Hypercholesterolemia is a significant risk factor in the development of cardiovascular disease and is associated with chronic leukocyte adhesion in the microvasculature. While the underlying mechanisms behind this have yet to be determined, it may be possible that hypercholesterolemia impairs the fluid shear stress (FSS) inactivation of neutrophils through the rigidifying effect of cholesterol on membrane fluidity. FSS restricts surface expression of CD18 integrins through cathepsin B (ctsB) proteolysis, which minimizes neutrophil adhesivity. If hypercholesterolemia blocks FSS mechanotransduction, then the inhibition of CD18 cleavage may link pathologic blood cholesterol elevations with dysregulated neutrophil adhesion. We hypothesized that elevated cholesterol contributes to dysregulated neutrophil adhesion by impairing ctsB FSS-induced CD18 cleavage through membrane fluidity changes. In the first part of this study, we demonstrated that FSS-induced CD18 cleavage is a robust response of neutrophils and involves selective cleavage of macrophage 1-antigen (Mac1) through ctsB proteolysis. The second part of this study confirmed that ctsB regulates neutrophil adhesion through its proteolytic actions on Mac1, an important integrin involved in adhesion and chemotaxis. Specifically, ctsB accelerated neutrophil motility through an effect on Mac1 integrins during pseudopod retraction. Furthermore, by using a flow-based assay to quantify the mechanoregulation of neutrophil adhesivity, we demonstrated that FSS-induced ctsB release promoted neutrophil detachment from platelet-coated substrates and unstimulated endothelium. For the third part of this study, we linked cholesterol-related membrane fluidity changes with the ability of FSS to restrict neutrophil adhesion through Mac1. We also determined that pathologic cholesterol elevations associated with hypercholesterolemia could block FSS-induced Mac1 cleavage and were linked to disrupted tissue blood flow. This was accomplished using low-density lipoprotein receptor deficient (LDLR-/-) mice fed a high-fat diet. Ultimately, the results provided in the present study confirmed that cholesterol-related changes in membrane fluidity blocked the ability of ctsB to regulate neutrophil adhesion through FSS-induced Mac1 cleavage. This implicates an impaired neutrophil FSS mechanotransduction response in the dysregulation of neutrophil adhesion associated with hypercholesterolemia. Since dysregulated adhesion may be one of the earliest upstream features of cardiovascular disease associated with hypercholesterolemia, the present study provides a foundation for identifying a new mechanobiological factor in the pathobiology of microcirculatory dysfunction.

Cancer cells traveling to distant tissues during metastasis must survive passing through the circulation. However, the influence of this fluid microenvironment on these cells is poorly understood. It was previously viewed that exposure to the hemodynamic shear forces within circulation was inhospitable to cancer cells, causing the cells to be destroyed. Recent evidence indicates that transformed cells are markedly more resistant to fluid shear stress when compared to non-transformed epithelial cells. Furthermore, these cells selectively adapt following exposure to fluid shear stresses and become more resistant to subsequent exposures to shear stress. The mechanisms behind this difference in phenotype and induced resistance are investigated. The elastic modulus, a measure of stiffness, may play a role in resistance and is shown to be altered upon exposure to fluid shear forces. Additionally, plasma membrane repair is a critical process in the resistance phenotype as cells sustain damage but are able to maintain viability. Cytoskeletal dynamics are also shown to play a role in resistance to fluid shear forces.

Ph.D. / Biomedical Engineering / The cardiovascular disease atherosclerosis is directly linked to the functions of the endothelium, the monolayer of endothelial cells (ECs) that line the lumen of all blood vessels. EC functions are affected by fluid shear stress (FSS), the tangential force exerted by flowing blood. In vivo FSS is determined by vascular geometry with relatively straight vessels producing high, unidirectional FSS and vessel branch points and curvatures producing low, oscillatory FSS. While these distinct FSS conditions differentially regulate EC functions, they also dramatically affect EC shape and cytoskeletal structure. High and unidirectional FSS induces EC elongation and cytoskeletal alignment, while concurrently promoting EC functions that are atheroprotective. In contrast, low and oscillatory FSS induces cobblestone-shaped ECs with randomly oriented cytoskeletal features, while simultaneously promoting EC functions that create an athero-prone vascular environment. Whether these distinct EC shapes and cytoskeletal structures influence EC functions, independent of FSS, is largely unknown. The overall hypothesis of this study is that cell shape and cytoskeletal structure regulate EC functions through mechanisms that are independent of FSS. Due to advances in surface engineering in the field of micropatterning, EC shape can be controlled independent of external forces by creating spatially localized surface cues. In this research, lanes of protein were micropatterned on glass surfaces to induce EC elongated shape in the absence of FSS. In Aim 1, micropattern-elongated EC (MPEC) shape and cytoskeletal structure were fully characterized and determined to be comparable to FSS-elongated ECs. Thus, inducing EC elongation on micropatterned lanes provides a platform for studying the functional consequences of EC shape, independent of FSS. Using this model, the following important markers of EC functions related to atherosclerosis were evaluated to determine the influence of EC shape and cytoskeletal alignment: extracellular matrix deposition (Aim 2), inflammatory function(Aim 3), and thrombotic potential (Aim 4). The results indicate that EC-elongated shape and cytoskeletal alignment participate in promoting selected EC functions that are protective against atherosclerosis, independent of FSS. Since EC shape is governed by the cytoskeleton, this data suggests that the cytoskeleton plays an active role in the regulation of EC functions that promote cardiovascular health.

Cardiovascular disease (CVD) is the leading cause of death internationally. Efforts to decrease CVD death has been explored through stem cell technology, specifically organoid formation. Current cardiac organoid models lack the vascular networks for nutrient supply and maturation. In this study, pillar perfusion technology is used to fabricate cardiac organoids and induce vascularization via dynamic culturing and the addition of vascular promoting growth factors (GFs). In addition to this study, a millifluidic chip is engineered for shear stress application via flow simulations and experimental flow analysis. We successfully optimized the millifluidic chip to achieve fluid shear stress of 20mPa and validated through particle tracking velocimetry using 0.1um diameter beads under flow. The results of cardiac organoids displayed contraction and growth of endothelial cells (ECs) under dynamic flow with GFs. In addition, smooth muscle cells (SMCs) displayed growth via GFs in both dynamic and static culturing. Although vascular networks were not present in all conditions of this experiment, this thesis can serve a basis for searching other methods of inducing vascularization.

Im 1997 wurden der Orphan GPR30 aus HUVECs kloniert, die FSS ausgesetzt waren. In dieser Studie konnte gezeigt werden dass die Expression von GPR30 durch die FSS-Behandlung im Vergleich zu unbehandelten HUVEC-Zellen deutlich induziert wurde. Daraufhin wurde in einer Studie von Isensse et al. die zelluläre und gewebsspezifische Expression von GPR30 in GPR30-LacZ Reportergen-Mäusen untersucht. Es konnte eine Expression von GPR30 vorwiegend in den Endothelzellen der kleinen Arterien verschiedenster Gewebetypen nachgewiesen werden. GPR30 war postuliert dass E2 direkt binden kann und dadurch rasche nicht-genomische Signale vermittelt. Im Gegensatz dazu haben verschiedene andere Veröffentlichungen gezeigt dass E2 nicht spezifisch an GPR30 bindet. Trotz der Kontroverse ob es sich bei GPR30 um einen Östrogenrezeptor oder nicht ist bislang nichts über seine Interaktion zu anderen Proteinen und deren Wechselwirkung bekannt. Deswegen war ein Ziel dieser Arbeit, Interaktionspartner von menschlichen GPR30 zu identifizieren und folglich ein humanes vaskuläres in vitro Modell zu etabliren, um die potentiellen Interaktionen von GPR30 sowie die downstream-Effekte der Wechselwirkung zwischen GPR30 und den neuen Interaktionspartner des vaskulären Modells auf Transkriptebene zu evaluieren. Ein Screening einer humanen kardiovaskulären cDNA-Bibliothek mit Hilfe des Y2H-Systems führte zur Identifizierung mehrerer Interaktionspartner für GPR30 darunter PATJ und FUNDC2. Durch anschließende CoIP konnte die Interaktion von GPR30 mit PATJ validiert werden. Des Weiteren konnte in dieser Arbeit die Wirkung von FSS auf die Expression von GPR30 in HUVECs bestätigt und ebenfalls in weiteren anderen Endothelzellen gezeigt werden. Abschließend wurde die Rolle von GPR30 und PATJ bei der Reaktion auf FSS auf transkriptioneller Ebene in HMEC-1-Zellen genomweit untersucht. Interessanterweise war eine Gruppe von Genen aufgrund von FSS in Zellen die GPR30 überexprimierten dereguliert als alleine durch FSS. / In 1997, the orphan G protein-coupled receptor 30, GPR30, was cloned using HUVECs exposed to FSS. It was shown that the level of GPR30 expression was up-regulated in response to FSS. Subsequently, in a study performed in the laboratory where the work for this thesis was carried out, the cellular and tissue distribution of GPR30 were investigated in GPR30-LacZ reporter mice and the expression was found predominantly in the endothelial cells of small arteries in several tissue types. GPR30, was also claimed to bind 17-β-estradiol (E2) directly and to mediate rapid non-genomic signalling. In contrast, various reports have indicated that E2 fails to bind GPR30 in a specific manner. Despite the controversy on whether GPR30 is an estrogen receptor or not, nothing is known at present about its relation and interaction with other proteins. Therefore, the aim of the work described in this thesis was to identify human GPR30 protein interaction partners and to establish a human vascular in vitro model in order to evaluate the potential role of GPR30 and the downstream effects of the interaction between GPR30 and new interaction partners in a vascular model at transcript level. The screening of a human heart cDNA library using the yeast two-hybrid assay led to the identification of several interaction partners for GPR30, among them PATJ and FUNDC2. These interactions were verified by CoIP experiments and the interaction of GPR30 with PATJ could be confirmed. The effect of FSS on the expression of GPR30 was confirmed in HUVECs and was detected in other endothelial cell types. In HUAECs, HAoECs and HMEC-1 cells GPR30 was also found up-regulated upon FSS, suggesting that GPR30 may indeed play a key role in vascular physiology. Finally, the role of GPR30 and PATJ in the FSS response was investigated at the genome-wide transcript level in HMEC-1 cells. Interestingly, a different panel of genes was deregulated owing to FSS in cells over-expressing GPR30 compared to FSS alone.

Osteocytes make up the largest cell population in bone and are believed to be the main mechanosensory bone cells. During mechanical disuse and overuse, osteocyte viability is compromised and is found to be co-localized with increased osteoclastic bone resorption. Osteoclasts are recruited to remodel sites of apoptosis or bone microdamage; however, it is unclear whether the apoptotic or neighbouring healthy osteocytes are responsible for targeted bone remodeling. I hypothesized that apoptotic osteocytes are: (a) directly responsible for initiating bone remodeling by recruiting osteoclast precursors and directing osteoclast differentiation, and (b) indirectly responsible by signaling to nearby healthy osteocytes that, in turn, regulate osteoclastogenesis. In this in vitro study, apoptotic osteocytes were found to increase osteoclast precursor migration and osteoclast formation. Inhibition of the osteoclastogenic protein, receptor activator of nuclear factor kappa B ligand (RANKL), in conditioned medium abolished the osteoclastogenic effect of apoptotic osteocytes. Healthy osteocytes surrounded by apoptotic regions were modeled by applying apoptotic osteocyte conditioned medium to healthy osteocytes. These cells also promoted osteoclastogenesis, and had increased expression of macrophage colony stimulating factor (M-CSF) and vascular endothelial growth factor (VEGF). Inhibition of these factors abrogated the pro-osteoclastic effect of healthy osteocytes conditioned by apoptotic osteocytes. These findings support the hypothesis that apoptotic osteocytes directly and indirectly, by signaling to nearby healthy osteocytes, initiate osteoclastogenesis. One limitation of our and other conventional in vitro models is the lack of real-time cell communication and physiologically-relevant mechanical environment. Using a microfluidics approach, a miniature fluid shear delivery system was created for in vitro osteocyte cultures. The purpose of this microsystem was to increase control of the cell microenvironment for subsequent integration into scalable screening platforms or co-culture systems for studying osteocyte mechanobiology under physiological loading conditions. Fluid shear stress was periodically applied without external pumping using a deflecting elastomer membrane, where up to 2 Pa of oscillating shear stress was possible by manipulating membrane dimensions. Osteocyte culture, viability and calcium response were demonstrated in the microdevice. Further studies should attempt to characterize calcium signaling in osteocytes which, using a conventional macro-scale system, was found to dependent on cell-cell communication.

Osteocytes make up the largest cell population in bone and are believed to be the main mechanosensory bone cells. During mechanical disuse and overuse, osteocyte viability is compromised and is found to be co-localized with increased osteoclastic bone resorption. Osteoclasts are recruited to remodel sites of apoptosis or bone microdamage; however, it is unclear whether the apoptotic or neighbouring healthy osteocytes are responsible for targeted bone remodeling. I hypothesized that apoptotic osteocytes are: (a) directly responsible for initiating bone remodeling by recruiting osteoclast precursors and directing osteoclast differentiation, and (b) indirectly responsible by signaling to nearby healthy osteocytes that, in turn, regulate osteoclastogenesis. In this in vitro study, apoptotic osteocytes were found to increase osteoclast precursor migration and osteoclast formation. Inhibition of the osteoclastogenic protein, receptor activator of nuclear factor kappa B ligand (RANKL), in conditioned medium abolished the osteoclastogenic effect of apoptotic osteocytes. Healthy osteocytes surrounded by apoptotic regions were modeled by applying apoptotic osteocyte conditioned medium to healthy osteocytes. These cells also promoted osteoclastogenesis, and had increased expression of macrophage colony stimulating factor (M-CSF) and vascular endothelial growth factor (VEGF). Inhibition of these factors abrogated the pro-osteoclastic effect of healthy osteocytes conditioned by apoptotic osteocytes. These findings support the hypothesis that apoptotic osteocytes directly and indirectly, by signaling to nearby healthy osteocytes, initiate osteoclastogenesis. One limitation of our and other conventional in vitro models is the lack of real-time cell communication and physiologically-relevant mechanical environment. Using a microfluidics approach, a miniature fluid shear delivery system was created for in vitro osteocyte cultures. The purpose of this microsystem was to increase control of the cell microenvironment for subsequent integration into scalable screening platforms or co-culture systems for studying osteocyte mechanobiology under physiological loading conditions. Fluid shear stress was periodically applied without external pumping using a deflecting elastomer membrane, where up to 2 Pa of oscillating shear stress was possible by manipulating membrane dimensions. Osteocyte culture, viability and calcium response were demonstrated in the microdevice. Further studies should attempt to characterize calcium signaling in osteocytes which, using a conventional macro-scale system, was found to dependent on cell-cell communication.

At the present study we examined the effect of fluid shear stress on two different cell types. The cells studied were the Human Umbilical Vein endothelial cells and Smooth Muscle cells. For that purpose, a device which was simulating the arterial circulation was used. Shear stress is the hemodynamic force of blood. We show that this mechanical stress can efficiently parallelize the cellular morphology and induce changes at a gene transcription level. Specifically, we proove that shear stress is responsible for the upregualation of specific endothelial markers whereas can mediate the downregulation of smooth muscle cells markers in both cell types examined. / Στην παρούσα εργασία μελετήθηκε η επίδραση της διατμητικής τάσης ρευστού επάνω σε δυο διαφορετικούς τύπους κυττάρων. Τα κύτταρα που μελετήθηκαν ήταν τα Ανθρώπινα Ενδοθηλιακά κύτταρα απο φλέβα του Ομφάλιου λώρου και τα Λεία Μυικά κύτταρα. Χρησιμοποίηθηκε μια συσκευή η οποία προσομοίωνε την αρτηριακή κυκλοφορία του αίματος. Η διατμητική τάση ρευστού είναι η αιμοδυναμική δύναμη του αίματος. Στην εργασία δείχτηκε πως η δύναμη αυτή μεταβάλει τη μορφολογία των κυττάρων παραλληλίζοντας τα με τη ροή ενώ αυξάνει τα ενδοθηλιακά γονιδία και μειώνει τα λεία μυικά γονίδια και στους δυο τύπους κυττάρων που εξετάστηκαν.

Most women die of ovarian metastasis rather than the effects of the primary tumor. However, little is known about the factors that support the survival and secondary outgrowth of exfoliated ovarian cancer cells. In addition to genetic and molecular factors, the unique environment of the peritoneal cavity exposes ovarian cells to biophysical forces, particularly fluid shear stress (FSS). These biomechanical forces, only recently identified as a hallmark of cancer, induce rapid signaling events in attached and aggregated cells, a process termed mechanotransduction. The cellular responses to these forces and their impact on tumor initiation, progression, and metastasis are not understood. In order to delineate these phenomena, dynamic and syngeneic cell models are needed that represent the development of the disease and can be used in relevant engineered testing platforms. Thus, in an interdisciplinary approach, this work bridges molecular and cancer biology, device engineering, fluid mechanics, and biophysics strategies. The results demonstrated that even a low level of continual FSS significantly and differentially affected the viability of epithelial ovarian cancer cells of various stages of progression over time, and enhanced their aggregation, adhesion, and cellular architecture, traits of more aggressive disease. Furthermore, benign cells that survived FSS displayed phenotypic and genotypic changes resembling more aggressive stages of the disease, suggesting an impact of FSS on early stages of tumor development. After identifying a biological affect, we designed an in vitro testing platform for controlled FSS investigations, and we modeled the system fluid mechanics to understand the platform's performance capability. A cylindrical platform divided into annular sections with lid-driven flow was selected to allow continuous experiments sustainable for long durations. Tuning of the lid speed or fluid height resulted in a wide range of FSS magnitudes (0- 20 N/m2) as confirmed by analytical and numerical modeling. Further, detailed numerical modeling uncovered that FSS magnitudes experienced by cell aggregates were larger than previously observed, suggesting an even larger role of FSS in ovarian cancer. Finally, we built and engineered the designed platform to investigate changes in benign and cancer cells as a function of time and FSS magnitude. Device precision was balanced with biological consistency needs, and a novel platform was built for controlled FSS investigations. This work provides a foundational understanding of the physical environment and its potential links to ovarian cancer progression and metastatic potential. / Ph. D. / Most women die of ovarian metastasis rather than the effects of the primary tumor. However, little is known about the factors that support the survival and secondary outgrowth of exfoliated ovarian cancer cells. In addition to genetic and molecular factors, the unique environment of the peritoneal cavity exposes ovarian cells to biophysical forces, particularly fluid shear stress (FSS). These biomechanical forces, only recently identified as a hallmark of cancer, induce rapid signaling events in attached and aggregated cells, a process termed mechanotransduction. The cellular responses to these forces and their impact on tumor initiation, progression, and metastasis are not understood. In order to delineate these phenomena, dynamic and syngeneic cell models are needed that represent the development of the disease and can be used in relevant engineered testing platforms. Thus, in an interdisciplinary approach, this work bridges molecular and cancer biology, device engineering, fluid mechanics, and biophysics strategies. The results demonstrated that even a low level of continual FSS significantly and differentially affected the viability of epithelial ovarian cancer cells of various stages of progression over time, and enhanced their aggregation, adhesion, and cellular architecture, traits of more aggressive disease. Furthermore, benign cells that survived FSS displayed phenotypic and genotypic changes resembling more aggressive stages of the disease, suggesting an impact of FSS on early stages of tumor development. After identifying a biological affect, we designed an in vitro testing platform for controlled FSS investigations, and we modeled the system fluid mechanics to understand the platform’s performance capability. A cylindrical platform divided into annular sections with lid-driven flow was selected to allow continuous experiments sustainable for long durations. Tuning of the lid speed or fluid height resulted in a wide range of FSS magnitudes (0 − 20 N/m² ) as confirmed by analytical and numerical modeling. Further, detailed numerical modeling uncovered that FSS magnitudes experienced by cell aggregates were larger than previously observed, suggesting an even larger role of FSS in ovarian cancer. Finally, we built and engineered the designed platform to investigate changes in benign and cancer cells as a function of time and FSS magnitude. Device precision was balanced with biological consistency needs, and a novel platform was built for controlled FSS investigations. This work provides a foundational understanding of the physical environment and its potential links to ovarian cancer progression and metastatic potential.

Les interactions cellule-cellule et cellule-matrice extracellulaire (MEC) sont cruciales pour entretenir la cohésion tissulaire. Ces deux types d'adhésions sont fonctionnellement interconnectés par un dialogue permanent qui met en jeu des voies de signalisation convergentes régulant notamment l'architecture et la contractilité du cytosquelette d'acto-myosine sous-jacent. Ce dialogue permet d'établir un équilibre de forces intracellulaires en réponse à la tension appliquée par le milieu extérieur. L'endothélium des vaisseaux sanguins est un tissu soumis à des conditions mécaniques particulières. En plus des compressions intercellulaires subies par tout épithélium, les cellules endothéliales (CEs) doivent également subir et résister aux forces hémodynamiques du flux sanguin et à la rigidité de la lame basale – deux signaux mécaniques agissant de part et d'autre de l'endothélium. Les Cerebral Cavernous Maformations (CCM) ou encore angiomes caverneux sont des lésions vasculaires hémorragiques d'origine génétique qui se développent au niveau des capillaires du système nerveux central et qui se caractérisent par des défauts dans l'environnement proche des CEs. La perte des jonctions intercellulaires et du recouvrement par les cellules murales, l'organisation aberrante de la membrane basale aussi que la stagnation du flux sanguin sont les caractéristiques des CCM. C'est pourquoi nous avons choisi cette pathologie comme modèle intéressant de mécanotransduction mettant en jeu le dialogue entre les intégrines et les cadhérines. En effet, les trois gènes indifféremment mutés dans cette pathologie codent pour des protéines, CCM1-3, qui s'associent en un complexe ternaire et qui sont reconnues comme des acteurs importants de la régulation des jonctions adhérentes. Des études moléculaires et protéomiques montrant que le complexe CCM interagit avec la protéine ICAP-1, un régulateur négatif de l'intégrine β1, nous ont conduit à formuler l'hypothèse selon laquelle ce complexe jouerait un rôle pivot dans la signalisation croisée entre ces intégrines et cadhérines. Les études effectuées pendant ma thèse ont démontré que les protéines CCM régulent l'homéostasie tensionnelle médiée par les structures d'adhérence intercellulaires et à la MEC par leur action inhibitrice sur l'intégrine β1 et en controlant une balance d'activité entre les deux isoformes de ROCK, ROCK1 et ROCK2. Nous avons montré que, suite à la perte des protéines CCMs, la suractivation de l'intégrine β1 augmente la sensibilité des CEs aux signaux mécaniques comme la rigidité de la MEC ou les forces hémodynamiques du flux sanguin. Il en résulte une suractivation de la contractilité cellulaire dépendante de ROCK1 déclenchant une boucle de rétrocontrôle mécanique conduisant à l'amplification des tensions intra- et extracellulaire et brisant ainsi l'homéostasie tensionnelle pour favoriser le phénotype malin. / Cell-cell or cell-matrix interactions have crucial roles in the maintenance of the physical cohesion of any tissue. In addition, growing body of evidence indicates that these two adhesion systems do not act independently, but rather are functionally interconnected by a permanent crosstalk. This dialog usually operates via common molecules that trigger convergent signaling as well as by actomyosin network which, by providing physical link, contributes to establishment of intracellular force counterbalancing tension applied by extracellular surrounding. Blood vessels endothelium is a particular tissue in term of mechanical conditions. Apart from intracellular compression, endothelial lining needs to resist hemodynamic forces as well as rigidity of the basal membrane - two mechanical inputs acting from opposite sides of the endothelial layer. Cerebral Cavernous Malformation (CCM) is a sporadically acquired or inherited disease of venous capillaries within neuro-vascular unit characterized by defects in all aspects of local microenvironment. Loss of intra-endothelial junctions and mural cell coverage, aberrant organization of basal lamina as well as stagnant blood flow are features of CCM lesions. Thereby, CCM became for us an interesting model to study mechanotrasduction process and in this context, the cross-talk between integrin and cadherin mediated adhesion structures. Indeed, CCM proteins are well recognized players involved in a control of VE-cadherin mediated intracellular junctions. In addition, CCM1 was found to interact with ICAP-1, a negative regulator of β1 integrin, raising the possibility that this complex most likely acts as molecular node regulating β1 integrin/ VE-cadherin convergent signaling pathways.Studies performed during this thesis have demonstrated that CCM complex coordinates cadherin- and integrin-mediated tensional homeostasis by repressing β1 integrin activation and maintaining a balance of activity between the two isoforms of RhoA-associated kinases ROCK1 and ROCK2. We have found that β1 integrin sustained over-activation upon CCM proteins loss contributes to increased ECs sensitivity to mechanical cues, such as ECM physical reorganization or hemodynamic force that in turn activates ROCK1-dependent contractility. This establishes a positive feedback mechanical loop that breaks tensional homeostasis and switches on the malignant phenotype.