Steady-state and time-resolved fluorescence spectroscopy was used to follow the local and global changes in structure and dynamics during chemical and thermal denaturation of unlabeled human serum albumin (HSA) and HSA with an acrylodan moiety bound to Cys34. Acrylodan fluorescence was monitored to obtain information about unfolding processes in domain I, and the emission of the Trp residue at position 214 was used to examine domain II. In addition, Trp-to-acrylodan resonance energy transfer was examined to probe interdomain spatial relationships during unfolding. Increasing the temperature to less than 50 degrees C or adding less than 1.0 M GdHCl resulted in an initial, reversible separation of domains I and II. Denaturation by heating to 70 degrees C or by adding 2.0 M GdHCl resulted in irreversible unfolding of domain II. Further denaturation of HSA by either method resulted in irreversible unfolding of domain I. These results clearly demonstrate that HSA unfolds by a pathway involving at least three distinct steps. The low detection limits and high information content of dual probe fluorescence should allow this technique to be used to study the unfolding behavior of entrapped or immobilized HSA.

The steady-state and time-resolved fluorescence of Trp-214 was used to examine the conformation, dynamics, accessibility, thermal stability, and degree of ligand binding of human serum albumin (HSA) after entrapment of the protein in sol−gel processed glasses. The bioglasses were derived from tetraethyl orthosilicate and were aged in air without washing (dry-aged), in air after a washing step (washed), or in buffer (wet-aged). In all cases, significant changes were observed in the structure and dynamics of HSA, consistent with adsorption of the protein onto the silica surface combined with partial unfolding of the protein. Significant changes in the thermal stability and degree of ligand binding of the entrapped protein were also observed, with both stability and ligand binding capacity decreasing as aging continued. All proteins showed full accessibility to neutral quenchers over 2 months of aging but only partial accessibility to negatively charged quenchers, even at early aging times, indicating electrostatic repulsion of such analytes by the negatively charged matrix. Taken together, the results indicated that the reduced ligand binding for entrapped HSA was caused by a combination of protein denaturation and partial inaccessibility of the protein to negatively charged species. After 2 months of aging the entrapped proteins retained less than 15% of their binding ability in solution, regardless of which method was used to age the material. In light of these results, it is clear that improved sol−gel processing methods will be needed to overcome the time-dependent changes in the structure and function of proteins entrapped in silicate-based glasses.

We report the development of a fluorometric detection strategy for Ca(2+) based on induced changes in the conformation of cod III parvalbumin entrapped within a sol-gel processed glass. The detection scheme utilizes a fluorescent allosteric signal transduction (FAST) strategy wherein conformational changes induced by Ca(2+) binding result in alterations in the intrinsic fluorescence from the single tryptophan residue at position 102. Intrinsic fluorescence was also used to examine chemically induced changes in protein structure to ascertain the effects of entrapment on the conformational motions and stability of the protein. Fluorescence analysis indicated that the behavior of the protein depended on the entrapment protocols used. The entrapped protein retained conformational flexibility similar to that observed in solution and remained accessible to analytes such as Ca(2+). Entrapment also caused improvements in protein stability against chemical denaturants. However, entrapment caused the apparent affinity constant for binding of Ca(2+) to decrease substantially with aging time. Even so, in optimum cases, fluorometric detection of Ca(2+) could be done over a 600 μM range with a limit of detection of 3 μM and with no interference from divalent ions such as Mg(2+), Sr(2+), or Cd(2+), indicating the viability of using sol-gel entrapped FAST proteins for the detection of Ca(2+).

Recent years have seen a dramatic increase in the use of fluorescence-signaling DNA aptamers and deoxyribozymes as novel biosensing moieties. Many of these functional single-stranded DNA molecules are either engineered to function in the presence of divalent metal ion cofactors or designed as sensors for specific divalent metal ions. However, many divalent metal ions are potent fluorescence quenchers. In this study, we first set out to examine the factors that contribute to quenching of DNA-bound fluorophores by commonly used divalent metal ions, with the goal of establishing general principles that can guide future exploitation of fluorescence-signaling DNA aptamers and deoxyribozymes as biosensing probes. We then extended these studies to examine the effect of specific metals on the signaling performance of both a structure-switching signaling DNA aptamer and an RNA-cleaving and fluorescence-signaling deoxyribozyme. These studies showed extensive quenching was obtained when using divalent transition metal ions owing to direct DNA-metal ion interactions, leading to combined static and dynamic quenching. The extent of quenching was dependent on the type of metal ion and the concentration of supporting monovalent cations in the buffer, with quenching increasing with the number of unpaired electrons in the metal ion and decreasing with the concentration of monovalent ions. The extent of quenching was independent of the fluorophore, indicating that quenching cannot be alleviated simply by changing the nature of the fluorescent probe. Our results also show that the DNA sequence and the local secondary structure in the region of the fluorescent tag can dramatically influence the degree of quenching by divalent transition metal ions. In particular, the extent of quenching is predominantly determined by the fluorophore location with respect to guanine-rich and duplex regions within the strand sequence. Examination of the effect of both the type and concentration of metal ions on the performance of a fluorescence-signaling aptamer and a signaling deoxyribozyme confirms that judicious choice of divalent transition metal ions is important in maximizing signals obtained from such systems.

Contact with hydrophobic silicones frequently leads to protein denaturation. However, it is demonstrated that albumin in water-in-silicone oil emulsions retains its native structure in the presence of a functional, triethoxysilyl-terminated silicone polymer, TES-PDMS. Both HSA and TES-PDMS were essential for the formation of stable water-in-silicone oil emulsions: attempts to generate stable emulsions using independently either the protein or the functionalized silicone as a surfactant failed. Confocal microscopy indicated that the human serum albumin (HSA) preferentially adsorbed at the oil/water interface, even in the presence of another protein (glucose oxidase). A variety of experiments demonstrated that the hydrolysis of the Si-OEt groups on the functional silicone occurred only to a limited extent, consistent with the absence of a covalent linkage between the silicone and protein, or of cross-linked silicones at the interface. The fluorescence spectra of HSA extracted from the emulsions, front-faced fluorescence experiments on the HSA/silicone emulsion itself, and HSA/salicylate binding studies all demonstrated that the stability of the water/oil interface decreased as the protein began to unfold: unfolding of the protein in the emulsion was slower than in aqueous solution. The experimental evidence indicated that the interaction between HSA and TES-PDMS is not associated with either homomolecular (HSA/HSA; TES-PDMS/TES-PDMS) interactions or with covalent linkage between two the polymers. Rather, the data is consistent with the direct binding of unhydrolyzed Si(OEt) 3 groups to native HSA. The nature of these interactions is discussed.