This dissertation presents the basic principles of the magnetic protein chip immunoassay and applications of the magnetic protein chip for the detection of protein biomarkers. The magnetic protein chip is based on the giant magnetoresistive (GMR) spin-valve sensors, and electrical resistance changes of the sensors are measured as signals. Immunoassays were developed using the magnetic protein chips to quantitate protein biomarkers with high sensitivity. The magnetic protein chip immunoassay has several advantages over conventional non-magnetic techniques, such as multiplex capability, smaller sample volume requirement, enhanced signal-to-noise ratio, and facile integration with electronics. Ionizing radiation is high energy radiation that can remove electrons from atoms, and it causes various cellular damages some of which are lethal. Using an in vivo mouse radiation model, we developed protocols for measuring fms-related tyrosine kinase 3 ligand (Flt3lg) and serum amyloid A1 (Saa1) in small amounts of blood collected during the first week after X-ray exposures of sham, 0.1, 1, 2, 3, or 6 Gy. Flt3lg concentrations showed excellent dose discrimination at >= 1 Gy in the time window of 1 to 7 days after exposure except 1 Gy at day 7. Saa1 dose response was limited to the first two days after exposure. A discriminant analysis using both proteins could show improved dose classification accuracy. Our magnetic protein chip immunoassay demonstrated the dose and time responses and low-dose sensitivity that have important advantages in radiation triage biodosimetry. Necrotizing enterocolitis (NEC) is an inflammatory bowel disease often observed in pre-term babies and has high mortality rate. However, timely diagnosis of NEC has been hampered due to its unspecific symptoms and ineffective clinical tests currently available. We developed a magnetic protein chip immunoassay for the validation of NEC biomarkers. Three biomarkers, C-reactive protein (CRP), matrix metalloproteinase-7 (MMP7), and epithelial cell adhesion molecule (EpCAM) were quantitated using a small amount of blood samples. Receiver operating characteristic (ROC) curve analysis combined with bootstrapping technique showed excellent discrimination of NEC from healthy control and NEC from sepsis. Given the generality of the detection scheme used in the magnetic protein chip immunoassay, the magnetic protein chips are expected to hold great potential for medical diagnosis and clinical research.

Recent successes in the pharmaceutical industry have yielded a variety of new drugs based on proteins and nucleic acids. While these drugs are very promising, they are only effective once they are inside cells. Unfortunately, transporting drugs into cells tends to be challenging because cell membranes provide a protective barrier. However, there is hope for these drugs since researchers have identified many drug delivery agents that can shuttle drugs past this barrier. Cell penetrating peptides (CPPs) are particularly promising delivery agents due to their low toxicity and ability to deliver a wide number of therapeutic agents. However there are challenges to using CPPs, because their delivery mechanisms cannot yet be controlled. Engineering new CPPs that use specific, known translocation mechanisms would be a key achievement that could increase delivery efficiency and prevent unwanted side effects. Accomplishing this goal requires new experimental methods for determining the factors that control a CPP's translocation mechanisms. In this thesis I present new atomic force microscopy (AFM) methods for studying CPPs and other cell membrane active species. I first present a theoretical model that shows how results from AFM can indicate whether CPPs change the energy barrier to bilayer penetration. I then describe new experimental AFM methods we developed for examining CPP transduction mechanisms.

Cells communicate through direct contact and soluble chemical signals. Mimicking an extracellular environment requires controlling these signals at micron length scales. Integrated circuits make electronic control at these scales trivial, but fluidic control at these length scales requires very different principles. Standard microfluidic devices can finely control flowing fluids, but fluid flow affects cells in a myriad of ways. Alternatively, diffusion based chemical delivery methods tend to be crude, ill defined systems that offer very limited control. This thesis describes three distinctive platforms that combine the active spatial and temporal control of microfluidic systems with a delivery system that relies purely on diffusion. First, we detail a silicon based array of nanoreservoirs underneath the cell culture surface which are used to store and release bioactive molecules. These reservoirs are opened and closed with electrochemical dissolution and deposition at a narrow reservoir opening. Next, we describe an adaptation of traditional, elastomer based microfluidics. In these devices the cell culture area is separated from a microfluidic channel located directly underneath the chamber by a nanoporous membrane. The desirable microfluidic properties, including temporal and spatial control, are preserved, while fluidic flow over the cells is eliminated. Finally, we demonstrate a novel "nanostraw" culture surface, which is combined with the previous device to offer fluidic access directly to the cell cytosol, creating a powerful tool with implications for cell delivery and sampling. Additional work on probing the assembly of protein structures is also detailed. Clathrin 2-dimensional lattice assembly on lipid monolayers, serving as cell membrane mimics, was monitored and studied through surface rheological techniques. Rheological measurements elucidated important network properties, and the formation process was compared to various models for clathrin network assembly.

Abstract: The small scale of nano-materials makes them one of the best man-made candidates to interact with biological systems at subcellular or even molecular level. It has been the focal point of the research interests to interfacing live cells with one dimensional nanostructures, such as nanowires and nanopillars. In my Phd research, I have utilized nanopillar based structures and devices to interface biological cells electrically, optically and mechanically. 1. We achieve improved electric interface between biological cells and solid state device by using arrays of vertically aligned nanopillar electrodes. Their tight attachment to the cell membrane allows us to acquire intracellular-like action potential signals non-destructively from cultured cardiomyocytes, which is responsible for various important cellular functions. 2. We demonstrate below-the-diffraction-limit observation volume in vitro and inside live cells by using vertically aligned silicon dioxide nanopillars. With a diameter much smaller than the wavelength of visible light, a transparent silicon dioxide nanopillar embedded in a nontransparent substrate restricts the propagation of light and affords evanescence wave excitation along its vertical surface. This effect creates highly-confined illumination volume that selectively excites fluorescence molecules in the vicinity of the nanopillar. We show that this nanopillar illumination can be used for in vitro single molecule detection with high fluorescence background. In addition, we demonstrate that vertical nanopillars interface tightly with live cells and function as highly localized light sources inside the cell. Furthermore, chemical modification of the nanopillar surface provides a unique way to locally recruit proteins of interest and simultaneously observe their behavior within the complex, crowded environment of the cell. 3. We engineer and fabricate vertically nanopillar arrays, and culture various types of cells atop. We study the cell growth pattern in presence of nanopillar arrays, including attachment, migration, etc. We also design patterned nanopillar arrays and utilized them to guide and control cell growth via cell-nanopillar interaction.

Electrophysiological tools and biologic delivery systems generally rely on non-optimal methods for gaining access through cellular membranes. Electrophysiological techniques that provide intracellular access, such as patch clamping, result in membrane holes and cell death in a matter of hours, while the delivery of bioactive materials are hampered by low bioavailability following passage through the endosomal pathways. In each case, the lipid bilayer backbone of the cellular membrane presents a formidable barrier to intracellular access. As biological gatekeepers, cell membranes not only physically define everything from whole organisms to individual organelles, they also prevent unobstructed flow of molecules between the inner and outer regions of the membrane. This occurs since the hydrophobic lipid acyl tails form a narrow hydrophobic layer a few nanometers thick, which is highly unfavorable for the passage of most hydrophilic molecules. It is this region that is one of the greatest obstacles to the dream of biotechnology seamlessly and non-destructively integrating synthetic components with biological systems. This thesis contributes to the understanding of how to rationally design devices that interact specifically with this hydrophobic region. In turn, this work begins to establish design guidelines for creating non-destructive, membrane-penetrating bio-inorganic interfaces. The beginning chapters focus on the development of the "stealth" probe platform. In nature, there exist specialized transmembrane proteins capable of incorporating into lipid bilayers by replicating the lipid hydrophilic-hydrophobic-hydrophilic structure. The stealth probe design mimics this structure by creating 2-10nm hydrophobic bands on otherwise hydrophilic structures. However, since current lithographic methods do not possess the necessary resolution, a new fabrication technique using a combination of top-down fabrication with bottom-up self-assembly methods was developed. This approach uses an evaporated chrome-gold-chrome stack and focused ion beam (FIB) milling, where the exposed edge of the embedded gold layer can be specifically functionalized with a hydrophobic thiol-mediated self-assembled monolayer. Chapter 3 explores the propensity for insertion and specific interaction of the stealth probe hydrophobic band with the hydrophobic lipid bilayer core. In order to gain quantitative insight into the interaction behavior, atomic force microscopy was used in conjunction with a new, stacked lipid bilayer testing platform. By using stacks of 100's to 1000's of lipid bilayers, substrate-probe interaction artifacts can be removed while simultaneously allowing precise determination of probe location within a lipid bilayer. It was found that completely hydrophilic probes reside in the hydrophilic hydration region between bilayers, whereas hydrophobically functionalized stealth probes preferred to reside in the bilayer core. This behavior was found to be independent of hydrophobic functionalization, with butanethiol and dodecanethiol both displaying preferential localization. The subsequent chapters explore how the molecular structure of the hydrophobic band and the band thickness affect membrane-probe interface stability. The lipid stack platform provides an easy method of force-clamp testing, which enabled quantitative extrapolation of the unstressed interface strength. A series of tests with various length alkanethiols found that the crystallinity of the molecules in the hydrophobic band is the dominant factor influencing interfacial stability. Surprisingly, hydrophobicity was found to be a secondary factor, although necessary to drive spontaneous membrane integration. Molecular length was also found to play a role in determining the ultimate interfacial strength, with short chain molecules similar in length to amino acid side chains promoting the most stable interfaces. The thickness of the hydrophobic band was found to regulate the interface structure. Bands with thicknesses comparable to that of the host lipid bilayer core likely promote a fused interface geometry, similar in structure to that of transmembrane protein-lipid bilayer interfaces. Thicker bands began to transition to a 'T-junction' interface that is characterized by a lower interface stability. Interestingly, the behavior of 10nm bands were indistinguishable from completely hydrophobic probes, reinforcing the importance of nanoscale patterning for stable membrane integration. Chapter 6 builds on the results of the previous chapters by exploring how various stealth probe geometries influence adhesion behavior. In agreement with force clamp testing, short disordered monolayers displayed strong integration into the bilayer core, while crystalline monolayers displayed extremely weak integration. Preliminary adhesion testing results with human red blood cells demonstrate that the stealth probe geometry holds promise for in vitro and in vivo platforms, expanding the results of this work from simply a biophysical test system to a real world example. Finally, the behavior of two hydrophobic bands either commensurately spaced with the hydrophobic core spacing in the bilayer stack, or incommensurately spaced in order to force one band to reside in the hydrophilic hydration layer, is explored. It was found that the commensurately spaced bands display superior strength to single band tips, which is attributed to the necessity to simultaneously rupture both membrane-hydrophobic band interfaces. Conversely, the incommensurately spaced probes display a significant destabilization of the interface. This is thought to be due to the forced residence of one hydrophobic band in a hydrophilic hydration layer. This result is intriguing for biologic delivery systems, as the nuclear double membrane presents a unique barrier geometry, and a double band system may provide a facile means for penetration.