Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Studying cancer and metastasis: how multiphoton microscopy contribute to this Christa Eekhuis, University of Utrecht Abstract Cancer is worldwide a deadly disease and results in many deaths. Tumor formation can be induced by several factors, such as mutations in genes and viruses. Genes affected by mutations are mainly tumor suppressor genes and oncogenes, which results in the up- or downregulation of cellular processes. During tumor formation metastasis can occur as well and this result in many deaths due to cancer. Metastasis is a complex and dynamic process, consisting of many steps, like cell detachment and intravasation. Immunohistochemistry is mainly used to study metastasis, but nowadays intravital microscopy techniques are used as well. With intravital microscopy we are able to answer the questions, which still remain about metastasis. We are now able to follow metastasis in vivo over time with intravital microscopy. Multiphoton microscopy is one of the techniques which is used to study metastasis. With the use of fluorescent proteins, such as GFP and photoswitchable proteins, it is possible to follow single cells over time during metastasis to gain more knowledge about metastasis. This can help in developing treatments for cancer, thereby preventing the formation of secondary and tertiary tumors. Introduction Cancer Cancer is worldwide a deadly disease and results in many deaths. Almost 90% of the deaths are caused by metastasis of tumors(Gupta & Massague, 2006). Cancer development is a process in which cellular processes are altered, like cell proliferation and apoptosis. Most of the time mutations occur in genes involved in these processes, resulting in the up- or downregulation of genes. The genes mutated are mainly oncogenes or tumor suppressor genes(Wrzeszczynski et al., 2011). A variety of factors can cause the formation of tumors. One of these causes are viruses, like the human papillomavirus (HPV), which can cause cervix cancer(Faridi, Zahra, Khan, & Idrees, 2011). Second, mutations in the germline can occur as well, resulting in hereditary diseases. The most studied hereditary diseases are breast cancer (BRCA1/2 mutations) and HNPPC (APC, β-catenin, K-ras mutations)(Jass, 2004; Shulman, 2010) BRCA1/2 are tumor suppressor genes, which are involved in DNA repair, but when mutated they can cause breast cancer, but ovarian cancer too(Shulman, 2010). Hereditary non-polyposis colorectal cancer (HNPPC) is responsible for a small percentage of colorectal cancer. Females with this syndrome can develop cervix and ovarian cancer as well(Jass, 2004). Beside BRCA1/2 and HNPPC, other syndromes, like VHL, can cause cancer too(Jass, 2004; Moore et al., 2011; Shulman, 2010). Most tumors arise from somatic mutations caused by internal and external factors(Wrzeszczynski et al., 2011). External factors, like UV and chemicals, and internal factors, like mitochondria and peroxisomes, induce the formation of reactive oxygen species (ROS) in your body(Valko et al., 2007; Veal, Day, & Morgan, 2007). Most ROS will be scavenged by antioxidants, but when the ROS levels are too high to be scavenged, it will cause damage to DNA, lipids and proteins. In addition, ROS can induce chromosomal instability as well. The DNA damage can result in gene mutations or altered gene expression, which contributes to tumor formation(Valko et al., 2007). Furthermore, chromosomal rearrangements, like translocations and deletions, and epigenetic alterations, like DNA methylation, contribute to tumor formation too(Wrzeszczynski et al., 2011). Metastasis of tumors One of the hallmarks of cancer is metastasis (figure 1) and this is the cause of most deaths due to cancer. Metastasis is a very dynamic process and during metastasis cancer cells spread throughout the whole body via blood vessels, lymphatic vessels and serosal surfaces, before new colonies can be formed in other organs(Bacac & Stamenkovic, 2008; Hanahan & Weinberg, 2011). Most tumors spread throughout the body via the lymph nodes and blood vessels. Only a small group of tumors, like ovarian cancer, has the ability to metastasis transcoelomic(Bacac & Stamenkovic, 2008). Before metastasis can occur, cancer cells have to overcome several barriers in your body. Therefore metastasis can be seen as an evolutionary process, because evading barriers results in selection of cancer cells, which are able to survive during this process(Gupta & Massague, 2006). Metastasis is a dynamic process, which can be divided into two phases: the translocation of cancer cells to other organs (1) and the ability of these cells to adapt to the new environment and develop a secondary or tertiary tumor (2)(Bacac & Stamenkovic, 2008; Chaffer & Weinberg, 2011). These two phases consists of many steps, which are essential for the formation of secondary or even tertiary tumors. Figure 1| The different steps of metastasis. First there is detachment of tumor cells from the tumor, resulting in breaking through the basal membrane into the surrounding tissue stroma (invasion). After invasion, the cells migrate towards collagen fibers towards the blood and/or lymphatic vessels and intravasation occurs. Tumor cells move via the circulation to other organs, where they extravasat and start proliferating, inducing angiogenesis en form a secondary or tertiary tumor(Bacac & Stamenkovic, 2008). Page | 2 Cell detachment The first step of metastasis, is the loss of cell adhesion from the tumor(Bacac & Stamenkovic, 2008; Gupta & Massague, 2006). Most tumors consist of epithelial cells, which are strongly zipped to each other, via membranes like tight junctions and desmosomes(Garber, 2008; Thiery & Sleeman, 2006). Epithelial cells polarize apical-basolateral and are only able to move within the epithelial layer(Thiery & Sleeman, 2006). If epithelial cells want to metastasize from the tumor throughout the body they have to undergo a transformation into mesenchymal cells. This is called the epithelial-to-mesenchymal transition (EMT)(Garber, 2008; Mani et al., 2008; Thiery & Sleeman, 2006). When EMT occurs, the epithelial cell loses most of its characteristics and transform into a mesenchymal cell (figure 2). The transformed cell also acquire characteristics from a mesenchymal cell, resulting in the change of the architecture of the cell and its behavior(Garber, 2008; Thiery & Sleeman, 2006). The architecture of mesenchymal cells look like fibroblasts, they are not tightly zipped to each other and these cells are more motile then epithelial cells(Thiery & Sleeman, 2006). Figure 2| The plasticity cycle of the epithelial cell. It shows which events occur during the transformation of epithelial cells into mesenchymal cells and vice versa. EMT and MET are induced and regulated by different effectors, such as growth factors and adhesion. The MET and EMT effectors can inhibit each other. Furthermore, important processes, like the formation of tight junctions during MET, are indicated. During EMT and MET, several markers have been found, which are typical for mesenchymal and epithelial cells(Thiery & Sleeman, 2006). EMT is normally an important key player in embryogenesis(Garber, 2008; Mani et al., 2008; Thiery & Sleeman, 2006), because without EMT development stops at the blastula stage and further development cannot occur anymore. During embryogenesis, EMT is involved in several processes, like invagination and multi-layering. Due to EMT the three layered embryo can be formed, but also other structures, like vertebrae and craniofacial structures. Furthermore, EMT mediates the loss of the mullerian duct in males. Extracellular signals can induce EMT, during embryogenesis (figure 3)(Thiery & Sleeman, 2006). Mesenchymal to Page | 3 epithelial transition (MET) occurs during embyrogenesis as well, especially during the development of kidneys(Mani et al., 2008), the formation of coelomiccavity and during somitogenesis(Thiery & Sleeman, 2006). There are a lot of transcription factors, like Twist and Snail, which are involved in triggering EMT in cells during embryogensesis. These transcription factors are found to be involved in processes related to cancer formation and metastasis, like invasiveness and resistance to apoptosis(Mani et al., 2008). Not only transcription factors trigger EMT, but also growthfactors, like TGFβ, and parts of the extracellular matrix (ECM), like collagen(Thiery & Sleeman, 2006). When EMT is induced, desmosomes, β-catenin and E-cadherin are repressed, which results in the loss of cell-cell adhesion and changes in the cytoskeleton (Buda & Pignatelli, 2011; Mani et al., 2008). Other components are downregulated as well, especially those who are involved in cell-cell adhesion like vinculin and plakoglobin(Thiery & Sleeman, 2006). Figure 3| Overview of molecular pathways involved in EMT. Several pathways are able to activate EMT and there is crosstalk between some pathways. When receptor tyrosine kinases (RTKs) are activated, activation of Rac occurs as well to induce cytoskeleton rearrangement. The TGFβ pathway activates a lot of players, from which Twist and Snail are the proteins which induce EMT, by repressing E-cadherin or directly influence the EMT programme. Furthermore, MMPs are able to induce EMT too, but it is not known which receptors are involved(Thiery & Sleeman, 2006). E-cadherin is a member of the cadherin family, which are transmembrane glycoproteins involved in cell-cell adhesion. One of the groups of the cadherin family is the classical type I cadherins. The classical type I cadherins consist of five cadherins, including E-cadherin and N-cadherin(Van den Bossche, Malissen, Mantovani, De Baetselier, & Van Ginderachter, 2011). The cytoplasmic region of Page | 4 the E-cadherin is bound to β-catenin, which is essential for cadherin-based cellcell adhesions, and p120 and this complex binds to the actin cytoskeleton. The extracellular part of E-cadherin binds to another E-cadherin to form dimers, resulting in cell-cell adhesion(Van den Bossche et al., 2011; Yip & Seow, 2011). When EMT occurs in tumors, E-cadherin is downregulated by transcription factors like Snail and Twist, but also by hypermethylation of the promoter. Due to the loss or downregulation of E-cadherin and β-catenin cell detachment can occur, which can result in metastasis(Yip & Seow, 2011). The down regulation of E-cadherin and β-catenin, results in the up regulation of other proteins, like N-cadherin and vimentin(Mani et al., 2008). Ncadherin is an important player in EMT and is present in neural tissue, where it is highly expressed and localizes at adherens junctions. It plays a role in cell-cell and cell-matrix contacts and the cytoplasmic part is involved in signaling pathways. In many tumors E-cadherin is decreased, while N-cadherin is increased, probably playing an important role in cell motility. Studies have shown that N-cadherin can promote metastasis and invasiveness of tumor cells in different cancer models(Jennbacken et al., 2010). Beside N-cadherin other proteins are up regulated, which are characteristic for mesenchymal like cells(Mani et al., 2008). Although a lot of the researchers are convinced that EMT is involved in metastasis, there are still some researchers which are sceptic about this. Studies have shown that Snail and Twist are highly expressed in human tumors, which are highly invasive. However, pathologist David Tarin, MD. PhD. FRCPath is sceptic about EMT. He thinks that EMT cannot occur in tumors, because it is never detected in tumor samples, it is not shown in animal models for cancer, the markers used are not specific enough for mesenchymal cells and due to the genetic instability of tumors these markers are unreliable. Taken this together, there must be a change in epithelial cells to become more motile and invasive, but it is not shown yet that this occurs due to a total EMT(Garber, 2008). Invasion of the tumor cell When the cancer cell is detached from the tumor mass, the tumor cell can invade the surrounding tissue stroma or the blood vessels within the tumor(Thiery & Sleeman, 2006). Lymphatic vessels are not found in tumors, so to metastasize via the lymphatic system, tumor cells have to invade the underlying stroma(Pepper, Tille, Nisato, & Skobe, 2003). If a tumor wants to grow, angiogenesis is essential and this is mainly driven by the growthfactor VEGF(Amin et al., 2011), but also by other factors like matrix metalloproteinases (MMPs), which remodeling the basement membrane of the blood vessels. These growthfactors and proteases are expressed and secreted by tumor cells(Kalluri, 2003). When VEGF binds to its receptor, this induces chemotaxis and mitogenesis of endothelial cells and the permeability of vessels increases(Zhang et al., 2011). The permeability of vessels increases, so new vessels can sprout from these excisting vessels. Tumor vessels show already morphological abnormalities, like the lack of a monolayer of endothelial cells on the blood vessels. Furthermore, the cells lose focal intercellular gaps and interconnections, probably resulting in leaky vessels within the tumor. Due to the leakiness of the vessels, tumor cells are able to Page | 5 invade the blood vessels within the tumor to spread throughout the body(McDonald & Baluk, 2002). Beside invading blood vessels, tumor cells can invade the surrounding tissue stroma to metastasis. Matrix metalloproteinases (MMPs) are key players in invading the tissue stroma by tumor cells. MMPs are involved in biological processes, like migration(), tissue remodeling(Kessenbrock, Plaks, & Werb, 2010), angiogenesis(Kalluri, 2003) and the degradation and cleavage of ECM proteins(Egeblad & Werb, 2002). One of the proteins cleaved by MMPs is Ecadherin and this results in more invasive cells(Egeblad & Werb, 2002). In tumors, carcinoma and stromal cells secrete and express MMPs, resulting in the degradation of basement membrane (BM) components(Al-Raawi, Abu-El-Zahab, El-Shinawi, & Mohamed, 2011). The BM is a ECM, which is specialized, because it forms a barrier between the epithelial cells and the underlying stroma. Signals are transmitted in the ECM via cell-surface adhesion receptors and the ECM maintains cell-cell adhesion. The ECM contains a lot of molecules, including collagens, glycoproteins and proteoglycans(Egeblad & Werb, 2002). The BM is mainly made of networks of type IV collagen and laminin, which are interwoven and form a platform. Furthermore, the BM has pores within the network of type IV collagen, with a size of 50nm. Normally cells cannot migrate through pores which are smaller than 2,0µm, so the BM probably acts as a physical barrier to prevent cells from invasion into the underlying stroma(Rowe & Weiss, 2009). Breakdown of the basement membrane makes invasion possible for carcinoma cells, within the underlying stroma, blood vessels and lymphatic vessels. However, it is not known if invasion of tumor cells, in the underlying stroma, relies on the breakdown of the BM by MMPs(Al-Raawi et al., 2011). Figure 4| Multiphoton microscopy to visualize the migration of tumor cells. (a) The use of multiphoton microscopy resulted in imaging the ECM (purple) and tumor cells (green) at low magnification. Cells are moving along the ECM away from the tumor by invading the surrounding tissue. (b) A higher magnification shows that the tumor cells are slightly attached to the collagen fibers to move towards blood vessels and so on. This picture is a result of the generation of SHG signals(Condeelis & Segall, 2003). When the tumor cells have reached the underlying stroma, cell migration occurs. Cancer cells can move as groups in the form of a chain or as single cells, resulting in an amoeboid movement or extension-adhesion-retraction movement, like mesenchymal cell movement(Thiery & Sleeman, 2006). Invasion of groups tumor cells, occurs in epithelial tumors, like breast cancer, and there are three hallmarks which characterize this. To migrate in groups of cells, the cell-cell Page | 6 junctions within the group have to be intact during migration(1), remodeling occurs of the ECM and the BM has to be rearranged(2). And last, there must be polarity and the cytoskeleton must be activated to generate the traction force, resulting in cell movement. To move, the leading cells express integrins, necessary for the production of adhesion complexes. These adhesion complexes can connect to fibronectin and collagen and cell movement occurs. Collective cell movement has advantages, such as the production of promigratory factors in high concentrations and the inner cells are protected against external factors during the migration(Friedl & Wolf, 2003). Single cell movement can occur as well and the invasion of mesenchymal cells excist out of five steps. These steps include protrusion of pseudopods, focal contact development, proteolysis, contraction of actomyosin and the trailing edge have to detach. Mesenchymal cells derived from epithelial cells, due to EMT, are partially polarized, because rearrangement of F-actin occurs. The focal adhesions of the cell attach the ECM and the tail of the cell moves due to the retraction fibers, which contract. TGF-β is a keyplayer in the dedifferentiation of epithelial cells into mesenchymal cells, resulting in an increased invasion. Beside mesenchymal cell migration there is amoeboid cell movement. Amoeboid cells, compared to mesenchymal cells, do not have cell polarity, chemotaxis cannot occur anymore and the attachment to the ECM is loose. These cells move faster than mesenchymal cells and collective cancer cells. Furthermore, amoeboid cells migration is protease independent and does not result in the breakdown of ECM components, like fibrils. Migration occurs by displacing the fibrils with the use of mechanical forces, which are actomyosin based. However, this can only take place when there is no stabilization of the formed pores in the collagen network and thus no ECM stiffness occurs. Normally there will be protease activity of MMP14 in collagen networks, which provide amoeboid cell movement, because the pores within the collagen network are too small to move through. Amoeboid cell movement is probably a consequence of the use of protease inhibitors or integrin blocking drugs, which are used to treat cancer. Clinical trials have shown that MMP inhibitors induce cancer progression via amoeboid cells instead of preventing it. Moreover, in vitro studies about amoeboid cell invasion were done under conditions not equal to the in vivo collagen composition and this raises the debate if amoeboid cell movement is protease indepent(van Zijl, Krupitza, & Mikulits, 2011). Intravasation, circulation and extravasation When the tumor cells are in the underlying stroma, intravasation of blood vessels and lymphatic vessels can take place. Normally the lymphatic network collects immune cells and macromolecules from extravasated fluid, before sending them back to the blood system. If tumor cells want to enter the lymphatic system, they have to move through the wall of a lymphatic vessel. Tumor cells probably migrate through the wall of lymphatic capillairies. The wall of these capillairies lacks pericytes and a basement membrane and the junctions between the endothelial cells are more sparse, than the junctions in blood vessels. This could explain why it is easier for tumor cells to enter the lymphatic vessels, but this is not proved yet and it is still unknown if these junctions make it possible for tumor cells to enter. When the tumor cell has entered the lymphatic vessel it can Page | 7 stay there to grow or return to the blood circulation via the thoracic duct(Pepper et al., 2003). As mentioned above, MMPs are involved in angiogenesis and are able to remodel the basement membrane of blood vessels, so tumor cells can enter the blood circulation. With the use of the chick CAM model, researchers have shown that MMP9 is involved in intravasation of tumor cells. The blood vessels which are mainly intravasated by tumor cells, are probably angiogenic blood vessels instead of mature blood vessels. Intravasation of tumor induced angiogenic blood vessels versus mature blood vessels is shown with specialized mouse models. Although MMPs are involved in intravasation of tumor cells in animal models, this has not been shown yet in cancer patients, raising the question if MMPs are involved in metastasis in humans(Deryugina & Quigley, 2006). When the tumor cell enters the lymphatic circulation, it does not have to overcome barriers, so this could promote survival and dissemination of these cells. Most studies about metastasis focused on the behavior of tumor cells in the blood circulation, but not how tumor cells behave in the lymphatic circulation. Perhaps tumor cells have to overcome certain barriers when entering the lymphatic circulation, but this still has to be studied(Pepper et al., 2003). However, when tumor cells enter the blood circulation they have to survive to be able to form secondary tumors. One way to try to survive is to interact with blood platelets. Blood platelets facilitate migration and invasion of tumor cells and the arrest in blood vessels. Tumor cells are able to mimic receptor expression from blood platelets, resulting in the expression of the GPIIb/IIIa receptor. Tumor cells and blood platelets interact via bridging proteins, like fibrinogen, which can bind the GPIIb/IIIa receptor and this protect tumor cells. Tumor cells use the blood platelets as shields to protect them against phsycial stressors in the blood circulation and to evade the effector cells of the immune system(Bambace & Holmes, 2011). Tumor cells go with the flow of the blood stream, before they arrest in capillaries(Chambers, Groom, & MacDonald, 2002). When tumor cells arrest, which is facilitated by blood platelets, they can stay and grow or extravasate out of the capillary into the surrounding tissue(Bambace & Holmes, 2011; Chambers et al., 2002). There are several factors involved in extravasation, such as blood platelets. Blood platelets can help tumor cells to extravasate into the underlying tissue, because activated platelets can release proteolytic enzymes, like heparanase and MMPs, which can remodel the BM of blood vessels. In addition, blood platelets can also activate other proteinases and/or stimulate the activation of proteinases by endothelial and tumor cells(Bambace & Holmes, 2011). When tumor cells enter the underlying stroma, it is important to form a new tumor. To develop a new tumor, there must be growth of the cells and to induce this an reciprocal signaling network will be made by tumor cells. These signaling networks activate stromal cells to grow, resulting in a secondary tumor(Langley & Fidler, 2011). Furthermore, angiogenesis has to occur to sustain the growth of the tumor(Chambers et al., 2002). Another possibility is that the tumor cells enter the dormant state. This can occur in both the primary and secondary tumor. Dormancy has advantages and disadvantages. The disadvantages are that these tumors are too small to be detected and when the escape the dormant state they grow fast and are most of the time lethal. However, an advantage can be that tumor progression is paused Page | 8 and this can take years. When tumor cells are in the dormant state, they are asymptomatic and occult for a period. Although there is not much known about tumor dormancy, there are some mechanisms, which are probably involved in this process. One of these is impaired angiogenesis is associated with dormancy. In different experimental models tumor cells could not grow, because of impaired angiogenesis, resulting in asymptomatic small lesions. To escape the dormant state only one angiogenic factor is enough, which probably affects the angiogenic balance by inducing a shift. Other mechanisms which are most likely involved in dormancy are mechanisms such as the ECM, the environment of the surrounding tissue and the interaction between the normal stromal cells and the tumor itself. Even though there are many mechanisms, which can induce dormancy, escaping dormancy is poorly understood(Almog, 2010). Studying cancer metastasis It is important to study cancer and especially metastasis, not only to find out how tumor cells behave in vivo, but also the underlying molecular mechanisms of this process are important. The knowledge gained about metastasis can be used in the clinics to develop drugs and the right treatment program for patients. For many years researchers were only able to study metastasis with techniques, such as immunohistochemistry, which could only provide static pictures. These static pictures have advantages such as providing information about the expression of antigens (e.g. proteins)(Fritschy, 2008) and it quickly generates results(RamosVara, 2005). Nevertheless, there are disadvantages which make it hard to study metastasis. Metastasis is a highly dynamic process, which can take months and different time point have to be chosen to study this. At every time point an animal will be sacrificed to do IHC and due to the diversity within the species, a lot of animals have to be sacrificed before a valid conclusion can be made(Ramos-Vara, 2005). To overcome these problems, tissues of living animals are studied with microscopes, known as intravital microscopy, which was first done in the 19th century(Beerling, Ritsma, Vrisekoop, Derksen, & van Rheenen, 2011; Ntziachristos, 2010). During first years of intravital microscopy, most studies only focused on the microcirculation and vasculature. Imaging other tissues was limited, because there was no contrast and the optics available at the moment were not optimal yet. Later on in the 1950s with the use of an ear chamber, in a rabbit, metastasis could be visualized. The biggest breakthroughs for intravital microscopy occurred in the 1990s, when imaging techniques really improved and animal models were available, which were genetic cancer models and they could express fluorescent proteins. From that moment on, intravital microscopy become very important as tool for studies about the underlying mechanisms of cancer and metastasis(Beerling et al., 2011). Nowadays, there are several intravital microscopy techniques, which are used to study cancer and metastasis. One of these techniques is multiphoton microscopy and in this review I will show why multiphoton microscopy is a good alternative to study metastasis compared to the other techniques available. Page | 9 The advances of Multiphoton microscopy in cancer studies Immunohistochemistry versus intravital microscopy In 1941 Coons et al. published a paper describing a new technique, which can detect cellular antigens with immunofluorescence in different tissues. This was the beginning of a technique called immunohistochemistry(Ramos-Vara, 2005). Since the development of IHC, it has become an important tool for both the research and the diagnostics to study neoplastic and infectious diseases in animal models. IHC links three disciplines, which are immunology, chemistry and histology. Aim of IHC is to detect antigens within tissues using specific antibodies, resulting in the binding between antigen and antibody. This interaction can be visualized with ultraviolet light, using fluorochromes or with light microscopy, a histochemical reaction has been performed to color the antigen-antibody interaction. During the years, IHC became a more complex technique, because the specificity and sensitivity had to be increased(Mighell, Hume, & Robinson, 1998; Ramos-Vara, 2005). Nowadays it is possible to observe one or multiple antigens together or to detect one specific antigen in the same section of multiple tissues(Ramos-Vara, 2005). Before IHC can be performed, the tissue has to be fixated. Fixation is essential for IHC and there are four reasons which support this. First of all, with fixation cellular components are protected and this includes the structural and soluble proteins. Secondly, displacement of cell components and autolysis cannot occur anymore. Third, cellular materials are stabilized and protected against harmful effects of the following procedures. Last, fixation promote immunostaining and conventional staining. Although, IHC has advantages such as the detection of multiple antigens, gaining a lot of information and that fixation does not affect cellular materials, but there are disadvantages as well. As mentioned before, metastasis is a process which can take months, so to study metastasis a lot of time points have to be chosen. For every time point an animal has to be sacrificed and because there is diversity between the species, even more animals have to be sacrificed before a statistically valid conclusion can be made. In addition, cross reaction of antibodies between the species can occur during IHC, which can affect the results(Ramos-Vara, 2005). To overcome these obstacles, intravital microscopy techniques are developed. With these techniques we are able to follow metastasis over time in vivo, without sacrificing many animals. Furthermore, we can follow one cell or a group of cells over time and study the behavior of the single cell or group cells in different environments(Beerling et al., 2011). Taken together, intravital microscopy is very use full for studying processes, such as metastasis, in vivo to gain more knowledge to understand such processes better. Multiphoton microscopy versus other intravital microscopy techniques There are many intravital imaging techniques, which differ in imaging depth, resolution, cost and contrast reagents. Ultrasound, MRI and CT are techniques which are also used in the clinic. Although these techniques have a high imaging depth, the resolution is very low and only in some parts of the body there will be a high contrast, but still less details are visible{{3 Reviewed in Beerling,E. 2011}}. Page | 10 Techniques like positron emission tomography and whole body bioluminescence are important tools to study metastasis as a process. PET and whole body bioluminescence both facilitate the discovery of secondary tumors, which are tissue specific, and primary tumors in several animal models. Furthermore, it is possible to follow tumor growth in vivo, but these techniques are not able to follow these processes on the level of single cells(Beerling et al., 2011; Provenzano, Eliceiri, & Keely, 2009). Figure 5| Different intravital techniques to visualize metastasis. With these techniques we can visualize different stages of metastasis. Whole body imaging, PET and Biofluorescence/Bioluminescence are techniques used in the clinical and cannot follow metastasis on single cell level. To detect single cells techniques like confocal and multiphoton microscropy are used(Provenzano et al., 2009). There are other intravital microscopy techniques, which are able to study processes, such as metastasis, on the level of a single cell. This can only be done with the use of contrast reagents(Beerling et al., 2011). These days, there are animal cancer models, which are able to express fluorescent and photoswitchable proteins(Beerling et al., 2011; Kedrin et al., 2008). The expression of these proteins is necessary to visualize tumors or tumor cells, with the use of confocal laser scanning microscopy and multiphoton microscopy(Beerling et al., 2011). Confocal microscopy uses light, ranging from ultra violet to infrared, which can be focused on a specific part of the specimen. The light of the focused beam is necessary to excite a photon, which is absorpted by a fluorophore, resulting in the emission of fluorescence in different directions in the tissue. The fluorescent signals are collected by the pinhole aperture and the signals which are from outside the field are blocked(Smith, 2001). The spinning disk confocal microscopy uses multiple pinholes to block signals from out of the field(Beerling et al., 2011; Smith, 2001), which results in the enhancement of contrast and an increased Z-resolution and this make it easier to visualize the subcellular structues(Beerling et al., 2011). In addition, there is less photobleaching with the use of spinning disk confocal microscopy(Smith, 2001), so long term imaging of tumor cells can be performed as well(Beerling et al., 2011). However, there Page | 11 confocal microscopy has its disadvantages too. There is more photobleaching with the use of another confocal microscopy, the laser scanning confocal microscope(Smith, 2001). More important is that the confocal microscopy uses light to excite photons to induce fluorescence. These photons scatter in the tissue, which can contribute to blurry pictures(Dunn & Young, 2006; Ntziachristos, 2010; Provenzano et al., 2009) and this limits the imaging depth in specimen to 100µm(Beerling et al., 2011). By contrast, multiphoton microscopy makes use of near infrared (NIR) light, ranging from 700nm to 1100nm, to excite photons(Konig, 2000). In this techniques photons are used with half the energy as one photon used with confocal microscopy. Two photons has to be excited, before absorption by a fluorophore can occur, resulting in fluorescence(Beerling et al., 2011). Because this is a rare event, it can only occur at the focal point and photon scattering cannot occur. This results in a deeper penetration of the light within the specimen to 1000µm(Konig, 2000). With the use of multiphoton microscopy there is less photobleaching and phototoxicity, making this technique also useful for long term studies following tumor cells. In addition, multiphoton microscopy can generate second harmonic generation (SHG) signals. These second harmonic generation signals can provide essential information about the interaction between the cell and its surrounding environment, such as collagen(Provenzano et al., 2009). The generation of these signals do not interfere with the activation of photons, to induce fluorescence(Beerling et al., 2011). Taken together, both techniques can be very useful in studying metastasis, but due to the imaging depth and the high detection effenciency multiphoton microscopy is a good alternative to study metastasis in vivo (Beerling et al., 2011; Provenzano et al., 2009). The role of fluorescent proteins in research Using multiphoton microscopy or other intravital microscopy techniques, makes it possible to visualize interactions between different molecules and cell types in vivo en in vitro. For years, studies using intravital microscopy, were limited in studying processes such as cell-cell interactions, because cells were labeled with vital dyes and they could only study the vasculature of the tumor. With the discovery of the green fluorescent proteins (GFP) and the production of the three other colors, red fluorescent protein (RFP), cyan fluorescent protein (CFP) and yellow fluorescent protein (YFP) it is possible to study different cell types in vivo, specifically tumor cells(Condeelis & Segall, 2003; Hoffman, 2009). Farina et al. was the first to characterize the behavior of carcinoma cells during migration with the use of GFP and confocal microscopy(Condeelis & Segall, 2003). The use of GFP to label tumor cells for a longer period, resulted in questions about the long term effects. It was unknown if GFP affected the behavior of tumor cells during a time period. Several studies with GFP models have shown that the tumor is not affected in processes, like growth, when used for a longer period(Condeelis & Segall, 2003). A couple of years ago photoswitchable proteins were discovered, which are fluorescent proteins as well. With photoswitchable proteins, different cell types can be labeled and they can be combined with the other fluorescent proteins to study for example the behavior of tumor cells(Beerling et al., 2011). Page | 12 The advantage of the use of fluorescent proteins is that it makes it possible to study the cell fate over a longer period of time(Beerling et al., 2011; Ottobrini, Martelli, Trabattoni, Clerici, & Lucignani, 2011). Beside labeling with fluorescent proteins, cells can be labeled with fluorescent trackers(Beerling et al., 2011; Ottobrini et al., 2011). Cells are isolated and labeled, before they are transferred back into the animal(Beerling et al., 2011). Labeling cells with fluorescent trackers has the advantage that fluorescent trackers easily penetrate the membrane and that they bind specific targets. The drawbacks of this method are that the rate of successful binding of the label depends on the cell type (the membrane composition, the phagocytic capacity of the cells etc.), but more important, these trackers can only be used for cells which cannot dedifferentiate anymore, such as macrophages and dendritic cells. If proliferating cells are labeled with these trackers, the label in the daughter cells might be diluted or lost due to cell cycle events or cell death(Ottobrini et al., 2011). The only problem using most fluorescent proteins and dyes simultaneously, results in the overlap of the emission spectra. To discriminate between these fluorescent proteins and dyes, is using fluorescence lifetime imaging microscopy (FLIM). With FLIM the lifetime of the fluorescent signal will be measured. Furthermore, FLIM is very useful in distinguishing between tissue types, by measuring the autofluorescence. Autofluorescence occurs when cells are excitated, resulting in the emission of fluorescence by endogenous components, like melanin and tryptophan. Measuring the lifetime of the autofluorescence makes it possible to discriminate between tumor and healthy tissues(Beerling et al., 2011). Imaging processes over a longer period Many studies involving intravital microscopy have focused on the visualization of tumor cells in vivo. A lot of these studies followed tumor cells for a short period of time, from 4 hours up to 24 hours, thereby lacking information about tumor cells which are able to invade the underlying stroma. The technique used to follow cells in such a small time span is surgical dissection, but this technique has several disadvantages. The consequences of surgical dissection of the imaging site are dehydration of the dissected tissue, survival and/or thermoregulatory control of the animal, side effects of the use of anesthesia for a longer time period and the site to be imaged can be limited(Kedrin et al., 2008). Following tumor cells for a longer period, to gain more knowledge about cell behavior resulted in the development of observation chambers and Sandison (1924) was the first to implant a chamber into animals. In 1993 Lehr et al. developed the skin fold chamber, which were implanted on the dorsal side of the animals. These chambers combined with intravital microscopy are used to study the microvasculature and growth of non-neoplastic and neoplastic tissue for several days(Lehr, Leunig, Menger, Nolte, & Messmer, 1993). After the skin fold chamber, orthotopic chambers are developed to study tumors in their natural environment. Kedrin et al. developed a mammary imaging window (MIW) to track subpopulations of tumor cells within different tumor microenvironments with the use of photo-switchable fluorescent proteins. Cell migration can be followed for more than one day and the imaging site can be Page | 13 studied up to 21 days post implantation. With the use of the MIW Kedrin et al. found that in the tumor there are different microenvironments and that these environments affect intravasation and invasion of the tumor cells in breast tumors(Kedrin et al., 2008). So, imaging windows are very useful in studies about metastasis, to acquire more information about tumor cell behavior in vivo combined with intravital microscopy. The role of multiphoton microscopy in metastasis: lymphatic and/or hematogenous? When tumor cells are detached from the tumor, they can escape the tumor via lymphatic or blood vessels. However, there are studies which have shown with intravital imaging that lymphatic vessels are lacking within the tumor itself or are non functional(Beerling et al., 2011; Pepper et al., 2003). When the tumor cells enter the underlying stroma, they are able to metastasis via the lymphatic system. After entering the lymphatic system, the tumor cell first ends up in the lymph node where it can stay and grow or go to the heart via the efferent vessels of the lymphatic system(Beerling et al., 2011; Pepper et al., 2003). From the heart the cells go with the blood flow to other parts of the body, where secondary tumors can be formed(Beerling et al., 2011). The detection of tumor cells within the axillary node is used as an marker for metastasis(Patani & Mokbel, 2011), but it is still unknown if tumor cells start to metastasize via the lymphatic system or that this only shows the invasiveness of tumor cells within the lymph nodes(Beerling et al., 2011). Thus, it is still unknown via which route tumor cells metastasize to other parts of the body. To figure out if tumor cells metastasize via the hematogenous route, lymphatic route or both routes, Giampieri et al. used the transforming growth factor β (TGFβ) to study intravasation of tumor cells in lymph nodes or blood vessels. TGFβ can change the cell behavior, by the activation of Smads and Twist, which results in EMT and a more invasive and mesenchymal character of these cells. With the use of multiphoton microscopy, Giampieri et al. studied the intravasation process of MTLn3 cells in mice models. When TGFβ is expressed in these cells, Smad2, which is bound to GFP, translocates to the nucleus and accumulate. Within the tumor there was a difference in TGFβ signaling, resulting in quick moving single cells and slow moving cells, which are moving in groups. The cells which had high TGFβ signaling, moved toward blood vessels to intravasate them, before moving to other organs. Furthermore they studied the behavior of MLTn3 cells when there is no TGFβ activity at all. Overexpression of the TGFβ type 2 receptor, a dominant negative receptor, resulted in cohesively cell movement and intravasation into the blood vessels could not occur anymore. More important to notice is that these cells are still able to intravasate the lymph nodes to cause lymphatic metastasis and to spread throughout the body(Giampieri, Pinner, & Sahai, 2010). So, TGFβ is an important player for intravasation of single tumor cells into the blood vessels and this suggest that the tumor models used, tumor cells metastasis via the blood circulation. However, this has to be studied in other cancer models before it is proven that tumors start metastasizing via the blood circulation and not via the lymphatic system (Beerling et al., 2011; Giampieri et al., 2010). Page | 14 The role of multiphoton microscopy in metastasis: invasion and intravasation As mentioned before, until the development of intravital imaging, most studies were done with the use of techniques which provides static pictures. These pictures showed where the tumor cells are located in the tissue or organ and this provided information to determine whether the patient had a good or poor prognosis(Beerling et al., 2011). However, tumor cells which are escaping the tumor via blood vessels within the tumor are not marked in the pictures, because it was overlooked or not studied yet. Furthermore, it is not possible to visualize the intravasation and motility of these cells within the tumor(Beerling et al., 2011). During tumor formation there will be angiogenesis, so tumor cells can escape the tumor and move to other tissues and organs. The escape via blood vessels in the tumor mass is easier than the intravasation of mature blood vessels, because these vessels are very leaky due to the poor structure. The blood vessels within the tumor lack a monolayer of endothelial cells at the outside of the vessel and there is loss of interconnections between cells as well. In contrast the blood vessels in the underlying tissue have a dense basement membrane, so intravasation of these vessels is harder(McDonald & Baluk, 2002). Intravital microscopy have shown that tumor cells within the tumor end up in the blood circulation accidently and some delibaretly, but that there are tumor cells too, which migrate toward blood vessels in the surrounding microenvironment. Kedrin et al. showed with the integration of a mammary imaging window, that activated cells labeled with the photoswitchable protein Dendra2 show migration and invasion in the surrounding tissue stroma of the tumor. Futhermore, they showed that the microenvironment does affect the behavior of tumor cells. In microenvironments without blood vessels surrounding the photoswitched cells there is barely cell migration and invasion of the surrounding stroma, but the red cells are increased. Nevertheless, tumor cells in microenvironments near blood vessels, showed a decrease in tumor cells and a higher migration towards the blood vessels, resulting in intravasation of the blood vessels(Kedrin et al., 2008). Figure 6| Following tumor cell movement, within 24 hours, in two different environments: with and without blood vessels surrounding tumor cells. (a) The pictures show MLTn3 breast tumor cells, nonphotoswitched(green) and photoswitched(red) in an environment without blood vessels nearby. There is migration of the tumor cells, but this is limited. Nonetheless, there are more photoswitched cells visible. (b) In this environment there is a blood Page | 15 vessel next to the photoswitched tumor cells and most of the cells migrate towards the blood vessel or to other areas outside the field. The cells next to the blood vessel are decreasing, suggesting that intravasation occurs into the blood vessel and spreading throughout the body(Kedrin et al., 2008). Taken together, with the use of intravital microscopy we are able to visualize the intravasation of active and inactive tumor cells into blood vessels, which are localized in the tumor mass. This shows that intravasation of tumor cells into blood vessels not only occurs during the invasion of the underlying tissue stroma. So, escaping the tumor mass via the blood vessels within the tumor, makes this a very good escape route, which can be important to form secondary tumors elsewhere in the body(Beerling et al., 2011). Discussion In this review, I have tried to show that multiphoton microscopy is a good alternative compared to other intravital micropscopy techniques for studying cancer and especially metastasis. With this technique we are now able to follow tumor cells for longer periods in vivo with the use of imaging windows, resulting in gaining more knowledge about important processes like invasion, intravasation and tumor cell spread via the lymphatic or hematogenous system. Although a lot of studies focused on the early steps of metastasis, fewer studies focused on the behavior of tumor cells within organs, which are targets for metastatic lesions. If we are able to visualize tumor cells within in organs, such as the lungs, and to follow these over time with intravital microscopy, this will solve many questions, like why tumor cells specifically go to the lungs to form metastasic lesions or why tumor cells enter the dormant state and how to escape this state(Beerling et al., 2011). Although multiphoton microscopy has overcome most disadvantages of the other intravital imaging techniques, this technique also has disadvantages. Some disadvantages are that the costs are very expensive and the imaging depth is only <1000µm. For the future it is important to try to overcome these limitations, so studying processes like metastasis is easier, especially if you want to look at tumors which are located at >1000µm depth(Provenzano et al., 2009). Unfortunately, the use of intravital imaging techniques cannot provide information about the molecular pathways, which are involved in these processes. Studying metastasis or tumor formation is now mainly observational instead of providing information about molecular pathways and it is important for the future to use multiphoton microscopy in experiments, which give more information about the molecular processes. To study the molecular mechanisms intravital microscopy can be combined with other techniques like FRET and FLIM. These techniques are used to study processes, like intracellular signaling and communication between cells and this can contribute a lot to metastasis research(Beerling et al., 2011; Provenzano et al., 2009). The combination with other techniques can also help solving questions about other processes, not involved in metastasis or tumor formation. Although there are some good results with the use of FRET/FLIM together with multiphoton microscopy, still there are some disadvantages. It is known that a lot of FRET biosensors are not sensitive enough for the use in multiphoton microscopy experiments and the Page | 16 incorporation of these probes. Another problem is the low signal-to-noise ratio and the ambition is to increase the sensitivity and signal-to-noise ratio for the new probes, so that the results of the experiments are better. These probes are now made for studying e.g. cell motility, but the aim is to make new probes for other processes like adhesion or proliferation(Beerling et al., 2011). Multiphoton microscopy is a key player in understanding the process of tumor formation and especially metastasis. This technique can give us more information about essential steps during metastasis, like invasion, migration and extravasation. Nowadays, the technique is only used to observe how metastasis takes place, but does not give any information about the molecular pathways involved. In addition, most models used for studying cancer make use of xenograft techniques, so the environment of the tumor differs from the natural environment. In the tumor environment there are many key players, like immune cells, which are involved in the process of tumor formation and metastasis. Fortunately there are some cancer models, which are mimicking the development of human tumors. Taken together, multiphoton microscopy is a good alternative compared to other intravital techniques for studying metastasis, to understand the process better and to provide treatment techniques to prevent the formation of secondary and/or tertiary tumors(Beerling et al., 2011). References Almog, N. (2010). Molecular mechanisms underlying tumor dormancy. Cancer Letters, 294(2), 139-146. doi:10.1016/j.canlet.2010.03.004 Al-Raawi, D., Abu-El-Zahab, H., El-Shinawi, M., & Mohamed, M. M. (2011). Membrane type-1 matrix metalloproteinase (MT1-MMP) correlates with the expression and activation of matrix metalloproteinase-2 (MMP-2) in inflammatory breast cancer. International Journal of Clinical and Experimental Medicine, 4(4), 265-275. Amin, E. M., Oltean, S., Hua, J., Gammons, M. V., Hamdollah-Zadeh, M., Welsh, G. I., et al. (2011). WT1 mutants reveal SRPK1 to be a downstream angiogenesis target by altering VEGF splicing. Cancer Cell, 20(6), 768-780. doi:10.1016/j.ccr.2011.10.016 Bacac, M., & Stamenkovic, I. (2008). Metastatic cancer cell. Annual Review of Pathology, 3, 221247. doi:10.1146/annurev.pathmechdis.3.121806.151523 Bambace, N. M., & Holmes, C. E. (2011). The platelet contribution to cancer progression. Journal of Thrombosis and Haemostasis : JTH, 9(2), 237-249. doi:10.1111/j.1538-7836.2010.04131.x; 10.1111/j.1538-7836.2010.04131.x Beerling, E., Ritsma, L., Vrisekoop, N., Derksen, P. W., & van Rheenen, J. (2011). Intravital microscopy: New insights into metastasis of tumors. Journal of Cell Science, 124(Pt 3), 299310. doi:10.1242/jcs.072728 Buda, A., & Pignatelli, M. (2011). E-cadherin and the cytoskeletal network in colorectal cancer development and metastasis. Cell Communication & Adhesion, doi:10.3109/15419061.2011.636465 Chaffer, C. L., & Weinberg, R. A. (2011). A perspective on cancer cell metastasis. Science (New York, N.Y.), 331(6024), 1559-1564. doi:10.1126/science.1203543 Page | 17 Chambers, A. F., Groom, A. C., & MacDonald, I. C. (2002). Dissemination and growth of cancer cells in metastatic sites. Nature Reviews.Cancer, 2(8), 563-572. doi:10.1038/nrc865 Condeelis, J., & Segall, J. E. (2003). Intravital imaging of cell movement in tumours. Nature Reviews.Cancer, 3(12), 921-930. doi:10.1038/nrc1231 Deryugina, E. I., & Quigley, J. P. (2006). Matrix metalloproteinases and tumor metastasis. Cancer Metastasis Reviews, 25(1), 9-34. doi:10.1007/s10555-006-7886-9 Dunn, K. W., & Young, P. A. (2006). Principles of multiphoton microscopy. Nephron.Experimental Nephrology, 103(2), e33-40. doi:10.1159/000090614 Egeblad, M., & Werb, Z. (2002). New functions for the matrix metalloproteinases in cancer progression. Nature Reviews.Cancer, 2(3), 161-174. doi:10.1038/nrc745 Faridi, R., Zahra, A., Khan, K., & Idrees, M. (2011). Oncogenic potential of human papillomavirus (HPV) and its relation with cervical cancer. Virology Journal, 8, 269. doi:10.1186/1743422X-8-269 Friedl, P., & Wolf, K. (2003). Tumour-cell invasion and migration: Diversity and escape mechanisms. Nature Reviews.Cancer, 3(5), 362-374. doi:10.1038/nrc1075 Fritschy, J. M. (2008). Is my antibody-staining specific? how to deal with pitfalls of immunohistochemistry. The European Journal of Neuroscience, 28(12), 2365-2370. doi:10.1111/j.1460-9568.2008.06552.x Garber, K. (2008). Epithelial-to-mesenchymal transition is important to metastasis, but questions remain. Journal of the National Cancer Institute, 100(4), 232-3, 239. doi:10.1093/jnci/djn032 Giampieri, S., Pinner, S., & Sahai, E. (2010). Intravital imaging illuminates transforming growth factor beta signaling switches during metastasis. Cancer Research, 70(9), 3435-3439. doi:10.1158/0008-5472.CAN-10-0466 Gupta, G. P., & Massague, J. (2006). Cancer metastasis: Building a framework. Cell, 127(4), 679695. doi:10.1016/j.cell.2006.11.001 Hanahan, D., & Weinberg, R. A. (2011). Hallmarks of cancer: The next generation. Cell, 144(5), 646-674. doi:10.1016/j.cell.2011.02.013 Hoffman, R. M. (2009). Imaging cancer dynamics in vivo at the tumor and cellular level with fluorescent proteins. Clinical & Experimental Metastasis, 26(4), 345-355. doi:10.1007/s10585-008-9205-z Jass, J. R. (2004). HNPCC and sporadic MSI-H colorectal cancer: A review of the morphological similarities and differences. Familial Cancer, 3(2), 93-100. doi:10.1023/B:FAME.0000039849.86008.b7 Jennbacken, K., Tesan, T., Wang, W., Gustavsson, H., Damber, J. E., & Welen, K. (2010). N-cadherin increases after androgen deprivation and is associated with metastasis in prostate cancer. Endocrine-Related Cancer, 17(2), 469-479. doi:10.1677/ERC-10-0015 Kalluri, R. (2003). Basement membranes: Structure, assembly and role in tumour angiogenesis. Nature Reviews.Cancer, 3(6), 422-433. doi:10.1038/nrc1094 Page | 18 Kedrin, D., Gligorijevic, B., Wyckoff, J., Verkhusha, V. V., Condeelis, J., Segall, J. E., et al. (2008). Intravital imaging of metastatic behavior through a mammary imaging window. Nature Methods, 5(12), 1019-1021. doi:10.1038/nmeth.1269 Kessenbrock, K., Plaks, V., & Werb, Z. (2010). Matrix metalloproteinases: Regulators of the tumor microenvironment. Cell, 141(1), 52-67. doi:10.1016/j.cell.2010.03.015 Konig, K. (2000). Multiphoton microscopy in life sciences. Journal of Microscopy, 200(Pt 2), 83104. Langley, R. R., & Fidler, I. J. (2011). The seed and soil hypothesis revisited--the role of tumorstroma interactions in metastasis to different organs. International Journal of Cancer.Journal International Du Cancer, 128(11), 2527-2535. doi:10.1002/ijc.26031; 10.1002/ijc.26031 Lehr, H. A., Leunig, M., Menger, M. D., Nolte, D., & Messmer, K. (1993). Dorsal skinfold chamber technique for intravital microscopy in nude mice. The American Journal of Pathology, 143(4), 1055-1062. Mani, S. A., Guo, W., Liao, M. J., Eaton, E. N., Ayyanan, A., Zhou, A. Y., et al. (2008). The epithelialmesenchymal transition generates cells with properties of stem cells. Cell, 133(4), 704-715. doi:10.1016/j.cell.2008.03.027 McDonald, D. M., & Baluk, P. (2002). Significance of blood vessel leakiness in cancer. Cancer Research, 62(18), 5381-5385. Mighell, A. J., Hume, W. J., & Robinson, P. A. (1998). An overview of the complexities and subtleties of immunohistochemistry. Oral Diseases, 4(3), 217-223. Moore, L. E., Nickerson, M. L., Brennan, P., Toro, J. R., Jaeger, E., Rinsky, J., et al. (2011). Von hippellindau (VHL) inactivation in sporadic clear cell renal cancer: Associations with germline VHL polymorphisms and etiologic risk factors. PLoS Genetics, 7(10), e1002312. doi:10.1371/journal.pgen.1002312 Ntziachristos, V. (2010). Going deeper than microscopy: The optical imaging frontier in biology. Nature Methods, 7(8), 603-614. doi:10.1038/nmeth.1483 Ottobrini, L., Martelli, C., Trabattoni, D. L., Clerici, M., & Lucignani, G. (2011). In vivo imaging of immune cell trafficking in cancer. European Journal of Nuclear Medicine and Molecular Imaging, 38(5), 949-968. doi:10.1007/s00259-010-1687-7 Patani, N., & Mokbel, K. (2011). Clinical significance of sentinel lymph node isolated tumour cells in breast cancer. Breast Cancer Research and Treatment, 127(2), 325-334. doi:10.1007/s10549-011-1476-4 Pepper, M. S., Tille, J. C., Nisato, R., & Skobe, M. (2003). Lymphangiogenesis and tumor metastasis. Cell and Tissue Research, 314(1), 167-177. doi:10.1007/s00441-003-0748-7 Provenzano, P. P., Eliceiri, K. W., & Keely, P. J. (2009). Multiphoton microscopy and fluorescence lifetime imaging microscopy (FLIM) to monitor metastasis and the tumor microenvironment. Clinical & Experimental Metastasis, 26(4), 357-370. doi:10.1007/s10585-008-9204-0 Ramos-Vara, J. A. (2005). Technical aspects of immunohistochemistry. Veterinary Pathology, 42(4), 405-426. doi:10.1354/vp.42-4-405 Page | 19 Rowe, R. G., & Weiss, S. J. (2009). Navigating ECM barriers at the invasive front: The cancer cellstroma interface. Annual Review of Cell and Developmental Biology, 25, 567-595. doi:10.1146/annurev.cellbio.24.110707.175315 Shulman, L. P. (2010). Hereditary breast and ovarian cancer (HBOC): Clinical features and counseling for BRCA1 and BRCA2, lynch syndrome, cowden syndrome, and li-fraumeni syndrome. Obstetrics and Gynecology Clinics of North America, 37(1), 109-33, Table of Contents. doi:10.1016/j.ogc.2010.03.003 Smith, C. L. (2001). Basic confocal microscopy. Current Protocols in Neuroscience / Editorial Board, Jacqueline N.Crawley ...[Et Al.], Chapter 2, Unit 2.2. doi:10.1002/0471142301.ns0202s00 Thiery, J. P., & Sleeman, J. P. (2006). Complex networks orchestrate epithelial-mesenchymal transitions. Nature Reviews.Molecular Cell Biology, 7(2), 131-142. doi:10.1038/nrm1835 Valko, M., Leibfritz, D., Moncol, J., Cronin, M. T., Mazur, M., & Telser, J. (2007). Free radicals and antioxidants in normal physiological functions and human disease. The International Journal of Biochemistry & Cell Biology, 39(1), 44-84. doi:10.1016/j.biocel.2006.07.001 Van den Bossche, J., Malissen, B., Mantovani, A., De Baetselier, P., & Van Ginderachter, J. A. (2011). Regulation and function of the E-cadherin/catenin complex in cells of the monocytemacrophage lineage and DC. Blood, doi:10.1182/blood-2011-10-384289 van Zijl, F., Krupitza, G., & Mikulits, W. (2011). Initial steps of metastasis: Cell invasion and endothelial transmigration. Mutation Research, 728(1-2), 23-34. doi:10.1016/j.mrrev.2011.05.002 Veal, E. A., Day, A. M., & Morgan, B. A. (2007). Hydrogen peroxide sensing and signaling. Molecular Cell, 26(1), 1-14. doi:10.1016/j.molcel.2007.03.016 Wrzeszczynski, K. O., Varadan, V., Byrnes, J., Lum, E., Kamalakaran, S., Levine, D. A., et al. (2011). Identification of tumor suppressors and oncogenes from genomic and epigenetic features in ovarian cancer. PloS One, 6(12), e28503. doi:10.1371/journal.pone.0028503 Yip, W. K., & Seow, H. F. (2011). Activation of phosphatidylinositol 3-kinase/Akt signaling by EGF downregulates membranous E-cadherin and beta-catenin and enhances invasion in nasopharyngeal carcinoma cells. Cancer Letters, doi:10.1016/j.canlet.2011.12.018 Zhang, L., Wang, J. N., Tang, J. M., Kong, X., Yang, J. Y., Zheng, F., et al. (2011). VEGF is essential for the growth and migration of human hepatocellular carcinoma cells. Molecular Biology Reports, doi:10.1007/s11033-011-1304-2 Page | 20