To my parents Und für Gerda, Margarete, Ida und Maria

Everything has been said already, but not yet by everyone. Karl Valentin

List of Papers

This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Zeller1, K.S., Idevall-Hagren1, O., Stefansson1, A., Velling, T., Jackson, S.P., Downward, J., Tengholm, A., Johansson, S. (2010) PI3-kinase p110α mediates β1 integrin-induced AKT activation and membrane protrusion during cell attachment and initial spreading. Cellular Signalling 22:1838–1848 1Equal contribution

II Riaz, A., Zeller, K.S., Johansson, S. (2012) Receptor-specific mechanisms regulate phosphorylation of AKT at Ser473: Role of RICTOR in β1 integrin-mediated cell survival. PLoS ONE 7(2): e32081

III Zeller, K.S., Riaz, A., Sarve, H., Johansson, S. (2012) The role of mechanical force and ROS in integrin-dependent signals. Manuscript

Reprints were made with permission from the respective publishers.

The cover picture shows fibroblast-like BJ hTERT cells.

Contents

Introduction ...... 11 Background ...... 13 Integrin structure and activation...... 13 Integrin activating proteins...... 15 Integrin signaling...... 17 The classics – The FAK-SRC-MAPK and PI3K-AKT pathways ...... 18 The FAK-SRC-MAPK pathway ...... 18 The PI3K-AKT pathway ...... 19 RHO GTPases...... 20 Adhesion, focal adhesion dynamics and spreading ...... 21 Mechanosignaling...... 24 Substrate stiffness...... 24 Mechanosensitive extracellular proteins ...... 25 (Intra)-cellular mechanosensitive proteins ...... 25 Reactive oxygen species ...... 27 Present investigations...... 29 Aim ...... 29 Paper I...... 29 Paper II ...... 31 Paper III ...... 32 Future plans...... 35 Acknowledgements ...... 36 References ...... 38

Abbreviations

ADMIDAS adjacent metal ion-binding site AKT AKR mouse T-cell lymphoma-derived oncogene product, also protein kinase B (PKB) ARP2/3 actin-related protein 2/3 BAD Bcl-2-associated death promoter DRF diaphanous family of formins DUOX dual oxidase ECM extracellular matrix EGF(R) epidermal growth factor (receptor) ENA/VASP enabled/vasodilator-stimulated phosphoprotein (fami- ly) ERK extracellular signal regulated kinase FA focal adhesion FAK focal adhesion kinase FAT focal adhesion targeting (domain) FERM band four point one, ezrin, radixin, and moesin (domain) FN fibronectin GAP GTPase activating protein GDI guanine dissociation inhibitor GDP guanosine diphosphate GEF guanine nucleotide exchange factor GRB2 growth factor receptor-bound protein 2 GSK3 glycogen synthase kinase 3 GTP guanosine triphosphate ILK integrin-linked kinase JNK c-JUN N-terminal kinase LIM kinase LIM (Lin11, Isl-1 & Mec-3) domain kinase LIMBS ligand-induced metal ion-binding site LOX 5-lipogygenase LPA lysophosphatidic acid MAPK mitogen-activated protein kinase MEF mouse embryonic fibroblast MIDAS metal-ion dependent adhesion site mTOR mammalian target of rapamycin

NF-κB nuclear factor kappa-light-chain-enhancer of activated B cells NOX NADPH (nicotinamide adenine dinucleotide phos- phate) oxidase p130CAS p130-CRK-associated substrate PAK p21-activated kinase PDGF platelet-derived growth factor PDK 3-phosphoinositide-dependent kinase PDZ post synaptic density protein, drosophila disc large tumor suppressor, zonula occludens-1 protein (domain) PH pleckstrin homology (domain) PI3K phosphatidylinositol 3-kinase β-PIX p21-activated kinase interacting exchange factor-β PKB protein kinase B (=AKT) PKC protein kinase C PSI plexin-semaphorin-integrin (domain) PTB phosphotyrosine binding (domain) PtdIns-3,4,5-P3 phosphatidylinositol-3,4,5-triphosphate PtdIns-4,5-P2 phosphatidylinositol-4,5-bisphosphate PTEN phosphatase and tensin homolog deleted on chromo- some 10 PTP phosphotyrosine phosphatase PYK2 proline-rich kinase 2 RAC1 RAS-related C3 botulinum toxin substrate 1 RAP RAS-related protein RGD L-arginine, glycine, L-aspartic acid (Arg-Gly-Asp) (peptide) RHOA RAS homolog gene family, member A RIAM RAP1-GTP-interacting adaptor molecule ROCK RHO-associated coil-coiled containing protein kinase ROS reactive oxygen species RT-PCR reverse transcription polymerase chain reaction SFK SRC family kinases SH SRC homology (domain) SOD superoxide dismutase SRC c-SRC (short for cellular SRC), rous sarcoma virus oncogene TGF-β transforming growth factor β TM transmembrane (domain) VN vitronectin WASP Wiscott-Aldrich syndrome protein (family) WAVE WASP verprolin-homologous protein (family)

Introduction

Every organism has to understand its environment and to access whether the surroundings are suited to stay and to propagate. This ability is also crucial for every cell within that organism and if it is missing things go wrong. One such example of things going wrong is the uncontrolled proliferation of cells during tumor formation and metastasis. Animals have thus evolved control mechanisms that tightly regulate critical processes such as cell survival, differentiation, proliferation, and migration. The information that cells receive through their adhesion sites to the extracellular matrix and through cell-cell contacts is of great importance for these processes. Integrins are highly conserved receptor molecules that can transmit such information. They connect (“integrate”, hence the name) the extracellular matrix outside the cell with the cytoskeleton inside the cell, which provides mechanical anchorage and is also part of their signaling function. The fact that integrins are expressed in animals from sponges to humans, and that they can be found in all cells of our body except for red blood cells, illustrates their importance and gives an idea of the variety of reactions that integrins can participate in.

11

12 Background

Integrin structure and activation Integrins are transmembrane, glycosylated cell surface receptors. They are heterodimers composed of subunits from two different, highly conserved protein families. In mammals, 18 α- and 8 β-subunits can together form at least 24 distinct heterodimers (1,2). Most of them link cells to the extracellular matrix (ECM), and some are part of cell-cell contacts. The extracellular domains are rather large, made up of around 800 amino acids. β4 is an exceptional integrin subunit in various aspects and is comprised of almost 1100 extracellular amino acids. The 20 to 30 amino acid long transmembrane (TM) domains form α-helices and are longer than necessary to span the membrane, which allows for the structural arrangements involved in affinity state changes. The cytoplasmic tails are in general a maximum of 70 amino acids in length, which is surprisingly short considering all described interacting proteins, especially with regard to the β- tails. β4 is an exception again – its cytoplasmic part is made up of more than 1000 amino acids. Although α- and β-subunits do not show any homology, there are several common motifs shared by the members of each subunit type. I domains are for example present in 9 of the α-subunits, in which they mediate ligand binding. Also a conserved metal-ion dependent adhesion site (MIDAS) containing a coordinating Mg2+ ion is located in the I domain (3). The β - subunit comprises a βI domain, which contains a MIDAS and two adjacent Ca2+-binding sites, called ADMIDAS (adjacent metal ion-binding site) and LIMBS (ligand-induced metal ion-binding site). A schematic picture of the α- and β-integrin subunit domain structure is presented in Figure 1. Ligand binding of a heterodimer without αI domain occurs in a cleft in the α- and β- subunit interface involving the MIDAS in the activated β-subunit and the so- called propeller domain in the α-subunit (4). Several conserved motifs are present in the cytoplasmic tails. Close to the membrane, the α-chains contain the GFFKR motif, which together with the HDR(R/K)E motif in the membrane-proximal part of the β-subunit is important for the regulation of integrin conformation. Furthermore, at least one NxxY/F motif is expressed in the majority of β-tails, which is the canonical recognition sequence for phosphotyrosine binding (PTB) domains.

13 Most β-subunits contain two NxxY/F motifs, one membrane-proximal and one membrane-distal.

Figure 1. Integrin domain structure. The plexin, semaphorin, and integrin (PSI) domain is actually located at the N-terminus of the β-subunit, but disulphide bonds link it to more C-terminal residues. The β-subunit contains furthermore four epidermal growth factor (EGF) repeats that are not depicted here. Reprinted by permission from Macmillan Publishers Ltd: Nature Reviews, Molecular Cell Biology (Shattil et al., 11(4):288-300). Copyright (2010)

Integrins can be present in different activation states on the cell surface and the activation process needs to be tightly controlled. This is most obvious in blood cells, where inappropriate integrin activation would lead to thrombus formation. How the transition from an inactive to an active integrin occurs and how this is regulated, is a field of intense research and certain controversies. However, a general concept is that integrins can adopt at least two conformations: an inactive (closed, bent) and an active (open, extended, ligand-binding). In addition, “primed” or intermediate states of certain integrins have been described (5-8).

14 The activation process is characterized by a pronounced conformational change involving the separation of the previously associated TM regions of the α- and β-subunit (see Fig.1). The inactive integrin conformation is mediated by an “inner clasp” in the cytosolic, membrane-proximal part and an “outer clasp” in the TM domain (9). Hydrogen-bonding occurs in the TM membrane-proximal part between a conserved lysine in the β-subunit and the oxygen of carbonyl residues in the α-chain. Additionally, the inactive ectodomain seems to stabilize the dimer association. The TM separation then leads to the extension of the ectodomains, involving a large conformational change allowing ligand binding and subsequent intracellular signaling events. The conformational change includes downward displacement of the α7 helix in the β I domain and a “swing-out” of the hybrid domain. These movements are possibly ligand-induced, although the exact timing is not yet clear (7,8,10). Depending on the circumstances, the activation process could be triggered “outside-in” induced by ligand binding or “inside-out” by integrin-activating proteins regulated through intracellular signals. Some reports claim, that TM domain separation does not always take place in outside-in activation events (11,12), and there are doubts, if pure outside-in activation actually occurs in a physiological context. Many studies about outside-in activation employed high-affinity ligands derived from e.g. bacteria, or artificial ligands such as activating antibodies or manganese ions that cause integrin activation by binding to the ADMIDAS site (13). Additionally, numerous studies have focused on αIIbβ3 integrins. Due to high homology, the results obtained with this heterodimer can most likely be transferred to other integrin dimers, but certain differences remain.

Integrin activating proteins Several proteins have been shown to bind to integrin cytoplasmic tails and to contribute to the regulation of their activity (7,14-16). Two proteins seem to be crucial for the activation: talin and kindlin. Talin, which has two isoforms in vertebrates, consists of a rod and a globular head domain. The head domain contains a band four point one, ezrin, radixin, and moesin (FERM) domain with three typical subdomains (F1, F2, and F3) and an additional F0 subdomain. The FERM domain is commonly found in proteins linking the cytoskeleton with the cell membrane. The PTB-like domain in F3 binds to the membrane-proximal NPxY motif in the β-tail and an additional site in F3 binds to the part of the β-cytoplasmic tail involved in the inner membrane clasp, thereby competing with the α-tail to form a salt bridge. A positively charged “membrane orientation patch” provides further interactions with the membrane (17,18). The talin head domain alone has been suggested to be sufficient to activate

15 αIIbβ3 integrins (19), but the full-length protein, which forms homodimers, seems to be needed for integrin clustering and focal adhesion (FA) assembly (20,21). A recent publication provides evidence though, that talin is dispensable in nascent adhesions (22), which is contradictory to previous results. Talin’s rod domain links integrins directly to F-actin. Talin itself is a tightly regulated molecule. Its activation from the autoinhibited state in which it resides in the cell is likely to involve locally produced lipid second messengers such as phosphatidylinositol-4,5- bisphosphate (PtdIns-4,5-P2) (23). An additional molecule involved in integrin activation and other aspects of cell adhesion is the small GTPase RAP1 (24-27). It cycles between the active GTP-bound and the inactive GDP-loaded state, which is controlled by guanine nucleotide exchange factors (GEFs) and GTPase activating proteins (GAPs). One known RAP1 effector is RAP1-GTP-interacting adaptor molecule (RIAM) in hematopoietic cells or its paralog lamellipodin in other cells such as fibroblasts (28). An appealing model has been presented, where RIAM (and possibly lamellipodin) recruits talin to the membrane by binding both RAP1 that contains membrane-targeting sequences and talin, which together lead to integrin activation (29,30). However, so far no direct interaction between lamellipodin and RAP1 has been shown (28). Both lamellipodin and RIAM promote actin polymerization probably though interactions with proteins of the enabled/vasodilator-stimulated phosphoprotein (ENA/VASP) family. Kindlins are a conserved protein family with three homologs in humans (31), and they have a unique structure. Their centrally located FERM domain is disrupted by a pleckstrin homology (PH) domain, and a PTB domain resembling talin’s PTB domain binds β-integrin cytoplasmic tails at their membrane-distal NxxY motif (32,33). Other conserved residues also contribute to the binding of kindlins to the β-tails (34). Kindlins are involved in both integrin activation and integrin downstream functions. For integrin activation however, kindlin seems to depend on talin. Also talin cooperates with kindlin and both proteins have to be expressed at a balanced level in order to synergize as shown for αIIbβ3 activation (21). There might be integrin-specific differences though, since the coexpression of kindlin-1 or kindlin-2 with talin head fails to activate α5β1 (33). Kindlins are recruited to the membrane by PtdIns-3,4,5-P3 and they are also part of FAs binding the adaptor migfilin and integrin-linked kinase (ILK), which points to a role of kindlins in spreading. Since kindlin-2 is found in cell-cell contacts and at least kindlin-1 and kindlin-2 have been detected in the nucleus (35), they might carry out functions outside cellular matrix adhesion sites as well (34). Several proteins can interfere with integrin activation, for example sharpin (36), ICAP1 (37), DOK1 (38), and filamin A. Filamin A binds to the membrane-proximal NPxY motif in β-integrin subunits and thus blocks talin

16 binding (39). Interestingly, the kindlin-binding protein migfilin can in turn associate with filamin A and it has been suggested that it is recruited by kindlin in order to prevent filamin A binding to the β-tail (40). A recent publication supports the view of migfilin and filamin A as proteins regulating integrin activation (41). In summary, integrin activation is complex and tightly regulated, and still the whole complexity of this process is not yet discovered, especially with regard to heterodimer-specific differences that are likely to exist.

Integrin signaling Integrins are key components of newly formed cell adhesion structures termed nascent adhesions that link the extracellular matrix to the cytoskeleton. Around these connections, many additional molecules assemble and they can mature into focal complexes and in a mechanosensitive manner into larger FAs. Integrins do not only mechanically link outside and inside of the cell but also transfer signals into both directions. A few years ago, an “integrin adhesome interaction map” was published (42), which contained 156 components at that time and more have emerged since then. Thus, integrins are involved in a multitude of molecular interactions and intracellular pathways that are translated into complex responses beginning with adhesion and migration, proceeding with differentiation, cell survival and proliferation to just name a few. Which components actually join to build up the adhesome, depends for example on cell and ECM type, the involved integrin heterodimers, matrix compliance, and mechanical forces. Furthermore, integrins are not the only receptors transmitting signals. Several classes of receptors exist in a cell that signal in response to certain stimuli and these can be soluble or insoluble, such as secreted or matrix- bound growth factors. Importantly, numerous signaling pathways are utilized both by integrins and by other receptor types and many of these pathways can converge. Integrins and other receptors also posses the ability to “cross- talk” with each other by modulating each other’s signaling responses or by activating each other. Integrins can for example activate platelet-derived growth factor (PDGF) receptors in the absence of PDGF (43) and chemokines are capable of activating integrins. Integrins also cooperate with proteoglycans, such as syndecans (44,45). In the following section, a few central integrin signaling pathways and modulating molecules will be presented.

17 The classics – The FAK-SRC-MAPK and PI3K-AKT pathways

The FAK-SRC-MAPK pathway So far, it is not known in which order FA components assemble. It is clear though that clustering of integrins, an increase in tyrosine phosphorylation and elevated levels of lipid second messengers are early events upon integrin ligation (20,46). Integrin clustering can be induced by multivalent extracellular ligands or ligands in close proximity to each other, or by multivalent intracellular proteins recruited to the cytoplasmic tails of integrins. The integrins are indirectly connected to actin and various signaling pathways are triggered and modulated, resulting eventually in changes of gene expression. Activation of the focal adhesion kinase (FAK)/SRC complex is a central event following integrin ligation. FAK, a protein expressed in most cells, contains a FERM domain, a C-terminal focal adhesion targeting (FAT) domain, a central kinase domain and proline-rich regions. It autophosphorylates itself at Tyr397 upon activation, probably by transphosphorylation. This can occur not only downstream of integrins, but also downstream of growth factors (47). The triggering event is still unknown, but FAK activation is initiated by unfolding, breaking the interaction between the FERM and the kinase domain of the molecule. For integrin-induced FAK activation, the two conserved NxxY motifs in the β- subunit have been shown to be required for autophosphorylation (48). The phosphorylated Tyr397 is recognized by SRC-homology 2 (SH2) domain- containing proteins such as SRC family kinases (SFK) e.g. cellular SRC (c- SRC, hereafter termed SRC). These proteins are non-receptor tyrosine kinases, of which SRC binds constitutively to β3 integrin cytoplasmic tails, whereas other SFKs coimmunoprecipitate e.g. with β1 and β2 subunits (49,50). SFKs are autoinhibited by a phosphorylated tyrosine in the C-terminal part and in response to integrin ligation, the autoinhibitory site (Tyr527) gets dephosphorylated and an activating autophosphorylation at Tyr416 occurs. The SRC-FAK interaction stabilizes SRC’s active conformation and allows SRC to phosphorylate more sites on FAK. SFK’s SH3 domain is critical for the localization of activated SRC to FA. It binds to a region between FAK’s FERM and kinase domain (the “linker” region) (51-53), whereas the FAT domain in FAK can bind paxillin, another important FA protein. The catalytically fully active FAK/SRC complex provides then docking sites for further adaptor and signaling molecules at FA sites. It should be noted that functional FAs form even in the absence of FAK or SFK family proteins, but as a result cell motility is impaired, which implies a crucial role for the FAK/SRC complex in resolving and turnover of focal adhesions (47,54).

18 Active FAK/SRC allows p130CAS to bind FAK with its SH3 domain and to get phosphorylated. Binding of the adaptor protein CRK to phosphorylated p130CAS can via DOCK180 activate the RHO-GTPase RAC1 (55), which then may induce lamellipodia formation. Also the interaction between phosphorylated paxillin and CRK downstream of the active FAK/SRC complex can lead to RAC1 activation. Furthermore, the phosphorylation of FAK Tyr925 enables binding of growth factor receptor-bound protein 2 (GRB2), which in turn can activate RAS leading to the activation of a serine/threonine kinase cascade, namely the RAF-MEK-ERK pathway (56). The extracellular signal regulated kinase (ERK) pathway is also activated downstream of several growth factors integrating adhesion and growth factor signals. Integrin-dependent adhesion signaling has a strong impact on several cellular growth factor responses (57), and the here mentioned ERK pathway together with the phosphatidylinositol 3-kinase (PI3K)-AKT pathway described in the next section are two main examples. The FAK/SRC complex can furthermore activate the c-JUN N-terminal kinase (JNK) pathway and in summary, the complex regulates cell survival, differentiation and proliferation. On another regulation level, activated FAK can translocate to the nucleus and directly bind to p53, thereby inhibiting the p53 functions as an inducer of apoptosis and cell cycle arrest (58,59). Proline-rich kinase 2 (PYK2) is a kinase resembling FAK in domain architecture and they can activate common signaling pathways, although there are differences (60). PYK2 is for example predominantly found perinuclearly (61) and although it has been proposed that PYK2 can compensate at least partly for FAK deficiencies (62,63), it is not currently apparent how this would work considering the different subcellular localization of these two proteins.

The PI3K-AKT pathway Another important pathway activated downstream of integrin ligation is the PI3K-AKT pathway. This pathway has a strong impact on cell growth and survival and has experienced most attention downstream of other types of membrane receptors, such as receptor tyrosine kinases and G-protein- coupled receptors. However, not so much work has been performed on PI3K signaling downstream of integrins in cell adhesion and spreading, though its role in (directed) cell motility has been appreciated (64). Importantly, PI3K signaling can also act upstream of integrins and cause their activation e.g. in blood cells experiencing inflammation- or coagulation-related stimuli (65- 68). An important function of PI3K is to regulate GEFs and GAPs and to thereby control small GTPase protein activity. In turn, RAS can activate several PI3K isoforms, which may indicate the existence of a feedback loop (69). Other small GTPases have been suggested to regulate PI3K as well. The PI3K proteins can be divided into three classes. Class I contains PI3Ks primarily associated with cell membrane receptors and these PI3Ks

19 are made up by one of four 110 kDa catalytic subunits associating with one of several regulatory subunits. This association negatively regulates PI3K. The inhibition can be released for example by binding of the regulatory subunit to pTyr in a receptor or to an adaptor protein. Concurrently, this brings the catalytic subunits close to the membrane, where they generate PtdIns-3,4,5-P3, preferentially by phosphorylating PtdIns-4,5-P2 (for review see (70)). One possible mechanism for how integrins recruit PI3K is via FAK. The SH2 domain of the PI3K regulatory subunit p85 can bind to phosphorylated Tyr397 in FAK (71). PtdIns-3,4,5-P3 affects numerous molecules and those are often recruited to and activated at the membrane via their PH domains. Upon PI3K activation, 3-phosphoinositide-dependent kinase 1 (PDK1) phosphorylates Thr308 in the activation loop of protein kinase B (PKB, also named AKT) (72-74). This reaction is dependent on an unfolding of AKT mediated by AKT binding to PtdIns-3,4,5-P3. The so-called PDK2 additionally phosphorylates Ser473 in the hydrophobic motif in the AKT C-terminus, which allows full activation and substrate specificity of AKT (72,75,76). The identity of this PDK2 downstream of integrins has been controversial, but our studies have recently shed some more light on the mechanisms regulating Ser473 phosphorylation ((77), see Paper 2 in this thesis). The activity of PI3K can be counteracted by lipid phosphatases of the PTEN (phosphatase and tensin homolog deleted on chromosome 10) family (78,79). Since AKT has many downstream effectors such as mTOR, BAD, NF-κB, and GSK3, regulating cell proliferation, survival and metabolism, deregulation of this pathway is associated with diseases. In cancer, activating mutations of PI3K and/or inactivating mutations of PTEN are frequently found (80,81). We hypothesize that there is an alternative, FAK-independent way of integrin-induced PI3K activation. This is based on the observations that Ser473 AKT phosphorylation can occur in response to β1 integrin ligation even in FAK-/- cell lines. It can also be detected in cells expressing and engaging only a mutant β1 integrin devoid of the two NxxY motifs needed for FAK activation (48). This reaction could involve a yet unidentified tyrosine kinase, which is neither a SFK member nor a PDGF- or EGF- receptor, upstream of PI3K (82).

RHO GTPases RHO GTPases constitute a subgroup within the family of small GTPases and regulate a variety of cellular processes. Their activity can be regulated on at least four levels. They cycle between an active GTP-loaded state, promoted by GEFs and an inactive GDP-bound state, promoted by GAPs at the plasma membrane. Furthermore, they can be sequestered by guanine nucleotide dissociation inhibiting factors (GDIs) in the cytosol (83). The subcellular

20 localization of the GTPases is crucial for their function as well and it is regulated by several interactions and posttranslational modifications. An early event upon integrin ligation is the inactivation of RHOA, which reduces contractility and at the same time increases the activity of RAC1 and CDC42 promoting nascent adhesion and lamellipodia formation, and spreading. The inhibition of RHOA can be mediated by the active FAK/SRC complex. SRC can increase p190RHO-GAP activity directly by phosphorylation (84) and FAK forms a complex with p120RAS-GAP and p190RHO-GAP, which antagonizes RHOA activity (85). Downstream of paxillin, p21-activated kinase (PAK)-interacting exchange factor-β (β-PIX) can recruit and activate RAC1. PAK1 controls this reaction by competing with RAC1 for β-PIX binding (86). In a second spreading phase or in migration, a complex of FAK with RGNEF (p190RHO-GEF) is involved in integrin-mediated RHOA reactivation in order to allow FA formation, maturation, and stabilization (87). Interestingly, RHOA activity inhibits RAC1 activity and vice versa. RHOA suppresses RAC1 by RHO-associated coil-coiled containing protein kinase (ROCK)-mediated activation of a RAC-GAP, named FIL-GAP (88). RAC1 on the other hand leads to reactive oxygen species-mediated inhibition of phosphatases that inhibit p190RHO-GAP. This leads to increased GAP activity and thereby suppresses RHOA activity (89). Furthermore, lipids generated by PI3K can also target GEFs and GAPs, which all contain at least one PH domain, to the plasma membrane where they in turn regulate RAC1 and other RHO-GTPases (90). As mentioned previously, PI3K itself is regulated by RHO-GTPases possibly creating a feedback loop for protrusion formation (70). RHO guanine dissociation inhibitors (RHO-GDIs) provide another level of regulation. They usually bind RHO-GTPases in their GDP-loaded state preventing activation by GEFs, although it is likely that they can also sequester GTP-loaded RHO-GTPases, which inhibits effector binding as e.g. shown for RAC1 (91). GDIs can also interact with several cytoskeletal proteins, possibly localizing RHO-GTPase activity within the cell (92), while GDIs themselves can be regulated e.g. through phosphorylation. RHO- GDI phosphorylation by for example PAK reduces GDI affinity for RAC1 (93,94). Additionally, protein kinase Cα (PKCα) has been shown to phosphorylate RHO-GDIα on Ser34 in a PtdIns4,5-P2-dependent mechanism, resulting in the release of RHOA, its GTP loading and subsequent exertion of its functions (95).

Adhesion, focal adhesion dynamics and spreading Cell adhesion and spreading are crucial processes in development and tissue homeostasis (96). Although much work has been performed in the 2D environment, FA formation also occurs in a 3D matrix (97). One central

21 process is the connection of the cytoskeleton with the ECM. Talin is able to create this linkage by binding integrins and actin. It contains cryptic binding sites for vinculin, another filamentous (F)-actin-binding protein involved in FA growth and maturation (98). Vinculin itself possesses cryptic actin and talin binding sites (99). Actin crosslinking can be carried out by several proteins, including α-actinin (100), which is involved in FA maturation together with myosin II, and also by filamins. An important regulator of integrin-actin interactions is integrin-linked kinase (ILK), which is recruited to the membrane by PtdIns-3,4,5-P3 (101,102). The pseudo-kinase ILK binds PINCH and the actin-binding proteins parvin and paxillin (103). It stabilizes integrin clustering and thereby influences adhesion strength. Paxillin is found already early in adhesions and acts as a “molecular platform” (20), whereas tensin, another actin cross-linker, is recruited rather late to FAs. Together with the ILK- pinch-parvin complex, tensin is involved in the formation of even larger fibrillar adhesions (104). Another scaffold protein, zyxin, can remodel actin stress fibers in a force-dependent manner (105,106). The formation and maturation of FA is closely associated with several forces acting on the involved molecules during cell adhesion and spreading. For example, the so-called retrograde actin flow in a cell from the periphery to the cell center, is much faster in the lamellipodium, a thin protruding membrane extension, than in the zone right adjacent to it, the lamellum. In the lamellum, the retrograde flow is driven by myosin II, whereas it is powered by the actin polymerization in the lamellipodium. Nascent adhesions form in the lamellipodium, which, simplified, slows down the actin flow by friction and moves the boundary between lamellum and lamellipodium forward (105). Subsequently, the adhesions (linked to actin via talin) grow and elongate in the direction of applied tension in a myosin II-dependent process (107). The ability of adhesions to grow and strengthen in response to force, and to even transmit force at the same time as the components are constantly turned over, is difficult to understand. Quite recent work shows that nascent adhesions move for a few seconds towards the center of the cell before they start to grow (108). The softer the substrate, the more the adhesions move, and on a glass substrate, there is no visible displacement. It is suggested, that this corresponds to a “frictional slip” of integrin molecules on compliant surfaces (108). Integrins are connected both with the ECM and the cytoskeleton in nascent adhesions, but cannot yet bear the contractile forces. Therefore, integrin-ligand bonds may break and reform during this phase. Tension seems to build up during the slipping and leads by a yet unknown mechanism to a more stable linkage to the ECM. After this, integrins in the FAs are stable in relation to the ECM in a mature adhesion. However, other cytosolic FA components still move with the actin filaments in relation to the ECM, although with lower speed

22 (109,110). This “clutch” mechanism could influence signaling by controlling the time the FA components have to interact with each other in relation to the experienced substrate compliance and forces. Interestingly, the composition and density of the actin network differs in lamellipodia and lamella (111). In lamellipodia, high concentrations of the actin-related protein 2/3 (ARP2/3) complex can be found, where it may nucleate actin polymerization after activation by Wiscott-Aldrich syndrome protein (WASP/Scar) or WASP-family verprolin-homologous protein (WAVE) family proteins. In contrast, the nucleation in lamella is probably mediated by the diaphanous family of formins (DRFs) (112). The actin network is less dense in the lamella and they contain myosin II, whereas lamellipodia do not (112-114). Local actin polymerization at FAs is also regulated by the ENA/VASP actin-regulating protein family (106) and can be facilitated by zyxin, probably in concert with ENA/VASP. Another protein involved in actin polymerization is the actin-severing protein cofilin (115). It produces free barbed ends by “cutting” the filaments, which may allow actin polymerization. It is also reported to depolymerize actin (116), and which process is actually dominating seems to vary dependent on the concentration of free cofilin and free globular (G)-actin monomers (117). Two main mechanisms regulating cofilin activity have been described: cofilin phosphorylation at Ser3 (118) and binding of cofilin to PtdIns(4,5)P2 or cortactin (117,119). Notably, both mechanisms lead to inactive cofilin by inhibiting cofilin binding to F- or G-actin. Dephosphorylation of cofilin can be triggered by several surface receptors, and various kinases and phosphatases have been shown to act on cofilin. The LIM kinase family (activated for example downstream of RHOA/ROCK and RAC1/CDC42/PAK) and the phosphatases slingshot, PP1, and PP2A/PP2B are prominent examples (117,120). Phospholipase C on the other hand hydrolyzes PtdIns(4,5)P2 e.g. downstream of EGF and releases the probably dephosphorylated (active) cofilin. Interestingly, the total amount of phosphorylated cofilin increases in response to EGF in the investigated carcinoma cells (121), though cofilin still seems to sever actin filaments. This illustrates the complex regulation of cofilin activity, especially with regard to possibly differentially regulated separate subcellular pools. Additionally, cofilin acts in concert with other actin polymerization- regulating proteins. Syndecans, another type of transmembrane adhesion receptors, are also involved in cellular adhesion (recently reviewed e.g. in (44,122)). Syndecans are proteoglycans consisting of a core protein and covalently bound heparan sulfate chains, and they can bind ECM proteins such as fibronectin, vitronectin, and laminin. In mammals, the family consists of four members (syndecan 1-4). Syndecans act as coreceptors for growth factor signaling and

23 they are involved in numerous signaling pathways, although they, as integrins, lack intrinsic kinase activity. The divergent syndecan ectodomains are able to interact with a multitude of extracellular molecules including ECM proteins via their heparan sulfate chains, but they also regulate other cell surface receptors including integrins, for example the activation of αvβ3 and αvβ5 (123,124). The ectodomain of syndecan 1 together with these integrins were shown to recruit insulin-like growth factor 1 receptor (IGF1R) into a complex, which leads to autophosphorylation of IGF1R and inside-out activation of the integrins (125). Notably, syndecan ectodomains are trypsin-sensitive, which is important when performing and interpreting adhesion experiments. The intracellular domains of syndecans are conserved and can bind PDZ domain-containing proteins via their C2 region. The cytoplasmic parts can also interact with kinases such as FAK, with cytoskeletal proteins (for example α-actinin (126)) and various other signaling molecules. Syndecan 4 for example has been implicated in focal adhesion formation and in the activation of PKCα (127,128).

Mechanosignaling As discussed earlier, cells may experience several types of forces. Apart from the contractile force generated inside the cell, there are external mechanical stimuli such as shear stress, stretching, and pressure. Research on how different, specialized cell types sense and respond to these stimuli is a field of growing interest. Since work presented in this thesis (Paper III) deals with signals induced by stretching of fibroblasts, this part of the introduction focuses mostly on stretch-induced signaling.

Substrate stiffness Mechanosensing starts with cells reacting to the compliance of their substrate. Loose connective tissue is an example for a compliant and bone tissue for a stiff substrate. The substrate compliance can determine basic cellular behavior. Mesenchymal stem cells for example differentiate into different lineages according to the ridigity of their matrix, as elegantly shown by Engler et al. (129). The compliance influences also the cytoskeletal organization: a stiff substrate leads to pronounced FAs and stress-fiber formation, whereas these structures are less distinct when the substrate is more compliant (130). Similarly, FAs and stress fibers are less pronounced if the area available for cell spreading is restricted (131,132). This implies that cells adjust their own force generation according to the conditions they experience, and this in turn influences the already described signaling pathways downstream of integrins (133,134).

24 Mechanosensitive extracellular proteins Several proteins residing in the extracellular matrix can “react” to mechanical stimuli. Soluble fibronectin dimers for example are assembled into insoluble fibrils through interactions with syndecans and integrins (α5β1) on the cell surface (135). Myosin-dependent cell contraction leads to the exposure of cryptic binding sites of fibronectin that are necessary for its polymerization (fibrillogenesis) (136). Cells also remodel collagen fibrils in an active force-dependent manner (137). Another example is the potent and multifaceted cytokine transforming growth factor β (TGF-β), which is secreted to and stored in the ECM in form of the so-called large latent complex. The dissociation of TGF-β from this complex can be controlled by several mechanisms involving integrins and proteases (138). All αv integrins and α8β1 can bind latent TGF-β1 and TGF- β3 in vitro via their RGD motif; notably this motif is not present in latent TGF-β2. Only αvβ6, αvβ5, αvβ3, and αvβ8 have been actually shown to release the cytokine, and for αvβ6 and αvβ5 integrins this seems to be mediated by cellular contractility in combination with resistance from the ECM (139). Integrin αvβ8 does not link to the actin cytoskeleton and metalloproteinase MT1-MMP activity is required to create active TGF-β together with αvβ8 (140). In turn, TGF-β is able to modulate ECM production in addition to carrying out a multitude of other important functions. TGF-β, being activated by force and regulating ECM production, illustrates one principle of how cells can adjust their matrix according to the mechanical burden. Cells sense and exert force and these forces in turn regulate ECM-associated gene expression. The following changes in matrix composition and compliance are again detected by the cells; they react accordingly and so keep the forces in balance (134).

(Intra)-cellular mechanosensitive proteins It has become clear that cell stretching can trigger a multitude of signaling pathways. Depending on the cell type, these could comprise the PI3K/AKT and the MAPK and JNK pathways, responses mediated by reactive oxygen species (ROS), by mechanosensitive ion channels and Ca2+-dependent signaling. Mechanosensitive ion channels can be classified by their ion selectivity and sensitivity and if they are stretch-activated or -inactivated (141). Force transmission via certain cytoskeletal or cytoskeleton-associated proteins (142) as well as direct responses to tension in the membrane lipid bilayer have been suggested as “gating” mechanism. Most data describing the latter mechanism has been generated in prokaryotes (141). In non-sensory cell types, research focuses mainly on specialized cell types such as cardiac, smooth muscle, bladder, and endothelial cells.

25 In a way surprising, there is no formal proof so far that integrin conformations are directly regulated by a force-sensitive mechanism. It has been suggested though, that forces could induce or stabilize the active conformation of integrins (143,144). Figure 2 summarizes schematically a few suggested mechanosensitive components in and outside a cell that could react to internal or external mechanical force.

Figure 2. Overview about compartments and molecules possibly involved in mechanosignaling triggered by internal forces (exerted by the cell) or external forces such as mechanical stretching.

While there is no clear evidence for direct force-regulated integrin conformational changes, stretch-sensitive intracellular molecules have been described that seem able to translate force at adhesion sites and adherence junctions into biochemical signaling events. Two examples are talin and p130CAS. p130CAS has been shown to be extended by force, thereby providing better access for SRC to phosphorylate p130CAS, which subsequently leads to the activation of RAP1 (145,146). Similarly, as described above, it has been shown that force recruits vinculin to focal complexes (147,148) and that talin’s buried vinculin binding sites in the rod need to be in an active conformation to allow vinculin binding (149). Later, it was described that stretching of talin rods could expose the vinculin binding sites (150). The recruitment and integration of vinculin into focal complexes is likely to be important for reinforcing the connection of (integrin-)talin-actin, since vinculin can link talin to actin, and thereby it contributes to the stability and strengthening of focal adhesions (151).

26 Zyxin is another protein that might respond to force directly since it is found in contractile focal adhesions but not in focal complexes lacking connection to myosin II. When contractility is reduced, zyxin’s dissociation from the focal adhesion increases rapidly (152,153) and it has been implicated in stretch-induced actin polymerization at focal adhesions (106). Zyxin contains α-actinin and ENA/VASP binding sites and it is necessary to recruit ENA/VASP to FAs (154).

Reactive oxygen species Reactive oxygen species (ROS) are important molecules involved in numerous processes from host defense (“the oxidative burst”) to activation and regulation of cellular signaling pathways. In addition, ROS are part of the “oxidative stress” response and can exert toxic effects by non-specific reaction with molecules in our body. This implies, that both too high and too low concentrations of ROS can lead to disease depending on the context. ROS molecules comprise a range of molecules with different chemical properties, for example superoxide that can be converted enzymatically and non-enzymatically into hydrogen peroxide, the hydroxyl radical, and singlet 1 1 oxygen ( O2). Hydrogen peroxide and O2 can give rise to radical species, and hydrogen peroxide, in contrast to most other ROS, is able to pass cell membranes due to its neutral nature (155). Mitochondria are claimed to be the major source of ROS in a cell, where superoxide is created as a byproduct of ATP production. Most relevant for this ROS production seem to be complex I and complex III in the mitochondrial respiratory chain, although other sites in mitochondria may also produce superoxide (156). When the respiratory complexes leak electrons to oxygen, the resulting superoxide and its reaction products can reside in the mitochondrial matrix or in the intermembrane space. Mitochondrial superoxide dismutase (SOD) can convert superoxide into hydrogen peroxide. Since different cell types seem to vary with regard to their “equipment” with antioxidant defense systems (e.g. ROS converting enzymes such as SOD, catalase, glutathione peroxidase and additional scavengers), the exact consequences of mitochondrial ROS production are difficult to predict (156,157). Even other sources such as the mostly plasma membrane-bound NADPH oxidase (NOX)/dual oxidase (DUOX) family, and 5-lipogygenase (LOX) at intracellular membranes contribute to ROS production. How much they contribute to the overall ROS generation probably depends substantially on specific triggers and localization. The family of NOX enzymes comprises 7 members, NOX1 to NOX5 and DUOX1 and DUOX2. Whereas the latter are regulated by calcium (158), NOX1 to NOX4 require the regulatory subunit p22PHOX and further supportive molecules such as NOX organizers, NOX activators, and RAC1

27 or RAC2 in order to function (158,159). gp91PHOX is the catalytic subunit of NOX2, which was first found in phagocytic cells. All cell membrane- bound NOXes secrete superoxide into the extracellular space. There it is rapidly converted to hydrogen peroxide, which can then enter the cell via aquaporins (160) or by diffusion and exert signaling functions. The DUOX enzymes and NOX3 and NOX5 show rather restricted expression patterns, whereas NOX4 can be found in various tissues and cells, and also on intracellular membranes (161,162). NOX1 is highly expressed in colon, but also in certain other tissues (163,164). It has been shown that integrins are able to trigger ROS production and appropriate ROS levels affect many signaling pathways downstream of growth factors, cytokines and integrins in order to regulate e.g. cell survival and differentiation. ROS are promiscuous molecules, and in particular they react with several phosphatases (165), certain kinases and matrix metalloproteinases (155,166), transcription factors (155,167) and even actin (168,169) and actin-associated molecules (170-172). This often involves redox-sensitive residues such as cysteines. A classical example for the influence of ROS on a signaling pathway is the inhibition of phosphatases, e.g. PTEN, which allows sustained phosphorylation and activity of the signaling pathway, in this case PI3K (165). Integrin-induced ROS production has been proposed to derive mainly from mitochondria (173,174), NOXes, and LOX (175,176). Mitochondria are likely to be the dominant source during early cell spreading and LOX during the later spreading phase (173). Also downstream of mechanical cell stretching, NOXes (177) and mitochondria (178,179) have been proposed as ROS sources. Intriguingly, ROS can regulate RHO-GTPases in two directions. They can on one hand directly activate RHOA independently of GEFs by oxidation of a certain cysteine residue (180). On the other hand, RAC1-induced ROS production indirectly down-regulates RHOA activity by inhibiting a phosphatase responsible for the dephosphorylation of p190RHO-GAP, which leads to increased hydrolysis of GTP-bound active RHOA (89).

28 Present investigations

Aim The overall aim of this work was to understand in more detail how integrins signal and which downstream effectors are involved in their signaling. We focused particularly on adherent, often fibroblast-like cells and β1 integrins, since it is increasingly acknowledged that the different integrin heterodimers differ not only in their ligand binding specificity but also in their signaling functions. An important response triggered by integrins and several other receptors is the activation of the PI3K/AKT pathway, which was studied in Paper I and II. We aimed to identify the isoforms of the two enzymes activated by β1 integrins and some of the downstream reactions. In Paper III, we investigated the contribution of reactive oxygen species to integrin signaling under various conditions in order to find out if these molecules can explain some missing pieces of the integrin signaling puzzle.

Paper I PI3-kinase p110α mediates β1 integrin-induced AKT activation and membrane protrusion during cell attachment and initial spreading

The main aim of this study was to identify the PI3K isoform involved in early cell spreading and AKT activation downstream of integrins. We investigated the early spreading reaction using total internal reflection fluorescence (TIRF) microscopy, which allows visualization of the membrane close to the substrate. We showed that β1 integrins alone, (without growth factor contribution), can induce spreading on a β1 integrin- selective ligand, and that the spreading was indeed triggered by the β1 integrin subunit, since cells with a truncated β1 integrin tail could attach, but remained round. The spreading process also required actin polymerization and similar results were obtained with primary mouse embryonic fibroblasts (MEFs) and the GD25β1 cell line. Recording the cell spreading in the presence of general PI3K inhibitors revealed that the spreading reaction was dependent on PI3K. After determining the PI3K catalytic isoforms expressed by GD25β1 cells with

29 RT-PCR, we used isoform-selective inhibitors in order to identify the catalytic subunit responsible for spreading. The p110α inhibitor YM024 reduced the spreading rate significantly, while the other inhibitors had no effect. YM024 also slowed down spreading of the primary MEF cells, even when seeded on fibronectin, which is able to bind several integrin heterodimers. This points to a role of p110α downstream of β1 and possibly of additional integrins. Inhibiting p110α also reduced the activation of AKT measured by the phosphorylation of AKT-Ser473 in response to adhesion mediated by β1 integrins. We confirmed this using another p110α inhibitor (PIK75), but failed to do so with isoform-specific siRNA; however, the complications with knock-down approaches targeting these enzymes are known and discussed in the literature (69,181,182). Briefly, compensatory mechanisms could arise since in mammals theoretically 15 PI3K class I isoforms could form (69). Although certain isoforms are likely activated preferentially by distinct receptors (183-185), it is still not well understood what determines this specificity. Other unexpected effects might occur due to altered ratios of regulatory to catalytic subunits after knock-down or knock-out (182,186,187). In SYF cells, which lack the SRC family kinases SRC, YES, and FYN, spreading proceeded almost identically to SYF cells transfected to reexpress SRC. Also the use of an SFK inhibitor (SU6656) had little effect on the spreading rate of GD25β1 cells. Another SFK inhibitor (PP2) strongly slowed down spreading; however, it had a similar inhibitory effect on SYF cells, indicating that it targets additional (non-SFK). Similarly, FAK knock-out cells and FAK-expressing cells displayed very similar spreading rates and the phosphorylation of AKT Ser473 was not impaired in the FAK knock-out cells. FAK seems therefore dispensable in the early spreading reaction and for the activation of PI3K. It has previously been reported by our group that SRC and FAK are not necessary for the activation of AKT downstream of β1 integrins (82). It is known that PI3K/p110α-mediated functions are influenced by RAS, therefore we were interested in investigating the role of p110α-RAS interactions downstream of integrins. Gupta et al. (188) generated transgenic mice, which carry a double mutation targeting two amino acids in p110α that are critical for p110α interaction with RAS. These mice were surprisingly resistant to oncogenic RAS-induced tumor formation. Furthermore, MEFs derived from these mice show reduced or almost abrogated AKT activation downstream of EGF and FGF2 respectively, whereas PDGF-induced AKT phosphorylation is not affected (188). Interestingly, we found that AKT activation downstream of β1 integrins was strongly impaired in the mutant MEFs although the early spreading rates of wild-type and p110α-mutant cells were indistinguishable. Considering the high frequency p110α mutations in cancer, we suggest that these mutations could underlie the

30 ability of cancer cells to survive independent of integrin- and adhesion- induced signaling.

Paper II Receptor-specific mechanisms regulate phosphorylation of AKT at Ser473: Role of RICTOR in β1 integrin-mediated cell survival

In this project our central question was, which kinase is responsible for phosphorylating the hydrophobic motif of AKT (Ser473 in AKT1) downstream of β1 integrins and the PI3K subunit p110α. The identity of this so-called phosphoinositide-dependent protein kinase 2 (PDK2) has been debated for long and it has been particularly controversial for integrins. A large body of literature has implicated ILK as the integrin-activated PDK2 (189,190), while more recent reports argue that ILK is actually not a kinase (191,192). Even the function of the Ser473 phosphorylation site is not entirely clear (193), but most likely, it contributes to full activation (194) and the stabilization of the active conformation of AKT (195). Notably, the regulatory phosphorylations of Thr308 in the catalytic domain and of Ser473 may occur independently of each other (196), or alternatively, the phosphorylation of Thr308 may depend on prior phosphorylation of Ser473 (197) rather than the other way around. It has also been suggested that phosphorylation of the hydrophobic motif could give rise to AKT substrate specificity. Furthermore, we set out to identify, which isoform(s) of AKT do β1 integrins activate and if the phosphorylation of Ser473 is functionally linked to survival. Additionally, we were interested in the role of PAK downstream of β 1 integrins, since PAK has been shown to be required as a scaffold downstream of growth factor-induced AKT phosphorylation. Immunoprecipitations using MCF7 cells on the β1 integrin-selective ligand invasin revealed that in these cells both AKT1 and AKT2 were activated downstream of β1 integrin-mediated adhesion. Although western blotting can only be regarded as a semi-quantitative approach, in MCF7 cells AKT2 seemed to be the preferred activated isoform. This is interesting since AKT2 is reported to localize to the mitochondrial membrane, which is further discussed in this chapter during the presentation of Paper III. In addition we showed that the RICTOR-mTOR complex (TORC2) is the principal kinase phosphorylating AKT Ser473 downstream of β1 integrin- mediated adhesion. RICTOR knock-down by siRNA clearly impaired this reaction and also the phosphorylation of targets downstream of AKT including FOXO Thr24 and BAD Ser136. The phosphorylation of AKT Ser473 was shown to promote cell survival since less viable (Trypan blue exclusion method) and more apoptotic cells (TUNEL staining) were detected

31 after RICTOR knock-down than in control siRNA-treated cell cultures. Adhesion and spreading in general were not influenced by RICTOR siRNA treatment, indicating that RICTOR knock-down did not influence basic integrin functions and signaling; therefore, it selectively interfered with the generation of survival signals. Furthermore, ILK was dispensable for the phosphorylation of AKT Ser473 after β1 integrin stimulation, although ILK is clearly required for several other integrin functions. PAK1 and PAK2 could also be efficiently knocked-down without impairing the phosphorylation of the hydrophobic motif of AKT downstream of β1 integrins. We also investigated the importance of the mentioned kinases and adaptors for AKT Ser473 phosphorylation following stimulation of certain growth factor receptor types. In HeLa cells, the TORC2 complex was necessary for this reaction downstream of the tyrosine kinase receptors EGFR and PDGFR, but not downstream of the G-protein coupled lysophosphatidic acid (LPA) receptor. However, in MCF7 cells knock-down of RICTOR decreased the phosphorylation of AKT Ser473 in response to LPA, which might be explained by differences in the LPA receptor subtypes expressed by these cells. This is in line with our finding that the LPA- induced response in HeLa cells follows slower kinetics compared with the response in MCF7 cells. Thus, certain LPA receptors may engage an alternative PDK2 apart from TORC2, which would be consistent with the observation that even in RICTOR knock-out mouse embryos, a certain level of AKT Ser473 phosphorylation remains (198). Interestingly, ILK knock-down did not affect AKT Ser473 phosphorylation by any of these receptor types in HeLa or MCF7 cells. In contrast, PAK1 and PAK2 knock-down reduced AKT Ser473 phosphorylation stimulated by LPA and PGDF, but not by EGF as tested in both HeLa and MCF7 cells. These results were confirmed using the PAK inhibitor IPA3.

Paper III The role of mechanical force and ROS on integrin-dependent signals

Our aim in this study was to compare the integrin signaling induced by uniaxial, cyclic stretching with early adhesion-induced signaling. We chose to focus on AKT, ERK, cofilin, and p130CAS phosphorylation levels as read-outs, because these proteins have been reported to be prototype integrin signaling molecules, or their phosphorylation is indicative of actin dynamics or mechanosensitivity. Furthermore, we used subconfluent cell layers in order to minimize cell-cell contact-induced signaling, since some of our

32 experiments (data not shown) indicated that less dense cell layers reacted differently than confluent cell layers. When we compared 30 minutes adhesion to ECM protein-coated cell culture plastic with 10 minutes stretching on ECM protein-coated silicon, ERK phosphorylation was the only event that occurred for both stimuli, and was more pronounced after stretching. Increased phosphorylation levels for the other proteins were only clearly observed in response to adhesion. Notably, an increase in phosphorylation of cofilin indicates that it is less active, although its regulation is more complicated than that. The comparisons were performed using two different fibroblast-like cell lines (GD25β1 and BJ hTERT cells) and vitronectin (VN) or fibronectin (FN) as ligands. VN should primarily engage αvβ3 and FN should bind α5β1 integrin in the presence of the αvβ3-blocking RGD peptide. Not many cell line- or integrin-specific differences were seen, although in general AKT was weakly activated in BJ hTERT cells and AKT phosphorylation was even decreased following stretching. Interestingly, most of the signals detected after 30 minutes of spreading were not dependent on intracellular contraction, because neither the presence of a ROCK inhibitor nor of an inhibitor of myosin II influenced them. The FAK phosphorylation level was an exception, since it was reduced by the myosin II inhibitor. Another question we wanted to answer was if ROS were involved both in adhesion- and mechanical stretch-induced integrin signaling and if so, what their origin was. Since the administration of Rotenone, an inhibitor of complex I in the mitochondrial electron transport chain, efficiently reduced the spreading rate of GD25β1 cells on the β1-selective ligand invasin and inhibited AKT phosphorylation in response to spreading in BJ hTERT and GD25β1cells, we concluded, that mitochondrial ROS were involved in mediating these reactions. In Paper II we described that β1 integrins may preferentially activate AKT2, and AKT2 is known to colocalize with mitochondria (162). We speculate that β1 (and possibly other) integrins specifically activate an AKT2 pool that upon phosphorylation translocates from the plasma membrane to the mitochondria. Alternatively, AKT2 could be phosphorylated at the mitochondrial membrane. In contrast to the dramatic inhibition of AKT phosphorylation by Rotenone during cell spreading, the drug did not have any marked effect on stretch-induced signals. These results suggest that integrin ligand binding affects mitochondria. Furthermore, Rotenone affected phospho-ERK levels differently in the two cell lines: during spreading, phospho-ERK was essentially abolished in BJ hTERT while it was strongly elevated in GD25β1 cells. Possibly, these results relate to the more complex regulation of the ERK pathway compared to the AKT pathway. These cell line-specific observations illustrate the challenges that are associated with elucidating the roles of ROS in biological systems.

33 Knocking-down p22PHOX, a necessary subunit of several NOXes, in BJ hTERT cells, impaired ERK phosphorylation both after spreading and stretching on VN. Interestingly, the same experiments on FN+RGD did not reveal any consistent effect of p22PHOX knock-down. This could point to integrin subunit-specific signaling to NOXes but more experiments are needed to confirm and explain this finding. NOX4 was previously found to colocalize with intracellular membranes, e.g. mitochondrial membranes (162), which could be consistent with a model where superoxide (leaking from mitochondria or produced by NOX) acts locally and inhibits the activity of certain phosphatases, possibly after being converted into hydrogen peroxide. Decreased ROS amounts would lead to more phosphatase activity and thus to less phosphorylation of target proteins, e.g. AKT and ERK. Surprisingly, the presence of the antioxidant vitamin C, which we added in the form of a stable derivative (Asc-2P) in order to avoid reported problems resulting from the short half-life and possible prooxidant effects of “normal” vitamin C, had no clear effect on any integrin-induced signaling in BJ hTERT cells. However, Asc-2P markedly enhanced ERK phosphorylation in GD25β1 cells in response to stretching, albeit with a certain degree of variation. As mentioned, Rotenone present during adhesion had a similar effect on ERK phosphorylation in these cells. When we selectively removed extracellular ROS (i.e. hydrogen peroxide and/or superoxide) by adding catalase with or without SOD to the culture medium, we saw a similar pronounced increase in stretch-induced ERK phosphorylation as with Asc-2P, but no effect on the other protein phosphorylation levels we investigated. Although the mechanisms are far from clear, this is opposite to the results we obtained by p22PHOX knock- down. The addition of Asc-2P and the removal of extracellular hydrogen peroxide by catalase would both affect a variety of intracellular, and spatially not restricted reactions, which could increase the number of possible interactions and downstream effects. The ERK pathway for example is known to be triggered and regulated at many levels. As described above, ROS from mitochondria and the possibly mitochondria-associated NOX4 may in contrast exert local effects. Additionally, although we used Asc-2P instead of vitamin C, we can still not exclude the possibility that the ascorbate generated from Asc-2P causes prooxidant effects, e.g. by producing hydrogen peroxide. The similar effect of catalase administration, which has been also seen by others (199), argues against this explanation, unless the enzyme contains contaminants that could cause the observed ERK stimulation. Since we did not see any effect of Asc- 2P when we stretched BJ hTERT cells, the results could also be explained by differences in cell and antioxidant systems.

34 Future plans

In Paper I, we identified the catalytic PI3K subunit downstream of β1 integrins. The next objective would be to identify the regulatory subunit but attempts to do so using specific siRNA have so far failed, probably for reasons similar to those cited for the siRNA experiments which attempted to confirm the identity of the catalytic subunits (discussed above). In addition to the described complications, free catalytic class IA PI3K subunits in cells will be degraded, which of course will interfere with the interpretation of the knock-down effects of their dimerization partner. We are also still interested in the alternative signaling route from integrins to PI3K that is independent of FAK, SRC, YES, or FYN. Our published and unpublished results using siRNA knock-down and inhibitors leave a few possible candidates, which will be interesting to investigate further. Furthermore, the TIRF-based assay we have established and used in Paper I and Paper III could be used to study the mechanism of integrin- induced actin polymerization. It could be worthwhile to e.g. knock-down RIAM, lamellipodin, and/or different GEFs with siRNA and investigate their detailed roles in spreading. Also the preference of β1 integrins to activate AKT2, which we describe in Paper II, would be interesting to follow up. First it should be checked, if other cell lines or cell types show the same AKT activation pattern downstream of β1 integrins as MCF7 cells. The subcellular localization of AKT2 should be confirmed, both before and after integrin stimulation to determine where AKT phosphorylation takes place. As we discuss in Paper III, the widely expressed NOX4 can be found at mitochondrial membranes, which could possibly be linked to integrin signaling through a direct mechanical connection (200), or via the “cross-talk” with mitochondrial ROS or other messenger molecules. The results in Paper III add several questions to an already complicated research area. To give a few examples of possible further investigations, we would like to confirm the p22PHOX knock-down data in another cell line and additionally identify the involved NOX enzyme(s), e.g. by knock-down of the respective catalytic subunits. Certain useful inhibitors might exist as well that could be worthwhile testing after careful evaluation. We hope that we could then better distinguish which signals are induced downstream of integrin ligation and which are downstream of a mechanical stimulus, be it cellular tension or externally applied stretch.

35 Acknowledgements

First of all, ett stort tack till min handledare Staffan Johansson! It has been a great and inspiring time working with you! You are one of the most know- ledgeable persons I know and incredibly kind. Thank you for all you taught me, for your support and the countless good discussions (some more about science, some less)!

I furthermore would like to thank the following people:

- Kristofer Rubin, my second supervisor. Thank you for your support and all the interesting (and funny) discussions!

- Nils-Erik Heldin, my “second second” supervisor. Thanks for all help, especially in the beginning.

- My examiner Lars Rask. Thanks for your good advice. I am very grateful to have you as my examiner.

- All my co-authors and collaborators, especially Anders Tengholm and his group for good discussions and practical help at the TIRF microscope.

- Past and present members of my group: Birgitte Wärmegård, Anne Stefansson-Carnwall, Anjum Riaz and Rajesh Kumar Gupta. Birgitta, jag kan inte tackar tillräckligt! Tack också för alla gånger du lyssnade! Anne, saknar dig här! Anjum, it has been a pleasure to work with you; you are so polite and careful! Rajesh, I learnt a lot from you, mostly about myself. Though you can make me mad sometimes, I am happy you are my friend! Thanks also to Asif (nice food!) and to Tanveer for being an ambitious and talented student.

- Andrew Hamilton and Paul O’Callaghan for helpful comments on my thesis!

- The “northern” part of B9:3: Ananya for good company and for listening, Julia for many good times, and your help with the cover picture (Tadaa!). Else for being a good lab mate and sharing the thesis stress, Jessica and Andrew for nice chats and Anna-Karin for many helpful discussions and

36 VEGF! Thanks to Maria R. and Magnus for advice and support and to Kais, Maria G., Nashwan, Pegah, Per, and Yanyu for being nice colleagues! - Other past and present members of B9:3: Åsa, Lena P., Osama, Sara B. Susanne T.; Anette and Lena Möller for support in the (sometimes annoying) chemical room. Thanks also to Alexis, Anna T., Ellenor, Heidi, Monika, Sara Ö., Wael, and to all the other nice “virus people”! - Former D11:3 members: Alejandro, Chi, Christian, Jakob, Jonny, Marisa, Nathalie, Sandra, Tomas, and my former group members Stanley, Tjis, and Vahid. Thanks for all your help and nice company!

- Annika Hermansson för allt stöd i början!

- Olav Nordli for his help with so many things! Special thanks to Barbro, Erika, Eva G., Kerstin, Malin, Marianne, Rehné and Susanne, to the ladies at diskavdelningen, and to the gentlemen taking care of mail and other small and big things.

- Past and present IPhA people, who made it fun to meet and to organize things!

- All other people at IMBIM and neighboring departments for creating a nice atmosphere and for all the small encouraging chats in the corridors and in the coffee room.

- Leo Gros, Anna Tudos, Nicole Hauser, and Sven Laarmann - you brought me here by sharing your knowledge and inspiration!

- Other people here in Uppsala who have contributed to this work by supporting me with friendship, discussions and even food during the busy time at the end: thanks to Anne, Antonia, Ina, and Lothar for your long- lasting friendship! Thanks also to Carolina, Amira, and Amany and the other old Friday’s beer club people: Anders, Martin, Martina, Ute (danke auch für gute NachbarschaftJ), Pavel, and Stefan. Danke Mirjam och tack till Kerstin Avalder för alla roliga samtal! And thanks to Chi again: this time for your own view on this world and for your smiles!

Finally, I would like to thank my family and friends in Germany. Freue mich riesig, daß so viele kommen wollen, um mit mir zu feiern. Danke für Eure Freundschaft und Unterstützung! Ein großes Dankeschön meiner Familie: für Eure Geduld, Unterstützung, das Zuhören und Ablenken! Der größte Dank gilt meinen Eltern, die immer alle meine Pläne unterstützt haben, was für meine Mutter die letzten Jahre sicher nicht immer leicht war. Danke, Mama, und ich bin stolz auf Dich!

37 References

1. Barczyk, M., Carracedo, S., and Gullberg, D. (2010) Cell and tissue research 339, 269-280 2. Hynes, R. O. (2002) Cell 110, 673-687 3. Lee, J. O., Bankston, L. A., Arnaout, M. A., and Liddington, R. C. (1995) Structure 3, 1333-1340 4. Luo, B. H., Carman, C. V., and Springer, T. A. (2007) Annu Rev Immunol 25, 619-647 5. Adair, B. D., Xiong, J. P., Maddock, C., Goodman, S. L., Arnaout, M. A., and Yeager, M. (2005) The Journal of cell biology 168, 1109-1118 6. Kim, C., Ye, F., and Ginsberg, M. H. (2011) Annual review of cell and developmental biology 27, 321-345 7. Shattil, S. J., Kim, C., and Ginsberg, M. H. (2010) Nature reviews. Molecular cell biology 11, 288-300 8. Springer, T. A., and Dustin, M. L. (2012) Current opinion in cell biology 24, 107-115 9. Lau, T. L., Kim, C., Ginsberg, M. H., and Ulmer, T. S. (2009) The EMBO journal 28, 1351-1361 10. Emsley, J., Knight, C. G., Farndale, R. W., Barnes, M. J., and Liddington, R. C. (2000) Cell 101, 47-56 11. Wegener, K. L., and Campbell, I. D. (2008) Mol Membr Biol 25, 376-387 12. Luo, B. H., Springer, T. A., and Takagi, J. (2004) PLoS Biol 2, e153 13. Ye, F., Kim, C., and Ginsberg, M. H. (2012) Blood 119, 26-33 14. Liu, S., Calderwood, D. A., and Ginsberg, M. H. (2000) Journal of cell science 113 ( Pt 20), 3563-3571 15. Tadokoro, S., Shattil, S. J., Eto, K., Tai, V., Liddington, R. C., de Pereda, J. M., Ginsberg, M. H., and Calderwood, D. A. (2003) Science 302, 103-106 16. Larjava, H., Plow, E. F., and Wu, C. (2008) EMBO Rep 9, 1203-1208 17. Campbell, I. D., and Ginsberg, M. H. (2004) Trends in biochemical sciences 29, 429-435 18. Anthis, N. J., Wegener, K. L., Ye, F., Kim, C., Goult, B. T., Lowe, E. D., Vakonakis, I., Bate, N., Critchley, D. R., Ginsberg, M. H., and Campbell, I. D. (2009) The EMBO journal 28, 3623-3632 19. Calderwood, D. A., Yan, B., de Pereda, J. M., Alvarez, B. G., Fujioka, Y., Liddington, R. C., and Ginsberg, M. H. (2002) J Biol Chem 277, 21749- 21758 20. Legate, K. R., Wickstrom, S. A., and Fassler, R. (2009) Genes Dev 23, 397- 418 21. Moser, M., Legate, K. R., Zent, R., and Fassler, R. (2009) Science 324, 895- 899 22. Lawson, C., Lim, S. T., Uryu, S., Chen, X. L., Calderwood, D. A., and Schlaepfer, D. D. (2012) The Journal of cell biology 196, 223-232

38 23. Critchley, D. R., and Gingras, A. R. (2008) Journal of cell science 121, 1345- 1347 24. Bos, J. L. (2005) Current opinion in cell biology 17, 123-128 25. Boettner, B., and Van Aelst, L. (2009) Current opinion in cell biology 21, 684-693 26. Caron, E., Self, A. J., and Hall, A. (2000) Current biology : CB 10, 974-978 27. Reedquist, K. A., Ross, E., Koop, E. A., Wolthuis, R. M., Zwartkruis, F. J., van Kooyk, Y., Salmon, M., Buckley, C. D., and Bos, J. L. (2000) The Journal of cell biology 148, 1151-1158 28. Krause, M., Leslie, J. D., Stewart, M., Lafuente, E. M., Valderrama, F., Jagannathan, R., Strasser, G. A., Rubinson, D. A., Liu, H., Way, M., Yaffe, M. B., Boussiotis, V. A., and Gertler, F. B. (2004) Developmental cell 7, 571- 583 29. Lee, H. S., Lim, C. J., Puzon-McLaughlin, W., Shattil, S. J., and Ginsberg, M. H. (2009) J Biol Chem 284, 5119-5127 30. Lafuente, E. M., van Puijenbroek, A. A., Krause, M., Carman, C. V., Freeman, G. J., Berezovskaya, A., Constantine, E., Springer, T. A., Gertler, F. B., and Boussiotis, V. A. (2004) Developmental cell 7, 585-595 31. Siegel, D. H., Ashton, G. H., Penagos, H. G., Lee, J. V., Feiler, H. S., Wilhelmsen, K. C., South, A. P., Smith, F. J., Prescott, A. R., Wessagowit, V., Oyama, N., Akiyama, M., Al Aboud, D., Al Aboud, K., Al Githami, A., Al Hawsawi, K., Al Ismaily, A., Al-Suwaid, R., Atherton, D. J., Caputo, R., Fine, J. D., Frieden, I. J., Fuchs, E., Haber, R. M., Harada, T., Kitajima, Y., Mallory, S. B., Ogawa, H., Sahin, S., Shimizu, H., Suga, Y., Tadini, G., Tsuchiya, K., Wiebe, C. B., Wojnarowska, F., Zaghloul, A. B., Hamada, T., Mallipeddi, R., Eady, R. A., McLean, W. H., McGrath, J. A., and Epstein, E. H. (2003) American journal of human genetics 73, 174-187 32. Shi, X., Ma, Y. Q., Tu, Y., Chen, K., Wu, S., Fukuda, K., Qin, J., Plow, E. F., and Wu, C. (2007) J Biol Chem 282, 20455-20466 33. Harburger, D. S., Bouaouina, M., and Calderwood, D. A. (2009) J Biol Chem 284, 11485-11497 34. Bottcher, R. T., Lange, A., and Fassler, R. (2009) Current opinion in cell biology 35. Frame, M. C., Patel, H., Serrels, B., Lietha, D., and Eck, M. J. (2010) Nature reviews. Molecular cell biology 11, 802-814 36. Rantala, J. K., Pouwels, J., Pellinen, T., Veltel, S., Laasola, P., Mattila, E., Potter, C. S., Duffy, T., Sundberg, J. P., Kallioniemi, O., Askari, J. A., Humphries, M. J., Parsons, M., Salmi, M., and Ivaska, J. (2011) Nature cell biology 13, 1315-1324 37. Millon-Fremillon, A., Bouvard, D., Grichine, A., Manet-Dupe, S., Block, M. R., and Albiges-Rizo, C. (2008) The Journal of cell biology 180, 427-441 38. Oxley, C. L., Anthis, N. J., Lowe, E. D., Vakonakis, I., Campbell, I. D., and Wegener, K. L. (2008) J Biol Chem 283, 5420-5426 39. Kiema, T., Lad, Y., Jiang, P., Oxley, C. L., Baldassarre, M., Wegener, K. L., Campbell, I. D., Ylanne, J., and Calderwood, D. A. (2006) Molecular cell 21, 337-347 40. Ithychanda, S. S., Das, M., Ma, Y. Q., Ding, K., Wang, X., Gupta, S., Wu, C., Plow, E. F., and Qin, J. (2009) J Biol Chem 284, 4713-4722 41. Das, M., Ithychanda, S. S., Qin, J., and Plow, E. F. (2011) PloS one 6, e26355 42. Zaidel-Bar, R., Itzkovitz, S., Ma'ayan, A., Iyengar, R., and Geiger, B. (2007) Nature cell biology 9, 858-867

39 43. Sundberg, C., and Rubin, K. (1996) The Journal of cell biology 132, 741-752 44. Couchman, J. R. (2010) Annual review of cell and developmental biology 26, 89-114 45. Streuli, C. H., and Akhtar, N. (2009) The Biochemical journal 418, 491-506 46. Miyamoto, S., Teramoto, H., Coso, O. A., Gutkind, J. S., Burbelo, P. D., Akiyama, S. K., and Yamada, K. M. (1995) The Journal of cell biology 131, 791-805 47. Mitra, S. K., Hanson, D. A., and Schlaepfer, D. D. (2005) Nature reviews. Molecular cell biology 6, 56-68 48. Wennerberg, K., Armulik, A., Sakai, T., Karlsson, M., Fassler, R., Schaefer, E. M., Mosher, D. F., and Johansson, S. (2000) Molecular and cellular biology 20, 5758-5765 49. Arias-Salgado, E. G., Lizano, S., Sarkar, S., Brugge, J. S., Ginsberg, M. H., and Shattil, S. J. (2003) Proceedings of the National Academy of Sciences of the United States of America 100, 13298-13302 50. Shattil, S. J. (2005) Trends in cell biology 15, 399-403 51. Fincham, V. J., Unlu, M., Brunton, V. G., Pitts, J. D., Wyke, J. A., and Frame, M. C. (1996) The Journal of cell biology 135, 1551-1564 52. Playford, M. P., and Schaller, M. D. (2004) Oncogene 23, 7928-7946 53. Hall, J. E., Fu, W., and Schaller, M. D. (2011) International review of cell and molecular biology 288, 185-225 54. Ilic, D., Furuta, Y., Kanazawa, S., Takeda, N., Sobue, K., Nakatsuji, N., Nomura, S., Fujimoto, J., Okada, M., and Yamamoto, T. (1995) Nature 377, 539-544 55. Chodniewicz, D., and Klemke, R. L. (2004) Biochimica et biophysica acta 1692, 63-76 56. Schlaepfer, D. D., Hanks, S. K., Hunter, T., and van der Geer, P. (1994) Nature 372, 786-791 57. Howe, A. K., Aplin, A. E., and Juliano, R. L. (2002) Current opinion in genetics & development 12, 30-35 58. Ilic, D., Almeida, E. A., Schlaepfer, D. D., Dazin, P., Aizawa, S., and Damsky, C. H. (1998) The Journal of cell biology 143, 547-560 59. Golubovskaya, V. M., and Cance, W. G. (2011) Anti-cancer agents in medicinal chemistry 11, 617-619 60. Avraham, H., Park, S. Y., Schinkmann, K., and Avraham, S. (2000) Cellular signalling 12, 123-133 61. Sieg, D. J., Ilic, D., Jones, K. C., Damsky, C. H., Hunter, T., and Schlaepfer, D. D. (1998) The EMBO journal 17, 5933-5947 62. Klingbeil, C. K., Hauck, C. R., Hsia, D. A., Jones, K. C., Reider, S. R., and Schlaepfer, D. D. (2001) The Journal of cell biology 152, 97-110 63. Lakkakorpi, P. T., Bett, A. J., Lipfert, L., Rodan, G. A., and Duong le, T. (2003) J Biol Chem 278, 11502-11512 64. Hawkins, P. T., Anderson, K. E., Davidson, K., and Stephens, L. R. (2006) Biochem Soc Trans 34, 647-662 65. Puri, K. D., Doggett, T. A., Douangpanya, J., Hou, Y., Tino, W. T., Wilson, T., Graf, T., Clayton, E., Turner, M., Hayflick, J. S., and Diacovo, T. G. (2004) Blood 103, 3448-3456 66. Jackson, S. P., Schoenwaelder, S. M., Goncalves, I., Nesbitt, W. S., Yap, C. L., Wright, C. E., Kenche, V., Anderson, K. E., Dopheide, S. M., Yuan, Y., Sturgeon, S. A., Prabaharan, H., Thompson, P. E., Smith, G. D., Shepherd, P. R., Daniele, N., Kulkarni, S., Abbott, B., Saylik, D., Jones, C., Lu, L.,

40 Giuliano, S., Hughan, S. C., Angus, J. A., Robertson, A. D., and Salem, H. H. (2005) Nature medicine 11, 507-514 67. Ferreira, A. M., Isaacs, H., Hayflick, J. S., Rogers, K. A., and Sandig, M. (2006) Microcirculation 13, 439-456 68. Schoenwaelder, S. M., Ono, A., Sturgeon, S., Chan, S. M., Mangin, P., Maxwell, M. J., Turnbull, S., Mulchandani, M., Anderson, K., Kauffenstein, G., Rewcastle, G. W., Kendall, J., Gachet, C., Salem, H. H., and Jackson, S. P. (2007) J Biol Chem 282, 28648-28658 69. Vanhaesebroeck, B., Guillermet-Guibert, J., Graupera, M., and Bilanges, B. (2010) Nature reviews. Molecular cell biology 11, 329-341 70. Vanhaesebroeck, B., Leevers, S. J., Ahmadi, K., Timms, J., Katso, R., Driscoll, P. C., Woscholski, R., Parker, P. J., and Waterfield, M. D. (2001) Annu Rev Biochem 70, 535-602 71. Chen, H. C., Appeddu, P. A., Isoda, H., and Guan, J. L. (1996) J Biol Chem 271, 26329-26334 72. Stephens, L., Anderson, K., Stokoe, D., Erdjument-Bromage, H., Painter, G. F., Holmes, A. B., Gaffney, P. R., Reese, C. B., McCormick, F., Tempst, P., Coadwell, J., and Hawkins, P. T. (1998) Science 279, 710-714 73. Alessi, D. R., James, S. R., Downes, C. P., Holmes, A. B., Gaffney, P. R., Reese, C. B., and Cohen, P. (1997) Current biology : CB 7, 261-269 74. Vanhaesebroeck, B., and Alessi, D. R. (2000) The Biochemical journal 346 Pt 3, 561-576 75. Stokoe, D., Stephens, L. R., Copeland, T., Gaffney, P. R., Reese, C. B., Painter, G. F., Holmes, A. B., McCormick, F., and Hawkins, P. T. (1997) Science 277, 567-570 76. Alessi, D. R., Caudwell, F. B., Andjelkovic, M., Hemmings, B. A., and Cohen, P. (1996) FEBS letters 399, 333-338 77. Riaz, A., Zeller, K. S., and Johansson, S. (2012) PloS one 7, e32081 78. Stambolic, V., Suzuki, A., de la Pompa, J. L., Brothers, G. M., Mirtsos, C., Sasaki, T., Ruland, J., Penninger, J. M., Siderovski, D. P., and Mak, T. W. (1998) Cell 95, 29-39 79. Maehama, T., and Dixon, J. E. (1998) J Biol Chem 273, 13375-13378 80. Suzuki, A., de la Pompa, J. L., Stambolic, V., Elia, A. J., Sasaki, T., del Barco Barrantes, I., Ho, A., Wakeham, A., Itie, A., Khoo, W., Fukumoto, M., and Mak, T. W. (1998) Current biology : CB 8, 1169-1178 81. Carnero, A. (2010) Current pharmaceutical design 16, 34-44 82. Velling, T., Nilsson, S., Stefansson, A., and Johansson, S. (2004) EMBO Rep 5, 901-905 83. Olofsson, B. (1999) Cellular signalling 11, 545-554 84. Arthur, W. T., Petch, L. A., and Burridge, K. (2000) Current biology : CB 10, 719-722 85. Tomar, A., Lim, S. T., Lim, Y., and Schlaepfer, D. D. (2009) Journal of cell science 122, 1852-1862 86. ten Klooster, J. P., Jaffer, Z. M., Chernoff, J., and Hordijk, P. L. (2006) The Journal of cell biology 172, 759-769 87. Lim, Y., Lim, S. T., Tomar, A., Gardel, M., Bernard-Trifilo, J. A., Chen, X. L., Uryu, S. A., Canete-Soler, R., Zhai, J., Lin, H., Schlaepfer, W. W., Nalbant, P., Bokoch, G., Ilic, D., Waterman-Storer, C., and Schlaepfer, D. D. (2008) The Journal of cell biology 180, 187-203 88. Ohta, Y., Hartwig, J. H., and Stossel, T. P. (2006) Nature cell biology 8, 803- 814

41 89. Nimnual, A. S., Taylor, L. J., and Bar-Sagi, D. (2003) Nature cell biology 5, 236-241 90. Innocenti, M., Frittoli, E., Ponzanelli, I., Falck, J. R., Brachmann, S. M., Di Fiore, P. P., and Scita, G. (2003) The Journal of cell biology 160, 17-23 91. Del Pozo, M. A., Kiosses, W. B., Alderson, N. B., Meller, N., Hahn, K. M., and Schwartz, M. A. (2002) Nature cell biology 4, 232-239 92. Takahashi, K., Sasaki, T., Mammoto, A., Takaishi, K., Kameyama, T., Tsukita, S., and Takai, Y. (1997) J Biol Chem 272, 23371-23375 93. DerMardirossian, C., Schnelzer, A., and Bokoch, G. M. (2004) Molecular cell 15, 117-127 94. Garcia-Mata, R., Boulter, E., and Burridge, K. (2011) Nature reviews. Molecular cell biology 12, 493-504 95. Dovas, A., Choi, Y., Yoneda, A., Multhaupt, H. A., Kwon, S. H., Kang, D., Oh, E. S., and Couchman, J. R. (2010) J Biol Chem 285, 23296-23308 96. Huveneers, S., and Danen, E. H. (2009) Journal of cell science 122, 1059- 1069 97. Kubow, K. E., and Horwitz, A. R. (2011) Nature cell biology 13, 3-5; author reply 5-7 98. Wehrle-Haller, B. (2011) Current opinion in cell biology 99. Gilmore, A. P., and Burridge, K. (1996) Nature 381, 531-535 100. Choi, C. K., Vicente-Manzanares, M., Zareno, J., Whitmore, L. A., Mogilner, A., and Horwitz, A. R. (2008) Nature cell biology 10, 1039-1050 101. Delcommenne, M., Tan, C., Gray, V., Rue, L., Woodgett, J., and Dedhar, S. (1998) Proceedings of the National Academy of Sciences of the United States of America 95, 11211-11216 102. Pasquali, C., Bertschy-Meier, D., Chabert, C., Curchod, M. L., Arod, C., Booth, R., Mechtler, K., Vilbois, F., Xenarios, I., Ferguson, C. G., Prestwich, G. D., Camps, M., and Rommel, C. (2007) Molecular & cellular proteomics : MCP 6, 1829-1841 103. Wu, C. (2004) Biochimica et biophysica acta 1692, 55-62 104. Stanchi, F., Grashoff, C., Nguemeni Yonga, C. F., Grall, D., Fassler, R., and Van Obberghen-Schilling, E. (2009) Journal of cell science 122, 1800-1811 105. Alexandrova, A. Y., Arnold, K., Schaub, S., Vasiliev, J. M., Meister, J. J., Bershadsky, A. D., and Verkhovsky, A. B. (2008) PloS one 3, e3234 106. Hirata, H., Tatsumi, H., and Sokabe, M. (2008) Journal of cell science 121, 2795-2804 107. Geiger, B., Spatz, J. P., and Bershadsky, A. D. (2009) Nature reviews. Molecular cell biology 10, 21-33 108. Aratyn-Schaus, Y., and Gardel, M. L. (2010) Current biology : CB 20, 1145- 1153 109. Hu, K., Ji, L., Applegate, K. T., Danuser, G., and Waterman-Storer, C. M. (2007) Science 315, 111-115 110. Verkhovsky, A. B. (2010) Current biology : CB 20, R669-671 111. Shemesh, T., Verkhovsky, A. B., Svitkina, T. M., Bershadsky, A. D., and Kozlov, M. M. (2009) Biophysical journal 97, 1254-1264 112. Hotulainen, P., and Lappalainen, P. (2006) The Journal of cell biology 173, 383-394 113. Ponti, A., Machacek, M., Gupton, S. L., Waterman-Storer, C. M., and Danuser, G. (2004) Science 305, 1782-1786 114. Cai, Y., Biais, N., Giannone, G., Tanase, M., Jiang, G., Hofman, J. M., Wiggins, C. H., Silberzan, P., Buguin, A., Ladoux, B., and Sheetz, M. P. (2006) Biophysical journal 91, 3907-3920

42 115. DesMarais, V., Ghosh, M., Eddy, R., and Condeelis, J. (2005) Journal of cell science 118, 19-26 116. Carlier, M. F., Laurent, V., Santolini, J., Melki, R., Didry, D., Xia, G. X., Hong, Y., Chua, N. H., and Pantaloni, D. (1997) The Journal of cell biology 136, 1307-1322 117. Oser, M., and Condeelis, J. (2009) Journal of cellular biochemistry 108, 1252-1262 118. Arber, S., Barbayannis, F. A., Hanser, H., Schneider, C., Stanyon, C. A., Bernard, O., and Caroni, P. (1998) Nature 393, 805-809 119. Gorbatyuk, V. Y., Nosworthy, N. J., Robson, S. A., Bains, N. P., Maciejewski, M. W., Dos Remedios, C. G., and King, G. F. (2006) Molecular cell 24, 511-522 120. Nagata-Ohashi, K., Ohta, Y., Goto, K., Chiba, S., Mori, R., Nishita, M., Ohashi, K., Kousaka, K., Iwamatsu, A., Niwa, R., Uemura, T., and Mizuno, K. (2004) The Journal of cell biology 165, 465-471 121. Song, X., Chen, X., Yamaguchi, H., Mouneimne, G., Condeelis, J. S., and Eddy, R. J. (2006) Journal of cell science 119, 2871-2881 122. Choi, Y., Chung, H., Jung, H., Couchman, J. R., and Oh, E. S. (2011) Matrix biology : journal of the International Society for Matrix Biology 30, 93-99 123. Beauvais, D. M., Burbach, B. J., and Rapraeger, A. C. (2004) The Journal of cell biology 167, 171-181 124. McQuade, K. J., Beauvais, D. M., Burbach, B. J., and Rapraeger, A. C. (2006) Journal of cell science 119, 2445-2456 125. Beauvais, D. M., and Rapraeger, A. C. (2010) Journal of cell science 123, 3796-3807 126. Greene, D. K., Tumova, S., Couchman, J. R., and Woods, A. (2003) J Biol Chem 278, 7617-7623 127. Mostafavi-Pour, Z., Askari, J. A., Parkinson, S. J., Parker, P. J., Ng, T. T., and Humphries, M. J. (2003) The Journal of cell biology 161, 155-167 128. Lim, S. T., Longley, R. L., Couchman, J. R., and Woods, A. (2003) J Biol Chem 278, 13795-13802 129. Engler, A. J., Sen, S., Sweeney, H. L., and Discher, D. E. (2006) Cell 126, 677-689 130. Pelham, R. J., Jr., and Wang, Y. (1997) Proceedings of the National Academy of Sciences of the United States of America 94, 13661-13665 131. Tolic-Norrelykke, I. M., and Wang, N. (2005) Journal of biomechanics 38, 1405-1412 132. Reinhart-King, C. A., Dembo, M., and Hammer, D. A. (2005) Biophysical journal 89, 676-689 133. Klein, E. A., Yin, L., Kothapalli, D., Castagnino, P., Byfield, F. J., Xu, T., Levental, I., Hawthorne, E., Janmey, P. A., and Assoian, R. K. (2009) Current biology : CB 19, 1511-1518 134. Schwartz, M. A. (2010) Cold Spring Harbor perspectives in biology 2, a005066 135. Mao, Y., and Schwarzbauer, J. E. (2005) Matrix biology : journal of the International Society for Matrix Biology 24, 389-399 136. Huveneers, S., Truong, H., Fassler, R., Sonnenberg, A., and Danen, E. H. (2008) Journal of cell science 121, 2452-2462 137. Canty, E. G., Starborg, T., Lu, Y., Humphries, S. M., Holmes, D. F., Meadows, R. S., Huffman, A., O'Toole, E. T., and Kadler, K. E. (2006) J Biol Chem 281, 38592-38598

43 138. Wipff, P. J., and Hinz, B. (2008) European journal of cell biology 87, 601- 615 139. Worthington, J. J., Klementowicz, J. E., and Travis, M. A. (2011) Trends in biochemical sciences 36, 47-54 140. Mu, D., Cambier, S., Fjellbirkeland, L., Baron, J. L., Munger, J. S., Kawakatsu, H., Sheppard, D., Broaddus, V. C., and Nishimura, S. L. (2002) The Journal of cell biology 157, 493-507 141. Sukharev, S., and Corey, D. P. (2004) Science's STKE : signal transduction knowledge environment 2004, re4 142. Matthews, B. D., Overby, D. R., Mannix, R., and Ingber, D. E. (2006) Journal of cell science 119, 508-518 143. Puklin-Faucher, E., Gao, M., Schulten, K., and Vogel, V. (2006) The Journal of cell biology 175, 349-360 144. Friedland, J. C., Lee, M. H., and Boettiger, D. (2009) Science 323, 642-644 145. Sawada, Y., Tamada, M., Dubin-Thaler, B. J., Cherniavskaya, O., Sakai, R., Tanaka, S., and Sheetz, M. P. (2006) Cell 127, 1015-1026 146. Tamada, M., Sheetz, M. P., and Sawada, Y. (2004) Developmental cell 7, 709-718 147. Galbraith, C. G., Yamada, K. M., and Sheetz, M. P. (2002) The Journal of cell biology 159, 695-705 148. Ziegler, W. H., Liddington, R. C., and Critchley, D. R. (2006) Trends in cell biology 16, 453-460 149. Papagrigoriou, E., Gingras, A. R., Barsukov, I. L., Bate, N., Fillingham, I. J., Patel, B., Frank, R., Ziegler, W. H., Roberts, G. C., Critchley, D. R., and Emsley, J. (2004) The EMBO journal 23, 2942-2951 150. del Rio, A., Perez-Jimenez, R., Liu, R., Roca-Cusachs, P., Fernandez, J. M., and Sheetz, M. P. (2009) Science 323, 638-641 151. Peng, X., Nelson, E. S., Maiers, J. L., and DeMali, K. A. (2011) International review of cell and molecular biology 287, 191-231 152. Zaidel-Bar, R., Ballestrem, C., Kam, Z., and Geiger, B. (2003) Journal of cell science 116, 4605-4613 153. Lele, T. P., Pendse, J., Kumar, S., Salanga, M., Karavitis, J., and Ingber, D. E. (2006) Journal of cellular physiology 207, 187-194 154. Hoffman, L. M., Jensen, C. C., Kloeker, S., Wang, C. L., Yoshigi, M., and Beckerle, M. C. (2006) The Journal of cell biology 172, 771-782 155. Droge, W. (2002) Physiological reviews 82, 47-95 156. Murphy, M. P. (2009) The Biochemical journal 417, 1-13 157. Turrens, J. F. (2003) The Journal of physiology 552, 335-344 158. Lambeth, J. D., Kawahara, T., and Diebold, B. (2007) Free radical biology & medicine 43, 319-331 159. Martyn, K. D., Kim, M. J., Quinn, M. T., Dinauer, M. C., and Knaus, U. G. (2005) Blood 106, 3962-3969 160. Miller, E. W., Dickinson, B. C., and Chang, C. J. (2010) Proceedings of the National Academy of Sciences of the United States of America 107, 15681- 15686 161. Martyn, K. D., Frederick, L. M., von Loehneysen, K., Dinauer, M. C., and Knaus, U. G. (2006) Cellular signalling 18, 69-82 162. Block, K., Gorin, Y., and Abboud, H. E. (2009) Proceedings of the National Academy of Sciences of the United States of America 106, 14385-14390 163. Krause, K. H. (2004) Japanese journal of infectious diseases 57, S28-29 164. Lambeth, J. D. (2004) Nature reviews. Immunology 4, 181-189 165. Tonks, N. K. (2005) Cell 121, 667-670

44 166. Svineng, G., Ravuri, C., Rikardsen, O., Huseby, N. E., and Winberg, J. O. (2008) Connect Tissue Res 49, 197-202 167. Morgan, M. J., and Liu, Z. G. (2011) Cell research 21, 103-115 168. Lassing, I., Schmitzberger, F., Bjornstedt, M., Holmgren, A., Nordlund, P., Schutt, C. E., and Lindberg, U. (2007) Journal of molecular biology 370, 331-348 169. Fiaschi, T., Cozzi, G., Raugei, G., Formigli, L., Ramponi, G., and Chiarugi, P. (2006) J Biol Chem 281, 22983-22991 170. Fiaschi, T., Cozzi, G., and Chiarugi, P. (2012) Journal of signal transduction 2012, 754964 171. Kim, J. S., Huang, T. Y., and Bokoch, G. M. (2009) Molecular biology of the cell 20, 2650-2660 172. Klemke, M., Wabnitz, G. H., Funke, F., Funk, B., Kirchgessner, H., and Samstag, Y. (2008) Immunity 29, 404-413 173. Taddei, M. L., Parri, M., Mello, T., Catalano, A., Levine, A. D., Raugei, G., Ramponi, G., and Chiarugi, P. (2007) Antioxidants & redox signaling 9, 469- 481 174. Werner, E., and Werb, Z. (2002) The Journal of cell biology 158, 357-368 175. Chiarugi, P., Pani, G., Giannoni, E., Taddei, L., Colavitti, R., Raugei, G., Symons, M., Borrello, S., Galeotti, T., and Ramponi, G. (2003) The Journal of cell biology 161, 933-944 176. Dib, K., Melander, F., Axelsson, L., Dagher, M. C., Aspenstrom, P., and Andersson, T. (2003) J Biol Chem 278, 24181-24188 177. Ali, M. H., Mungai, P. T., and Schumacker, P. T. (2006) American journal of physiology. Lung cellular and molecular physiology 291, L38-45 178. De Keulenaer, G. W., Chappell, D. C., Ishizaka, N., Nerem, R. M., Alexander, R. W., and Griendling, K. K. (1998) Circulation research 82, 1094-1101 179. Zhang, Y., Peng, F., Gao, B., Ingram, A. J., and Krepinsky, J. C. (2010) Antioxidants & redox signaling 13, 959-973 180. Aghajanian, A., Wittchen, E. S., Campbell, S. L., and Burridge, K. (2009) PloS one 4, e8045 181. Foukas, L. C., Berenjeno, I. M., Gray, A., Khwaja, A., and Vanhaesebroeck, B. (2010) Proceedings of the National Academy of Sciences of the United States of America 107, 11381-11386 182. Weiss, W. A., Taylor, S. S., and Shokat, K. M. (2007) Nature chemical biology 3, 739-744 183. Foukas, L. C., Claret, M., Pearce, W., Okkenhaug, K., Meek, S., Peskett, E., Sancho, S., Smith, A. J., Withers, D. J., and Vanhaesebroeck, B. (2006) Nature 441, 366-370 184. Janas, M. L., Hodson, D., Stamataki, Z., Hill, S., Welch, K., Gambardella, L., Trotman, L. C., Pandolfi, P. P., Vigorito, E., and Turner, M. (2008) J Immunol 180, 739-746 185. Papakonstanti, E. A., Zwaenepoel, O., Bilancio, A., Burns, E., Nock, G. E., Houseman, B., Shokat, K., Ridley, A. J., and Vanhaesebroeck, B. (2008) Journal of cell science 121, 4124-4133 186. Ueki, K., Fruman, D. A., Brachmann, S. M., Tseng, Y. H., Cantley, L. C., and Kahn, C. R. (2002) Molecular and cellular biology 22, 965-977 187. Luo, J., and Cantley, L. C. (2005) Cell Cycle 4, 1309-1312 188. Gupta, S., Ramjaun, A. R., Haiko, P., Wang, Y., Warne, P. H., Nicke, B., Nye, E., Stamp, G., Alitalo, K., and Downward, J. (2007) Cell 129, 957-968

45 189. Persad, S., Attwell, S., Gray, V., Mawji, N., Deng, J. T., Leung, D., Yan, J., Sanghera, J., Walsh, M. P., and Dedhar, S. (2001) J Biol Chem 276, 27462- 27469 190. Troussard, A. A., McDonald, P. C., Wederell, E. D., Mawji, N. M., Filipenko, N. R., Gelmon, K. A., Kucab, J. E., Dunn, S. E., Emerman, J. T., Bally, M. B., and Dedhar, S. (2006) Cancer research 66, 393-403 191. Wickstrom, S. A., Lange, A., Montanez, E., and Fassler, R. (2010) The EMBO journal 29, 281-291 192. Fukuda, K., Knight, J. D., Piszczek, G., Kothary, R., and Qin, J. (2011) J Biol Chem 286, 21886-21895 193. Dong, L. Q., and Liu, F. (2005) American journal of physiology. Endocrinology and metabolism 289, E187-196 194. Yang, J., Cron, P., Good, V. M., Thompson, V., Hemmings, B. A., and Barford, D. (2002) Nature structural biology 9, 940-944 195. Huang, X., Begley, M., Morgenstern, K. A., Gu, Y., Rose, P., Zhao, H., and Zhu, X. (2003) Structure 11, 21-30 196. Alessi, D. R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice, N., Cohen, P., and Hemmings, B. A. (1996) The EMBO journal 15, 6541-6551 197. Scheid, M. P., Marignani, P. A., and Woodgett, J. R. (2002) Molecular and cellular biology 22, 6247-6260 198. Shiota, C., Woo, J. T., Lindner, J., Shelton, K. D., and Magnuson, M. A. (2006) Developmental cell 11, 583-589 199. Ulrich-Merzenich, G., Zeitler, H., Panek, D., Bokemeyer, D., and Vetter, H. (2007) European journal of nutrition 46, 87-94 200. Wang, N., Tytell, J. D., and Ingber, D. E. (2009) Nature reviews. Molecular cell biology 10, 75-82

46