Label-free mapping of microstructural organisation in self-aligning cellular collagen hydrogels using image correlation spectroscopy

Abstract

Hydrogels have emerged as promising biomaterials for regenerative medicine. Despite major advances, tissue engineers have faced challenges in studying the complex dynamics of cell-mediated hydrogel remodelling. Second harmonic generation (SHG) microscopy has been a pivotal tool for non-invasive visualization of collagen type I hydrogels. By taking into account the typical polarization SHG effect, we recently proposed an alternative image correlation spectroscopy (ICS) model to quantify characteristics of randomly oriented collagen fibrils. However, fibril alignment is an important feature in many tissues that needs to be monitored for effective assembly of anisotropic tissue constructs. Here we extended our previous approach to include the orientation distribution of fibrils in cellular hydrogels and show the power of this model in two biologically relevant applications. Using a collagen hydrogel contraction assay, we were able to capture cell-induced hydrogel modifications at the microscopic scale and link these to changes in overall gel dimensions over time. After 24h, the collagen density was about 3 times higher than the initial density, which was of the same order as the decrease in hydrogel area. We also showed that the orientation parameters recovered from our automated ICS model match values obtained from manual measurements. Furthermore, regions axial to cellular processes aligned at least 1.5 times faster compared with adjacent zones. Being able to capture minor temporal and spatial changes in hydrogel density and collagen fibril orientation, we demonstrated the sensitivity of this extended ICS model to deconstruct a complex environment and support its potential for tissue engineering research. Statement of significanceIt is generally accepted that looking beyond bulk hydrogel composition is key in understanding the mechanisms that influence the mechanical and biological properties of artificial tissues. In this manuscript, we performed label-free non-invasive imaging and extended a robust automated analysis method to characterize the microstructural organisation of cellular hydrogel systems. We underpin the sensitivity of this technique by capturing minor changes in collagen density and fibril orientation in biologically relevant systems over time. Therefore, we believe that this method is applicable in fundamental cell-matrix research and has high-throughput potential in screening arrays of hydrogel scaffolds, making it an interesting tool for future tissue engineering research.