Design and characterization of materials with microphase-separated surface patterns for screening osteoblast response to adhesion

Combinatorial techniques have changed the paradigm of materials research by allowing efficient screening of complex materials problems with large, multidimensional parameter spaces. The focus of this thesis is to demonstrate combinatorial methods (CM) and high-throughput methods (HTM) applied to biomaterials design, characterization, and screening. In particular, this work focuses on screening the effects of biomaterial surface features on adherent bone cell cultures. Polymeric biomaterials were prepared on two-dimensional combinatorial libraries that systematically varied the size and shape of chemically-distinct microstructural patterns. These libraries were generated from blends of biodegradable polyurethanes and polyesters prepared with thickness, composition and temperature gradient techniques. Characterization and screening were performed with high-throughput optical and fluorescence microscopy. A unique advance of this work is the application of data mining techniques to identify the controlling structural features that affect cell behavior from among the myriad variety of metrics from the microscope images.;Libraries were designed to exhibit chemically-distinct cell-adhesive versus non-adhesive microstructural domains that improve library performance compared to previous implementations that had employed only modest chemical differences. Improving adhesive contrast should minimize combination of effects of chemistry and physical structure, making data interpretation simpler. To accomplish this, a method of blending and crosslinking cell-non-adhesive poly(ethylene glycol) (PEG) with cell-adhesive poly(·-caprolactone) (PCL) was developed. The behavior of MC3T3-E1 osteoblast cells cultured on the PCL-PEG libraries were observed, equivalent to thousands of distinct chemistries and microstructures.;Cell spreading area, shape, and density upon adhesion on surface patterns are observed in this study. Characterization of the surface library and screening of surface physical properties via HTM and PCA show that cell density is sensitive to the physical distribution, shape, solidity, and orientation of the PCL and PEG domains. Correlation is shown between surface pattern descriptors and the subsequent cellular adhesion responses. Certain spacing and shapes in surface pattern are preferred to others for distinct cellular states; circular pattern favors apoptotic cells, while elongated patterns favor viable cells---for both cases, cells preferred anchoring themselves to surface patterns. However, the effect of surface pattern's solidity and area did not show any conclusive trend in this dataset. This might be due to the existence of correlation between solidity and eccentricity as described in Chapter 1. Further improvement in the surface pattern library generation is necessary for future studies.;The results from this study demonstrated the potentials of CM/HTS to be applied to exploratory studies involving complex systems in life sciences. This study accomplishes the goal to demonstrate the efficient screening and exploration of vast and complex dataset, extracting important and meaningful information to narrow down the future path of study in this field.;Further study aimed to tuning cellular responses via signals from surface cues will be necessary to examine the causal relationships beyond the observed correlations shown in this exploratory study. It is recommended for further studies to narrow down the range for surface patterning around each of the three 'activation' ranges found in this study: apoptotic, viable, and one unknown state to be studied further. Different cellular-function staining methods will be necessary to be used in cellular imaging techniques in order to explore this unknown state further.