothelial Cells (CECs).
The identification of CECs has come a long way since Bouvier and Hladovec first described them in the 1900s. They employed techniques such as Giemsa staining, vital light microscopy and density centrifugation. These techniques were improved upon by use of Immunofluorescence with antibodies directed against endothelial markers such as von willibrand factor (48). To date, two main methodologies have been applied in the isolation and enumeration of CECs: immunomagnetic bead separation and flow cytometry with the former yielding more consistent data as it was originally devised to detect rare events in peripheral blood and hence performs well in detecting CECs which are rare in healthy blood (64).
The initial techniques established in the 1900s identified CECs from smears and solely on the basis of morphology and were far from ideal. The immunofluorescence technique also relied on smears and was mainly hindered by the lack of specific markers for endothelial cells (48). The recent advancements in the isolation and enumeration of CECs by immunomagntic bead separation and flow cytometry is still immature.  Enumeration of CECs by the immunomagnetic bead capture method is the most common and accepted technique. From published data, enumeration of CECs using this technique appears more grounded and consistent in comparison with that from flow cytometry. Both techniques however have their advantages and disadvantages. (See table 9).
For this project enumeration would be done by the immunmagnetic bead capture method. This method makes use of immunomagnetic polymer beads coated with antibodies directed against CD146 mainly S Endo 1 and P1H12.

Table 9: Comparison of Immunomagnetic bead capture method and Flow cytometry (64, 66).
Immunomagnetic isolation
 Flow cytometry
More time consuming requiring preliminary preparations. 
FACS approaches are less time
Consuming and more affordable for diagnostics
Consistent data
Varying cell numbers with broad range
Consensus protocol
Different protocols
Secondary stain needed
No need for secondary staining
Detects via a Single marker
Multi marker approach
Cannot differentiate between CEC and EPC
CECs can be differentiated from EPCs
Cells can only be enumerated without studying endothelial cell markers.
Endothelial cells markers can be studied.
Both positive and negative isolation (depletion of unwanted cells) can be achieved
Does not permit characterisation of the cell phenotype

1.6.2 Bioconjugation and the principle of Immunomagnetic Bead Isolation.
The technique of immunomagnetic separation makes use of small paramagnetic particles or in this case paramagnetic beads coated with a ligand usually antibodies which can be directed against a specific target usually a surface antigen on cells. (49). Isolation of target cells can be achieved from body fluids such as serum, ascites fluid or whole blood or from tissue digest or even cell/ tissue culture. Once the desired target cell type has been isolated, identification can then be carried out as per regular conventional methods such as immunohistochemistry.
Since the introduction of this separation technique it has proven vital in several research and medical applications (69). It has several advantages in comparison to other separation techniques such as chromatography. For example, the isolation of target cells can be achieved from a crude sample without the need of initial sample purification. Secondly most of the reaction takes place in a single tube reducing cell loss. Thirdly the process is simpler and less time consuming (70).
Another magnetic separation product on the market aside the dynal range is the magnetic activated cell sorting (MACS) system marketed by miltenyi. Both techniques are based on the same principle of isolation but differ in the sequence of events, the size of beads and the type of magnetic field. The dynal beads are of particulate size (4.5μM) and that of MACS are of colloidal size (50nm).   Due to the bigger size of the miltenyi MACS beads, a stronger magnetic field is required to attract the beads (71).
One challenge facing the use of both immunomagnetic separation techniques is the removal of the beads following positive isolation especially where further functional studies are required. To address this problem there is the availability of small biodegradable beads about 50nm in size and a recent development by dynal of incubating the bead bound cells in a release buffer (70, 71). This release buffer breaks the bond holding the two components together thereby releasing the cells. The introduction of this dynal product with the option of releasing the bead bound cells appears promising and will be tested out in this project.
The dynal magnetic beads used are based on the principle of bioconjugation between biotin and streptavidin. Streptavidin binds to four moles of biotin per mole of protein (73). This streptavidin –biotin bond is the strongest non-covalent biological interaction known, with a dissociation constant, Kd, in the order of 4 x 10-14M (73). The strength and specificity of this interaction has made it an attractive choice for use in molecular, immunological, and cellular assays where affinity pairing is desired.
The Dynal beads come coated in streptavidin. The antibody of your choice (CD146) is then labelled with biotin. When the cells are incubated with the biotinylated antibody, the antibody binds to the active sites on the cell surface forming a cell – biotin – antibody complex.
Addition of the Streptavidin coated dynal beads brings about the conjugation of the biotin to the streptavidin yielding the complex shown below. Passing the cell suspension through the magnet causes the bead bound cell to be attracted to the magnet. Upon addition of a release buffer the streptavidin – biotin bond is broken thereby releasing the cells. This methodology differs from that suggested by Woywodt et al (74) and that used but Clarke et al (44). (see table 10). (Refer to appendix C for the protocol using this method).
Figure 12: Principle of isolation using streptavidin coated beads.

Table 10: Differences between Imuunomagnetic bead separation methods.
Bead size
Principle of isolation
Streptavidin – biotin conjugate
Antibody – antigen conjugate
Bead preparation
Beads come coated in streptavidin and ready to use
Beads need to be prepared before use by coating with antibody of choice (anti-CD146)
Incubation time with cells
10 minutes
30 minutes
Blocking agent
No blocker used
20μl of FcR blocking agent added.
Detaching cells
Cells released
Cells not released
Bead concentration
15mg / ml
Volume of beads added
75μl per 2 ml of blood
50μl per 1 ml of blood
Starting sample
No volume of blood was discarded
First 7.5ml of whole blood was discarded.

  1.6.3 Therapeutic use of Circulating Endothelial Cells (CECs)
The varying heterogeneity of the endothelium and its ability to respond uniquely to different external influences makes it an attractive therapeutic target. To study its functional characteristics it is necessary to isolate primary endothelial cell lines and to establish in vitro models from which various parameters may be studied.  Most experiments carried out on endothelial cells make use of those from the large vessels such as HUVEC due to their ease of accessibility and their ability to be isolated and grown readily in culture. This source is however not a true representation of the adult endothelium as HUVECs are of foetal origin and express different phenotypic characteristics to that of the adult endothelium. These differences include the high level of CD95 (Fas) ligand expressed on HUVECs. Also the adult endothelium is amenable to changes due to potential prolonged exposure to cytokines, hormones and other stimuli (75). It is for these reasons that an alternative source of endothelial cells which is a better representation of the adult endothelium such as from whole blood will be most welcome.
CECs have promising therapeutic potential due to their role as a marker of endothelial injury. A breakthrough in determining their phenotype would help shed more light on the pathogenesis of most vascular disorders. An understanding of how they behave, how their behaviour varies between and within tumours, and how their behaviour is related to responsiveness to drugs is critical for the development of effective anti-angiogenic treatment strategies (76). An understanding of the endothelial response to pathogens can provide a foundation for therapeutic design.

1.7 The cell cycle
The cell cycle refers to the process whereby a single cell replicates into two daughter cells. The duration varies from on cell type to another with Mammalian cells having an average cycle of about 24hrs.  There are four known phases of the cell cycle namely G1, S, G2 and M. They can be further grouped under the division phase (M) and the interphase (G1-S). The G1 phase is the gap between the M phase and the S phase whereas the G2 phase is that between the S phase and the M phase. DNA replication occurs at the S phase and Mitosis (nucleus division) and cytokinesis (Cell splits in two) occurs at the M phase. During this phase cell growth ceases for the nucleus and subsequently the cell to divide. Continuous cell growth proceeds afterwards through the interphase (77). (Figure 13)
This is the phenomenon upon which growth of cells in culture is based. The extent of cell proliferation in culture is however dependent on certain factors including cell density, nutrient, and serum.  Other factors include temperature, surface area of flask/ well, volume of media and the rate of disturbance. Whether or not a cell remains in its quiescent or proliferative state depends on these factors. In vivo cell proliferation is controlled by protein kinases whose enzymatic activity is controlled by cyclins. The relationship between these protein kinases also known as cyclin dependent protein kinase and the cyclins, control the cell cycle (77)
Figure 13: Stages of the cell controlled by cylin dependent protein kinases and cyclins (77).

To mimic the process of cell replication in vitro, the medium which cells are cultured with have been enriched with the necessary proteins, enzymes and nutrients to promote proliferation.  However unlike unicellular microorganisms such as bacteria and yeast whose ability to proliferate depends mainly on the supply of nutrients, mammalian cells cannot grow on nutrients in medium alone. They concurrently feed off stimulatory chemical signals from neighbouring cells in their environment (80).  The rate of proliferation from the chemical signalling process is then enhanced by protein growth factors.
These growth factors form part of the ingredients found in culture medium. Some of these identified growth factors include cytokines and hormones. Growth factors represent a relatively large group of polypeptides which induce cell multiplication both in vivo and in vitro at very low concentrations of about 10-9 – 10-11 (80). They are usually synthesised in different cell types and tissues and tend to act locally within the tissue in which they are synthesised. They exert their biological effects on cells by interacting with specific cell surface receptors leading to the activation of a number of possible signal transduction pathways one of which promotes cell proliferation (80). The growth factors identified in the growth and differentiation of endothelial cells include vascular endothelial growth factor (VEGF), fibroblast growth factor (FGF), platelet –derived growth factor (PDGF) and endothelium derived growth factor (EDGF) (78).

1.7.1 Cell Density, Cell Signalling and Cell Proliferation.
Florey et al successfully described the endothelium after observing cells in culture (80). Culturing cells is vital for the successful observation and subsequent characterisation of any cell type. The idea is to be able to recreate the natural environment of cells within the body (78). In the quest to phenotype CECs, positive isolated cells will need to be cultured. Their ability to proliferate well in vitro requires amongst other things, for the cells to be in close proximity of each other as they feed of chemical signals generated from parent cells. This is usually mediated by a receptor protein (79). Other external factors required for proliferation include growth factors, nutrients, hormones and amino acids in the presence of serum. Studies have shown that an increase in the percentage of plasma derived serum present in growth media increases the rate of cell proliferation (81).  Endothelial cell growth in contrast to the growth of other cell types is characterised by the formation of an ordered monolayer. Once the cells have formed this ordered monolayer subsequent contact between cells results in the phenomenon known as contact inhibition of movement which results in cell death (82).
Cell signalling represents the communication system of the cells. This occurs in a numbers of ways depending on the distance between the signalling cell and the target cell. Signalling over long distances is via synapses which transmit electrical signals either directly through gap junctions or chemically by the release of neurotransmitters. This route is typical of neurotransmitters signalling to the brain. Signalling may also be by the release of a signalling molecule from once cell and its transport via a receptor to the target cell as seen with the endocrine system. Direct cell signalling is typical of cells that are in close proximity of each other. This is achieved either by the recognition of molecules or surface markers on each other’s surface, via gap junctions or by the release and detection of extra cellular signalling molecules such as hormones, cytokines and growth factors. These are usually carried in the blood stream for a long period of time without aiming for a specific target and can be the cause of some diseases such as growth of cancerous cells (79).
These extra cellular signalling molecules mainly growth factors are involved in the regulation of proliferation and cell motility. As there are no neurotransmitters involved in the growth of endothelial cells, chemical signalling across a distance via axons is not required. Signalling in the case of cells in culture is via the expression of surface markers or these extra cellular molecules and requires the cells to be in close contact. This is where cell density makes a difference in the rate of proliferation. The higher the cell density, the greater the probability of the  cells being in close contact and the higher the concentration of the extracellular molecules being generated and channelled between the cells. This results in an increase in cell proliferation. Where the cell density is low, the extracellular molecules secreted do not reach the target cell as they are too far away to be detected.

2.0 Aim
The exact phenotype of CECs is yet to be unveiled as culturing them has been challenging. This is mainly due to the cells being necrotic before reaching culture stage. There is evidence however of viable cells post extraction from whole blood with limited proliferation, although these have been suggested to be EPCs as supposed to CECs (46).
The concept of immunomagnetic extraction has been tested and approved for isolation in the past. There is however no publication reporting the efficiency of detaching positive isolated cells from beads. The initial aim of this project is to successfully extract bead free endothelial cells using this technique. The process will then be repeated for the extraction of CECs from whole blood. The cells will then be identified and confirmed as expressing endothelial cell markers.  Ultimately, attempts will be made to culture isolated bead free cells and their viability and growth assessed. If successful proliferation is observed, attempts can be made to phenotype the CECs.

2.1 Objectives
Before attempting to isolate bead free cells from whole blood, the technique will be optimised using HUVECs in culture. This will involve the following;
ü  To determine the optimal labelled antibody concentration required for Isolating HUVECs from up to 5 x 107 cells.
ü  To initially isolate bead free cells from HUVECs in culture using immunomagnetic beads.
ü  To determine the percentage yield obtained post isolation from the maximum labelled antibody concentration used.
ü  To determine the efficiency of bead detachment from positive isolated HUVECs.
ü  To successfully culture isolated bead free HUVECs and monitor growth / viability.

Once the technique has been optimised using HUVECs in culture, the process will be repeated for extracting CECs from whole blood. This will involve the following;
ü  To  determine the reproducibility of the immunomagnetic bead isolation technique
ü  To isolate and enumerate CECs from whole blood using immunomagnetic beads.
ü  To identify/ confirm endothelial cell markers on isolated CECs.
ü  To culture bead free cells enumerated from whole blood.
ü  To assess viability and proliferative rate of cells in culture.

2.2 Materials and method
2.2.1 Subjects
Ethical approval was obtained from the Kingston University Ethics committee prior to recruiting subjects. All subjects were given a copy of the document containing information on the use of their blood. Informed consent was obtained via signature from both parties and blood was treated as stated in the information leaflet. Approximately 2 - 10 ml of blood was taken per volunteer per experiment.  All subjects were healthy with no known vascular chronic medical condition.

2.2.2 Obtaining Whole blood
2 – 10 ml of Whole blood was obtained per healthy volunteer, transferred into EDTA coated tubes and mixed. Blood was obtained under sterile conditions using venepuncture free from trauma.

2.2.3 Tissue culture - Initiating culture
Cell growth media was prepared using a 1: 1 ratio of MCDB131 to EGM-2. This will be referred to as HUVEC media in this report. MCDB 131 was prepared by adding 20% foetal bovine serum (FBS) and 5 ml Penicillin/ Streptomycin (GIBCO®) to 500 ml MCDB 131 (GIBCO®). EGM-2(GIBCO®) comes with single quots of ready to use growth factors, serum and antibiotics, namely;
· hEGF (human epidermal growth factor)
· Hydrocortisone
· GA-1000 (Gentamicin, Amphotericin-B)
· FBS (Foetal Bovine Serum) 10 ml
· VEGF (Vascular endothelial growth factor)
· hFGF-B (Human fibroblast growth factor)
· R3-IGF-1 (Revitropin 3 – Insulin – like growth factor - 1)
· Ascorbic Acid
· Heparin
Hydrocortisone was omitted in preparing the EGM-2 media as it has been reported to hinder growth rate and viability of endothelial cells in culture (83, 84). A vial of frozen HUVECs was taken out of liquid nitrogen storage and allowed to thaw at room temperature. Cells were quickly transferred in to a T25 flask pre-coated with 5 ml 0.1% gelatin. 12ml of media was gently added to the cells in the flask. Flask was observed under a microscope before transferring to the incubator at 370C and 5% CO2/ 95% air humidified incubator.

2.2.4 Maintaining culture
Media was changed 24 hours after setting up culture and cells were passaged at a frequency dependent on the growth rate, usually 36 – 48 hours after media change.  At each passage cycle, media is discarded and cell monolayer was washed once with 1X phosphate buffered saline (PBS) without calcium and magnesium, (GIBCO®). 5 ml of trypsin/EDTA was added to the cells and incubated at room temperature for 5 minutes with gentle tapping to detach the cells whiles observing for roundness and detachment under the microscope. Trypsin/EDTA was neutralised with 5ml media and the cell suspension transferred into a 15 ml falcon tube.  Cells were centrifuged at 15000rpm for 5 minutes, supernatant poured off and pellet re-suspended in 1ml media. Re-suspended cells are either seeded and returned to the incubator or harvested for use in immunomagnetic isolation. Excess cells were frozen down in FBS containing 7% DMSO. Each confluent T75 flask was frozen down into four 1 ml cryo vials.

2.2.5 Cell Growth and Viability
Cell growth was monitored daily using microscopy to assess morphology and cell density. Cells were counted on a haemocytometer using a 1 in 2 dilution in trypan blue and viability assessed.

2.3 Preparing antibody conjugate
Anti CD146 antibody (Sigma, clone P1H12) was obtained at a concentration of 1.8mg/ml in 0.01M PBS, pH 7.4 and 15mM Sodium Azide. The final antibody volume was 0.2 ml. Based on the principle of immunomagnetic isolation previously discussed (1.6.2), the antibody was labelled with DSB Biotin as per manufacturers protocol (Appendix D). The protocol is outlined for labelling 0.2 ml of antibody at a concentration of 0.5-3 mg/ml. For this project the antibody volume used was 50µl. All volumes have therefore been scaled down in proportion to the antibody volume in order to maintain the appropriate concentrations of all reagents in the reaction mixture (See appendix E).The labelled antibody was then purified of the high Sodium Azide content to avoid interference and reduced activity.
The labelling and purification was carried out as follows for a 50µl antibody volume. 5µl of freshly prepared 1M NaHCO3, pH 8.3 was added to 50µl of Anti CD146 antibody. 40µl of DMSO was added to 200µg of DSB Biotin (Molecular Probes/ Invitrogen). The biotin solution in DMSO was prepared just before use to ensure maximum reactivity of the succinimidyl ester on the Biotin. 1.25µl of the freshly prepared biotin solution was then added to the antibody solution and mixed for 1hour on a magnetic stirrer at room temperature. The reaction conditions are designed by the manufacturer to obtain 3 – 8 DSB-X biotin molecules covalently bound per antibody molecule.
The labelled antibody was concentrated and purified of Sodium Azide by use of protein A agarose resin. All materials required for the purification process was supplied with the kit from the manufacturers. Again all volumes have been scaled in proportion to the antibody. Briefly, the Protein A agarose purification resin (supplied as a suspension in buffer with the kit) was mixed thoroughly by gentle swirling of the bottle. 500µl of the re-suspended purification resin was added to a spin column. The resin in the spin column was allowed to settle before spinning at 1100 x g for 3 minutes. This was to remove any residual buffer giving a final bed volume of about 400µl. The labelled antibody was then loaded onto the centre of the spin column that contained the resin and spun at 1100 x g for 5minutes. The collection tube that contained the concentrated purified antibody in about 50µl PBS, pH 7 with 2mM sodium azide was stored at 40C.  
The final antibody concentration was calculated based on the assumption that 85% of the antibody in the reaction is obtained as a purified DSB – X biotin conjugate. From a starting antibody concentration of 1.8mg/ml, the final labelled antibody concentration = 85/100 x 1.8mg, = 1.53mg/ml.

2.4 Isolation from Human Umbilical Vein Endothelial Cells.
A flask of HUVECs grown to confluence was harvested, re-suspended in 1ml of HUVEC media and counted. The isolation procedure was carried out as outlined in the manufactures protocol (Appendix A). The recommended labelled antibody concentration required for isolation from up to 5 x 107cells is 5 - 50µg. The maximum antibody concentration was used in this instance. The antibody at the assumed concentration of 1.53mg/ml was diluted down to the required concentration using PBS w/ 0.5% BSA. 
The cell solution was centrifuged at 15000rpm for 5 minutes. The Pellet was then resusupended in 500µl of isolation buffer (PBS containing 0.1% BSA and 2mM EDTA). Cells were incubated for 10mins at 40C with 25µl of labelled antibody. Cells were washed with 2ml cold isolation buffer, centrifuged for 8minutes at 350 x g and the pellet was re-suspended in 1ml isolation buffer.
FlowComp Dynabeads (Dynal / Invitrogen) was taken out of 40C and mixed. 75µl of the bead suspension was incubated with the cells for 15mins at 40C on a tilting roller. Tubes were then placed in a magnet (MPC-l, dynal/Invitrogen) for 1minute and the supernatant removed with the tube still in the magnet. Bead bound cells were washed twice thereafter with 1ml cold isolation buffer.
1ml FlowComp release buffer (supplied with the kit) was incubated with the cells for 10minutes at room temperature on a tilting roller. The cells were pipetted gently but efficiently after incubation to detach the cells from the beads. The Tube was returned to the magnet and the supernatant containing bead free cells was transferred to a new tube. The process was repeated to remove any residual beads.   
Cells were counted before, during and after the isolation procedure to assess the percentage loss.

2.4.1 Culturing bead free HUVEC cells
Bead free cells in release buffer were centrifuged at 3500rpm for 5minutes and pellet re-suspended in 100µl of EGM-2 media. Post isolated cells were plated out into a 24 well tissue culture plate pre-coated with 0.1% gelatin. Cells were transferred to the incubator and maintained as previously described.

2.4.2 Culturing bead bound HUVEC cells
Before incubating cells with release buffer, 500µl of bead bound cells in isolation buffer was aliquoted into an effendorf tube. The tube was spun at 35000rpm for 5 minutes and the supernatant discarded. The cells were re-suspended in 100µl of EGM-2 media and plated out into a 24 well tissue culture plates pre-coated with 0.1% gelatin. Cells were transferred to the incubator and maintained as previously described.

2.5 Adherence test
A time course test was set up using HUVECs cultured in a 6 well tissue culture test plate to assess the time frame within which cells begun to adhere to the plate. Plate was observed at hourly intervals for three hours.

2.6 Isolation from whole blood.
The protocol for isolating cells from whole blood was optimised based on the results obtained from isolating cells from HUVECs in culture. (See appendix B).
Whole blood was obtained from healthy volunteers as previously described (2.2.2). Blood was transferred into a 15ml falcon tube and centrifuged at 350 x g for 10minutes. Plasma was discarded leaving 1ml above the RBC pellet. The pellet was then transferred into a 6well tissue culture test plate and incubated for 2hours at 370C in EGM-2 media to pellet RBCs, platelets and monocytes whiles retaining endothelial cells. The Supernatant was removed after incubation and wells washed with isolation buffer at 350 x g for 10mins. 50µl of labelled antibody was added to whole blood, mixed and incubated at 40C for 30minutess. Cells were washed in isolation buffer as previously described. 50µl of FlowComp Dynal beads was added to the cells, mixed and incubated with agitation for 30minutes at 40C. Cells were washed with 4ml cold isolation buffer and tube transferred to a magnet for 3 minutes. The supernatant was carefully pipetted off and discarded without disturbing the tube wall with the tube in the magnet. The washing procedure was repeated twice. Cells were then re-suspended in 1ml release buffer and incubated with agitation for 10minutes at 40C. The cell suspension was then pipetted efficiently to release cells and tubes returned to the magnet for 1minute. The supernatant containing bead free cells was collected and 1ml of EGM-2 media added before centrifuging at 1500rpm for 10minutes. Pellet was re-suspended in 100µl of EGM-2 media, plated into a 96 well tissue culture plate and incubated overnight at 360C.

2.6.1 Identification and Quantification of Cells from whole blood
2ml of whole blood was obtained as previously described. 20µl of FcR block (Miltenyi Biotech, Bisley, UK) was added to blood followed by 25µl of labelled antibody. Tube was incubated with agitation for 60minutes at 40C and then placed in a magnet for 2minutes. The supernatant was discarded being careful not to disturb the tube wall. Labelled cells were re-suspended in 1ml isolation buffer and returned to the magnet for another minutes. Supernatant was discarded as before. The wash step is repeated twice before re-suspending cells in 90µl of isolation buffer and 10µl of 2mg/ml FITC labelled Ulex Europeus lectin (Sigma – Aldrich). Tube was incubated in the dark for 1hour at room temperature with agitation.  Cells were washed three times after incubation with isolation buffer and re-suspended in 200µl of buffer.
To count, 100µl of cell suspension was transferred to a nageotte chamber and all 40 lines counted under a fluorescent microscope. CECs were identified as bright yellow Ulex dyed cells about 10µm in size with beads attached (Figure 21).

3.0 Results
3.1 Culturing HUVECs
Figure 9 shows images of the proliferative stages of HUVEC from initiating culture from cryopreserved cells till they are harvested for use under the culture conditions previously described. A cryovial contains 1/4th of cells harvested from a confluent T75 which usually has cell numbers of about 1 x 106cells/ml. Cells reach 80% confluence within 36 hours and are fully confluent within 48hours thereafter. Once trypsined they become detached within 2 minutes and 80 – 90% of cells are harvested.   
A B  C  D     E
Figure 13: A) Cells initiated from cryopreserved cells. B) Appearance of cells 36hours after culture. C) Fully confluent cells within 5days of initiating culture. D) Detached trypsinsed cells. E) Left over cells after harvesting.

3.2 Percentage cell loss post Isolation
The number of HUVECs was counted on a haemocytomer during each stage of the isolation process. There were over 60% cell losses by the end of the isolation process. There were no cells collected in the flow through. Cells counted on a nageotte chamber over all 40 lines shows over 50% loss in cell numbers between bead bound cells and bead free cells.  Subsequent culture of bead bound cells and bead free cells show a higher proliferative rate with the bead free cells. Proliferation of bead bound cells increased with each media change. 

Figure 14/ Table 11: The table below shows the cell counts on a haemocytomer at each stage of the isolation process from harvesting HUVECs in culture from a confluent T75, Incubating with antibody in the presence of isolation buffer, washing off excess antibody after incubation, incubating with beads, flow through after passing bead bound cells through the magnet, and finally the cell number after incubating with the release bufferThe isolation process was repeated on two different occasions. With each experimental run there was a significant loss in cell numbers by the end of the isolation process. The percentage cell loss from the first run was about 80%. The cell loss from the second run was about 60%. The flask of HUVECs harvested for the first run was almost a third less than that harvested for the second run.


Figure 14
Table 11

Due to the signigficant drop in cell numbers observed post isolation and to further confirm the clear drop in numbers between two of the key stages of the isolation process, the total number of bead bound cells was counted on all 40 lines on a nageotte chamber before and after the addition of release buffer. Individual cell counts are plotted below followed by a plot of the mean cell counts.

Figure 15:

Figure 15: The count from all 40 lines on a nageotte chamber before and after detachment of beads. A percentage cell loss of about 50% is observed between the two stages confirming the cell counts achieved post isolation.
Figure 16:
The mean count for the bead bound cells is 286.7 ± 35. The standard error of the mean is 5.53. That of the eluted bead free cells was 97.7 ± 15.0 The standard error of the mean is 2.36.
A paired sample t-Test comparing the average cell counts before and after detachment of the beads is shown in table 12. This is based on the null hypothesis that there is no significant difference between the means of the two variables.  The mean of the bead bound cells is higher than that of the bead free cells. There is a positive correlation between the two variables such that the loss in cell numbers correlates with the counts on each of the 40 lines. The degree of freedom (df) is 39. Tcritical is 2.02. Assuming a 95% confidence interval of 0.05, there is a significant difference between the two variables.

Figure 16
Table 12: SPSS output of a Paired sample t-Test comparing
 the mean cell counts before and after detachment of beads.

Figure 17: A Q-Q Plot showing the distribution of the cell numbers. Two different populations are observed. The plot gives an s-shaped chart with variables distributed at either end of the central line.

Figure 18: 
Top left: Appearance of bead bound cells on a nageotte chamber.
Top right: Appearance of eluted bead free cells on a nageotte chamber. There is an obvious drop in the cell population between the two cell conditions.
Bottom left: Bead bound cells in culture after 5 days in a 24 well tissue culture plate. Bead free cells had a faster rate of proliferation post isolation compared to the bead bound cells in culture.
 Bottom right: Bead free cells in culture after 4 days in a 24 well tissue culture plate.

Figure 18


Once bead free cells were in culture their growth rate and characteristics was as expected of HUVECs in culture. Bead bound cells however were observed to have a slower growth rate. One interesting feature observed as the media was changed between days was the washing away of the bead with each wash. With time the number of beads observed declined slowly with an increase in percentage cell confluence.  
3.3 Number of beads per Human Umbilical Vein Endothelial Cell.
The number of beads attached per HUVEC was counted over a 100 cells selected at random.  The mean count was 14.23. All counts above 60 beads per cell were cut off at 60. About 25 cells were identified in this category from a sample size of 100. Most cells had an average bead number ranging between 50 -60 bound to them at the maximum antibody concentration of 50µg.
A)      B)

Figure 19: A) An average frequency distribution chart showing the number of Dynabeads attached per HUVEC counted over a 100 bead bound cells. B) Beads attached onto the cell surface at varying frequencies. C) x40 magnified image showing degree of binding on a cell.


3.4 Adherence Test
Before analysis on whole blood, an adherence test was carried out on HUVECs as described in 6.8. Whole blood was plated out on a 6well tissue culture plate and incubated for 2hours. The Adherence test performed on HUVECs showed that cells did not adhere to a gelatinised well until after 2.5 - 3hrs at 370C. In whole blood, after incubation on a plastic 6 well tissue culture test plate for 2 hours, a large number of cells though to be monocytes, leukocytes and platelets were observed to have adhered to the plate. 
A)                 B)
Figure 20: Monocytes, Leukocytes and Platelets adhered after plate was incubated for 2hrs at 370C.

3.5 UEA-1 Lectin Staining
CECs isolated from whole blood were stained with Ulex europeus for identification. In one subject, 64 CECs were identified per ml of whole blood. This number was expected as the all the blood obtained was used in the process without discarding the first 5ml of which a higher than usual amount of CECs is expected to be present. Leukocytes which also express the endothelial cell marker CD146 were also present as expected. These are distinguished from CECs by their miniature size in comparison to CECs.

Figure 21: Ulex staining of CEC after Immunomagnetic Isolation. A) Staining with bead bound cells. CEC stains bright green with Ulex europeus lectin B) Pre stained CECs observed after addition of release buffer and subsequent washing before plating in a 96 well tissue culture plate.

3.6 Reproducibility
Repeating the isolation process using HUVECs in culture consistently yielded bead free endothelial cells. A significant percentage cell loss was observed between bead bound cells and eluted bead free cells each time. These numbers however improved as the isolation process was repeated. The subsequent culturing of bead bound cells and bead free cells showed similar growth characteristics when the procedure was repeated. Bead free cells in culture constantly proliferated at a higher rate compared to bead bound cells.
Isolation from whole blood was not quantified after the initial staining and enumerating experimental run due to the low numbers of CECs. Batch to batch reproducibility can therefore not be compared quantitatively. No growth was observed in the post isolated bead free cell culture.

4.0 Discussion
The technique of Immunomagnetic isolation has been around for some time now. What has proven challenging is the detachment of beads from the positive isolated cells.  The results obtained from the positive isolation of bead free target cells shows the technique works. The efficiency of release is however questionable. When the isolation was performed using HUVECs in culture there was a cell loss of over 50% after elution of the bead free cell. Although a loss in cell numbers is expected with every wash step there was a significant drop in the cell numbers between the bead bound cells and the eluted bead free cells observed both quantitatively and visually. Also transfer of cells into a new tube with each wash step is bound to result in some cell loss. The drop in cell numbers was measurable in the case of the HUVECs as there were enough cells available to monitor cells numbers at each stage. In the case of whole blood however, there is very low numbers extracted from a 2ml sample and therefore performing a count at each stage which requires a 100μl loading volume would not be ideal.
In the experimental run where positive isolated bead bound CECs were stained with Ulex europeus and enumerated, 64 cells / ml of whole blood was counted in one subject. This is higher than the usual reported range but was expected as the initial few mls known to contain a high concentration of CECs from the venipuncture was not discarded.  The cells were still alive by this point indicating that the CECs extracted although believed to be matured and detached from the vessel wall is neither necrotic nor apoptotic at the point of extraction.   By the end of the isolated process, although the CECs were not quantified, they visually appeared sparse in a 96 well tissue culture plate inferring a further loss in cell numbers. This decline in cell numbers aside the obvious loss from washing, pipetting and changing tubes could be due to the efficiency of the release buffer.
A t –test performed on the cell counts before and after detachment of beads using HUVECs in culture showed a positive correlation between the two variables with a significant difference in the cell counts. This further confirms the lack of efficiency of the release buffer and the need for improvements on the technique.
The more efficient the beads attach to the cells, the more efficient the release buffer need to be at detaching the beads. The degree of beads attaching to the cells is dependent on how effective the tubes are kept agitated for.  The beads are incubated with the cells at 40c with constant rolling and tilting so all angles on the cell have a chance of exposure to the beads and to bind. The low temperature keeps the cells from being activated and maximizes the binding efficiency. The release buffer is also incubated with the beads on a rolling tilter so as to give all the cells the same probability of detachment as they were with the attachment. This is however carried out at room temperature as breaking of the streptavidin – biotin bond to release the beads would require some energy provided by the room temperature heat.
 The ratio of the labeled antibody concentration to that of the cell numbers and the amount of beads added is important for the positive isolation of the desired cell type. During the optimization studies using HUVECs, when the amount of labeled antibody used was too dilute in relation to the number of HUVECs harvested, the isolation procedure failed. This could be because there was very few labeled Anti – CD146 antibody present in the cell suspension for the cells to bind to. The few cells that may have competed and successfully detected the antibody to bind to could have been lost through the process. Also the streptavidin coated beads would have little or no biotin to bind to. When the eluted HUVECs were cultured, there was no evidence of beads present implying the process did not work.  
Subsequent experimental runs with the required antibody concentration range saw an excess of beads in the eluted bead free cells which washed out with extra wash steps.
The streptavidin –biotin bond is the strongest non-covalent biological interaction known, with a dissociation constant, Kd, in the order of 4 x 10-14M (73). The strength and specificity of this interaction has made it an attractive choice for use in molecular, immunological, and cellular assays where affinity pairing is desired. Holmberg et al reported efficiency in breaking this bond at high temperatures above 700C in the presence of non-ionic aqueous solutions mainly H2O and 10mM NaCl without denaturing the streptavidin tetramer and both molecules still remaining active (73).  Dynal also reported release efficiencies of up to 97% after incubation with 10 mM EDTA pH 8.2 and 95% formamide as well as 80 mM NaOAc pH 9 and 95% formamide. Both of these buffers were incubated with streptavidin- biotin conjugate at a high temperature of 900C (86).
The release buffer supplied by dynal for use in releasing the cells from the beads is assumed to be a salt solution of unknown concentration. Incubating the cells in this buffer is carried out at room temperature which was usually about 200C. Understandably, the high efficient temperatures reported by Holmberg et al cannot be employed in this instance as it will impede on the viability of the cells and denature the antibody. This will imply therefore that the final elution step which collects the supernatant containing bead free cells needs to be optimized as the beads being left behind may still have cells bound to them. Since bead bound cells in culture are known to grow, these beads left behind should have been washed and cultured to observe for cell viability and growth. This was not tested in this instance.
Since the rate of proliferation of bead free cells in culture is considerably higher than that of bead bound cells it is possible that the degree of growth seen with the bead bound cells will be dependent on the density of the beads in relation to cell numbers. The beads are denser than the cells and will settle to the bottom of a culture plate before the cells. If the cells have no surface onto which to attach or have no room for cell signaling then their growth will be impaired. Since the beads only settle to the bottom of a culture plate without any means of attachment, they can be washed off. This was observed with each media change for the bead bound cell in culture. The bead density reduced with each wash while the cell expanded in their formation of a monolayer. Some viable parent cells with the beads still attached were observed after 5 days in culture with daily media change but the number of beads per cell was minimal (<10). The daughter cells generated were obviously free of beads.
The lack of efficiency in breaking the streptavidin – biotin bond could be attributed to the length of the incubation. (10 minutes). Incubating for longer will hinder the viability of the cells. An alternative to using the release buffer supplied (thought to be a salt solution) could be trypsin/EDTA.  
The efficiency of this technique leaves much room for improvement.
Another key area to consider throughout this process is the length of time it takes from obtaining whole blood to putting the cells in culture. The longer the cells stay out of their desired environment in culture medium, the more susceptible they are to cell death. The whole process once familiar takes about 3 hours from obtaining whole blood to plating out isolated cells.  Due to the delicate nature of CECs it is important that they be kept in suitable growth conditions for as long as possible with minimal stress. The isolation buffer is spiked with serum which is ideal for maintaining cell haemostasis whiles out of culture.  The question remains however as to whether or not the serum alone at the concentration added is enough to sustain the cells for that length of time. 
The labelled antibody concentration suggested from the protocol is between 5- 50μg per 5 x 107cells. The average number of HUVECs harvested from a confluent T75 flask is 1 x 106cells.
Due to the low numbers of endothelial cells in whole blood, the maximum suggested labelled antibody concentration was used at 50μg. In theory, the successful proliferation of positive isolated cells is dependent on the cell numbers post isolation. Bearing in mind the other factors responsible for the growth of cells in culture such as cell proximity, surface area, serum and nutrients, culture conditions can be adjusted to suit the number of extracted cells post isolation. This may be feasible where HUVECs are concerned due to their general ease of handling and culture. CECs in comparison are fragile and very few cells are reported to be extracted from healthy whole blood. It is therefore important to have as many viable cells post isolation to increase chances of growth. They are also known to apoptose before reaching culture stage and the few cells that make it through the isolation process soon die off if conditions are not optimised. To combat this flaw, it was necessary to reduce the length of the isolation process so the cells are out of their preferred environment for as little time as possible. Secondly, excess wash steps which increase the chances of cell loss were eliminated.
The toxicity of the beads has not been established and so it cannot be concluded as having toxic effects on the cells. When bead bound HUVECs were cultured they continued to proliferate although this was at a slower than normal rate. This has been attributed to the beads impeding the cell signalling channels as supposed having a toxic effect on the cells.     
EGM-2 is modified MCDB131 containing containing 2% foetal bovine serum, VEGF, bFGF, IGF-1, EGF, heparin, hydrocortisone and ascorbic acid. MCDB131 was developed by Knedler and Ham as a reduced serum-supplemented medium for the culture of human microvascular endothelial cells. Nutrient and protein supplements are required for the culture of other cell types (86). The initial HUVEC culture was carried out in MCDB 131 media and the cells failed to grow to confluence. When the percentage serum in the MCDB 131 media was increased to 20% there was an improvement in the cell growth although this was still sluggish. The EGM-2 media with the bullet kit is the preferred choice of media for growing HUVECs due to its nutritional content. The hydrocortisone in the EGM-2 kit was excluded as it has been reported to hinder endothelial cell growth (84-85)
The antibody came purified of serum and other proteins but was supplied in buffer with a high azide content of 15mM. This had to be purified of the azide as it is known cause cell death.
CECs are widely accepted as a key marker of endothelial damage. Its isolation from whole blood using the immunomagnetic bead separation method makes use of antibody targeted against CD146 which is known to be expressed on their cell surface. This surface marker is also reported to be expressed on other cell types including activated T-cells, pericytes, trophoblasts, mesenchymal stem cells and malignant tissues and care should be taken not to mistaken CECs for these other cell types (74). Due to this there is the need for further identification using other fluorescent techniques. This is the one advantage that isolation and enumeration by flow cytometry has over the immunomagnetic separation technique as flow cytometry allows for a multi marker approach.
Where levels of CECs are being used to serve as a marker of vascular disease, it is necessary that the first few mls of blood is discarded usually 7.5 mls as the trauma caused from the venepuncture may dislodge endothelial cells from the vessel wall giving false positives. The collection of whole blood into EDTA coated tubes did not seem to affect the percentage yield of CECs extracted. When CECs were stained with Ulex from a sample of blood collected into EDTA coated tubes and with which the first 7.5 mls was not discarded, 64 cells / ml was counted. This is a reasonable number for CECs as the reported expected range is between 1 – 20 cells per ml of blood from which the first 7.5 ml has been discarded.  Unfortunately a comparative could not be done with citrate coated tubes although comparable data has been reported from whole blood collected into EDTA coated tubes and that collected into Citrate coated tubes (74). Enumeration of CECs by Ulex europeus staining is a skilful technique requiring experience especially in identifying CECs based on morphology and the ability to differentiate T cells from CECs both of which express CD146. This is one advantage that enumeration by flow cytometry has over immunomagtic bead isolation as secondary staining can be employed to distinguish between CECs and other cell sub sets.    
The isolation technique was generally reproducible but as expected of any experimental run, it gets more accurate and consistent with each practise. The technique is also cumbersome and time consuming if CECS are to be isolated, enumerated and cultured. Also identification requires familiarity and experience especially in the separation of T-cells from CECs. Based on morphology alone T-cells appear much smaller in size compared to CECs.
The ability to phenotype CECs will be of great medical benefit as it serves as a biomarker for several vascular diseases. Their enumeration provides will be useful in establishing endothelial damage, determining the extent of the damage and the progress of treatment. Establishing the phenotype of CECs can be achieved once isolated CECs from whole blood are successfully cultured. To enable the validation of comparing enumerated CECs numbers before, during and after therapeutic intervention it is important that a consensus protocol is reached. The consensus protocol suggested by Woywodt et al (74) although reproducible has since been updated to include the detachment of the beads from cells. Also as pointed out in section 3.6.2 table 10, there are significant differences in the two protocols. Detaching the beads from the cells is a more appealing concept as it gives more scope for further functional studies. As seen during the culture of Bead bound HUVECs in comparison to that of bead free cells, the presence of beads on the cells can hinder cell proliferation in culture. There is yet to be published data on the isolation of CECs from whole blood which includes the detachment of the beads from the cells. The observations made from this project in terms of the validity of the isolation process to include the detachment of the cells from the beads are yet to be verified. Further experimental runs including optimisation of the isolation process still need to be carried out.
The whole isolation process and subsequent culture of the cells had to be carried out under sterile conditions to avoid contamination.

5.0 Suggestion and Further work
The keys stages of this isolation process with a chance of improvement are;
1.         The initial concentration of the antibody,
2.         The efficiency of labelling the antibody to the biotin,
3.         The cell numbers in relation to the labelled antibody concentration,
4.         The amount of beads added to the cells
5.         The efficiency of the release buffer.
Other factors to consider are the incubating times, washing steps and handling times. Possible suggestions to improve some of these parameters are given below.

The main challenge faced with isolating circulating endothelial cells from whole blood is maintaining the viability of the cells. The viability of the CECs extracted needs to be established and if possible monitored over the isolation stages to determine the point of death. Bearing in mind the numbers available per ml of whole blood this may not be feasible. A possible way around this is to initially extract the blood from a volunteer after high cardiovascular activity as this causes shedding off of CECs from the vascular wall. If the initial CEC numbers are higher than usual, there is a greater chance of some cells surviving through the process.
Once the cells have been isolated, an MTT assay can be set up to establish the viability of the cell. If the cells are viable before incubating with the release buffer then culture could be initiated by that point in the presence of the beads. The presence of the beads seems to slow the growth of the cells but the beads have been seen to come off with every wash step. If the bead concentration is just enough such that there is room in the culture wells for the CECs to adhere, then there is a probability of cell growth. The objective here will be to optimise the culture conditions so the cells proliferate probably by increasing the percentage serum and other growth factors in the EGM-2 media.
Familiarity with the procedure is necessary to obtain optimal result mainly due to improvements in handling techniques. Incubating times and wash steps should be kept as minimal as possible to avoid cell death and cell loss.
Incubating the blood for 2 hours to pellet out monocytes, leukocytes and platelets is not ideal as the cells have a higher chance of being distressed form being out of their normal environment for too long. An alternative to separate CECs from the other cell types present in whole blood will be ideal. Red blood cells can easily be eliminated by use of red blood cell (RBC) lysis buffer at the start of the isolation process. Ficoll density centrifugation is one of the methods used to isolate CECs in the past. This technique has its shortfalls but could be an alternative to pelleting out unwanted cells.
Alternatively, the cells could be cultured before incubating with the magnetic beads. Once the whole blood is obtained it will be incubated with RBC lysis buffer to lyse out the RBCs and centrifuged to remove the dead cell layer. The remainder of the cells can then be cultured under endothelial cell culture conditions. Assuming non endothelial cells fail to grow, the culture can then be labelled with anti CD146 coupled to biotin and the isolation process carried out as per isolating from HUVECS.
Another option is to label and stain the cells for CD146 with Ulex europeus and then pass through the magnet. The positively isolated bead free CECs will then be spiked with HUVECs and cultured. This should in theory assist with the proliferation of the CECs and the original parent cells can still be identified and viability assessed from the bright yellow Ulex europeus stain.  Another option is to spike the positive isolated bead free CECs with TNFα to stimulate the cells.
With regards to detaching the cells from the beads, an alternative is to incubate the bead bound cells with trypsin /EDTA instead of the release buffer provided. Alternatively the incubating temperature for releasing the beads could be at 370C as supposed to the room temperature used.
An alternative to determine the optimal antibody concentration required for the expected low CEC numbers is to carry out an ELISA. The assay can be run with constant number of HUVECs to mimic the low numbers of CECs in blood and the labelled antibody concentration varied. This could give a better indication of the optimal antibody concentration required as supposed to cells counts post isolation.

7.0 Conclusion
In conclusion, the process of Immunomagnetic bead isolation with the option of detaching the beads from the cells looks promising. The efficiency of the release buffer is however questionable as there appear to be a significant loss in cell numbers between the bead bound cells and the eluted bead free cells. It is important that the ratio of the labelled antibody to the cell numbers and subsequently the amount of beads added is in proportion for optimal results.
When HUVECs were isolated from culture using this method, there was a percentage cell loss of over 50% by the end of the isolation process. A significant drop in cell numbers was observed between bead bound cells and bead free cells inferring a lack of efficiency in the release buffer or in the technique of detaching the cells.
Enumerating CECs from whole blood after fluorescent staining with Ulex europeus lectin yielded a count of 64 cells/ ml of whole blood in one volunteer. Incubating the cells with release buffer to detach the beads from the cells again saw a significant drop in the cell numbers.
Repeating the isolation process using HUVECs in culture consistently yielded bead free endothelial cells. Post isolated bead free HUVCEs in culture proliferated. Excess beads in the culture washed out with each media change.  
Circulating endothelial cells are present in whole blood and they are viable at the time of obtaining the blood and before incubating with release buffer to detach the beads from the cells. This was confirmed from the Ulex europeus stain after CECs had been incubated with the beads. CECs in culture after the beads had been detached failed to grow. Culture conditions need to be optimised to aid the growth of the positive isolated bead free CECs. CECs need to be kept out of their preferred environment for as little time as possible minimise cell stress and death. Incubating times and wash steps also need to be kept to a minimum to avoid excessive cell loss.

11.0 References
1.      Aird, WC. Phenotypic Heterogenicity of the endothelium: I.Strucutre, Function and Mechanisms. Circ Res. (2007); 100; 158-173.
2.      Sumpio BE, Riley JT, Dardik A. Cells in focus: endothelial cell. Int J Biochem Cell Biol. (2002); 34(12); 1508-12.
3.      Limaye, V. Vadas, M. Mechanisms of Vascular Disease: (2007). A Textbook for Vascular Surgeons. Cambridge University Press. New York.
4.      Cines, DB. Pollak, ES. Buck, CA. Loscalzo, J. Zimmerman, GA. McEver, RP. Pober, JS.  Wick, TM.  Konkle, BA.  Schwartz, BS. Barnathan, ES.  McCrae, KR. Hug, BA. Schmidt, A-M. Stern, DM.  Endothelial cells in physiology and in the pathophysiology of vascular disorders. Blood. (1998); 91(10) 3527-3561.
5.      Lin, Yi. Weisdorf, DJ. Solovey A. Hebbel RP. Origins of circulating endothelial cells and endothelial outgrowth from blood. J Clin Invest. (2000); 105(1):71–77.
6.      Alberts B, Johnson A, Lewis J, Raff M, Roberts K, Walter P. (2002). Molecular Biology of the Cell. (2002). 4th edition. Garland Science New York.
7.      Dye, JF. Jablenska, R. Donnelly, L. Lawrence, L. Leach, L. Clark, P. Firth, JA. Phenotype of the endothelium in the human term placenta. Placenta (2001); 22; 32-43.
8.      Landmesser, U.  Hornig, B.Drexler, H. Endothelial Function: A Critical Determinant in Atherosclerosis. Circulation (2004); 109 (2); 27--33.
9.      Galley, HF. Webster NR.  Physiology of the endothelium. Br J Anaesth (2004); 93(1): 105 -13.
10.  Ward, JPT.  Ward, J.  Leach, RM. The respiratory system at a glance. (2010). 3rd Edition. John Wiley and Sons. England.
11.  Effros MR. Anatomy, development, and physiology of the lungs. (2006). GI Motility online. Accessed 8th August 2011.
12.  Budhiraja, R., Tuder, RM, Hassoun, PM. Endothelial Dysfunction in Pulmonary Hypertension. Circulation. (2004); 109: 159-165.
13.  Aird WC. Endothelial cell heterogeneity. Crit Care Med. (2003); 31: S221–S230.
14.  Block ER. Pulmonary endothelial cell pathobiology: implications for acute lung injury Am J Med Sci. (1992); 304(2):136-44.
15.  Lafranconi WM, Huxtable RJ. Changes in angiotensin-converting enzyme activity in lungs damaged by the pyrrolizidine alkaloid monocrotaline. Thorax (1983); 38:307-309.
16.  Choi K, Kennedy M, Kazarov A, Papadimitriou JC, Keller G. A common precursor for hematopoietic and endothelial cells. Development (1998) 125(4):725-32.
17.  Risau W. Differentiation of endothelium. The FASEB Journal (1995); 9 (10); 926-933.
18.  Seghezzi G. Patel S. Ren CJ. Gualandris A. Pintucci G. Robbins E. S. Shapiro RL. Galloway AC. Rifkin DB. Mignatti P. Fibroblast Growth Factor-2 (FGF-2) Induces Vascular Endothelial Growth Factor (VEGF) Expression in the Endothelial Cells of Forming Capillaries: An Autocrine Mechanism Contributing to Angiogenesis. (1998) J. Cell Biol. 141:1659–1673.
19.  Wagner DD. Olmsted JB. Marder VJ. Immunolocalization of von Willebrand protein in Weibel-Palade bodies of human endothelial cells. J. Cell Biol (1982); 95 (1); 355-360.
20.  Weibel ER. Palade GE. New cytoplasmic components in arterial endothelia. J. Cell Biol (1964); 23; 101-112.
21.  Thorin, E. Shatos, MA. Shreeve, M. Walters, C. Bevan, J. Human Vascular Endothelium Heterogeneity. Stroke. (1997); 28; 375-381.
22.  Langenkamp, E. Molema, G. Microvascular endothelial cell heterogeneity: general concepts and pharmacological consequences for anti-angiogenic therapy of cancer. Cell Tissue Res. (2009); 335(1):205-22.
23.  Dejana, E.  Endothelial cell–cell junctions: happy together. Nat Rev Mol Cell Bio (2004); 5; 261-270.
24.  Middleton, J. Americh, L. Gayon, R. Julien, D. Mansat, M Mansat, P Anract, P. Cantagrel, A Cattan, P.  Reimund, JM. Aguilar, L  Amalric, F. A comparative study of endothelial cell markers expressed in chronically inflamed human tissues: MECA-79, Duffy antigen receptor for chemokines, vonWillebrand factor, CD31, CD34, CD105 and CD146. J Pathol (2005); 206: 260–268.
25.  Aird, WC. Mechanisms of Endothelial Cell Heterogeneity in Health and Disease. Circulation Research. (2006); 98:159.
26.  Radu V S. Endocytosis pathways in endothelium: how many? Am J Physiol Lung Cell Mol Physiol. (2006). 290:806-808.
27.  Cooke, J P. Dzau, V J. Nitric oxide synthase: Role in the Genesis of Vascular Disease          Annu. Rev. Med. (1997). 48:489–50.
28.  Aird, W. The role of the endothelium in severe sepsis and multiple organ dysfunction syndrome. Blood (2003); 101(10); 3765-3777.
29.  Schouten, M. Wiersinga WJ. Levi M.  van der Poll T. Inflammation, endothelium, and coagulation in sepsis. Journal of Leukocyte Biology. (2008); 83:536-545.
30.  Mehta, D. Malik, AB. Signalling Mechanisms Regulating Endothelial Permeability. Phys rev. (2006); 86(1); 279-367.
31.  Muller, W A. Leukocyte–endothelial-cell interactions in leukocyte transmigration       and the inflammatory response. TRENDS in Immunology. (2003); 24(6):327-34.
32.  Wood P. (2006).Understanding Immunology. Second edition.  Pearson.
33. Accessed 10th August 2011.
34.  Luscinskas, FW. Ma, S. Nusrat, A. Parkos, CA Shaw, SK. The role of endothelial cell lateral junctions during leukocyte trafficking. Immu Rev. (2002). 186(1); 57-67.
35.  Parham, P. The Immune system. (2004). Garland publishing.
36.  Khatib, NE. Genieys, S. Volpert, V.  Atherosclerosis Initiation Modelled as an Inflammatory Process. Math. Model. Nat. Phenom. (2007); 2(2), 126-141.
37.  Davis, N E. Atherosclerosis—An Inflammatory Process. J Insur Med (2005); 37:72–75.
38.  Paoletti, R. Gotto, AM. Jr. Hajjar, DP. Atherosclerosis: Evolving Vascular Biology and Clinical Implications Inflammation in Atherosclerosis and Implications for Therapy. Circulation. (2004); 109(3):20-26.
39.  Carter, AM. Inflammation, thrombosis and acute coronary syndromes. Diabetes and Vascular Disease Research (2005) 2: 113.
40.  Gutstein, DE. Pathophysiology and clinical significance of atherosclerotic plaque rupture Cardiovasc Res (1999); 41 (2):323-333.
41.  Behrendt, D. Ganz, P. endothelial function: From Vascular Biology to Clinical Applications. Am J Cardiol (2002); 90:40 – 48.
42.  Libby,P. Ridker, PM. Maseri, A. Inflammation and Atherosclerosis. Circulation (2002). 105:1135-1143.
43.  Ross R. Mechanisms of Disease: Atherosclerosis— An Inflammatory Disease. N Engl J Med. (1999); 340:115–126.
44.  Clarke, LA. Shah, V. Arrigoni, F Eleftheriou, D. Hong, Halcox, N. Klein, NJ. Brogan, PA. Endothelial injury and repair in systemic vasculitis of the young. Arthritis & Rheumatism. (2010); 62 (6): 1770 - 1780.
45.  Boss, JC. Lip, GYH. Blann, AD. Circulating endothelial cells in cardiovascular disease. J.Am Coll Cardiol. (2006); 48; 1538 – 1547.
46.  Woywodt A, Erdbruegger U, Haubitz M. Circulating endothelial cells and endothelial progenitor cells after angioplasty: news from the endothelial rescue squad. J Thromb Haemost (2006); 4: 976–8.
47.  Uta, E. Haubitz, M. Woywodt, A. Circulating endothelial cells: A novel marker of endothelial damage. Clinica Chimica Acta 373. (2006); 17-26.
48.  Woydt, A. Streiber, F. De Groot, K. Regelsberger, H. Haller, H. Haubitz, M. Circulating endothelial cells as markers for ANCA-associated small-vessel vasculitis.  The lancet. (2003); 361(9353): 206-210.
49.  Mutunga M, Fulton B, Bullock R, et al. Circulating endothelial cells in patients with septic shock. Am J Respir Crit Care Med.(2001); 163:195–200.
50.  Mutin M, Canavy I, Blann A, Bory M, Sampol J, Dignat-George F. Direct evidence of endothelial injury in acute myocardial infarction and unstable angina by demonstration of circulating endothelial cells. Blood. (1999); 93:2951–8.
51.  Bonello P, Basire A, Sabatier F, Paganelli F, Dignat-George F. Endothelial injury induced by coronary angioplasty triggers mobilization of endothelial progenitor cells in patients with stable coronary artery disease. J Thromb Haemost. (2006); 4(5):979–81.
52.  Chong AY, Lip GY, Freestone B, Blann AD. Increased circulating endothelial cells in acute heart failure: comparison with von Willebrand factor and soluble E-selectin. Eur J Heart Fail. (2006); 8(2):167–72.
53.  Nakatani K, Takeshita S, Tsujimoto H, Kawamura Y, Tokutomi T, Sekine I. Circulating endothelial cells in Kawasaki disease. Clin Exp Immunol. (2003); 131:536–40.
54.  Nadar SK, Lip GY, Lee KW, Blann AD. Circulating endothelial cells in acute ischaemic stroke. Thromb Haemost. (2005); (94):707–12.
55.  McClung JA, Naseer N, Saleem M, Rossi GP, Weiss MB, Abraham NG, Kappas, A. Circulating endothelial cells are elevated in patients with type 2 diabetes mellitus independently of HbA. (1) c. Diabetologia. (2005); 48:345–50.
56.  Solovey A, Lin Y, Browne P, Choong S, Wayner E, Hebbel RP. Circulating activated endothelial cells in sickle cell anemia. N Engl J Med. (1997); 337:1584–90.
57.  Woywodt A, Schröder M, Gwinner W, Mengel, M, Jaeger M, Schwarz A, Haller H, Haubitz M. Elevated numbers of circulating endothelial cells in renal transplant recipients. Transplantation. (2003); 76:1–4.
58.  Woywodt A, Scheer J, Hambach L, Buchholz S, Ganser A, Haller H, Hertenstein B, Haubitz M. Circulating endothelial cells as a marker of endothelial damage in allogeneic hematopoietic stem cell transplantation. Blood. (2004); 103:3603–5.
59.  Beerepoot LV, Mehra N, Vermaat JS, Zonnenberg BA, Gebbink MF, Voest EE. Increased levels of viable circulating endothelial cells are an indicator of progressive disease in cancer patients. Ann Oncol (2004); 15:139–45.
60.  Lefevre P, George F, Durand JM, Sampol J. Detection of circulating endothelial cells in thrombotic thrombocytopenic purpura. Thromb Haemost. (1993); 69:522.
61.  Makin AJ, Blann AD, Chung NA, Silverman SH, Lip GY. Assessment of endothelial damage in atherosclerotic vascular disease by quantification of circulating endothelial cells. Relationship with von Willebrand factor and tissue factor. Eur Heart J. (2004); 25:371–6.
62.  Woywodt, A. Ferdinand, H. Bahlmann,FH. de Groot, K. Haller, H. Haubitz, M. Circulating endothelial cells: life, death, detachment and repair of the endothelial cell layer. Nephrol Dial Transplant. (2002) 17: 1728–1730.
63.  Goon, PKY. Boss, CJ. Stonelake, PS. Blann, AD. Lip, GYH. Detection and quantification of mature circulating endothelial cells using flow cytometry and immunomagnetic beads: a methodological comparison.Thromb Haemost. (2006); 96; 45-52.
64.  Goon, PKY.Lip, GYH. Boss, CJ. Stonelake, PS. Blann, AD. Circulating Endothelial Cells, Endothelial Progenitor Cells, and Endothelial Microparticles in Cancer. Neoplasia. (2006); 8(2): 79–88.
65.  Dignat-George and J. Sampol, Circulating endothelial cells in vascular disorders: new insights into an old concept, Eur J Haematol.  (2000); 62: 215–220.
66.  Percivalle, M.G. Revello, L. Vago, F. Morini and G. Gerna, Circulating endothelial giant cells permissive for human cytomegalovirus (HCMV) are detected in disseminated HCMV infections with organ involvement, J Clin Invest (1993) 92; 663–67.
67.  Blann, AD. Woywodt, A. Bertolini, F. Bull, TM. Buyon, JP. Clancy, RM. Haubitz, M. Hebbel, RP. Lip, GYH. Mancuso, P. Sampol, J. Solovey, A. Dignat – George. Circulating endothelial cells. A biomarker of vascular disease. Thromb Haemost. (2005); 93: 228 – 35.
68.  Andrä, W, Häfeli, U, Hergt, R. Misri, R. (2007). Application of Magnetic Particles in Medicine and Biology. Handbook of Magnetism and Advanced Magnetic Materials. John Wiley & Sons. London.
70.  Safarik, I.  Safarikova, M. Magnetic techniques for the isolation and purification of proteins and peptides (2004) BioMagnetic Research and Technology. 2:7. 
71.  Safarikova, M. Safarik , I. The application of magnetic techniques in bioscience. Magnetic and electrical separation. (2001); 10; (223 – 252).
73.  Holmberg, A., Blomstergren, A., Nord, O., Lukacs, M., Lundeberg, J. and Uhlén, M. The biotin-streptavidin interaction can be reversibly broken using water at elevated temperatures. Electrophoresis, (2005); 26: 501–510.
74.  Woywodt A, Blann AD, Kirsch T, Erdbruegger  U, Banzet N, Haubitz M, Digna- Georg F. Isolation and enumeration of circulating endothelial cells by immunomagnetic isolation: proposal of a definition and a consensus protocol.  J Thromb Haemost. (2006); 4: 671–7.
75.  Tan, PH. Chan, C. Xue, SA. Dong R., Ananthesayanan B., Manunta M.. Kerouedan C. Cheshireb N. J. W, Wolfeb J. H. Haskardc D. O. Taylord K. M.  George A. J. T. Phenotypic and functional differences between human saphenous vein (HSVEC) and umbilical vein (HUVEC) endothelial cells. Atherosclerosis. (2004); 173 (2); 171-183.
76.  Langenkamp, E Molem, G. Microvascular endothelial cell heterogeneity: general concepts and pharmacological consequences for anti-angiogenic therapy of cancer. Cell Tissue Res.  (2009); 335(1):205-22.
77.  Alberts, B. Bray, D. Hopkin, K.  Johnson, A. Lewis, J. Roberts, K.  Raff, M. Walter, P.  Essential Cell Biology (2003). Garland Science. New York.
79.  Heath, JK. Growth factors. (1993). Cambridge University press. NewYork.
80.  Hancock. JT. Cell signalling. (2010) 3rd edition. Oxford university press. NewYork.
81.  Florey, L. The endothelial cell. Brit. med. J. (1966); 2; 487-490.
82.  Engstrom, W. Differential effects of epidermal growth factor (egf) on cell locomotion and cell proliferation in a cloned human embryonal carcinoma-derived cell line in vitro. J. Cell Sci. (1986) 86, 47-55.
83.  Guelstein, VI. Ivanova, O. Margolis, LB. Vasilie, Ju. Gelfand, IM. Contact Inhibition of Movement in the Cultures of Transformed Cells. PNAS (1973);  70(7); 2011-2014.
84.  Samples JR, Nayak SK, Gualtieri C, Binder PS.  Effect of hydrocortisone on corneal endothelial cells in vitro. Exp Eye Res. (1985);41(4):487-95
85.  Tsunashima YKondo AMatsuda TTogari A. Hydrocortisone inhibits cellular proliferation by downregulating hepatocyte growth factor synthesis in human osteoblasts. Biol Pharm Bull. (2011); 34(5):700-3.

No comments: