Science Concentrates

Resistin's form hints at its control

The unusual crystal structures of resistin and its cousin--two hormones that may link obesity with type 2 diabetes--hint at how these proteins are regulated [Science, 304, 1154 (2004)]. A group led by Lawrence Shapiro, associate professor of biochemistry at Columbia University, found that most crystallized resistin proteins are made of six subunits, each with a β-sandwich head and an -helical tail. The subunits coordinate into two trimers bound through highly exposed disulfide bonds (yellow) between -helices. "This hexameric geometry is unlike the geometry of other proteins we know to be involved in signaling," Shapiro says. His group found a mutant trimer protein that lacked the disulfide-forming cysteines and was more bioactive than the hexameric form, suggesting that cleavage of the exposed disulfide bonds may be an important method of resistin regulation. Less active hexamers could circulate in the blood and be cleaved into trimers to activate the as-yet-undiscovered resistin receptors. Regulation of a hormone's activity through modification of its quaternary structure is a new possibility, Shapiro says, raised by the structure of resistin.

Mesopores go chiral

A strategy for synthesizing mesoporous silica produces rods of material with chiral pores. This pore chirality is of potentially great value, as it might allow enantioselective catalysis or pharmaceutical manufacturing. But synthesizing such materials has been difficult. Now, chemistry professor Shunai Che at China's Shanghai Jiao Tong University and colleagues report that the material, produced by self-assembly of chiral anionic surfactants and inorganic precursors, takes the form of twisted hexagonal rods 130–180 nm in diameter and 1–6 µm long [Nature, 429, 281 (2004)]. The rods are filled with chiral channels 2.2 nm in diameter that wind around the rod axis. Although the pores in individual rods are chiral, the collections of rods themselves are mixtures of left- and right-handed crystals, an issue the authors are studying.

New role for familiar compound

The food additive potassium ferrocyanide--K₄[Fe(CN)₆]--is a useful reagent for synthesizing aryl cyanides, a new study shows. The aryl cyanide (benzonitrile) motif is found in various therapeutic agents. One route to benzonitriles is palladium-catalyzed cyanation of aryl halides. The reactions, however, are plagued by catalyst deactivation. Low turnover numbers (20 to 400) are believed to be due to formation of highly stable palladium-cyano complexes. Furthermore, current sources of cyanide for the reaction, such as potassium cyanide, are highly toxic or require special handling. A new study, by Matthias Beller and coworkers at the Institute for Organic Catalysis at the University of Rostock, in Germany, indicates that potassium ferrocyanide, which is stable and nontoxic, would be an advantageous alternative cyanating agent [Chem. Commun., published online May 12, http://www.rsc.org/is/journals/<br > current/chemcomm/ccadvarts.htm]. With bromobenzene, at 25 mol % of potassium ferrocyanide and a catalyst loading of 0.01 mol %, the yield was 97% and the turnover number was 9,700--the best reported so far, according to Beller. The researchers suggest that potassium ferrocyanide releases cyanide ions slowly, thus preventing the formation of stable complexes with palladium, which kills the catalyst. Another advantage is the price: Potassium ferrocyanide costs only one-third as much as potassium cyanide.

An automated protocol has been designed for parallel synthesis of a ligand library and its rapid screening for asymmetric hydrogenation. Laurent Lefort, Johannes G. de Vries, and coworkers at DSM Pharma Chemicals, Geleen, the Netherlands, have prepared and screened 32 phosphoramidites, taking advantage of the straightforward stoichiometric synthesis involved (shown).

When the reaction is complete, the only impurity is a salt, which precipitates. One filtration yields ligands that are slightly impure but sufficiently clean for screening. Although enantioselectivities are about 5% lower than those obtained with purified ligands, they reflect the relative activities and enantioselectivities of library members accurately [Org. Lett ., 6, 1733 (2004)]. With the protocol, ligand screening can be accomplished in days rather than weeks, de Vries says. However, the protocol must be carried out carefully for results to be meaningful.

Tube shuttles protein into cell

Single-walled carbon nanotubes (SWNTs) ferry proteins across the cell membrane via endocytosis, according to a group led by Stanford University chemistry professors Hongjie Dai and Paul A. Wender [J. Am. Chem. Soc., published online May 13, http://dx.doi.org/10.1021/ja0486059]. They report that human promyelocytic leukemia cells die when they are incubated with biotin-functionalized SWNTs that noncovalently bind streptavidin--a protein with clinical anti- cancer applications. On their own, the biotin-functionalized SWNTs are not toxic to the cells, but they are essential for bringing the cytotoxic streptavidin protein into the cells. No toxicity was observed in cells incubated with streptavidin alone, even in highly concentrated solutions of the protein. Dai and Wender's team thinks that the SWNTs nonspecifically associate with hydrophobic regions on the cell's surface and then become internalized. They find that internalization does not occur when the endocytosis pathway is blocked.