[Show abstract][Hide abstract] ABSTRACT: Cell biology research is fundamentally limited by the number of intracellular components, particularly proteins, that can be co-measured in the same cell. Therefore, cell-to-cell heterogeneity in unmeasured proteins can lead to completely different observed relations between the same measured proteins. Attempts to infer such relations in a heterogeneous cell population can yield uninformative average relations if only one underlying biochemical network is assumed. To address this, we developed a method that recursively couples an iterative unmixing process with a Bayesian analysis of each unmixed subpopulation.
Our approach enables to identify the number of distinct cell subpopulations, unmix their corresponding observations and resolve the network structure of each subpopulation. Using simulations of the MAPK pathway upon EGF and NGF stimulations we assess the performance of the method. We demonstrate that the presented method can identify better than clustering approaches the number of subpopulations within a mixture of observations, thus resolving correctly the statistical relations between the proteins.
Coupling the unmixing of multiplexed observations with the inference of statistical relations between the measured parameters is essential for the success of both of these processes. Here we present a conceptual and algorithmic solution to achieve such coupling and hence to analyze data obtained from a natural mixture of cell populations. As the technologies and necessity for multiplexed measurements are rising in the systems biology era, this work addresses an important current challenge in the analysis of the derived data.
BMC Systems Biology 06/2015; 9(1):24. DOI:10.1186/s12918-015-0170-2 · 2.44 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: eLife digest
The space between the cells in a multi-cellular organism is filled with a structure called the extracellular matrix. The roles of this matrix vary from tissue to tissue, and include providing the cell with structural support and signalling cues.
Cells use large multi-protein structures along their plasma membrane, called cell-matrix adhesion sites, to attach to the extracellular matrix and sense its properties. These sites are rapidly self-assembled upon local cues and are dynamically maintained by exchanging their components with the cytosol of the cell. Given the large number of different proteins that can be used to build adhesion sites, and the large number of possible interactions between these proteins, a major question is—how are these structures self-assembled both rapidly and correctly?
Most research so far has concentrated on the adhesion sites themselves, as it has generally been assumed that their constituent proteins enter and leave these sites individually. However, using fluorescence techniques, Hoffmann et al. have now shown that these proteins are actually pre-assembled in the cytosol into small ‘building blocks’. These can make the construction of cell-matrix adhesion sites quicker and more accurate by reducing the number of steps in the self-assembly process. In addition, under most circumstances, these building blocks do not change when they are exchanged between the cytosol and the adhesion site: this means that they can be re-used repeatedly and accurately. The work of Hoffmann et al. also raises questions about whether similar principles govern the self-assembly of other large intracellular multi-protein structures.
[Show abstract][Hide abstract] ABSTRACT: Biochemical research has yielded an extensive amount of information about dependencies between protein interactions, as generated by allosteric regulations, steric hindrance and other mechanisms. Collectively, this information is valuable for understanding large intracellular protein networks. However, this information is sparsely distributed among millions of publications and documented as freely styled text meant for manual reading. Here we develop a computational approach for extracting information about interaction dependencies from large numbers of publications. First, keyword-based tokenization reduces full papers to short strings, facilitating an efficient search for patterns that are likely to indicate descriptions of interaction dependencies. Sentences that match such patterns are extracted, thereby reducing the amount of text to be read by human curators. Application of this approach to the integrin adhesome network extracted from 59,933 papers 208 short statements, close to half of which indeed describe interaction dependencies. We visualize the obtained hypernetwork of dependencies and illustrate that these dependencies confine the feasible mechanisms of adhesion sites assembly and generate testable hypotheses about their switchability.
[Show abstract][Hide abstract] ABSTRACT: A convenient way of modelling complex interactions is by employing graphs or networks which correspond to conditional independence structures in an underlying statistical model. One main class of models in this regard are Bayesian networks, which have the drawback of making parametric assump-tions. Bayesian nonparametric mixture models offer a possibility to overcome this limitation, but have hardly been used in combination with networks. This manuscript bridges this gap by introducing nonparametric Bayesian network models. We review (parametric) Bayesian networks, in particular Gaussian Bayesian networks, from a Bayesian perspective as well as nonparametric Bayesian mixture models. Afterwards these two modelling approaches are combined into nonparametric Bayesian networks. The new models are com-pared both to Gaussian Bayesian networks and to mixture models in a simula-tion study, where it turns out that the nonparametric network models perform favorably in non Gaussian situations. The new models are also applied to an example from systems biology, namely finding modules within the MAPK cascade.
[Show abstract][Hide abstract] ABSTRACT: Motivation: Protein interactions are fundamental building blocks of
biochemical reaction systems underlying cellular functions. The complexity and
functionality of such systems emerge not from the protein interactions
themselves but from the dependencies between these interactions. Therefore, a
comprehensive approach for integrating and using information about such
dependencies is required. Results: We present an approach for endowing protein
networks with interaction dependencies using propositional logic, thereby
obtaining protein hypernetworks. First we demonstrate how this framework
straightforwardly improves the prediction of protein complexes. Next we show
that modeling protein perturbations in hypernetworks, rather than in networks,
allows to better infer the functional necessity of proteins for yeast.
Furthermore, hypernetworks improve the prediction of synthetic lethal
interactions in yeast, indicating their capability to capture high-order
functional relations between proteins. Conclusion: Protein hypernetworks are a
consistent formal framework for modeling dependencies between protein
interactions within protein networks. First applications of protein
hypernetworks on the yeast interactome indicate their value for inferring
functional features of complex biochemical systems.
[Show abstract][Hide abstract] ABSTRACT: We extend the in vitro principle of co-immunoprecipitation to quantify dynamic protein interactions in living cells. Using a multiresolution implementation of fluorescence correlation spectroscopy to achieve maximal temporal resolution, we monitored the interactions of endogenous bait proteins, recruited by quantum dots, with fluorescently tagged prey. With this approach, we analyzed the rapid physiological regulation of protein kinase A.
[Show abstract][Hide abstract] ABSTRACT: Although much is known about the molecular components of cellular signaling pathways, very little is known about how these multicomponent biochemical machineries process complex extracellular signals to generate a consolidated cellular response. A newly developed theoretical approach for reverse engineering network structure analyzing how perturbations propagate in a network-can be combined with chemical perturbations and quantitative detection approaches to reveal the causal connections within protein networks in cells. This information indicates the dynamic capabilities of a network and thereby its potential function.
Nature Chemical Biology 12/2008; 4(11):643-7. DOI:10.1038/nchembio1108-643 · 13.00 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Cellular processes occur within dynamic and multi-molecular compartments whose characterization requires analysis at high spatio-temporal resolution. Notable examples for such complexes are cell-matrix adhesion sites, consisting of numerous cytoskeletal and signaling proteins. These adhesions are highly variable in their morphology, dynamics, and apparent function, yet their molecular diversity is poorly defined.
We present here a compositional imaging approach for the analysis and display of multi-component compositions. This methodology is based on microscopy-acquired multicolor data, multi-dimensional clustering of pixels according to their composition similarity and display of the cellular distribution of these composition clusters. We apply this approach for resolving the molecular complexes associated with focal-adhesions, and the time-dependent effects of Rho-kinase inhibition. We show here compositional variations between adhesion sites, as well as ordered variations along the axis of individual focal-adhesions. The multicolor clustering approach also reveals distinct sensitivities of different focal-adhesion-associated complexes to Rho-kinase inhibition.
Multicolor compositional imaging resolves "molecular signatures" characteristic to focal-adhesions and related structures, as well as sub-domains within these adhesion sites. This analysis enhances the spatial information with additional "contents-resolved" dimensions. We propose that compositional imaging can serve as a powerful tool for studying complex multi-molecular assemblies in cells and for mapping their distribution at sub-micron resolution.
PLoS ONE 02/2008; 3(4):e1901. DOI:10.1371/journal.pone.0001901 · 3.23 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The study of normal or malignant haematopoiesis requires the analysis of heterogeneous cell populations using multiple morphological and molecular criteria. Flow cytometry has the capacity to acquire multi-parameter information of large haematopoietic cell populations, utilizing various combinations of >200 molecular markers (clusters of differentiation, CD). However, current flow cytometry analyses are based on serial gating of two-parametric scatter plots--a process that is inherently incapable to discriminate all subgroups of cells in the data. Here we studied the cellular diversity of normal bone marrows (BM) using multi-dimensional cluster analysis of six-parametric flow cytometry data (four CD, forward scatter and side scatter), focusing mainly on the myeloid lineage. Twenty-three subclasses of cells were resolved, many of them inseparable even when examined in all possible two-parametric scatter plots. The multi-dimensional analysis could distinguish the haematopoietic progenitors according to International Society of Haematotherapy and Graft Engineering criteria from other types of immature cells. Based on the defined clusters, we designed a classifier that assigns BM cells in samples to subclasses based on robust six-dimensional position and extended shape. The analysis presented here can manage successfully both the increasing numbers of haematopoietic cellular markers and sample heterogeneity. This should enhance the ability to study normal haematopoiesis, and to identify and monitor haematopoietic disorders.
British Journal of Haematology 05/2005; 129(3):420-31. DOI:10.1111/j.1365-2141.2005.05471.x · 4.71 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Biological image analysis software packages offer tools to analyze microscope images of cells. Some of these tools allow quantitative analysis through interactive processing. High-throughput applications employing microscopy for cell-based assays require analysis of large number of images. We describe here acquisition and analysis of cell images in high throughput automated mode aiming to screen for effects in structure and molecular organization of cellular components recorded by high resolution cell images and in cell motility.
Biomedical Imaging: Nano to Macro, 2004. IEEE International Symposium on; 05/2004
[Show abstract][Hide abstract] ABSTRACT: Cadherins are a family of transmembrane glycoproteins mediating calcium-dependent, homophilic cell-cell adhesion. In addition, these molecules are involved in signaling events, regulating such processes as cell motility, proliferation, and apoptosis. Members of the cadherin subfamily, called either classical or type I cadherins, contain a highly conserved sequence at their homophilic binding site consisting of the three amino acids--histidine-alanine-valine (HAV). Previous studies have shown that peptides containing the HAV motif inhibit cadherin-dependent events such as cell aggregation, compaction, and neurite outgrowth. We report here that a cyclic peptide, N-Ac-CHAVC-NH2 can perturb cadherin-mediated endothelial cell interactions, resulting in a progressive apoptotic cell death. This effect depends on cell density, as it is only observed when dense cultures are treated with the peptide. Adherens junction (AJ)-associated cadherin and catenins are differentially affected by the N-Ac-CHAVC-NH2 treatment, as judged by double immunofluorescence labeling followed by immunofluorescence-ratio imaging. However, cell-cell adhesions are largely retained during the first few hours after addition of the peptide. It was also observed that following treatment, actin filaments partially lose their plasma membrane anchorage at AJs and translocate towards the cell center. Interestingly, addition of basic fibroblast growth factor to confluent, peptide-treated, endothelial cell cultures, completely blocks apoptosis and the inhibitory peptide reduce the phosphorylation of the FGF receptor target protein FRS2, suggesting that the peptide exerts its effect by inhibiting cadherin-mediated activation of fibroblast growth factor receptor signaling. We propose that cadherin-mediated signaling is essential for maintaining viability of confluent endothelial cells, and that its perturbation by N-Ac-CHAVC-NH2 drives these cells to apoptosis.
Experimental Cell Research 05/2004; 294(2):366-78. DOI:10.1016/j.yexcr.2003.11.033 · 3.25 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Heparanase is an endo-beta-D-glucuronidase involved in degradation of heparan sulfate (HS) and extracellular matrix (ECM) of a wide range of cells of vertebrate and invertebrate tissues. The enzymatic activity of heparanase is characterized by specific intrachain cleavage of glycosidic bonds with a hydrolase mechanism. This enzyme facilitates cell invasion and hence plays a role in tumor metastasis, angiogenesis, inflammation, and autoimmunity. Although the expression pattern and molecular properties of heparanase have been characterized, its subcellular localization has not been unequivocally determined. We have previously suggested that heparanase subcellular localization is a major determinant in regulating the enzyme's biological functions. In the present study we examined heparanase localization in three different cell types, utilizing immunofluorescent staining and electron microscopy. Our results indicate that heparanase is localized primarily within lysosomes and the Golgi apparatus. A construct composed of heparanase cDNA fused to green fluorescent protein, utilized in order to visualize the enzyme within living cells, confirmed its localization in acidic vesicles. We suggest that following synthesis, heparanase is transported into the Golgi apparatus and subsequently accumulates in a stable form within the lysosomes, where it functions in HS turnover. The lysosomal compartment may also serve as a site for heparanase confinement within the cells, limiting its secretion and uncontrolled extracellular activities associated with tumor metastasis and angiogenesis.
Experimental Cell Research 12/2002; 281(1):50-62. DOI:10.1016/S0014-4827(02)95651-X · 3.25 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Heparanase is a heparan-sulfate-degrading endoglycosidase that has important roles in various biological processes, including angiogenesis, wound healing and metastatsis. Human heparanase is synthesized as a 65 kDa latent precursor, which is proteolytically processed into a highly active 50 kDa form. Extracellular heparanase is found in various tissues and is utilized by both normal cells and metastatic cancer cells to degrade heparan sulfate moieties in basement membranes and extracellular matrices. This study characterizes the processing and trafficking events associated with cellular activation of extracellular heparanase. We show that primary human fibroblasts are capable of binding and converting the 65 kDa heparanase precursor into its highly active 50 kDa form, concomitantly with its cytoplasmic accumulation. Heparanase uptake depends on the actin cytoskeleton integrity, resulting in a prolonged storage of the enzyme, mainly in endosomal structures. Heparanase endocytosis and its proteolytic activation are independent processes, indicating that heparanase cleavage is a cell surface event. Heparin completely inhibits heparanase endocytosis but only partially inhibits its association with the cells, suggesting that cell surface heparan sulfate moieties play a specific role in its endocytosis. Cellular binding and uptake of extracellular heparanase control its activation, clearance rate and storage within the cells.
[Show abstract][Hide abstract] ABSTRACT: Currently >50 proteins have been reported to be associated with focal contacts and related ECM adhesions. Most of these contain multiple domains through which they can interact with different molecular partners, potentially forming a dense and heterogeneous protein network at the cytoplasmic faces of the adhesion site. The molecular and structural diversity of this 'submembrane plaque' is regulated by a wide variety of mechanisms, including competition between different partner proteins for the same binding sites, interactions triggered or suppressed by tyrosine phosphorylation, and conformational changes in component proteins, which can affect their reactivity. Indeed, integrin-mediated adhesions can undergo dynamic changes in structure and molecular properties from dot-like focal complexes to stress-fiber-associated focal contacts, which can further 'mature' to form fibronectin-bound fibrillar adhesions. These changes are driven by mechanical force generated by the actin- and myosin-containing contractile machinery of the cells, or by external forces applied to the cells, and regulated by matrix rigidity.
[Show abstract][Hide abstract] ABSTRACT: Modern light microscopy has become a most powerful analytical tool for studying molecular processes in live cells. Recent advances in sample preparation, microscope design and image processing allow the generation of "multidimensional" data, simultaneously reporting the three-dimensional distribution and concentrations of several different molecules within cells and tissues at multiple time points with sub-micron spatial resolution and sub-second temporal resolution. Thus, molecular interactions and processes that were approached by biochemical analyses in vitro can now be directly monitored in live cells. Here, we address different aspects of multidimensional microscopy and, in particular, image quantification and the characterization of molecular dynamics, as applied to the study of cell adhesion.
[Show abstract][Hide abstract] ABSTRACT: The transition of cell-matrix adhesions from the initial punctate focal complexes into the mature elongated form, known as focal contacts, requires GTPase Rho activity. In particular, activation of myosin II-driven contractility by a Rho target known as Rho-associated kinase (ROCK) was shown to be essential for focal contact formation. To dissect the mechanism of Rho-dependent induction of focal contacts and to elucidate the role of cell contractility, we applied mechanical force to vinculin-containing dot-like adhesions at the cell edge using a micropipette. Local centripetal pulling led to local assembly and elongation of these structures and to their development into streak-like focal contacts, as revealed by the dynamics of green fluorescent protein-tagged vinculin or paxillin and interference reflection microscopy. Inhibition of Rho activity by C3 transferase suppressed this force-induced focal contact formation. However, constitutively active mutants of another Rho target, the formin homology protein mDia1 (Watanabe, N., T. Kato, A. Fujita, T. Ishizaki, and S. Narumiya. 1999. Nat. Cell Biol. 1:136-143), were sufficient to restore force-induced focal contact formation in C3 transferase-treated cells. Force-induced formation of the focal contacts still occurred in cells subjected to myosin II and ROCK inhibition. Thus, as long as mDia1 is active, external tension force bypasses the requirement for ROCK-mediated myosin II contractility in the induction of focal contacts. Our experiments show that integrin-containing focal complexes behave as individual mechanosensors exhibiting directional assembly in response to local force.
The Journal of Cell Biology 07/2001; 153(6):1175-86. DOI:10.1083/jcb.153.6.1175 · 9.83 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Activation of tyrosine kinases during integrin-mediated cell-matrix adhesion is involved both in the regulation of focal contact assembly and in the initiation of signaling processes at the cell-matrix adhesive interface. In order to determine the role of pp60(c-src) and related kinases in these processes, we have compared the dynamic reorganization of phosphotyrosine, vinculin, focal adhesion kinase and tensin in cells with altered expression of Src-family kinases. Both null cells for pp60(c-src) and triple knockout cells for pp60(c-src), pp59(fyn), and pp62(c-yes) exhibited decreased phosphotyrosine levels in focal contacts when compared with wild-type cells. pp60(c-src)-null cells also exhibited faster assembly of cell-matrix adhesions and a more exuberant recruitment of FAK to these sites. Tensin, which normally segregates into fibrillar adhesions was localized in large focal contacts in the two mutant cell lines, suggesting involvement of pp60(c-src) in the segregation of focal contacts and fibrillar adhesions. Moreover, treatment of wild-type cells with tyrphostin AG1007, which inhibits both pp60(c-src) and FAK activity, induced accumulation of tensin in peripheral focal adhesions. These findings demonstrate that Src family kinases, and pp60(c-src) in particular, have a central role in regulating protein dynamics at cell-matrix interfaces, both during early stages of interaction and in mature focal contacts.
[Show abstract][Hide abstract] ABSTRACT: The transition of cell–matrix adhesions from the initial punctate focal complexes into the mature elongated form, known as focal contacts, requires GTPase Rho activity. In particular, activation of myosin II–driven contractility by a Rho target known as Rho-associated kinase (ROCK) was shown to be essential for focal contact formation. To dissect the mechanism of Rho-dependent induction of focal contacts and to elucidate the role of cell contractility, we applied mechanical force to vinculin-containing dot-like adhesions at the cell edge using a micropipette. Local centripetal pulling led to local assembly and elongation of these structures and to their development into streak-like focal contacts, as revealed by the dynamics of green fluorescent protein–tagged vinculin or paxillin and interference reflection microscopy. Inhibition of Rho activity by C3 transferase suppressed this force-induced focal contact formation. However, constitutively active mutants of another Rho target, the formin homology protein mDia1 (Watanabe, N., T. Kato, A. Fujita, T. Ishizaki, and S. Narumiya. 1999. Nat. Cell Biol. 1:136–143), were sufficient to restore force-induced focal contact formation in C3 transferase-treated cells. Force-induced formation of the focal contacts still occurred in cells subjected to myosin II and ROCK inhibition. Thus, as long as mDia1 is active, external tension force bypasses the requirement for ROCK-mediated myosin II contractility in the induction of focal contacts. Our experiments show that integrin-containing focal complexes behave as individual mechanosensors exhibiting directional assembly in response to local force.
The Journal of Cell Biology 06/2001; 153(6):1175-1186. · 9.83 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: This study establishes that the physical state of the extracellular matrix can regulate integrin-mediated cytoskeletal assembly and tyrosine phosphorylation to generate two distinct types of cell-matrix adhesions. In primary fibroblasts, alpha(5)beta(1) integrin associates mainly with fibronectin fibrils and forms adhesions structurally distinct from focal contacts, independent of actomyosin-mediated cell contractility. These "fibrillar adhesions" are enriched in tensin, but contain low levels of the typical focal contact components paxillin, vinculin, and tyrosine-phosphorylated proteins. However, when the fibronectin is covalently linked to the substrate, alpha(5)beta(1) integrin forms highly tyrosine-phosphorylated, "classical" focal contacts containing high levels of paxillin and vinculin. These experiments indicate that the physical state of the matrix, not just its molecular composition, is a critical factor in defining cytoskeletal organization and phosphorylation at adhesion sites. We propose that molecular organization of adhesion sites is controlled by at least two mechanisms: 1) specific integrins associate with their ligands in transmembrane complexes with appropriate cytoplasmic anchor proteins (e.g., fibronectin-alpha(5)beta(1) integrin-tensin complexes), and 2) physical properties (e.g., rigidity) of the extracellular matrix regulate local tension at adhesion sites and activate local tyrosine phosphorylation, recruiting a variety of plaque molecules to these sites. These mechanisms generate structurally and functionally distinct types of matrix adhesions in fibroblasts.
Molecular Biology of the Cell 04/2000; 11(3):1047-60. DOI:10.1091/mbc.11.3.1047 · 4.47 Impact Factor