(I-346) Multiplex Microfluidic-Based Investigation of Haptotaxis on Fibronectin1 to Model Skull Bone Morphogenesis

Introduction:: Haptotaxis is cell migration directed by concentration gradients of insoluble ligands and dependent on receptor-ligand affinity. Haptotaxis is a critical driver of morphogenesis, which tissue engineers exploit for spatial control of cellular events. Selection of the appropriate ligand and gradient for different cells is time-consuming and obscured by chemotaxis from cellular crosstalk. Here, we describe a microfluidic-based haptotaxis array to evaluate multiple ligand gradients in parallel (up to 32) with live imaging. Intermediary barriers segment each concentration gradient into three regions to evaluate low-density cell culture for haptotaxis. Skull morphogenesis is a complex dynamic process involving two different germ layers and coordinated growth of individual bones1,2. While osteochondral differentiation from the mesenchyme is known to involve spatiotemporal presentation of factors, mechanisms of directional growth toward the apex is not completely understood. Preliminary evidence in vivo suggests that fibronectin expression is graded in

the cranial mesenchyme along the direction of frontal bone primordia growth. Fibronectin-mediated haptotaxis has not been tested for calvarial osteoblast migration for rapid intramembranous bone growth. We hypothesize that primary calvarial osteoblasts undergo haptotaxis along a defined range of the fibronectin adhesion gradient. Using computer modeling and fluorescently labeling, we evaluated fibronectin gradients for patterns that match histological analysis of the developing embryonic mouse skull. Primary calvarial osteoblasts were plated in the chip at various positions along the selected gradient of adsorbed fibronectin. We observe that haptotaxis occurs solely over intermediate fibronectin concentrations with stationary or unbiased cell migration in other regions.




Materials and Methods:: We make use of an established automated microfluidic chip3 with live imaging to investigate fibronectin-mediated haptotaxis. The PDMS microfluidic chip consists of 32 independent devices (2.75mm x 500um) or repeat units for cell culture printed on a microscope slide. Pressure-regulated gates4 that are liquid impermeable segment each unit into quadrants (Figure 1A) generating 128 individual cell culture chambers. By chip design, all areas are individually addressable allowing spatial specificity of cell plating. We design the haptotaxis parameters with MATLAB simulations of the diffusion kinetics of fibronectin within the defined chip geometry and ranges of starting concentration, resupply frequency, and diffusion duration. Empirical testing of fluorescent labels (dextran, fibronectin) in the chip followed our MATLAB simulations. We maintain the starting fibronectin concentration by periodic resupply in one unit (gates closed) and during intervening periods allow for passive diffusion through the remainder of the unit (gates open). We evaluated gradients of adsorbed fibronectin with antibody and histological staining. Gradients are selected that best matched immunohistological fibronectin staining of the developing mouse skull. Primary osteoblasts were evaluated for haptotaxis (fibronectin gradient, uniform fibronectin, and blank control) in the chip over 24 hours. The migratory tracks for individual cells were collected by time-lapse imaging and analyzed using ImageJ tracking plugins. We evaluated multiple metrics of migration including forward migration index (FMI). One-way ANOVA with Tukey post-hoc tests was performed. Cell morphology and cytoskeleton organization of filamentous actin was labeled by phalloidin staining in the chip and imaged on a Leica SP8 confocal microscope.




Results, Conclusions, and Discussions:: The gradient that demonstrated the most similar gradation of fibronectin to the in vivo expression was selected (100 ug/ml; resupply every 30 mins; 24 hours diffusion). The line scan analysis of fluorescent dextran and fibronectin images showed a stable fibronectin gradient is generated by diffusion. We plated primary osteoblasts at low density and segmented the cell culture chamber into three regions of low (1°), intermediate (2°), and high (3°) ranges of fibronectin. The overlapping migratory tracks and phase-contrast timelapse images were completed and reviewed by 3 researchers with at least 150 tracks per condition (Figure 1B). We identified that calvarial osteoblasts undergo haptotaxis and a significant increase in Forward Migration Index (FMI) in the direction of higher Fibronectin1 over 18-24 hours within an intermediary region of the grade (Figure 1C). The velocity of haptotaxic calvarial osteoblasts was about 20 um/hr, which is consistent with previous reports for fibroblasts5. In-chip immunocytochemistry was performed (phalloidin-green; dapi-blue) before detaching the slide for confocal microscopy. The haptotaxic calvarial osteoblasts exhibited organized cortical actin, elongated morphology with multiple protrusions in the direction of graded fibronectin, indicating directional movement (Figure 1B). Cells plated higher in the gradient failed to migrate with a low FMI. The cells were large in morphology with enrichment of cortical filamentous actin, suggesting extensive adhesion. On uniform coating of low Fibronectin1 and in the absence of Fibronectin1, there was random directional migration. Consistently, the morphology of cells showed far less cortical actin and rounded morphology. Collectively the initial data supports that the microfluidic chip provides a platform to investigate the haptotaxis of extracellular matrix gradient. Future studies will look into how inhibition of lamellipodia actin nucleation or β-1 integrin signaling affects the calvarial osteoblast cell migration by haptotaxis along a Fibronectin1 gradient.


Acknowledgements (Optional): : We thank CWRU core facilities: Electronic Design Center (Laurie Dudik), Materials for Opto/Electronics Research and Education Center (Ina Martin), and Microfabrication Laboratory (Nichole Hoven) for assistance with microfabrication. This work was supported by the following grants: NIH R21DE029348 and R56DE030206.

References (Optional): : [1]. Flaherty, Kevin, Nandini Singh, and Joan T. Richtsmeier. “Understanding Craniosynostosis as a Growth Disorder.” WIREs Developmental Biology 5, no. 4 (2016): 429–59.

[2]. Ferguson, James W., and Radhika P. Atit. “A Tale of Two Cities: The Genetic Mechanisms Governing Calvarial Bone Development.” Genesis (New York, N.Y.: 2000) 57, no. 1 (January 2019): e23248.

[3]. Watson, Craig, Chao Liu, Ali Ansari, Helen C. Miranda, Rodrigo A. Somoza, and Samuel E. Senyo. “Multiplexed Microfluidic Chip for Cell Co-Culture.” Analyst 147, no. 23 (November 21, 2022): 5409–18.

[4]. Watson, Craig, and Samuel Senyo. “All-in-One Automated Microfluidics Control System.” HardwareX 5 (April 1, 2019): e00063.

[5]. King, Samantha J., Sreeja B. Asokan, Elizabeth M. Haynes, Seth P. Zimmerman, Jeremy D. Rotty, James G. Alb, Alicia Tagliatela, et al. “Lamellipodia Are Crucial for Haptotactic Sensing and Response.” Journal of Cell Science 129, no. 12 (June 15, 2016): 2329–42.