Number and organization of focal adhesion clustering determine mitotic dynamics and errors in fibrous microenvironments

Introduction:: Eukaryotic cell division occurs through spatial segregation of chromosomes via a tightly controlled biological assembly also referred to as the mitotic spindle. Spindle positioning determines the orientation of the cell division axis and thereby is crucial for tissue morphogenesis. Abnormalities in spindle positioning and assembly are often implicated in various developmental disorders and cancer progression. Mitotic spindle positioning is governed by external biophysical cues driven by actin retraction fibers that stabilize the rounded mitotic cell body by connecting them to the adhesion patterns. However, how the retraction fibers stabilize the mitotic bodies in fibrous extracellular matrix (ECM) and affect the regulation of mitotic dynamics remains partially described.

Fibrous ECM in vivo consists of nanoscale fibrils and bundled fibers, organized in various architectures ranging from aligned to crossing arrangement of fibers. Cells within loose connective tissues and interstitial fiber networks often interact with only a few fibers, which can dictate cell migration and cytoskeletal force generation. How mitosis proceeds and accrues errors in these isolated interactions is not clear.

Here, using suspended fibers generated by the non-electrospinning STEP (Spinneret Tunable Engineered Parameters) fiber manufacturing technique, we recover in vivo focal adhesion organization in diverse cell shapes. By systematically varying the number of adhesion patterns, we interrogate the arrangement of actin retraction fibers in regulating the cell division dynamics and the mitotic spindle errors. A theoretical and stochastic Monte Carlo-based computational approach accounting for dynamics of centrosome, chromosome, and adhesion patterns explains the observed outcomes and errors.


Materials and Methods:: Suspended nanofiber networks were generated from polystyrene solutions in xylene using our previously reported non-electrospinning STEP technique. Horizontal arrays of 250-nm fibers with interfiber spacings between 3 and 25 μm were deposited on large-diameter (~2 μm) vertical support fibers placed ~350 μm apart. Polystyrene nanofibers were coated with 4 µg/mL fibronectin to mediate cell attachment. Mitotic dynamics were tracked by performing timelapse imaging on HeLa cells expressing histone H2B GFP. Cell divisions were synchronized through a 20 h treatment of the Cdk1 inhibitor RO-3306, followed by drug washout. Cells in metaphase, found after ~ 1h from drug washout, were fixed, permeabilized, blocked and fluorescently co-stained for microtubules (tubulin), kinetochores (Hec1), actin cytoskeleton and DNA. Confocal microscopy was performed using a laser scanning confocal microscope (LSM 880, Carl Zeiss Inc.) with optimal imaging settings. Cells stained for microtubules and DNA were visually inspected under a 40× objective to categorize as bipolar, monopolar, or multipolar based on the number of spindle poles. At least 300 cells over 3 to 4 independent experiments were considered for each category, and the average proportion of observed bipolar, monopolar, and multipolar cells is reported. Statistical analysis was performed in GraphPad Prism software. Statistical comparisons among multiple groups were performed using one-way ANOVA along with Tukey’s test. By utilizing a Monte Carlo simulation within our computational model, we employed stochastic energy minimization and computational screening of forces between CSs, CHs, and cell membranes to determine the mechanical conditions for the formation of mono-, bi-, and multipolar spindles.


Results, Conclusions, and Discussions:: Here, we used suspended nanofiber networks in both aligned and crosshatch configurations. Within aligned fiber networks, we achieve three elongated cell shapes: i) one fiber-elongated (1F-elongated) shaped cells attached to single fibers and having two major focal adhesion (FA) clusters, ii) two fiber-elongated (2F-elongated) shaped cells attached to two fibers and having four FA clusters, and iii) multifiber-elongated (MF-elongated) shaped cells attached to ≥ three fibers with multiple aligned FA clusters (Figure 1A). Crosshatch geometries induced symmetric kite shapes (aspect ratio ~1), and 4 FA clusters. Confocal microscopy of actin-stained cells demonstrated increasing coverage of retraction fibers (RFs) with increased FA clustering (Figure 1B, 1C, 1D). Increased RF coverage (RFC, Figure 1D) directly affected the rounded cell body movements with 1F-elongated and 2F-kite categories displaying significantly higher movements as compared to flat 2D and multifiber categories where the cell body was essentially locked in place (minimal movement) (Figure 1E). Additionally, metaphase plate (MP) stability and inter-kinetochore stretching during metaphase were also influenced by RF coverage, with 1F-elongated cells demonstrating the lowest MP stability and inter-kinetochore separation distance. Such observations can explain why 1F-elongated cells take significantly longer (68.4±3 minutes) times to complete mitosis (NEBD-telophase) than other cell shapes (50-55 minutes, Figure 1F). Reduced orientational stability on single fibers correlates with increased mitotic spindle defects (Figure 1G) and increased occurrence of monopolar spindle defects, while higher RF coverage in flat 2D and multifiber categories, resulted in predominantly multipolar defects (Figure 1G). The stochastic Monte-Carlo simulations with matching RF coverage from experiments (Figure 1H), successfully recapitulate the observed propensity of monopolar and multipolar defects (Figure 1I). Overall, using a model system of extracellular matrix-mimicking suspended fibers, we expose the role of fiber geometry towards mitotic stability. We describe how the number density of adhesion-originated retraction fibers regulates the orientational stability of rounded cells and mitotic spindle and induces a dramatic switch in the character of the mitotic defects. Our study unravels a previously unknown link between ECM fiber geometry and mitotic fidelity that may occur in vivo.


Acknowledgements (Optional): :

References (Optional): :