Morphology and mechanical stability of amyloid-like peptide fibrils

Synthetic, amyloid-like peptide fibrils have recently attracted interest as a novel, potentially biocompatible material for applications in biotechnology and tissue-engineering. In this paper, we report atomic force microscopy (AFM) studies of the morphology and mechanical stability of fibrils self-assembled in vitro from the short peptide TTR(105-115), which serves as a model system for amyloid fibrils. It forms predominantly straight rods of approximately 1 microm in length and of diameters between 7 nm and 12 nm. We found polymorphism, with some fibrils exhibiting an unstructured morphology and others showing a regular, longitudinal surface pattern of 90 nm periodicity. Contact mode AFM-imaging in air was utilised to perform mechanical tests of individual fibrils on the nanometer scale with a defined, vertical force in the nN-range applied by the AFM-tip. Above 100 nN, all fibrils showed a permanent, mechanical deformation whereas below 40 nN, fibrils remained unaffected. Tapping-mode AFM-imaging in water led to fibril decomposition within 1.5 h whereas tapping-mode imaging in air left fibrils intact. Additional investigations by circular-dichroism spectroscopy showed that dispersed fibrils were structurally stable in aqueous solution between pH 3 and pH 8, and in sodium phosphate buffer of concentration between 50 mM and 1 M.

Antibody linking to atomic force microscope tips via disulfide bond formation

Covalent binding of bioligands to atomic force microscope (AFM) tips converts them into monomolecular biosensors by which cognate receptors can be localized on the sample surface and fine details of ligand-receptor interaction can be studied. Tethering of the bioligand to the AFM tip via a approximately 6 nm long, flexible poly(ethylene glycol) linker (PEG) allows the bioligand to freely reorient and to rapidly "scan" a large surface area while the tip is at or near the sample surface. In the standard coupling scheme, amino groups are first generated on the AFM tip. In the second step, these amino groups react with the amino-reactive ends of heterobifunctional PEG linkers. In the third step, the 2-pyridyl-S-S groups on the free ends of the PEG chains react with protein thiol groups to give stable disulfide bonds. In the present study, this standard coupling scheme has been critically examined, using biotinylated IgG with free thiols as the bioligand. AFM tips with PEG-tethered biotin-IgG were specifically recognized by avidin molecules that had been adsorbed to mica surfaces. The unbinding force distribution showed three maxima that reflected simultaneous unbinding of 1, 2, or 3 IgG-linked biotin residues from the avidin monolayer. The coupling scheme was well-reproduced on amino-functionalized silicon nitride chips, and the number of covalently bound biotin-IgG per microm2 was estimated by the amount of specifically bound ExtrAvidin-peroxidase conjugate. Coupling was evidently via disulfide bonds, since only biotin-IgG with free thiol groups was bound to the chips. The mechanism of protein thiol coupling to 2-pyridyl-S-S-PEG linkers on AFM tips was further examined by staging the coupling step in bulk solution and monitoring turnover by release of 2-pyridyl-SH which tautomerizes to 2-thiopyridone and absorbs light at 343 nm. These experiments predicted 10(3)-fold slower rates for the disulfide coupling step than actually observed on AFM tips and silicon nitride chips. The discrepancy was reconciled by assuming 10(3)-fold enrichment of protein on AFM tips via preadsorption, as is known to occur on comparable inorganic surfaces.

DNA molecular motor driven micromechanical cantilever arrays

The unique ability of living systems to translate biochemical reactions into mechanical work has inspired the design of synthetic DNA motors which generate nanoscale motion via controlled conformational change. However, while Nature has evolved intricate mechanisms to convert molecular shape change into specific micrometer-scale mechanical cellular responses, the integration of artificial DNA motors with mechanical devices presents a major challenge. Here we report the direct integration between an ensemble of DNA motors and an array of microfabricated silicon cantilevers. The forces exerted by the precise duplex to nonclassical i-motif conformational change were probed via differential measurements using an in-situ reference cantilever coated with a nonspecific sequence of DNA. Fueled by the addition of protons, the open to close stroke of the motor induced 32 +/- 3 mN/m compressive surface stress, which corresponds to a single motor force of approximately 11 pN/m, an order of magnitude larger than previous classical hybridization studies. Furthermore, the surface-tethered conformational change was found to be highly reversible, in contrast to classical DNA motors which typically suffer rapid system poisoning. The direction and amplitude of motor-induced cantilever motion was tuneable via control of buffer pH and ionic strength, indicating that electrostatic forces play an important role in stress generation. Hybrid devices which directly harness the multiple accessible conformational states of dynamic oligonucleotides and aptamers, translating biochemical energy into micromechanical work, present a radical new approach to the construction of "smart" nanoscale machinery and mechano-biosensors.

Quick measurement of protein sulfhydryls with Ellman's reagent and with 4,4'-dithiodipyridine

Since its introduction in 1959, Ellman's reagent (5,5'-dithio-bis(2-nitrobenzoic acid)) has been the favorite reagent for spectrophotometric measurement of protein sulfhydryls. Meanwhile however, evidence has accumulated that many protein sulfhydryls give an incomplete reaction with Ellman's reagent, even during prolonged assay times. In the present study, the kinetic problem was solved by including cystamine as a "mediator" between the protein sulfhydryl and Ellman's reagent, as previously applied in an enzymatic thiol assay [9]. As an alternative, 4,4'-dithiodipyridine (DTDP) was used in place of Ellman's reagent. Due to its small size, amphiphilic nature, and lack of charge, DTDP quickly reacts with poorly accessible protein sulfhydryls, without any catalysis by cystamine. The DTDP method and the Ellman/cystamine method were both optimized for maximal sensitivity, minimal sample consumption (detection limit 0.2 nmol mL(-1), determination limit 0.6 nmol mL(-1)), and minimal assay time (5 min). In validation experiments, both methods gave identical results and the measured sulfhydryls/protein matched the expected values. Electronic supplementary material to this paper can be obtained by using the Springer Link server located at http://dx.doi.org/10.1007/s00216-002-1347-2.

Anomalous fluorescence enhancement of Cy3 and cy3.5 versus anomalous fluorescence loss of Cy5 and Cy7 upon covalent linking to IgG and noncovalent binding to avidin

This study provides a critical examination of protein labeling with Cy3, Cy5, and other Cy dyes. Two alternate situations were tested. (i) Antibodies were covalently labeled with Cy dye succinimidyl ester at various fluorophore/protein ratios and the fluorescence of the labeled antibodies was compared to that of free Cy dye. (ii) Fluorescent biotin derivatives were synthesized by derivatizing ethylenediamine with one biotin and one Cy3 (or Cy5) residue. The fluorescence properties of these biotin-Cy dye conjugates were examined at all ligand/(strept)avidin ratios (0 </= n </= 4). The results showed an astounding discrepancy between Cy3 and Cy5: Cy3-labeled antibodies fluoresced very well, even at high Cy3/protein ratios, and the same applied to (strept)avidin with up to four bound biotin-Cy3 conjugates. In contrast, antibodies with six covalently bound Cy5 labels (obtained with the recommended procedure) were almost nonfluorescent, only at 2-3 Cy5 labels/IgG some moderate fluorescence was obtained. By analogy, the biotin-Cy3 conjugate fluoresced intensely, even at high ligand/avidin ratio, in contrast to the weakly fluorescing biotin-Cy5 conjugate. Three mechanisms are responsible for the discrepancy between Cy3 and Cy5. (i) Attachment of Cy3 to a protein's surface causes an anomalous enhancement in fluorescence (by 2-3-fold) while no enhancement occurs with Cy5. (ii) Mutual quenching of IgG-bound Cy dyes by resonance energy transfer is much more pronounced for Cy5 labels than for Cy3. (iii) In IgG with six bound Cy5 labels, about one-third of the labels adopt a nonfluorescent state which is characterized by a large UV-vis absorption maximum at 600 nm instead of at 650 nm. Cy3.5 was found to mimick the properties of Cy3, while Cy7, and to some extent also Cy5.5, were similar to Cy5. In conclusion the Cy dye series is divided into two groups: Antibodies with multiple Cy3 or Cy3.5 labels yield bright fluorescence while extensive quenching occurs in antibodies labeled with Cy5 and Cy7.