Nature Biotechnology pp 162 - 166

Going soft is not necessarily a bad thing for researchers attempting to improve cereal crops. Softer grain texture improves flour quality, reduces milling costs, enhances digestibility, and possibly improves nutritive value. Now two researchers, Konduru Krishnamurthy and Michael Giroux of Montana State University (MSU) have exploited two proteins from wheat to engineer rice (Oryzae sativa L.) with softer grains. The hardness of corn and rice kernels limits their use as ingredients in foods and other products. Without fine-textured flour from soft wheat, for example, angel food cakes and croissants would lose their lightness and be just too chewy. To see if it would be possible to improve the texture of rice, Krishnamurthy and Giroux investigated whether a pair of proteins—puroindolines a and b (PINA and PINB)—present in soft-textured wheat (but missing in hard wheat) could be introduced into rice to improve grain texture. When they studied plants containing PINA and PINB DNA, sure enough, the kernels they produced were of significantly softer texture, resulting in flour that was both finer and of higher quality than that from unaltered plants. The next step is to try it in other hard-grained crops like corn.

Nature Biotechnology pp 142 - 147

The T cells of the immune system play an important role in defending the body against invading microorganisms. However, sometimes subsets of these cells fail to distinguish friend from foe, wrecking havoc with the body's own tissues and leading to allergies or autoimmune diseases such as rheumatoid arthritis and multiple sclerosis. One strategy for treatment would be to suppress the activity of all T cells, but this can seriously compromise an individual's immunity. Now, Teodor-D Brumeneanu and colleagues have devised a way to selectively eliminate only those T cells that are causing the problem. T cells express protein structures (receptors) on their cell surface that permit them to spot and then bind other protein "flags"—called antigens—on a foreign target. Groups of T cells express different receptors that allow them to recognize various intruders. Using a model system that reflects what might occur in the body, the researchers exploited this property by engineering a molecule that could bind to a specific subset of T cells, delivering a toxic drug called doxorubicin to those cells. In the test tube, the engineered molecule selectively killed the cells. When tested in mice, it also reduced the concentration of the targeted T cells compared with the free drug. Although preliminary, the results suggest that targeted drug of this type could be useful in treating autoimmune diseases by eliminating specific T cell populations.

Nature Biotechnology pp 121 - 124

Its good to talk. But getting living cells to interact with nonbiological materials such as silicon is a difficult task. This has made the marriage of living cells with microelectronics problematic. Using a genetic engineering approach, Peter Fromherz and colleagues now show that it is possible to bring living cells close to a silicon surface and detect the tiny electrical currents they produce. Such living devices could one day serve as the basis for biosensors that measure cell viability or function.

For biomaterials engineering, the trick is getting cells and semiconductors sufficiently close to permit electrical conversation. This is not straightforward because cultured cells naturally keep their distance both from one another and the structure on which they are grown. Small electrical currents then become too faint for an electrical device to detect.

To get around this problem, Fromherz and his team expressed recombinant potassium channels—so-called maxi-K channels—in cultured human cells. These maxi-K channels conduct a sufficiently large electrical current to be detected and amplified by the transistor. The researchers showed that, at least in theory, cells and silicon can "talk"—something that, if further refined, could lead to much more sensitive biosensors.

However, guanine-guanine mismatches are the rarest DNA mutations, and scientists must design ligands that are selective for other mismatches before this assay can be used for large-scale gene scanning.

Nature Biotechnology pp 157 - 161 and pp 115 - 116

GM crop antibiotic marker genes, which encode proteins that confer resistance to certain antibiotics, are potentially hazardous because they may be transferred to bacteria, leading to environment damage and threats to human health. Methods currently available to excise marker DNA are either unreliable or not applicable to most economically important crop plants, which are propogated by seeds. But now, Nam-Hai Chua and his team at the Rockefeller University have developed a reliable way of generating marker-free transgenic plants that is precise, convenient, and applicable for a wide range of crops.

Antibiotic marker genes in GM plants are a byproduct of the process used to identify plants that have taken up foreign DNA. Once such genes have enabled scientists to identify plants that contain the foreign gene of interest, they remain in the newly created plants as extraneous foreign DNA.

Chua's team have developed a system that uses a chemical to switch on a set of molecular protein shears that precisely snips out marker DNA from GM plants. They engineered thale cress (Arabidopsis) plants with a DNA sequence that contains the antibiotic marker gene, a recombinase gene (the shears), and its chemically induced promoter sequence. The entire sequence was flanked on either side by sequences that serve as targets for the recombinase. Thus, when the scientists added the chemical, the resultant recombinase protein degraded the flanking target sequences and removed the entire foreign DNA sequence (marker gene, recombinase and all). As well as showing remarkably tight control and high efficiency, this system is applicable to vegetatively propagated plants, which includes many crops.