Stephanie I. Fraley

A distinctive role for focal adhesion proteins in three- dimensional cell motility

Focal adhesions are large multi-protein assemblies that form at the basal surface of cells on planar dishes, and that mediate cell signalling, force transduction and adhesion to the substratum. Although much is known about focal adhesion components in two-dimensional (2D) systems, their role in migrating cells in a more physiological three-dimensional (3D) matrix is largely unknown. Live-cell microscopy shows that for cells fully embedded in a 3D matrix, focal adhesion proteins, including vinculin, paxillin, talin, α-actinin, zyxin, VASP, FAK and p130Cas, do not form aggregates but are diffusely distributed throughout the cytoplasm. Despite the absence of detectable focal adhesions, focal adhesion proteins still modulate cell motility, but in a manner distinct from cells on planar substrates. Rather, focal adhesion proteins in matrix-embedded cells regulate cell speed and persistence by affecting protrusion activity and matrix deformation, two processes that have no direct role in controlling 2D cell speed. This study shows that membrane protrusions constitute a critical motility/matrix-traction module that drives cell motility in a 3D matrix.

Hypoxia-inducible factor 1 is a master regulator of breast cancer metastatic niche formation

Primary tumors facilitate metastasis by directing bone marrow-derived cells (BMDCs) to colonize the lungs before the arrival of cancer cells. Here, we demonstrate that hypoxia-inducible factor 1 (HIF-1) is a critical regulator of breast cancer metastatic niche formation through induction of multiple members of the lysyl oxidase (LOX) family, including LOX, LOX-like 2, and LOX-like 4, which catalyze collagen cross-linking in the lungs before BMDC recruitment. Only a subset of LOX family members was expressed in any individual breast cancer, but HIF-1 was required for expression in each case. Knockdown of HIF-1 or hypoxia-induced LOX family members reduced collagen cross-linking, CD11b+ BMDC recruitment, and metastasis formation in the lungs of mice after orthotopic transplantation of human breast cancer cells. Metastatic niche formation is an HIF-1–dependent event during breast cancer progression.

Three-dimensional matrix fiber alignment modulates cell migration and MT1-MMP utility by spatially and temporally directing protrusions.

Multiple attributes of the three-dimensional (3D) extracellular matrix (ECM) have been independently implicated as regulators of cell motility, including pore size, crosslink density, structural organization, and stiffness. However, these parameters cannot be independently varied within a complex 3D ECM protein network. We present an integrated, quantitative study of these parameters across a broad range of complex matrix configurations using self-assembling 3D collagen and show how each parameter relates to the others and to cell motility. Increasing collagen density resulted in a decrease and then an increase in both pore size and fiber alignment, which both correlated significantly with cell motility but not bulk matrix stiffness within the range tested. However, using the crosslinking enzyme Transglutaminase II to alter microstructure independently of density revealed that motility is most significantly predicted by fiber alignment. Cellular protrusion rate, protrusion orientation, speed of migration, and invasion distance showed coupled biphasic responses to increasing collagen density not predicted by 2D models or by stiffness, but instead by fiber alignment. The requirement of matrix metalloproteinase (MMP) activity was also observed to depend on microstructure, and a threshold of MMP utility was identified. Our results suggest that fiber topography guides protrusions and thereby MMP activity and motility.

Dimensional and temporal controls of three-dimensional cell migration by zyxin and binding partners

Spontaneous molecular oscillations are ubiquitous in biology. But to our knowledge, periodic cell migratory patterns have not been observed. Here we report the highly regular, periodic migration of cells along rectilinear tracks generated inside three-dimensional matrices, with each excursion encompassing several cell lengths, a phenotype that does not occur on conventional substrates. Short hairpin RNA depletion shows that these one-dimensional oscillations are uniquely controlled by zyxin and binding partners α-actinin and p130Cas, but not vasodilator-stimulated phosphoprotein and cysteine-rich protein 1. Oscillations are recapitulated for cells migrating along one-dimensional micropatterns, but not on two-dimensional compliant substrates. These results indicate that although two-dimensional motility can be well described by speed and persistence, three-dimensional motility requires two additional parameters, the dimensionality of the cell paths in the matrix and the temporal control of cell movements along these paths. These results also suggest that the zyxin/α-actinin/p130Cas module may ensure that motile cells in a three-dimensional matrix explore the largest space possible in minimum time.

Reply: reducing background fluorescence reveals adhesions in 3D matrices

1Departments of Chemical and Biomolecular Engineering, The Johns Hopkins University, Baltimore, MD 21218, USA. 2Johns Hopkins Physical Sciences in Oncology Centre, Johns Hopkins University, Baltimore, MD 21218, USA. 3Departments of Medicine and Cell Biology and Physiology, Washington University School of Medicine, St. Louis, MO 63110, USA. 4BRIGHT Institute, Washington University School of Medicine, St. Louis, MO 63110, USA. 5Correspondence should be addressed to D.W. or G.D.L. (wirtz@jhu.edu or glongmor@dom.wustl.edu) Reply: reducing background fluorescence reveals adhesions in 3D matrices

NAC1 Is an Actin-Binding Protein That Is Essential for Effective Cytokinesis in Cancer Cells

NAC1 is a transcriptional corepressor protein that is essential to sustain cancer cell proliferation and migration. However, the underlying molecular mechanisms of NAC1 function in cancer cells remain unknown. In this study, we show that NAC1 functions as an actin monomer-binding protein. The conserved BTB protein interaction domain in NAC1 is the minimal region for actin binding. Disrupting NAC1 complex function by dominant-negative or siRNA strategies reduced cell retraction and abscission during late-stage cytokinesis, causing multinucleation in cancer cells. In Nac1-deficient murine fibroblasts, restoring NAC1 expression was sufficient to partially avert multinucleation. We found that siRNA-mediated silencing of the actin-binding protein profilin-1 in cancer cells caused a similar multinucleation phenotype and that NAC1 modulated the binding of actin to profillin-1. Taken together, our results indicate that the NAC1/actin/profilin-1 complex is crucial for cancer cell cytokinesis, with a variety of important biologic and clinical implications.

Trainable high resolution melt curve machine learning classifier for large-scale reliable genotyping of sequence variants

High resolution melt (HRM) is gaining considerable popularity as a simple and robust method for genotyping sequence variants. However, accurate genotyping of an unknown sample for which a large number of possible variants may exist will require an automated HRM curve identification method capable of comparing unknowns against a large cohort of known sequence variants. Herein, we describe a new method for automated HRM curve classification based on machine learning methods and learned tolerance for reaction condition deviations. We tested this method in silico through multiple cross-validations using curves generated from 9 different simulated experimental conditions to classify 92 known serotypes of Streptococcus pneumoniae and demonstrated over 99% accuracy with 8 training curves per serotype. In vitro verification of the algorithm was tested using sequence variants of a cancer-related gene and demonstrated 100% accuracy with 3 training curves per sequence variant. The machine learning algorithm enabled reliable, scalable, and automated HRM genotyping analysis with broad potential clinical and epidemiological applications.

3D collagen architecture induces a conserved migratory and transcriptional response linked to vasculogenic mimicry

The topographical organization of collagen within the tumor microenvironment has been implicated in modulating cancer cell migration and independently predicts progression to metastasis. Here, we show that collagen matrices with small pores and short fibers, but not Matrigel, trigger a conserved transcriptional response and subsequent motility switch in cancer cells resulting in the formation of multicellular network structures. The response is not mediated by hypoxia, matrix stiffness, or bulk matrix density, but rather by matrix architecture-induced β1-integrin upregulation. The transcriptional module associated with network formation is enriched for migration and vasculogenesis-associated genes that predict survival in patient data across nine distinct tumor types. Evidence of this gene module at the protein level is found in patient tumor slices displaying a vasculogenic mimicry (VM) phenotype. Our findings link a collagen-induced migration program to VM and suggest that this process may be broadly relevant to metastatic progression in solid human cancers.Extracellular matrix plays a central role in driving cancer development. Here the authors using an in vitro approach show that confining collagen architectures induce fast and persistent cell migration and the formation of multicellular network structures linked to vascular mimicry observed in tumours from patients.

Nested Machine Learning Facilitates Increased Sequence Content for Large-Scale Automated High Resolution Melt Genotyping

High Resolution Melt (HRM) is a versatile and rapid post-PCR DNA analysis technique primarily used to differentiate sequence variants among only a few short amplicons. We recently developed a one-vs-one support vector machine algorithm (OVO SVM) that enables the use of HRM for identifying numerous short amplicon sequences automatically and reliably. Herein, we set out to maximize the discriminating power of HRM + SVM for a single genetic locus by testing longer amplicons harboring significantly more sequence information. Using universal primers that amplify the hypervariable bacterial 16 S rRNA gene as a model system, we found that long amplicons yield more complex HRM curve shapes. We developed a novel nested OVO SVM approach to take advantage of this feature and achieved 100% accuracy in the identification of 37 clinically relevant bacteria in Leave-One-Out-Cross-Validation. A subset of organisms were independently tested. Those from pure culture were identified with high accuracy, while those tested directly from clinical blood bottles displayed more technical variability and reduced accuracy. Our findings demonstrate that long sequences can be accurately and automatically profiled by HRM with a novel nested SVM approach and suggest that clinical sample testing is feasible with further optimization.

Massively parallel digital high resolution melt for rapid and absolutely quantitative sequence profiling

In clinical diagnostics and pathogen detection, profiling of complex samples for low-level genotypes represents a significant challenge. Advances in speed, sensitivity, and extent of multiplexing of molecular pathogen detection assays are needed to improve patient care. We report the development of an integrated platform enabling the identification of bacterial pathogen DNA sequences in complex samples in less than four hours. The system incorporates a microfluidic chip and instrumentation to accomplish universal PCR amplification, High Resolution Melting (HRM), and machine learning within 20,000 picoliter scale reactions, simultaneously. Clinically relevant concentrations of bacterial DNA molecules are separated by digitization across 20,000 reactions and amplified with universal primers targeting the bacterial 16S gene. Amplification is followed by HRM sequence fingerprinting in all reactions, simultaneously. The resulting bacteria-specific melt curves are identified by Support Vector Machine learning, and individual pathogen loads are quantified. The platform reduces reaction volumes by 99.995% and achieves a greater than 200-fold increase in dynamic range of detection compared to traditional PCR HRM approaches. Type I and II error rates are reduced by 99% and 100% respectively, compared to intercalating dye-based digital PCR (dPCR) methods. This technology could impact a number of quantitative profiling applications, especially infectious disease diagnostics.