This Week in Science

Science  21 Jun 2002:
Vol. 296, Issue 5576, pp. 2097
  1. In Brevia

    Canli et al. (p. 2191) show that the activation of the amygdala, a part of the brain involved in processing facial expressions, varied with the degree of extroversion in subjects viewing happy faces.

  2. Disordered but Coherent

    Diluting a ferromagnetic with non-magnetic ions usually results in the loss of the ferromagnetic ordering, and at low temperatures the system would be expected to freeze into a glasslike state with little long-range coherence. Ghosh et al. (p. 2195) instead find that a disordered magnet does exhibit coherence at low temperature with properties quite different from that of a normal glass. The relaxation rates actually sharpen up, which suggests that the system behaves as a distribution of coupled oscillators.

  3. Weather Caught Underground

    The dramatic cooling that occurred approximately 8200 years ago has been associated with increased melting of the Laurentide ice sheet of North America that then perturbed North Atlantic thermohaline circulation. Baldini et al. (p. 2203) present a high-resolution trace element record of the “8200-year” event from a stalagmite in southwest Ireland which shows that the event was characterized by rapid initiation, a brief intra-event relaxation, enhanced seasonality, and an abrupt termination.

  4. One-Way Trips

    The Oort cloud is a somewhat spherical region at the edge of the solar system that contains about a trillion comets but feeds them into the inner solar system at a rate of only about 12 per year. Dynamical models have predicted more returning comets than are actually observed. Levison et al. (p. 2212; see the Perspective by Bailey) combined observations with new simulations to show that most of these missing returning comets are destroyed, not darkened by space weathering (so that they are considered dormant and unobservable). Thus, returning Oort cloud comets are more prone to destruction than their returning Kuiper belt cousins, which have been observed in relatively greater numbers.

  5. Loading Condensates Continuously

    One of the long-standing goals in the field of Bose-Einstein condensates has been to realize a continuous-beam atom laser. Atom lasers have either been operated in a pulsed mode or run until the initial reservoir of atoms is depleted. The problem has been to overcome the technical barriers of topping off the reservoir without destroying the delicate condensate itself. Chikkatur et al. (p. 2193; see the Perspective by Pfau) can now maintain more than 1 million atoms in a reservoir by using an optically trapped condensate as the reservoir and optical tweezers to transfer a “scoop” of condensed atoms from another condensate into the reservoir.

  6. Antibodies Give Membranes a Handedness

    For pharmaceuticals, often only one of several possible chiral forms (enantiomers) is actually active, and thus there is an increasing demand for separating enantiomers. One approach that allows continuous separation is to use membranes. Lee et al. (p. 2198) now show that derivatizing alumina nanopore membranes with antibodies allows the separation of the RR from the SS forms of a drug molecule. The binding strength of the antibody and, hence, the flux through the membrane was tuned by adding an organic molecule, dimethyl sulfoxide, to the buffer solution.

  7. Martian Lake Runneth Over

    Using data from the Mars Global Surveyor Laser Altimeter, Irwin et al. (p. 2209) have determined that the 900- kilometer long and 8- to 15-kilometer wide Ma'adim Vallis was probably formed by overfilling of a large lake. A breach in the lake's boundary rapidly incised the valley and flooded several prominent impact craters down slope. The valley is older than 3 billion years, and the identification of a large lake, where surface water played a prominent role in the development of Ma'adim Vallis, indicates the possible intensity and complexity of the early martian hydrologic cycle.

  8. Shaking Molecules by Their Tails

    When vibrational energy is deposited in one part of a molecule, it will eventually redistribute, but the actual progress is difficult to observe. Wang et al. (p. 2201) have achieved sufficient resolution in ultrafast spectroscopy to observe the transfer of vibrational energy through various liquid-phase alcohols. They excite the O-H stretching vibrations and then can see energy transfer, both up and down, to CH2 or CH groups before reaching the CH3 group.

  9. Cutting Peptides Inside Membranes

    Signal peptides can have multiple functions. They are inserted into membranes and are required for the biosynthesis of membrane and secreted proteins. Some are enzymatically cleaved and release peptides that are functionally important. Weihofen et al. (p. 2215) have identified this elusive signal peptide protease (SPP). It joins a growing family of multipass intramembrane cleaving proteases, whose founding member is presenilin. In a Perspective, Wolfe and Selkoe discuss how characterization of SPP may quell the controversy over presenilin's proteolytic activity.

  10. RNAs Made Mainly in the Plane

    RNA replication of positive-strand RNA viruses such as poliovirus and hepatitis C virus occurs on the surface of cytoplasmic membranes of the host cell. Poliovirus RNA synthesis is catalyzed by a virally encoded RNA-dependent RNA polymerase. Lyle et al. (p. 2218) show that purified polymerase oligomerizes to form planar or tubular arrays that correlate with optimal RNA binding and elongation. Poliovirus infection of cells produced vesicles with structures consistent with sheets of polymerase coating the vesicle membranes. The two-dimensional enzyme arrays may confer the advantages of surface catalysis to viral RNA replication.

  11. Mixing It Up

    During male germ-cell meiosis, homologous chromosomes pair and exchange genomic content and allow for genetic variability in offspring. Lynn et al. (p. 2222) studied human spermatocytes with immunofluorescence microscopy and found substantial variations in the rates of recombination within and among males. In addition, they identified a relation between this variation and the differences in the length of the synaptonemal complex. This finding is consistent in both humans and mice and may allow for a better understanding of the mechanisms involved in aneuploidy.

  12. The Substructure of Human Variation

    With approximately 4 million of the estimated 10 million common single nucleotide polymorphisms of the human genome already in databases, it is increasingly practical to study common genetic variation. Combinations of such variants seen on individual chromosomes in populations are called haplotypes. Gabriel et al. (p. 2225) conducted a genome-wide survey in samples from Africans, Asians, Caucasians, and African Americans. They conclude that most of the human genome consists of blocks over which there was little historical recombination and within which an average of three to five common haplotypes accounted for 90% of all chromosomes in each population sample.

  13. Bacterium Versus Fungus

    Many of the bacteria that cause many animal diseases likely evolved in a broad ecological system, and by looking elsewhere we may find alternative tractable experimental models for investigating disease mechanisms. Hogan and Kolter (p. 2229) have observed a range of antagonistic interactions among a selection of mutants of two commonly coexisting opportunistic pathogens, the bacterium Pseudomonas aeruginosa and the fungus Candida albicans. The bacterium attaches by its poles to the filamentous form of the fungus, developing a biofilm over the filaments and finally killing the fungus. Bacterial attachment requires the adherent virulence factor known as type IV pili. When under assault, the fungus reverts into a yeastlike form that appears to be impervious to the bacteria.

  14. Building Up and Tearing Down

    Hormone induction can stop very rapidly once the hormone is removed, but it has not been clear why. Freeman and Yamamoto (p. 2232; see the news story by Marx) show that molecular chaperones can disassemble the large multisubunit complexes that form on promoters and that induce gene expression. The authors used chimeric constructs to increase the local concentration of the chaperones in vivo and showed that when the chaperone p23 is localized to a promoter region, transcription is down-regulated. Hence, chaperones may play dual roles in the assembly and disassembly of transcription complexes.

  15. Transcriptional Dysfunction in Huntington's Disease

    The neurodegeneration characteristic of Huntington's disease (HD) is caused by mutations that induce expansion of a polyglutamine tract in the huntingtin protein. Mutant huntingtin is believed to interfere with transcription of genes that may be important for neuronal survival. Dunah et al. (p. 2238; see the Perspective by Frieman and Tjian) report that mutant huntingtin interferes with transcription mediated by the transcriptional activator Sp1 and its coactivator TAFII130. Coexpression of Sp1 and TAFII130 in cultured striatal cells from HD transgenic mice reversed the transcriptional block caused by mutant huntingtin. Soluble mutant huntingtin prevented binding of Sp1 to DNA in postmortem brain tissue from patients with presymptomatic and symptomatic HD, which suggests that transcriptional dysfunction caused by mutant huntingtin is an early event in HD pathogenesis.

  16. Hippocampal Place Cells and Their Circuitry

    Pyramidal cells in hippocampal area CA1 can exhibit their location-specific activity; for example, they can fire preferentially when an animal recognizes that it is in a certain place in a maze. Which is the input that determines the specific firing properties of these place cells? Brun et al. (p. 2243) completely severed all connections from hippocampal area CA3 to CA1. Place fields and place recognition behavior were still largely intact, indicating that direct input from the entorhinal cortex is sufficient for spatial recognition memory. However, navigation learning in these animals was disrupted. Navigation memory may thus depend on intact interaction between areas CA3 and CA1.

  17. Single-Sourcing Granite

    Differences in the isotopic composition of some granitic rocks have been attributed to differences in the rocks that melted to produce the granitic magmas. In melting experiments from a common source rock, Knesel and Davidson (p. 2206) found that both Sr concentration and 87Sr/86Sr ratios varied in the melt depending on the pressure, temperature, and duration of the experiments, as well as on the presence of water. They apply their results to the variety of granitic rocks produced in the Himalayas and show that these varied rocks could have been produced from one host rock undergoing periods of fluid-rich and dry melting.

  18. X-Inactivation in Germ Cells

    Although much is known about the mechanism of dosage compensation (the reduction of gene expression) of the X chromosomes in somatic cells, little is known about the mechanisms that underlie the repressed and condensed state of the X chromosomes in germ cells. Fong et al. (p. 2235) show that the Caenorhabditis elegans protein MES-4, which is essential for germline development, encodes an evolutionarily conserved SET-domain protein that binds exclusively to the autosomes, protecting them from silencing. Exclusion of MES-4 from the X chromosomes requires the activity of MES-2 and MES-6, polycomb group proteins required for X inactivation, and MES-3.

  19. Changes in Tropical Clouds and Radiation

    Earth's climate is influenced by changes in the radiative energy budget, the balance between absorbed solar radiation and energy emitted back into space. Chen et al. and Wielicki et al. (Reports, 1 February 2002, p. 838 and p. 841) analyzed more than 20 years of satellite data and reported large decadal variations in the energy budget of the tropics that are not easily explained by existing climate models. Trenberth comments that the results presented “reveal the shortcomings of the current climate observing system,” noting that inaccurate satellite calibration, sampling error, and introduction of variability into the cloud record may have contributed to unreliable time series and a flawed analysis. In response, Wielicki et al. argue that calibration errors cannot explain their observations and that, although further study and more accurate climate observing and modeling systems are indeed needed, “real evidence exists for consistent decadal variation in [radiation] fluxes and independently observed climate parameters.”

    The full text of these comments can be seen at