Type I Collagen Self-Assembly: The Roles of Substrate and Concentration

• Published in: Langmuir 2013-02, Vol.29 (7), p.2330-2338
• Main Authors: Fang, Ming, Goldstein, Elizabeth L, Matich, Eryn K, Orr, Bradford G, Banaszak Holl, Mark M
• Format: Article
• Language: English
• Subjects: Collagen Type I - chemistry
• ISSN: 0743-7463; 1520-5827
• DOI: 10.1021/la3048104

Collagen molecules, self-assembled into macroscopic hierarchical tissue networks, are the main organic building block of many biological tissues. A particularly common and important form of this self-assembly consists of type I collagen fibrils, which exhibit a nanoscopic signature, D-periodic gap/overlap spacing, with a distribution of values centered at approximately 67 nm. In order to better understand the relationship between type I collagen self-assembly and D-spacing distribution, we investigated surface-mediated collagen self-assembly as a function of substrate and incubation concentration. Collagen fibril assembly on phlogopite and muscovite mica as well as fibrillar gel coextrusion in glass capillary tubes all exhibited D-spacing distributions similar to those commonly observed in biological tissues. The observation of D-spacing distribution by self-assembly of type I collagen alone is significant as it eliminates the necessity to invoke other preassembly or postassembly hypotheses, such as variation in the content of collagen types, enzymatic cross-linking, or other post-translational modifications, as mechanistic origins of D-spacing distribution. The D-spacing distribution on phlogopite mica is independent of type I collagen concentration, but on muscovite mica D-spacing distributions showed increased negative skewness at 20 μg/mL and higher concentrations. Tilted D-spacing angles were found to correlate with decreased D-spacing measurements, an effect that can be removed with a tilt angle correction, resulting in no concentration dependence of D-spacing distribution on muscovite mica. We then demonstrated that tilted D-spacing is uncommon in biological tissues and it does not explain previous observations of low D-spacing values in ovariectomized dermis and bone.