Silicon lights up imaging

Nanometer-sized spherical silicon crystals fluoresce brightly enough to serve as multicolored labels in biomedical imaging, say US researchers. The researchers treated silicon wafers with hydrogen fluoride and hydrogen peroxide, creating large numbers of nanocrystals, each containing only a few dozen silicon atoms, surrounded by an outer skin of hydrogen atoms (Appl. Phys. Lett. 80, 841–844, 2002). The hydrogen “cap” causes the nanocrystal to act as a fluorophore—often called a quantum dot—much brighter than organic dyes such as fluorescein. The hydrogen-capped crystals occur only in certain sizes, the four smallest of which fluoresce in ultraviolet blue, green, yellow, and red wavelengths. Munir Nayfeh and colleagues at the University of Illinois at Urbana-Champaign (Urbana, IL) and University of North Carolina (Durham, NC) then separated the different-colored crystals by centrifugation and sonication or by gel chromatography. Nayfeh says: “The particles are highly photostable, do not blink, and can be functionalized without compromising their brightness, making them suitable fluorescent tags for analysis of sensitive biological materials.” PM

An AGENT for angina relief

Gene therapy could be a viable option for the millions of people suffering from angina. Cardiovascular specialists Collateral Therapeutics (San Diego, CA) and Berlex Biosciences (Montville, NJ) report that injections of the gene for human fibroblast growth factor 4 (FGF4) can alleviate some of the symptoms of angina (Circulation 105, 1291–1297, 2002). Angina occurs when the blood supply to the heart is impaired, leading to pain during exertion. FGF4 is an angiogenic factor, triggering the formation of new blood vessels to help restore blood flow. However, injections of FGF4 itself did not benefit patients, presumably because the protein's half-life in the body is too short. Instead, Collateral engineered a replication-defective adenovirus (Ad5) vector containing the FGF4 gene, injecting this directly into the coronary arteries of 60 patients. In the so-called AGENT trial (for angiogenic gene therapy), about 87% of the vector, which the company has named Generx, remained in the heart tissue and no FGF4 was detected elsewhere in the body. Moreover, the therapy increased the time that patients could remain on a treadmill by 20–30%, the range of improvement seen after standard treatments for angina—angioplasty and coronary bypass surgery. In a statement, one of the study authors said that the results were “far in excess” of what had been expected. International phase 2b/3 trials are now planned. LF

Protein chips go micro

An atomic force microscope (AFM) has now been used to construct protein chips whose active features are only a few hundred nanometers wide, providing a way to monitor protein–analyte reactions at the molecular level. Chad Mirkin and Milan Mrksich, at Northwestern University (Evanston, IL) and Chicago University (Chicago, IL) respectively, used a technique called dip-pen lithography to mark out patterns of a protein-capturing organic acid, 16-mercaptohexadecanoic acid (MHA), onto gold-coated chips. The needle-like sensing arm of an AFM was first dipped in the desired reagent, and then scanned across the surface to be treated, depositing minute amounts of the reagent in 100–350-nm-wide dots in the process. The array substrates were then washed with a surfactant to make unmarked areas inert, helping to reduce non-specific protein binding (Science 295, 1702–1705, 2002). Proteins including lysozyme and rabbit immunoglobulin G were then adsorbed onto the chips, where they adhered only to the MHA features. Images of these "nanoarrays," also taken with an AFM, showed that only one or two layers of protein molecule adhered to each MHA site. When loaded with recombinant fibronectin, these arrays could also capture fibroblasts onto the protein dots. Microarrays created using atomic force microscopy could also be valuable for investigating cell adhesion and other protein–molecule interactions on a much finer scale than can be done with conventional microarrays. PM

Non-infecting pigs

News that Immerge BioTherapeutics (Charlestown, MA) has shown that inbred miniature swine do not transmit porcine endogenous retrovirus (PERV) to human cells (J. Virology 76, 3045–3048, 2002) might counter one of the most serious potential problems of xenotransplantation, cross-species infection. Unlike may other viruses, PERV is integral to the pig genome and so can be passed from one generation to the next. Three strains of PERV have been identified: PERV-A and PERV-B can be transmitted from pig to human cells in vitro, whereas PERV-C cannot. Now, Clive Patience and colleagues at Immerge have shown that some families of inbred miniature swine do not transmit any variant of PERV to human cells. The team cocultured a variety of different pig and human cells in vitro, on four separate occasions, to ensure that any virus present could be detected. They identified families of pigs that do not transmit PERV to human cells, and where transmission did occur, the infecting PERV was formed from a recombination between PERV-A and PERV-C during the culture procedure. The researchers must now determine whether or not such “PERV-free” pigs will remain so with further breeding, or if they can be “re-infected” by pigs carrying transmissible PERV. LF

Sliced mice on the web

Plans are afoot for a “visible mouse” project—a repository of images revealing the morphological changes arising in transgenic mice. In a report in Radiology (222, 789–793, 2002), Allan Johnson and colleagues at the Center for In Vivo Microscopy at Duke University (Durham, NC) explain how magnetic resonance microscopy (MRM), could be used to deliver high-resolution images of mice. The researchers claim that over 6 million transgenic animals are generated each year, and much of the valuable information embedded in these animals is costly and time consuming to extract. It can cost around $7 to carry out histology and staining of a single slice of animal tissue. Not only is MRM quicker, but it does not distort the tissue, making it easier for researchers to compare wild-type and knockout mice. The Duke Center will collaborate with the Jackson Laboratory (Bar Harbor, ME), a leading source of laboratory mice, to create MRM atlases of normal laboratory strains, which will be posted on the Web. A more detailed description of MRM will be published in June in the Journal of Magnetic Resonance Imaging. LF

Research News Briefs written by Liz Fletcher and Peter Mitchell.