Fly smells euclidian peach

How does the olfactory system dissect the world of odors? Elissa Hallem and John Carlson (Cell 125, 143–160; 2006) have now achieved systematic in vivo expression of 31 of the 32 odorant receptors of the fruit fly Drosophila melanogaster in a mutant neuron of the antenna, 'the empty neuron', which had previously been shown to lack odorant receptors by Dobritsa et al. (Neuron 37, 827–841; 2003).The authors introduced 100 odorants into the headspace of the fly and recorded extracellular electrophysiological activity from a series of single antennal sensilla. Twenty-four of the receptors gave responses to subsets of odorants, but several did not, leading the authors to speculate that these respond to specialized odorants such as pheromones. Strong responses (those eliciting many electrical spikes) were sparsely distributed; some receptors were broad in their specificity, responding to odorants in more than one chemical class, while others were much more narrowly restricted. The euclidian distances of odorants were determined within the 24-dimensional odor space defined by the receptors. Chemical class—such as lactone, alcohol or amine—was often a good predictor of similarity. Interestingly, chemically similar compounds such as esters could be widely separated. Peach, apricot and cherry, fruit odors from a single genus, clustered most closely together, while cherry and banana were most distant in odorant space. MA

Toward a therapy for Marfan syndrome

Marfan syndrome (MFS) is a disorder of connective tissue caused by mutations in the gene encoding fibrillin-1, a component of the extracellular matrix. It has been shown that fibrillin-1 regulates TGF-β activity and that increased TGF-β signaling is a primary cause of the many features of MFS. Jennifer Habashi and colleagues now report that losartan, an FDA-approved angiotensinogen II type 1 receptor antagonist that also antagonizes TGF-β, prevents aortic aneurysm in a mouse model of MFS (Science 312, 117–121; 2006). Habashi et al. used mice that are heterozygous for a fibrillin-1 missense mutation, which typically causes aortic dilatation and aortic root enlargement. Treatment of pregnant mice with losartan in the drinking water, or of postnatal mice by intraperitoneal injection, either prevented or resulted in a complete reversion of these major, life-threatening manifestations of MFS. Treated mice also showed a reduction in distal airspace of the lung. The authors suggest that individuals with other aortic aneurysm syndromes, such as Loeys-Dietz syndrome (Nat. Genet. 37, 275–281; 2005) and arterial tortuosity syndrome (Nat. Genet. 38, 452–457; 2006) may also benefit from losartan. A clinical trial is planned to test the drug in individuals with MFS. AP

Polycystic kidney disease and mTOR

Despite significant progress toward understanding the molecular and cellular mechanisms underlying polycystic kidney disease (PKD), effective treatment options remain limited. Thomas Weimbs and colleagues (Proc. Natl. Acad. Sci. USA 103, 5466–5471; 2006) now identify the signaling molecule mTOR as a promising therapeutic target in common inherited forms of the disease. The authors found evidence for a direct physical interaction between polycystin-1, a protein frequently disrupted in autosomal dominant PKD, and tuberin, a negative regulator of the mTOR pathway. They further showed that the mTOR pathway is inappropriately active in the epithelial cells lining cysts in individuals with PKD and in several mouse models of the disease. Treating two different mouse models with the mTOR inhibitor rapamycin caused a marked improvement in disease symptoms, including a reduction in cyst size and normalization of plasma blood urea nitrogen levels. Notably, in a retrospective study of individuals with autosomal dominant PKD who received kidney transplants and who were then treated with rapamycin to prevent transplant rejection, the authors found that the cystic kidneys that remained in these transplant recipients showed a significant reduction in kidney volume, suggesting a possible therapeutic benefit in humans. KV

Paths of protein evolution

Dan Hartl and colleagues present a new study on the evolution of proteins involved with drug resistance to β-lactams. (Science 312, 111–114; 2006) The simultaneous presence of five point mutations in β-lactamase is known to confer a high level of drug resistance. The authors now consider the range of possible evolutionary paths that could have led to this high level of drug resistance. They compare strains with each of the 32 combinations of these five mutations, testing the effect of each mutation individually and in all combinations for resistance to cefotaxime, a cephalosporin β-lactam, in E. coli. Four out of five of these mutations showed different effects on fitness (classified as positive, negative or neutral) depending on the combination with other mutations, and one mutation showed positive effects on fitness in combination with any of the other mutations. The authors then estimated the relative probability of the 120 possible routes to sequentially acquire these five mutations under selection, assuming that only a beneficial mutation that increased resistance would be fixed. These analyses found that only a few of the possible pathways had a probability of occurring. These studies also suggest how protein evolution may be constrained through interactions between mutations that jointly confer higher fitness. OB

Functional conservation

Proving pathogenicity of sequence changes in enhancer regions can be a challenging task. Part of the difficulty is showing that the region in which the sequence change occurs is a bona fide regulatory element. A high level of sequence conservation is thought to indicate functionality. Andy McCallion and colleagues now report an investigation of the regulatory potential of conserved noncoding sequences at the RET locus (Science advanced online publication 23 March 2006; doi:10.1126/science.1124070). The authors identified noncoding regions conserved between Fugu and zebrafish and between human and at least three non-primate mammals and found that the human conserved sequences could not be aligned with the zebrafish RET locus. Using a zebrafish transgenic system based on the Tol2 transposon, they tested the ability of each conserved region to drive reporter expression in vivo. Nearly all of the zebrafish conserved sequences drove reporter expression in a pattern consistent with endogenous ret expression. And interestingly, almost all of the human conserved sequences drove expression in an endogenous pattern, even in zebrafish cell types not present in mammals. The authors propose that there may be transcription factor binding sites within the functional regions that are not detectable by algorithms. This work suggests that noncoding sequences that escape detection by traditional sequence alignment methods may yet maintain conserved enhancer function. EN

Research Highlights written by Myles Axton, Orli Bahcall, Emily Niemitz, Alan Packer and Kyle Vogan.