More SNO in the Forecast

When it was discovered that proteins could be modified at their serine and tyrosine residues by the addition of phosphate molecules, an entire new field of signal transduction research opened up: the study of kinases and their role in disease, particularly cancer. There is little doubt that researchers studying S-nitrosylation—the covalent modification of Cys residues by addition of nitric oxide, or NO, to form an S-nitrosothiol (SNO)—hope for the same scientific impact. This won't be the case, however, until a more straightforward method for unequivocally determining the identity of those modified Cys sites becomes available. This goal may now be within reach, thanks to a recently published technique from Steven Gross's group at Cornell University. Their process, named SNOSID for “SNO site identification,” is based on a 2001 biotin-switch method which, when combined with PAGE and in-gel proteolysis with trypsin, was capable of identifying proteins that had undergone SNO modification. The procedure involves blocking non-SNO Cys residues in a complex protein mixture by methylthiolation, followed by reduction of the SNO moiety and covalent addition of biotin at the resulting reactive thiols. The novelty of the SNOSID method is the addition of a complete proteolysis step prior to capture of the biotinylated fragments with avidin. This allows for mass spectros-copy (MS) analysis, in this case using nanoflow liquid chromatography tandem MS, to characterize and identify each fragment previously carrying a SNO modification. The method, applied to the identification of S-nitrosylation in rat cerebellum proteins, is not only high-throughput, but represents the first unbiased assay for large-scale SNO-Cys residue identification. Although the authors hoped that the data derived from this work would enable elucidation of consensus motifs for SNO-Cys modification, results using a machine-learning approach showed that tertiary structure seems to be of greater influence that linear protein sequence. –SS

Hao et al. 2006. SNOSID, a proteomic method for identification of cysteine S-nitrosylation sites in complex protein mixtures. Proceedings of the National Academy of Sciences of the USA 103(4):1012-1017.

Strings Attached

Total internal reflection fluorescence microscopy (TIRFM) relies upon surface-associated evanescent fields. As a consequence, background signal is less of an encumbrance than in conventional epifluorescence microscopy, because only molecules that are within a couple of hundred nanometers of the surface are illuminated. Of course, this dependence upon surface proximity complicates matters too, since molecules under study must be tethered in a way that preserves their authentic biological behavior. In a recent paper, a group from Columbia University argues that strategies that have been used to date may not be broadly applicable or physiologically relevant. Instead, Granéli et al., propose using lipid bilayer-coated surfaces and show data indicating that DNA can be attached and imaged in such an environment. The authors’ raw materials include DOPC liposomes, YOYO1-stained and biotinylated λ DNA, and a microfluidics flow cell devised from silica slides. In a first approach, a dilute solution of neutravidin was placed in the flow cell, allowing scattered absorption of this streptavidin alternative. When liposomes are subsequently introduced, they rupture and coat the remaining areas of exposed silica. The authors found that injected DNA would become tethered at the neutravidin molecules without interacting nonspecifically with the liposome-derived bilayer. Under flow, the labeled DNA molecules would extend and become visible to TIRFM along their entire length; if the λ DNA was biotinylated at both ends, the stretched-out molecules could be maintained without constant flow. For greater spatial control, the authors also tested an alternate setup, in which barriers were etched in a slide at 10-µm intervals. Liposomes containing biotin-conjugated lipids were used to coat the subdivided slide, and neutravidin was added. After biotinylated DNA was injected, a flow was established, and the λ DNA would accumulate at the edges of the barriers. The end result is an array of DNA molecules whose spacing is tunable by the position of the barriers and the DNA concentration. This TIRFM-compatible system should be of immense benefit to highly parallel, single molecule studies of protein-nucleic acid interactions. –ND

Granéli et al. 2006. Organized arrays of individual DNA molecules tethered to supported lipid bilayers. Langmuir 22:292-299.

Addressing a Specificity Problem

Type II restriction endonucleases (REases) are enzymes that cut in a highly specific manner at sites in DNA molecules defined by the recognition sequence of the particular nuclease. They are currently ubiquitous in molecular biology labs around the world and are used in a multitude of applications. However, for certain experimental purposes, including physical genome mapping or the intentional creation of double-strand DNA break sites, their specificity might not be sufficiently high. For instance, a 6-bp cutter is expected to cut once every ∼4100 bp, while for an 8-bp cutter, this increases to every ∼65,500 bp (or over 48,000 times in the 3164.7 million base pairs in the human genome); if only a small number of cuts in the genome is required, a means to increase this specificity is needed. This has encouraged researchers to explore so-called programmable nucleases, in which sequence recognition modules are covalently attached to DNA-cutting enzymes to allow for more specialized splicing activity. These modules can be DNA-binding domains or triple helix-forming oligonucleotides (TFOs); the latter are made up of modified deoxynucleotides that bind in the major groove of DNA with exquisite specificity. In a recent Nucleic Acids Research paper from a group of collaborators in Europe, the authors covalently fused a TFO through a bivalent cross-linker to a single-chain version of the PvuII REase. Following optimization by adjusting several variables, including linker length and the distance between the restriction enzyme cleavage site and the triple helix-forming site (TFS), PvuII, normally a relatively frequent 6-bp cutter, was transformed into a chimeric enzyme that cut at its canonical site (CAGCTG) adjacent to the TFO-addressed TFS over 1000-fold more frequently than the unaddressed site. Preincubation of the PvuII-TFO conjugate with target DNA was required due to the slow kinetics of triple helix formation; this was done in the absence of divalent cations, required by PvuII for binding and endonuclease activity, such that addition of Mg^2 + initiated cleavage. Programmable endonucleases of this type clearly provide a very useful tool for researchers, and the principles used in their generation should be applicable to other DNA modifying enzymes. –SS

Eisenschmidt et al. 2005. Developing a programmed restriction endonuclease for highly specific DNA cleavage. Nucleic Acids Research 33(22):7039-7047.

In the Fold

RNA molecules may have fewer structural options than proteins, but their shape is equally critical to understanding their function. Although crystallography or NMR can be used to try and reveal tertiary structure, most RNA studies will begin by looking at secondary structure. Popular approaches include wet lab techniques, such as mapping regions of the RNA for differential sensitivities to chemical or enzymatic modification and computational methods based upon the inference of hydrogen bonding patterns from phylogenetic data and predictive folding models. These strategies, however, essentially give static information about a given RNA molecule and do not allow real-time analysis of folding pathways or responses to perturbations. As a result, a number of techniques employing molecular beacons, FRET donors and acceptors, and fluorescent base analogs have been described. In this vein, Tinsley and Walter report on the application of pyrrolo-C, a fluorescent analog of cytidine, in fluorescence spectroscopy of RNA molecules. Like 2-aminopurine, a fluorescent analog of adenine and guanine, pyrrolo-C is significantly less bulky than the organic dyes typically used to label nucleic acids. The analog has been previously used for structure probing, but only within the context of the dideoxynucleotide to monitor DNA local structure. The authors show that pyrrolo-C undergoes measurable changes in fluorescence upon incorporation into single- and double-stranded RNA and that its fluorescent signal is maintained over a broad range of buffer conditions. They go on to demonstrate the suitability of pyrrolo-C for probing dynamic RNA structure by using the analog to track both the binding and association of a DNA oligonucleotide to a single-stranded RNA and also the thermal melting of an RNA duplex. Thus, pyrrolo-C appears to be well-placed to offer a complementary means of probing RNA structure and dynamics via fluorescence spectroscopy. –ND

Tinsley and Walter. 2006. Pyrrolo-C as a fluorescent probe for monitoring RNA secondary structure formation. RNA 10.1261/rna.2165806.