PATIC: a conformationally constrained photoisomerizable amino acid

Abc Amino Acids: Design, Synthesis, and Properties of New Photoelastic Amino Acids

Photoisomerizable amino acids provide a direct avenue to the experimental manipulation of bioactive polypeptides, potentially allowing real-time, remote control of biological systems and enabling useful applications in nanobiotechnology. Herein, we report a new class of photoisomerizable amino acids intended to cause pronounced expansion and contraction in the polypeptide backbone, i.e., to be photoelastic. These compounds, termed Abc amino acids, employ a photoisomerizable azobiphenyl chromophore to control the relative disposition of aminomethyl and carboxyl substituents. Molecular modeling of nine Abc isomers led to the identification of one with particularly attractive properties, including the ability to induce contractions up to 13 A in the backbone upon trans --> cis photoisomerization. This isomer, designated mpAbc, has substituents at meta and para positions on the inner (azo-linked) and outer rings, respectively. An efficient synthesis of Fmoc-protected mpAbc was executed in which the biaryl components were formed via Suzuki couplings and the azo linkage was formed via amine/nitroso condensation; protected forms of three other Abc isomers were prepared similarly. An undecapeptide incorporating mpAbc was synthesized by conventional solid-phase methods and displayed characteristic azobenzene photochemical behavior with optimal conversion to the cis isomer at 360 nm and a thermal cis --> trans half-life of 100 min at 80 degrees C.

Photo-control of peptide helix content by an azobenzene cross-linker: steric interactions with underlying residues are not critical

Photo-control of protein conformation could prove useful for probing function in diverse biological systems. Recently, we reported photo-switching of helix content in a short peptide containing an azobenzene cross-linker between cysteine residues at positions i and i + 7 in the sequence. In the original sequence, underlying residues at positions i + 3 and i + 4 were made bulky as preliminary modelling suggested that this would enhance photo-control of helix content. To test this hypothesis, peptides with Val, Aib; Ile, Aib; and Ala, Ala at positions i + 3 and i + 4 were synthesized, cross-linked and characterized. Before cross-linking, the peptides show distinct conformational behaviours: two with differing helix/coil mixtures whereas the other has a circular dichroism (CD) spectrum characteristic of beta-sheet and a tendency to aggregate. However, upon cross-linking the peptides have very similar CD spectra: predominantly random coil in the dark but predominantly helical upon irradiation. These results refute the original hypothesis. Steric interactions between the linker and underlying residues do not appear to be critical for photo-switching behaviour. When the cross-linking bridge is lengthened by replacing the i, i + 7 cysteine residues with homocysteine, a lower degree of photo-control of helicity is observed. Furthermore, a non-cross-linking version of the azobenzene reagent is shown not to produce any photo-control of helicity. We conclude that the intramolecular cross-link is essential for photo-switching and that it should be applicable to a wide range of peptides and proteins.

Using an azobenzene cross-linker to either increase or decrease peptide helix content upon trans-to-cis photoisomerization

Reversible photocontrol of peptide and protein conformation could prove to be a powerful tool for probing function in diverse biological systems. Here, we report reversible photoswitching of the helix content in short peptides containing an azobenzene cross-linker between cysteine residues at positions i, i + 4, or i, i + 11 in the sequence. Trans-to-cis photoisomerization significantly increases the helix content in the i, i + 4 case and significantly decreases the helix content in the i, i + 11 case. These cross-linker designs significantly expand the possibilities for photocontrol of peptide and protein structure.

A photoregulated ligand for the nuclear import receptor karyopherin alpha

The ability to orchestrate the transport of proteins between nucleus and cytoplasm provides cells with a powerful regulatory mechanism. Selective translocation between these compartments is often used to propagate cellular signals, and it is an intimate part of the processes that control cell division, viral replication, and other cellular events. Therefore, precise experimental control over protein localization, through the agency of light, would provide a powerful tool for the study and manipulation of these events. To this end, a prototype photoregulated nuclear localization signal (NLS) was derived from a native NLS. A library of 30 mutants of the bipartite NLS from Xenopus laevis nucleoplasmin containing a novel, photoisomerizable amino acid was prepared by parallel, solid-phase synthesis and screened in vitro for binding to the nuclear import receptor karyopherin alpha, which mediates the nuclear import of cellular proteins. A single peptide was identified in which the cis and trans photoisomers bind the receptor differentially. The strategy used to obtain this peptide is systematic and empirical; therefore, it is potentially applicable to any peptide-receptor system.

Kinetic characterization of ribonuclease S mutants containing photoisomerizable phenylazophenylalanine residues

Incorporation of the photoisomerizable amino acid phenylazophenylalanine (PAP) into enzyme structures has been proposed as a strategy for photoswitching enzyme activity. To evaluate the strengths and limitations of this approach to enzyme photo-control, we performed a kinetic analysis of RNase S analogues containing PAP in positions 4, 7, 8, 10, 11 or 13. For an enzyme containing a single PAP group, the maximum extent of photoconversion (between approximately 96% trans/4% cis and 10% trans/90% cis under standard conditions) sets a limit on the maximum fold change in the initial rate of approximately 25-fold, if the cis form is the more active isomer, and approximately 10-fold if the trans form is more active. This extent of photoswitching was not realized in the present case because the effects of photoisomerization on kinetic constants were small and distributed among effects on S-peptide binding, substrate binding and the rate of the chemical step. These results suggest that photoisomerization could substantially alter enzyme kinetic constants but that a directed combinatorial approach might be required for realizing maximal photo-control in such systems. The limit set by the extent of photoconversion might be overcome by coupling multiple PAP groups to one enzyme or by altering the behaviour of a system that required oligomerization for activity.