Neurofilaments (NFs) are neuron-specific intermediate filaments composed of multiple subunit proteins including Neurofilament-Light (NFL), -Medium (NFM), -Heavy (NFH), α-internexin, and peripherin. These subunits have unique intrinsically disordered tail domains which protrude from the assembled filament, forming a dynamic protein layer which does not take on a well-folded structure. This layer can be viewed as a polymer brush made of intrinsically disordered protein, and it has been hypothesized to contribute to axonal mechanics via properties such as the height of the brush and its capacity to support inter-filament crosslinks. Yet, it remains unclear how the sequence features of the distinct subunit tail domains might differently contribute either to brush structure or to cell-level mechanical properties.In this dissertation, we examine the functions of the NF subunit proteins and their tail domains from a variety of perspectives. We first develop, express, and purify a set of recombinant proteins based on the tail domains of NFL, NFM, and NFH. These protein constructs are grafted onto a functionalized surface, forming a protein brush that can be phosphorylated in situ by a purified recombinant kinase. We demonstrate control over grafting density and composition of these multicomponent protein brushes, we measure the overall brush height by atomic force microscopy (AFM), and we employ a complementary self-consistent field theory to access brush morphology. We then use these platforms to explore how solution ionic strength, mixture composition, multisite phosphorylation, and sequence charge patterns govern overall brush height and the formation of multilayered structures within a protein brush. We find that NFM tails support a highly extended, multilayered brush regardless of their phosphorylation state and that they exhibit a nonlinear composition dependence in NFL:NFM mixtures. In contrast, we find that NFH tails are collapsed without phosphorylation and only expand to a relatively moderate degree after extensive multisite phosphorylation. We explore how specific negatively charged or charge-neutral regions in the NFH tail domain contribute to the brush height and multilayered structure. We also use charge-shuffled variants of the NFM tail to show that charge desegregation expands the NFM tail brush near physiological ionic strengths. We then probe the roles of specific NF subunits in contributing to the mechanical properties of live cells. Here we find that NFL knockdown in neuronal model cell line SH-SY5Y does not result in significantly different neurite stiffness. Transmission electron microscopy, inhibitor studies, Western blotting, and immunofluorescence microscopy suggest that this lack of effect may be due to an abundance of microtubules in SH-SY5Y neurites at the relatively early differentiation timepoint chosen, or functional compensation by other NF subunits present in the cell such as α-internexin or peripherin. Turning to a SW13−, a simpler model cell line lacking endogenous intermediate filaments, we implement a frequency sweep test via AFM oscillatory microrheology to access cell viscoelastic properties. Using this test, we show opposite changes to cell mechanical properties after treatment with cytoskeletal inhibitors targeting microtubules and actin microfilaments, respectively. However, we detect no significant changes to the mechanical properties of cells either stably expressing intermediate filament vimentin or transiently expressing filament-forming combinations of NF genes, when compared to non-filament-forming control conditions. Overall, this dissertation examines the functions of NF subunit proteins, both as complex biopolymers constituting a protein brush and as mechanical elements in living cells.