A Tiny Protein Plays a Big Role in DNA Repair

The human genome lies deep within the cell interior, sequestered behind the double-walled membrane of the nucleus, yet still remains vulnerable to a wide variety of hazardous materials. Two of DNA's worst enemies, ultraviolet light and chemical carcinogens, can wreak havoc on the molecule by mutating individual nucleotides or changing its physical structure. In most cases, genomic integrity is restored by specialized suites of proteins dedicated to repairing specific types of injuries. 
 
One mending mechanism, called nucleotide excision repair (NER), recruits and coordinates the services of roughly 25 proteins to recognize and remove structure-impairing lesions, including those induced by ultraviolet light. At the center of this effort is the ten-subunit transcription/repair factor IIH (TFIIH) complex. As its name suggests, in addition to its role in DNA repair, it also regulates transcription; how TFIIH coordinates these very different activities is still under investigation. Regardless, three TFIIH genes— XPB, XPD, and TTDA—have been implicated in the photosensitive form of a rare inherited premature aging syndrome called trichothiodystrophy (TTD), which is characterized by brittle hair and nails, scaly skin, and neurological degeneration. 
 
In a new study, Giuseppina Giglia-Mari, Jan Hoeijmakers, Catherine Miquel, Wim Vermeulen, and colleagues created a fluorescently tagged version of trichothiodystrophy group A (TTDA) to investigate its role in repair and transcription. By experimentally modifying the transcription function and by triggering DNA repair in human cell lines expressing the fluorescent TTDA protein, the researchers show that TTDA first of all dynamically associates with TFIIH, and that TTDAs become stably incorporated only while the complex is engaged in NER. 
 
NER steps in when damage interferes with the elongation of a newly transcribed gene, promoting cell survival. It also plays an anti-cancer role by surveying the genome for damage. In ultraviolet-induced NER, TFIIH unzips the double helix to access the injury, and then recruits four proteins to stabilize the open strand, evaluate the damage, and cleave the DNA on either side of the lesion. Other proteins produce new nucleotides to fill the gap. Cells collected from patients with TTD-A have TTDA mutations that produce either truncated, nonfunctional versions of the protein or no protein at all, resulting in reduced TFIIH levels. TTD-A cells also have NER defects and are hypersensitive to ultraviolet light. 
 
To study the role of the TTDA subunit in the TFIIH complex's functions, the researchers tagged TTDA and the XPD subunit with green fluorescent protein (GFP) to monitor and compare their movement and behavior using high-resolution confocal microscopy. Both TTDA-GFP and XPD-GFP could stably incorporate into TFIIH and function in DNA repair, proving reliable tools for studying TFIIH's spatial and temporal distribution and activity during transcription and DNA repair. 
 
Both TTDA-GFP and XPD-GFP were observed in the cytoplasm and nucleus, in contrast to the XPB subunit, which is known to localize only in the nucleus. To determine if TTDA and XPD assemble with TFIIH in the cytoplasm, the researchers used a customized version of a motility-monitoring technique called fluorescence recovery after photobleaching (FRAP). Applying a high-intensity laser to a specific region in a cell destroys fluorescence in that area; as unaffected GFP-tagged molecules from other regions replace the zapped molecules, fluorescence is recovered. Recovery is much faster for free-moving GFP-tagged proteins than for those caught up in interactions with other cellular proteins. 
 
In the cytoplasm, TTDA-GFP moved much faster than XPD-GFP, but did not interact with each other. Both proteins, however, were less motile in the nucleus, where they interacted with TFIIH. These results indicated that TTDA exists in two pools in non-irradiated cells, one free and one bound to TFIIH. After irradiation, however, fluorescence was not completely recovered for either TTDA-GFP or XPD-GFP, suggesting a lack of motility one would expect by participating in DNA repair. When the researchers inhibited a protein essential for NER initiation in the irradiated cells, they observed a reduced fraction of immobilized TTDA-GFP, confirming that TFIIH must incorporate TTDA-GFP (thus, reducing its mobility) for NER. By interfering with the transcription machinery, they go on to show that TFIIH does not stably incorporate TTDA when TFIIH is simply bound to DNA; TFIIH only enhances association with TTDA during NER. 
 
Altogether, these results show that NER-specific lesions induce TTDA to form a stable association with the TFIIH in living cells. Giglia-Mari et al. propose that once TTDA nestles into place after the core TFIIH complex attaches to a lesion, it triggers a conformational change that recruits the other subunits required for repair. TTDA may also help TFIIH fold properly, preventing it from being degraded and allowing it to accumulate to the levels necessary for NER function. TFIIH molecules spend far longer at NER sites (four to five minutes) than they do at transcription sites (two to ten seconds), suggesting why TFIIH concentrations are so important to NER—and why people that can't produce TTDA experience such debilitating symptoms.

To use FRET for picking apart the pathways involved in phagocytosis, the researchers made mouse phagocytes that expressed engineered proteinsversions of ADP ribosylation factors (ARFs) 1 and 6 had fl uorescent protein attached, and fl uorescent marker proteins that would bind to either ARF1 or ARF6, when they had been activated inside the cell. Binding of the activated fl uorescent ARFs to the fl uorescent markers could be detected via FRET. Thus, the researchers could see when and where ARF1 or ARF6 were active in the cell during phagocytosis.
The researchers found that pseudopods extending around a particle activated ARF6 at their tips. Then as the cell more fully enveloped a particle, ARF6 turned off, and activated ARF1 appeared throughout the pseudopod until it closed around the particle. The discovery of a role for ARF1 was unexpected, since previous studies suggested this protein is not involved in phagocytosis. Also, the researchers found that inhibiting PI-3K prevented cells from switching between these two phases of phagocytosis, and caused them to stop halfway without having closed around the particles.
It's still not entirely clear what role each of the ARFs plays. ARF6 may regulate how the cell supplies more membranes to the area of phagocytosis, so that the cell can expand its surface area there and engulf particles. ARF1, on the other hand, plays a key role in protein traffi cking, and so may regulate which proteins migrate to the site of phagocytosis. But fi nding that phagocytosis has distinct phases, regulated by phospholipids in the membrane, is an important step toward sorting out how thousands of receptors can be coordinated in time and space to make a phagosome.

Synopses of Research Articles
June 2006 | Volume 4 | Issue 6 | e190 | e206 Eating at an Ethiopian restaurant might give one an appreciation for the process of phagocytosis, a widespread method immune cells use to ingest particles. To dine using injera, the spongy Ethiopian fl atbread, you form a pocket of bread around the food that you pinch off and, with practice, stuff cleanly into your mouth. Similarly, when cells ingest small particles, viruses, or other smaller cells by phagocytosis, they extend protrusions called pseudopods around their target. The pseudopods then seal around the object, and a bit of the cell membrane that encases the object buds off and travels inside the cell.
Figuring out which molecules are involved in coordinating the fast-moving, dynamic process of phagocytosis has been diffi cult. But using a sensitive, high-resolution technique called fl uorescence resonance energy transfer (FRET), a new study reveals the patterns of protein activities that help regulate phagocytosis. Peter Beemiller, Adam Hoppe, and Joel Swanson show that two stages of phagocytosis-the extension of pseudopods and the closure around an object-each require activation of specifi c proteins, members of the ADP ribosylation factor (ARF) family, which are major regulators of the budding of vesicles, or membrane-bound sacs, from internal organelles. The researchers also found that an enzyme called phosphatidylinositol-3′-kinase (PI-3K) plays a key role in organizing these two stages of phagocytosis.
Key to the discovery was a technique called FRET. Though the technique was fi rst described nearly 50 years ago, it has recently been greatly refi ned and has experienced a renaissance. The technique relies on the ability of one fl uorescent molecule to excite another fl uorescent molecule, which emits light when the two molecules are very close to each other. Thus, when exciting the fi rst fl uorescent molecule lights up the second one, it signals that the two fl uorescent molecules are within tens of angstroms-practically on top of each other.

Inconspicuous Consumption: Uncovering the Molecular Pathways behind Phagocytosis
Mason Inman | DOI: 10.1371/journal.pbio.0040190 DOI: 10.1371/journal.pbio.0040190.g001 Fluorescent proteins localize the guanosine 5'-triphosphate hydrolase ARF in the Golgi apparatus of a living macrophage. FRET studies revealed ARF activation in the Golgi and in the formation of phagosomes.
The genetic code that dictates the translation of RNA triplets into amino acids is rife with synonyms. Both CUG and CUA encode the amino acid leucine, for instance, and CCC and CCU encode phenylalanine. Each member of these synonymous pairs works equally well-the use of any particular synonym has no effect on the incorporation of its corresponding amino acid. But in mammalian genes, there is a surprising cumulative effect-genes with a greater proportion of third-position Gs or Cs are expressed more than genes with third-position As or Us. This effect is so pronounced that some researchers use the GC content of a multigene segment of a chromosome as a rough measure of its proteinproducing activity.
There is a long path from gene to protein, and it is unclear at what step along that path the GC versus AU difference exerts its infl uence. A

More GC Means More RNA
Richard Robinson | DOI: 10.1371/journal.pbio.0040206 new study by Grzegorz Kudla, Leszek Lipinski, and colleagues show that the difference is neither in translation of RNA into protein nor in RNA stability, but in production, or transcription, of RNA from the DNA template.
The authors began by comparing two genes that encode protein-stabilizing heat shock proteins, Hsp70 and Hsc70. The coding regions of the two genes are similar in length, and the proteins are largely similar in sequence. The "GC3" for Hsp70 is 92%, meaning it uses a G or C at 92% of its third positions, while the GC3 for Hsc70 is 46%. Within the cell, however, the genes also differ in their location in the chromosome, their promoters (which bind the RNA-synthesizing machinery), and other important determinants of gene expression. To control for these potentially confounding factors, the authors introduced the genes for Hsp70 and Hsc70 into cells in chromosome-independent vectors. They found the GC-rich gene produced ten times as much protein as the GC-poor one, an effect that was independent of the type of cell. This effect was not due to the rate of protein synthesis, which was the same when equal amounts of the messenger RNA for each gene was present. Neither was it due to differences in RNA stability, which was also similar for the two genes. Instead, the difference was due to production of RNA.
This effect was not limited to the heat shock proteins. Similar differences were seen with GC-rich and -poor versions of genes for the mammalian immune system protein interleukin-2, as well as for green fl uorescent protein, from the very nonmammalian jellyfi sh. Neither was it limited to the extrachromosomal position of Light and temperature exert signifi cant control over the daily lives of plants and animals. Many nocturnal species feed during the cover of night to escape a bevy of potential predators that hunt during the day. The effect of photoperiod (the 24-hour daylight cycle) is so great that zoos can trick nocturnal animals into becoming active for visitors simply by keeping their exhibit dark in the daytime and illuminating it at night.
But the world is fi lled with cues that could infl uence the daily feeding patterns of an organism. Many plants, for example, respond to foraging damage by releasing specialized chemical signals-volatile organic compounds that evaporate in the air-that attract the forager's natural enemies. This strategy is obviously no use against a cow, but proves effective when the offender is a caterpillar and the summoned predator is a wasp. Just how much control such biotic factors exert over a forager's daily routine has remained an open question. But in a new study, Kaori Shiojiri, Rika Ozawa, and Junji Takabayashi show that plant signals can indeed regulate herbivore behavior.
When the larvae of beet armyworms (Spodoptera exigua) feed on corn, the plant releases volatile compounds that act as a magnet for parasitic wasps (Cotesia marginiventris), which deposit their eggs in the larvae. Production of volatile chemicals increases during the day (when wasps are active) and decreases at night, suggesting that variations in production might affect the daily activity patterns of foraging larvae, with low production sending the signal that the coast is clear.
To test this hypothesis, Shiojiri et al. exposed larvae of a corn-munching caterpillar, Mythimna separata, to volatile compounds from corn and varied the light and dark conditions for both corn and insect. Corn infested with M. separata releases volatiles that attract parasitic wasps (C. kariyai). Previous work had established that M. separata is nocturnal, based on experimental observations of the larvae's daily feeding and hiding behavior on corn plants, but the impact of variable volatile production had not been examined.
In this study, the researchers separated the effects of photoperiod from that of host plant volatiles to tease out their relative contributions to caterpillar behavior. First, they tested the effects of light. If larvae are diurnal, they should hide in "shelters" fashioned out of fi lter paper attached to the plastic cups they were kept in. When larvae were fed an artifi cial diet, however, different light conditions produced no changes in their hiding behavior.
But introducing plants changed larvae behavior under both day and night conditions. Six pots of three uninfested corn plants-plants that had never been grazed-were placed around the cups of larvae. After eight hours, about 20% more larvae went into hiding when the lights were on and plants were added. And when plants were introduced under dark conditions, about 30% less larvae were found hiding than were found in the dark without plants.
To test the effect of plant volatiles directly, the researchers exposed larvae-some in the light and some in the dark-to a fl ow of volatiles collected from both uninfested and infested corn plants in light and dark conditions. When larvae in the dark were exposed to volatiles from uninfested plants, they hid in far greater numbers when the volatiles came from plants in the light than when they came from plants in the dark. And when larvae were in the light, far more hid when exposed to volatiles taken from plants in the light. Larvae responded similarly to volatiles taken from infested plants, though volatiles from infested plants in the light sent even more larvae into hiding.
These results demonstrate that it is not light that's controlling larval diurnal and nocturnal activity but volatiles released by the corn. Volatile compounds released during the day encourage hiding while those released at night indicate that it's safe to come out and eat. Just as parasitic wasps use plant volatiles to home in on potential victims, caterpillars use variations in their host plant's volatile production to reduce the risk of unpleasant encounters with wasps. Now that they've established volatiles' importance in infl uencing foraging behavior, the researchers plan to determine which compounds are responsible-and just how common insectplant communication may be. Any kid who spent summers playing in the woods knew Mom wouldn't let you back in the house without a headto-toe search for ticks-vectors for a wide range of pathogens throughout the world's temperate regions. In the United States, Rocky Mountain spotted fever was the main concern until 1975, when Lyme disease was found in Connecticut. Since then, incidence has skyrocketed from 497 cases reported in 1982 (the fi rst year national statistics were collected) to a record 23,763 in 2002.
Lyme disease, like other zoonoses, is transmitted by a vector that picks up the pathogen during a blood meal from a vertebrate host. In the eastern and central United States, the spirochete bacterium Borrelia burgdorferi infects blacklegged ticks, Ixodes scapularis, which feed on a wide variety of birds, lizards, and mammals, including mice, deer, and humans. Since human risk is a function of the prevalence of infection among vectors, outbreak prevention depends in part on understanding what controls infection rates among the agents of transmission.
In a new study, Richard Ostfeld, Felicia Keesing, and colleagues examined the ecological determinants of Lyme disease over a 13-year period in southeastern New York, a hot zone for the disease. Combining fi eld data with computer simulations, they analyzed trends in interannual variation and found two powerful predictors of entomological risk of Lyme disease in a given year: abundance of tick hostswhite-footed mice and chipmunks-in the previous year and abundance of acorns-which sustain the rodentstwo years out. Their fi ndings upset the long-held view that deer and climate are the best indicators of disease risk.
I. scapularis larvae hatch in midsummer, and acquire infection after feeding on an infected mouse or other small animal. Larvae detach after several days of feeding, then molt into nymphs and enter a nearly year-long dormant stage. After another round of feeding, nymphs fall off and molt into adults, which prefer the blood of larger mammals. Larvae  density of infected nymphal ticks, which is a product of the total density of nymphs and the nymphal infection prevalence. (Humans can reduce their personal risk by using repellents and routinely checking for ticks when visiting high-risk areas.) Many studies have examined variations in climate and whitetailed deer population dynamics as determinants of tick abundance and disease risk. But few have investigated the impacts of fl uctuations in the abundance of hosts for larval ticks, and none have examined all of these variables-temperature, precipitation, deer, mice, chipmunks, and acornssimultaneously over such a long period.
From 1991 to 2004, the researchers collected temperature and precipitation data, and estimated the abundance of acorns and animals on six plots of land. From this 13-year dataset, they developed computer models to estimate how each of the 11 variables (including multiple climate and deer indexes) contributed to yearly variations in the density of infected ticks and thus risk of human exposure.
While none of the climate variables infl uenced nymphal infection prevalence, higher temperatures in the previous year and precipitation patterns in the current year had weak, though unexpected, effects on total density and density of infected nymphs. It's thought that higher temperatures keep tick populations down, but the models showed them increasing both total density and density of infected nymphs. And though tick survival is expected to rise with precipitation, the models found the highest tick numbers at intermediate precipitation levels. These inconsistencies can be explored by incorporating other variables with documented effects into the approach outlined here. Also surprising, the researchers found that even a 3-fold variation in deer numbers had no impact on subsequent nymph abundance.
Density of infected nymphs-the principal determinant of Lyme disease risk-varied signifi cantly from year to year, fueled mostly by large fl uctuations in total nymph density, which in turn depended mostly on fl uctuations in abundance of acorns, mice, and chipmunks. Interestingly, though chipmunk densities are generally lower than mice, their numbers were the best predictor of total nymph density in the subsequent year, likely refl ecting their inferior grooming skills. Overall, the results found that acorns were the best predictor of Lyme disease risk-stemming from their crucial role in supporting white-footed mice, chipmunks, and likely other small animals, which in turn provide large reservoirs for B. burgdorferi. Acorns will not be a universal predictor of risk, the researchers acknowledge, since the disease occurs in areas without oaks. But the strength of these fi ndings suggests that the observed link between increased Lyme disease risk and high rodent densities indicates that important food sources-or predators-of the rodent hosts of nymphs will be valuable predictors of disease risk. Whether you are asleep or awake, engaged in fl ights of fancy, or engrossed in intense concentration, your brain is a hotbed of electrical and chemical activity. Nerve cells have channels that regulate the fl ow of ions in and out of their membrane. In a resting neuron, the potential inside the cell is negative, averaging about −70 millivolts, compared with the outside. Signals generated by a sensory stimulus drive the membrane potential to a less negative, depolarized, threshold value (about −55 millivolts). The neuron then releases a stream of electrical activity-as series of action potentials, or spikesalong its membrane. After fi ring an action potential, the membrane potential becomes even more negative than when at rest (called hyperpolarization), and the neuron resists fi ring for a variable period of time. The timing and fi ring rate of such spikes are all the information the brain has to represent a given sensory stimulus.
All the while neurons should be conveying precise signals, they are exposed to various sources of electrical noise. This noise causes "jitter"-tiny fl uctuations in the timing of any single action potential with respect to the signal that can accumulate and affect the accuracy of a transmitted message. On top of this noise, a neuron's resting potential oscillates around the resting value, well below the action potential threshold. Interestingly, and counterintuitively, increasing evidence suggests that this additional variability may help neurons separate signal from noise. In a new study, Troy Margrie and colleagues have used a combination of experimental and theoretical approaches to investigate the A mean peri-stimulus time histogram for two stimuli (blue and green) in the presence of a background oscillation. The gray shading represents the variance of the green stimulus; oscillations permit the discrimination of stimulus-specifi c situations (red arrows). role of membrane potential oscillations (MPOs) in signal processing. They show that oscillations in a single cell improve the precision of an action potential by reducing the accumulation of jitter that is inherent during ongoing spiking.

Membrane Oscillations Keep Neurons on the Right Track
It has been suggested that spike sequences (called spike trains) in hippocampal and olfactory cells adjust their timing to the phase of the cells' MPOs, thereby providing action potentials with a frame of reference that enhances signal processing. MPOs in the olfactory system are tied to the breathing cycle. To artifi cially manipulate the electrical signals in individual neurons and determine how MPOs affect action potential precision, the researchers recorded intracellularly from individual neurons in double tracheotomized mice. Since the overall timing of action potentials was the same for tracheotomized as compared with free-breathing mice, this meant the researchers could use the tracheotomized animals to compare action potential precision under oscillatory and nonoscillatory (control) conditions. They found that oscillations "greatly enhanced" the overall precision of the action potentials-though precision decreases with the number of preceding action potentials within an oscillation cycle.
To further investigate the source of this precision, the researchers recorded from mitral cells in olfactory bulb slices and explored the mechanism of enhanced precision by introducing alternating pulses of depolarizing and hyperpolarizing currents to simulate the natural oscillation cycle. Jitter accumulated with long periods of depolarization, but when a hyperpolarizing current was delivered between two action potentials, the second spike recovered the precision of the fi rst. Longer hyperpolarizing periods "dramatically reduced" the variance of the membrane potential-and yielded the most precise action potentials. These results support their in vivo evidence that spike trains within an oscillation cycle lose their enhanced precision since the trough of the oscillation is absent during such a spike train.
With enhanced precision comes enhanced performance. Exposing different types of neurons to over 77 stimulus pairs showed that MPOs signifi cantly increased a neuron's ability to distinguish temporally distinct stimulus patterns. Using simulations to vary the frequency and amplitude of the oscillations and the kinetics of the stimulus currents, the researchers found enhanced stimulus discrimination over "physiologically realistic ranges" of fi ring rates, input kinetics, and oscillation parameters. For low oscillation frequencies, however, stimulus discrimination is most robust when synaptic inputs arrived at the trough and early rising phase of an oscillation cycle.
By establishing a hyperpolarizing period, oscillations prevent jitter accumulation, which improves action potential precision and stimulus discrimination. That this feature occurs in different types of neurons suggests a relatively simple and general underlying mechanism, the researchers argue. Any system where inputs must be integrated and transformed into discrete outputs, from circadian rhythms to gene expression programs, would likely benefi t from any reset mechanism that minimizes the accumulation of noise.

Schaefer AT, Angelo K, Spors H, Margrie TW (2006) Neuronal oscillations enhance stimulus discrimination by ensuring action potential precision. DOI: 10.1371/journal.pbio.0040163
Molecular biology often seems like it leaves nothing to chance: signal A triggers reaction B, which activates enzyme C and silences gene Dleading cells to divide, differentiate, or die. But chance does sometimes meddle with molecular biology, as shown for instance by the apparently random inactivation (silencing) of the X chromosome. Indeed, in mammals, one copy of the X chromosome becomes silent early in the development of female embryos. As a result, female cells, which have two X chromosomes, produce the same amount of X-encoded molecules as male cells, which have only one. But while male cells have no choice as to which X to keep active, female cells do. That they leave that decision to chance is illustrated by the fact that female tissues are mosaics of cells with either the paternal or the maternal X chromosome in the silent state.
Precisely when chance comes into play during X inactivation is unclear. The prevailing theory is that the X chromosomes of female embryos are equivalent until the onset of inactivation, at which point the chromosomes' fates as active (Xa) or inactive (Xi) become sealed through chance selection. But Susanna Mlynarczyk-Evans, Barbara Panning, and colleagues challenge this theory with a new study that shows that X chromosomes oscillate between two states with different potentials for inactivation before they commit to the Xa or Xi fate. Their observations suggest that random X inactivation refl ects a much earlier randomization of X chromosome states than previously thought.
Mlynarczyk-Evans, Panning, and colleagues studied mouse embryonic stem (ES) cells, which mimic in culture the early growing stages of embryonic development, before differentiation triggers X inactivation. Using fl uorescent probes specifi c to various X-encoded genes, the researchers observed that in approximately half of female ES cells, a gene would appear as a single spot (S) on one X and a doublet (D) on the other. Moreover, nearby genes on the same chromosome
most often displayed the same S or D status. In any culture of actively dividing cells, one might expect a gene to appear as a singlet before DNA replication, and a doublet after replication. But using a variety of tests, the researchers demonstrated that S signals persisted after X chromosome replication. On the other hand, S signals were sensitive to experimental procedure, and split into two when cells underwent a staining protocol known to disrupt chromosome architecture. The researchers concluded that the X chromosomes of proliferating, undifferentiated ES cells exist in two different physical states.
Curious about what these different states might portend, the researchers repeated their experiments in ES cells carrying various mutations that bias X inactivation. For instance, in cells that harbor a wild-type X and an X mutant for the Xist gene, the wild-type X always becomes the Xi. In a large proportion of such cells, the researchers saw that the wild-type X had mostly D signals, and the mutant mostly S signals. Similar biases were observed with other mutations, suggesting that S and D signals (and the different physical states they reveal) might be early markers of the Xi and Xa fate.
But if wild-type X chromosomes are already predisposed for a future Xi or Xa fate before inactivation begins, how is randomness achieved? A possible answer to that question is that preinactivation states are unstable. To address this possibility, the researchers established cultures from single ES cells that harbored two wild-type chromosomes, one of which was easily distinguished from the other because it had fused to Chromosome 2. Each founder cell presumably carried either the normal or the fused X in the pre-Xi or pre-Xa state. Yet, in the progeny of each founder, both chromosomes were found in either state in equal proportions. Hence, X chromosomes seem able to switch from one state to the other during cell divisions, quickly achieving a random distribution of the pre-Xi and pre-Xa states.
Mlynarczyk-Evans, Panning, and colleagues do not yet understand the biochemical mechanisms that allow X chromosomes to switch states coordinately in dividing ES cells. But their observations suggest that the differentiation signal that triggers X inactivation might not initiate randomness, but simply stabilize the state of already randomized, shapeshifting chromosomes. In some ways, the developmental passage of a fruit fl y from a ball of cells through discrete morphological stages-embryo, pupa, and larva-to a fully formed organism really showcases the potential of biology. Organism development (in this case of a fl y) requires a temporally synchronized regulation of a myriad of genes and pathways across the Drosophila genome.

Mlynarczyk-Evans
No single component of these developmental processes can be considered more important than another-a mistake anywhere can prove fatal. One set of participants in Drosophila development are the Polycomb group (PcG) proteins, which bind to regions of chromatin (the genomic DNA twisted around histone proteins that helps structure the chromosomes) and silence the transcription of nearby genes. A handful of PcG response elements (PREs-the regions of the genome where PcGs bind) have been characterized to date, but until now not much was known about their broader distribution across the genome. Nicolas Nègre, Giacomo Cavalli, and colleagues provide just such a map of PcG binding locations across chromosomes X and 2 of Drosophila melanogaster. They observe a wide distribution that varies from one developmental stage to the next.
PcG proteins actually form multiprotein complexes, one of which-the PRC1 complex-contains two specifi c proteins, PC and PH. PC and PH have in the past been seen to co-localize with another protein called the GAGAfactor, or GAF. Nègre and colleagues mapped the binding locations of individual PC, PH, and GAF proteins (separately) to DNA from embryos, pupae, and adult male and female fl ies. They did this with a technique called "ChIP on chip," which identifi es matches between the region of chromatin where the protein of interest has bound (via chromatin immunoprecipitation, or ChIP) and the corresponding genomic region on a tiled microarray (or chip).
To double-check that their matches were sensible, some known PREs were placed on the microarrays (acting as positive controls), along with some randomly selected DNA fragments unrelated to any known PREs (negative controls). To confi rm that the chromosomal locations of

Polycombing Drosophila for Dynamic Developmental Footprints
Emma Hill | DOI: 10.1371/journal.pbio.0040177 DOI: 10.1371/journal.pbio.0040177.g001 Mapping the chromosomal distribution of Polycomb group proteins in Drosophila to study the developmental maintenance of cell identity.
the ChIP-on-chip matches corresponded with the correct region of the Drosophila genome, the authors also undertook some fl uorescence mapping experiments and, reassuringly, observed perfect co-localization of sites mapped by ChIP on chip with sites mapped on the chromosomes by cytology (observing the band on the chromosome that fl uoresces).
So how do these proteins distribute over the genome? As expected, PC and PH showed almost identical distribution, which confi rms their presence together in the PRC1 complex. GAF binding did not overlap perfectly with the PRC1 proteins and actually bound to more sites in total. GAF also tended to bind to narrower chromatin regions, while PC and PH were more often spread over larger regions. These results indicate that PC and PH binding interactions do not depend on GAF.
Interestingly, the PRC1 proteins frequently bound in clusters, and so it seems that in some genomic regions, more than one PRE must exist in proximate locations. The genes nearest to the PREs, as expected, are most frequently biased toward regulatory and developmental functions. Their distances from the PREs, however, are variable, and it is likely that the PcGs must use a variety of mechanisms to turn off transcription of these nearby genes (perhaps by binding on top of the genes' promoter region, or maybe by linking multiple PREs to interfere with more distant promoters).
How did the binding of these proteins differ across the different developmental time points? While the positions of GAF localization remained quite constant, the PcG protein binding sites did not. Although some PREs were maintained at all times, other PcG protein locations varied considerably, implying that PcGs employ dynamic regulation over time. In fact, adult male fl ies had the most divergent profi le of PREs, suggesting that some male-specifi c development might be regulated by these PcG proteins.
Nègre, Cavalli, and colleagues show that these PcG proteins do indeed constitute a critical component of Drosophila development and are likely to instruct a variety of developmental processes through their regulation of genes across the genome. The next exciting steps in better understanding the molecular construction of a fl y will be to extend this analysis to the remaining Drosophila chromosomes, across other developmental stages, and in different tissue types. From there, scientists can similarly probe what makes a mammal, and the extent to which these mechanisms are conserved across distant species. The notion that bacteria and other single-celled "prokaryotes" carry only one genome copy per cell is as ingrained as the notion that they lack a distinct nucleus. (Organisms with one genome copy are "haploid"; organisms with two, including humans, are "diploid"; and organisms with more than two, such as plants, are "polyploid.") This assumption is based on decades of research on Escherichia coli and a select group of other bacterial species.
But researchers have been fi nding indications that some bacteria, under certain conditions, have more genetic material in a single cell than would be expected for a single genome. And now, using a combination of cell sorting, fl uorescent microscopy, microarray, and quantitative PCR techniques, Deborah Tobiason and Hank Seifert present strong evidence that the bacterium that causes gonorrhea, (Neisseria gonorrhoeae), is indeed polyploid.
N. gonorrhoeae successfully infects at-risk individuals by reshuffl ing the genetic code for proteins expressed on its hairlike "pili" protrusions through a process called gene conversion. During gene conversion, one gene incorporates DNA from a variant copy that carries slightly different information. The high frequency of gene conversion events enhances pili protein diversity, which increases the likelihood that the human immune system won't detect the pathogen. It is conceivable that N. gonorrhoeae might briefl y express two copies of the pilin gene in one cell after DNA replication, before cell division. But Tobiason and Seifert thought that the high frequency of gene conversion required for antigenic diversity might instead be supported by the presence of extra genome copies.
To explore this possibility, the researchers had to distinguish the amount of DNA associated with genome duplication prior to cell division from that found in a nondividing cell. They labeled the DNA of N. gonorrhoeae with fl uorescent markers and, as a source of comparison, did the same for E. coli, which has provided a wellspring of knowledge on chromosomal DNA replication. Cells were either left untreated, allowing exponential growth, or were treated with an antibiotic that inhibits new rounds of replication but allows any DNA replication already in progress to fi nish. Cells in "stationary phase"-during which cell division stops and no new DNA replication begins, but active replication forks complete their task-were also labeled. Both antibiotic treated cells and stationary cells produce fully replicated chromosomes.
Using a standard cell sorting technique called fl ow cytometry, the researchers separated cells based on their DNA content. Treated and stationary phase E. coli cells produced DOI: 10.1371/journal.pbio.0040212.g001 The obligate human pathogen, Neisseria gonorrhoeae, has multiple genome copies per cell, which may be necessary for antigenic diversity and survival within the human host.
individual cells with the equivalent of two, four, or eight genomes per cell. Since multiple initiation events on the chromosome can occur per round of cell division, the number of replicated genomes refl ects the number of replication forks active when cell division stops. Untreated E. coli cells, however, revealed a wide range of DNA content per cell, likely refl ecting the presence of multiple replication forks at various stages of replication and cell division. The DNA content of untreated gonococcal cells also varied, with most cells having between four and six genome equivalents.
Gonococcus exists as single (monococcal) and double spheres (diplococcal)-it is these two forms, Tobiason and Seifert show with fl uorescence microscopy, that explains the range of DNA content in different cell populations found with fl ow cytometry. Monococci had, on average, about three to six genome equivalents, while diplococci averaged about four to ten per cell. To explore the reason for the expanded DNA content, the researchers treated gonococcal cells with antibiotics to generate fully replicated chromosomes. These experiments suggested that actively dividing cells have two or four pairs of active replication forks. They used independent measures to confi rm that both monococci and diplococci contain fully replicated chromosomes with a single pair of active replication forks on each chromosome-which means that the DNA content they saw in gonococcal cells is not the result of multiple replication initiation events on a single chromosome, as occurs with E. coli, but represents multiple, fully replicated chromosomes. Future studies will have to determine whether polyploidy underlies pilin antigenic variation and the bacterium's success as a human pathogen. But this study leaves little doubt that it's time to retire the notion that all prokaryotes are haploid. Starting at around three months old, children can fi nally reach for the countless toys their parents have been dangling before them since birth. These attempts often involve a good deal of fl ailing about, as motor skills, like anything else, require cultivation. Motor control depends on executing the proper musculoskeletal force to reach the desired object.
Prior learning facilitates motor control (all that fl ailing serves a purpose), which is aided by a fundamental feature of memory, called savings. When a novel response to a stimulus is learned in one set of trials, then "washed out" in an unlearning phase, subsequent relearning proceeds faster. Neuroscientists have been puzzled by savings and other features such as interference and rapid unlearning reported in adaptation studies because standard models of short-term motor adaptation couldn't explain them. But now Maurice Smith, Ali Ghazizadeh, and Reza Shadmehr have solved this puzzle by combining experimental and computational approaches, and show that two adaptive brain components with different learning rates and retention capacities can account for these seemingly disparate properties of learning.
In a recent study of savings in eyesaccade adaptation (saccades are rapid eye movements that shift gaze direction) in monkeys, Yoshiko Kojima et al.

A New Model of Short-Term Motor Adaptation
Liza Gross | DOI: 10.1371/journal.pbio.0040205 DOI: 10.1371/journal.pbio.0040205.g001 Two learning processes with distinct time courses contribute signifi cantly to short-term motor skill acquisition, and the interactions between these processes explain a host of puzzling phenomena in motor adaptation. explained the savings they observed with a model in which two distinct mechanisms increased or decreased the amplitude gain of saccades based on a target's direction. This "two-state, gainspecifi c" model explained the savings and washout effects the researchers had seen. But Kojima et al. found that when adaptation was extinguished, and then time passed with the monkey receiving no feedback on its behavior, there was a sudden recovery of the animal's saccade gains toward the initial learning state. This spontaneous recovery could not be explained by the gain-specifi c model. Nor, Smith et al. explain, can the gain-specifi c model explain a property of learning called anterograde interference, in which initial motor learning is faster than subsequent adaptation to an opposing task.
Smith et al. focused on the spontaneous recovery phenomenon: since behavior changed even when there was no error feedback as a guide, they reasoned that motor error could not be the only factor affecting motor learning and that passage of time must have also played a role. They considered a system in which two states both learned from error: one learned rapidly but tended to quickly forget, while the other learned slowly and tended to remember.
To test their hypothesis, they fi rst designed a variation on a standard paradigm for studying how humans learn to compensate for an imposed force while reaching for a target. Participants held a joystick-like, robotcontrolled "manipulandum," which measures hand position and velocity, and applies well-controlled counter forces to the hand. Participants' hand position was visible on a computer monitor in front of them, and they were told to quickly reach toward small circular targets spaced ten centimeters apart along a horizontal plane. After practicing without the robot motors, participants adapted their reaching movements to a force-fi eld perturbation pushing their hand in one direction; then learning was extinguished by adaptation to an opposing force fi eld. Error feedback was then removed with an "error clamp" that forces the hand along a straight line to eliminate lateral errors during movement. Even though participants were kept from experiencing lateral errors, the manipulandum recorded their lateral force resistance to the error clamp, so the researchers could infer participants' motor plans and track the evolution of learned changes in motor output.
As predicted by the model, the observed spontaneous recovery suggested that two different learning states-one with a fast time scale and another with a slower time scale-combined to produce motor output. Rebound occurs because the fast-learning module decays quickly during the no-feedback block, but the slow-learning module decays gradually, allowing a transient recovery of motor output.
Altogether, these results argue for a learning system based on two time courses and explain a wide range of phenomena associated with short-term motor adaptation. The multirate model accounts for anterograde interference, by showing that the slow learning component is initially biased against learning the second adaptation. And it explains "rapid unlearning"-in which the rate of de-adaptation after a period of learning can be much faster than initial learning-and additionally predicts that this de-adaptation is fastest after short adaptation blocks, then declines as the amount of training in the initial adaptation increases.
It may well be that different functional units in the brain learn motor control at different rates-a possibility that researchers can explore with the help of this unifying model of motor adaptation. And by combining behavioral experiments with targeted brain lesions and functional imaging in the cerebellum-which is required for normal motor adaptation-researchers can investigate what regions of the brain control these learning modules.

Smith MA, Ghazizadeh A, Shadmehr R (2006)
Interacting adaptive processes with diff erent timescales underlie short-term motor learning. DOI: 10.1371/journal.pbio.0040179 The main argument for planting transgenic, pest-resistant crops rests on their potential to reduce widespread pesticide spraying. Such benefi ts require that agricultural land-use strategies impede the evolution of pest resistance. But such strategies-including the high-dose refuge (HDR) strategy mandated in the United States-are based on key assumptions about pest biology. Insect fi tness depends on fi nding the optimal balance between when to mate and when to set off for greener pastures-mating before dispersal increases the risk of inbreeding, but dispersing before mating increases the risk of never fi nding a mate. Pest management depends on understanding and using this balance to prevent the evolution of insect resistance.
As important as mating and dispersal are to thwarting resistance, empirical evidence on these behaviors in major crop pests-including the European corn borer (Ostrinia nubilalis), targeted by Bacillus thuriengiensis (Bt) toxins produced by transgenic insecticidal corn-is limited. While European corn borers can disperse over half a mile-the maximum distance allowed between a Bt fi eld and the corresponding refuge-and even up to 20 miles, it's unclear whether they do so routinely, take fl ight before or after mating, or show any preference for resident or immigrant mates.
To gain a better understanding of the European corn borer's mating and dispersal habits, Ambroise Dalecky, Sergine Ponsard, Richard Bailey, Céline Pélissier, and Denis Bourguet released and recaptured moths in French crop fi elds over three breeding seasons in 2004 and 2005. They found that females mate randomly with resident and immigrant males-an important HDR assumption, since random mating dilutes resistance genes. But they also discovered that some males and females mate before dispersing-suggesting a possibly more restricted mating range than previously assumed.
The HDR strategy assumes that a patchwork of Bt cornfi elds interspersed with untreated refuges will facilitate gene fl ow between the areas. Resistance gene variants, or alleles, are found in either one (heterozygotes) or two (homozygotes) copies in individuals who carry them. Only homozygotes are resistant to (that is, survive the Bt toxins), while heterozygotes die as larvae if they feed on such plants. HDR management aims to eliminate as many resistance alleles from each generation as possible by maximizing the proportion of heterozygotes (where resistance alleles are available for purging) and maximizing the proportion of these heterozygotes developing in Bt fi elds (where resistance alleles are indeed purged), rather than in refuges. If substantial mating takes place before dispersal, there will be more homozygotes, which means fewer purging opportunitiesand a potentially less effective HDR strategy. If crops are not rotated and if post-mating dispersal is low, this deleterious effect of low pre-mating dispersal can be offset by having a higher proportion of heterozygous eggs laid on Bt corn plants and dying at larval stage. In other cases, however, low

Dispersal and mating patterns in European corn borer moths: a color-marked female and a biogeochemically marked spermatophore (shown at about the same scale). (Image: Dalecky et al., Laurent Pélozuelo, and Laurent Soldati)
pre-mating dispersal is predicted to reduce the effi cacy of the HDR strategy.
To track moth mating and dispersal habits, the researchers relied on biochemical markers-consisting of different carbon signatures produced by wheat and corn, which are in turn evident in lab-raised, wheat-fed moths, and corn-fed fi eld moths-and color-coded ink spots applied to adults before release. Male and female pupae were separated to prevent mating before they were released less than 24 hours after they emerged from their pupal case.
In 17 sessions over three weeks, nearly 8,800 virgin moths were released during the June 2004 breeding season into four different cornfi elds. Soybean and wheat-crops that aren't targeted by corn borers-grew in the fi elds the year before, so no pupae were present as if the fi eld had been planted with Bt corn. Refuge facsimiles were provided by adjacent fi elds planted with corn in 2003, the most likely source of wild moths at the 2004 study sites.
Thirty-six hours after release, marked moths and unmarked moths were captured along the fi eld borders, where the moths typically mate and rest. The proportion of marked individuals that were recaptured varied signifi cantly between sites and sessions but not between sexes, averaging about 4%. About 97% of recaptured females had mated-indiscriminately with resident and immigrant males, the researchers determined, by analyzing the carbon content of the sperm packets, or spermatophores, they acquired during mating. Resident males rarely mated with immigrant females, perhaps because immigrant females tended to arrive already mated.
Because wild moths might disperse sooner than 24 hours, the researchers placed corn borer pupae directly in a cornfi eld the following season, then watched and waited. Moths fi rst took wing about 3.5 to 9.5 hours after emerging, and their fl ight movements were unlikely to take them farther than 220 feet within the fi rst 24 hours-indicating that high rates and high distances of dispersal were unlikely to occur before mating and that far more than 4% were still present and hence likely to mate within their natal range.
The researchers go on to confi rm that a "substantial proportion"-up to about 57% for females-of newly emerged corn borers mate at a very local scale before dispersing. These fi ndings suggest that the HDR strategy may not ensure substantial gene fl ow between susceptible and resistant moths. This disconnect between the evidence and the HDR model indicates that other assumptions-including whether females who mate close to home also lay their eggs there-should be rigorously tested before counting on HDR to circumvent the resistance machinery of its targets. In most cases, a fertilized egg with two X chromosomes will become a girl and an XY egg will become a boy. Molecular analysis of the genomes of sex-reversed individuals identifi ed the sex-determining region Y gene (Sry), the primary sex-determining gene in mammals. The presence of an Sry gene normally drives male development. But when downstream elements of the sex-determining pathway are disrupted, chromosomal sex and sexual phenotype can be discordant. This phenomenon revealed a fatal fl aw in a now-defunct Olympics rule requiring female athletes to submit to "gender verifi cation" procedures that labeled XY females "imposters." Every embryonic gonad-at this point, neither ovary nor testis-harbors cells with the potential to choose either fate (male or female). Refl ecting this noncommittal approach, both XX and XY gonads initially display similar expression patterns for the sex-determining genes Sox9, Fgf9, and Wnt4. If Sry does not intervene during a critical window in development, the cells will default to the ovarian pathway. Once Sry expression begins, other genes start to choose sides: Sox9 and Fgf9 become active only in male gonads, while Wnt4 becomes active only in female gonads. Loss of Fgf9 in XY mice leads to sex reversal; in XX mice, loss of Wnt4 leads to partial testis development. Only when the protein products of Sry, Sox9, and Fgf9 are all expressed together do male-specifi c Sertoli cells developand prod the developing gonad toward full male differentiation.
Clearly, Fgf9 and Wnt4 must somehow infl uence sex determination-but how? In a new study, Yuna Kim, Blanche Capel, and colleagues explored how Sry, Sox9, Fgf9, and Wnt4 contribute to the initial events of sex determination in mice. By manipulating the activity of these signals, they show that Wnt4 and Fgf9 play a sexual tug of war. If Wnt4 wins, the gonad differentiates into an ovary. If Fgf9 wins, it turns into a testis. Living up to its name, Sry tips the balance in favor of Fgf9, by triggering expression of Sox9, and a male fate.
The researchers started by examining Fgf9 expression during normal gonad development in the mouse. 11.5 days after conception, Fgf9 was evenly distributed in the gonads of both sexes. By day 12, it was found only in XY gonads, and only in Sertoli cells within testis cords and cells near the surface of the gonad.
Seeing Fgf9 expressed so early in testis differentiation suggested it might trigger Sry activity or expression, but the researchers found that this wasn't the case. If Fgf9 doesn't regulate Sry, they reasoned, maybe it affects the malespecifi c up-regulation of Sox9, which is absent in XY mice lacking Fgf9. Adding

Fgf9 is critical to repress Wnt4 and establish the testis pathway during mammalian sex determination. Sertoli cells (red), germ cells, and endothelial cells (green).
Fgf9 to XX cells isolated from 11.5-dayold gonads triggered increased Sox9 expression. These fi ndings showed that artifi cial expression of Fgf9 can induce Sox9 expression in single XX cells, but what about in the intact gonad?
In normal and mutant Fgf9 XY mice, Sox9 expression followed similar patterns for the fi rst 24 hours after Sry was expressed. By 12.5 days, SOX9 proteins were no longer expressed in the mutant mice, and the cells that would normally have developed into Sertoli cells failed to organize into normal testis cords. These results, the researchers conclude, indicate that Fgf9 isn't necessary to trigger Sox9 expression, but instead is required to maintain Sox9 expression in Sertoli precursor cells.
Fgf9 expression depends on Sox9, and Fgf9 returns the favor by maintaining Sox9 expression in a positive feedforward loop. The feed-forward loop between these genes allows the proliferation and expansion of Sertoli and other somatic cells in XY gonads. Without Fgf9, XY gonads showed a rapid drop in SOX9 protein levels and expressed few or no Sertoli cell markers. Wnt4, on the other hand, increased, suggesting it was repressed by Fgf9. Without the antagonistic effects of Fgf9, Wnt4 predominates, disrupting the Fgf9-Sox9 pathway and establishing the infrastructure necessary to become female. And the reverse holds true: reducing Wnt4 levels in XX gonads allows FGF9 and SOX9 levels to increase-even in the absence of Srythough other male factors appear to be required to establish testis development.
Altogether, these results suggest that sex is not determined by the fl ip of a genetic switch but by a dose-dependent interplay between opposing signals. The researchers propose that Sry initiates the male fate by triggering Sox9 expression, which activates Fgf9, and sets off a feed-forward loop that produces a male, barring the interference of antagonistic signals. Whether these signals work within or between cells is not clear. It may be, for example, that Fgf9 recruits other cells to the Sertoli pathway by triggering increased Fgf9-Sox9 expression, or that Fgf9 causes Sertoli cells to proliferate until they reach the critical mass needed to fend off Wnt4 and secure the male pathway. Future studies can investigate the molecular mechanisms of Fgf9 and Wnt4-and whether these competing pathways regulate the battle of the sexes in a wide range of vertebrates. It's not easy coming out ahead in the virus-host battle: the mechanisms living things have for seeking and destroying invaders, as elaborate as they might be, often fi nd themselves up against equally elaborate viral mechanisms for evading them.
Such is the case when it comes to humans versus herpes simplex virus type-1 (HSV-1), a highly contagious virus that is virtually impossible to get rid of once it becomes established. When our bodies detect HSV-1 inside us, they mass produce antibodies called immunoglobulin G (IgG). These Y-shaped defense molecules latch onto antigens on the surface of the virus and virus-infected cells with their two arms, disabling them and marking them for pickup by the immune-system equivalent of roving trash trucks. But HSV-1 has a cagey way of turning the tables. A protein pair that it stations on its surface (and on the surface of cells it has taken over) known as gE-gI can grab IgG by its tail (known formally as its Fc region), incapacitating it instead.
Elizabeth Sprague, Pamela Bjorkman, and colleagues explored this interaction by focusing their attention on CgE, a portion of gE. Using an assay that measures how well two substances stick to each other, they showed that CgE binds Fc even in the absence of gI or other gE domains. They then used a technique that compares how crystals of CgE diffract X rays of various wavelengths to determine the threedimensional structure of CgE.
Next the researchers attempted to create a threedimensional picture of the gE-gI/Fc complex. Using the new information about CgE's structure, low-resolution images of gE-gI/Fc crystals, and techniques that make it possible to infer complete structure from partial information, they proposed a structure for the gE-gI/Fc complex in which two CgE components of gE-gI hold Fc like two hands holding a basketball. To substantiate that surmised structure, they applied a computational approach called protein docking, using what they knew about the structure of unbound CgE and Fc to calculate what complex would be energetically most favorable. Of the fi ve plausible models they came up with, one was remarkably similar to the structure derived through the previous process, confi rming its likelihood as the correct structure.
The researchers then used the newfound structural information to further explore the gE-gI/Fc interaction. They compared CgE with other proteins and a peptide of known structure that bind with the same region of Fc. They also used the structural information to explain on a molecular level the previously known inability of an Fc

Trapped: A Molecular-Level Look at How Herpes Disables Its Immune Destroyer
Mary Hoff | DOI: 10.1371/journal.pbio.0040189 DOI: 10.1371/journal.pbio.0040189.g001 The three-dimensional structure of a gE-gI/Fc complex suggests that an anti-HSV IgG bound by gE-gI is in an upright orientation, positioning it to also bind its antigen on the surface of the same cell. mutant, another human immunoglobulin, and rodent IgG to bind with gE-gI.
One fascinating mystery the structures helped the researchers address is the fact that gE-gI binds well with Fc at neutral or slightly basic pH, but not in an acidic environment. The researchers showed that CgE-Fc binding is pH dependent, and attributed this trait to four histidine amino acids at the CgE/Fc interface that would likely alter the complex's chemistry in the presence of spare protons. They also speculated that the pH sensitivity was part of a viral strategy for attacking IgG in which gE-gI/IgG is drawn into the cell where the acidity causes the antibody to dissociate and eventually to be destroyed.
The structural studies also shed light on previous fi ndings regarding the effects of mutations in gE on its ability to bind IgG and on cell-to-cell spread of HSV-1. Using their knowledge of CgE and gE-gI/Fc structures, the researchers were able to identify the structural and functional implications of various mutations and to pinpoint specifi c regions of CgE implicated in cell-to-cell spread.
Finally, the researchers used the structures to determine that gE-gI/IgG complexes likely orient upright in the cell membrane, leaving the arms of the gE-gI-bound IgG available to attach to a nearby HSV-1 antigen-a previously hypothesized state known as bipolar bridging. If bipolar bridging does indeed occur, subsequent endocytosis would then swallow up not only IgG but also any antigens involved in bridging, further hindering efforts to win the battle against this tricky viral invader. Long before anyone realistically entertained the notion of engineering transgenic corn, cloned dogs, or designer babies, scientists looked for clues to the fundamental properties of life in much simpler organisms. Studies in one such organism-bacteriainfecting viruses called bacteriophages, or phages-produced some of the most important discoveries of molecular biology: identifying DNA as the genetic material, elucidating the mechanism of genome replication, and revealing the power of recombinant DNA, the backbone of genetic engineering.
Nearly 60 years after phage's golden era, scientists are still using the virus to illuminate the inner workings of the cell. And now in a new study, Pallavi Ghosh, Laura Wasil, and Graham Hatfull describe a novel phage protein that uses an unusual mechanism to induce the infectious stage of the phage life cycle.
Lacking the means of self-replication, phages inject their genome into a bacterial cell and co-opt its molecular machinery to mass produce new viruses, ultimately destroying the cell to release its progeny. Temperate phages can reproduce through this lytic life cycle or enter a lysogenic life cycle during which they integrate their DNA into the bacteria's genome as a noninfectious "prophage." The prophage remains harmless until it is excised from the bacterial genome.
Both integration and excision depend on the combined efforts of integrases and recombination directionality factors (RDFs) in a cut-and-connect operation called site-specifi c recombination. During integration, integrases mediate DNAstrand exchange between phage DNA-at the phage attachment site, attP-and bacterial DNA-at the bacterial attachment site, attB. Successful integration creates junction sites-attL and attR-on both sides of the integrated prophage that mark future excision sites. RDFs can catalyze integration or excision, typically by binding to DNA near the respective recombination sites and forming a complex with integrase that remodels the DNA and creates macromolecular structures that promote one process while inhibiting the other.
While much is understood about how tyrosine integrases (so-called because the amino acid tyrosine mediates the catalytic reaction) effect site-specifi c recombination, Ghosh et al. were interested in the more cryptic serine integrases. Serine integrases lack many of the features other enzymes use to regulate site-specifi c recombination, such as DNA-binding cofactors or DNA remodeling, and little is known about the few RDFs that have been linked to them. To learn more about serine integrase activity, the researchers worked with a phage called Bxb1 that uses a serine integrase to insinuate itself into the genome of the bacterium Mycobacterium smegmatis.
They fi rst analyzed the Bxb1 genome to identify candidate RDFs using standard bioinformatic methods, but RDFs are too diverse to allow accurate sequence comparisons. So they turned to an experimental method to isolate the RDF, using a phage strain with genes organized akin to an integrated prophage, complete with recombination junctions (attL and attR) and integrase (int). From this strain, they engineered an "excision tester strain" with one stretch of sequence conferring antibiotic

A Novel Phage Protein Mediates the Virus's Removal from Bacterial Chromosomes
Liza Gross | DOI: 10.1371/journal.pbio.0040213 DOI: 10.1371/journal.pbio.0040213.g001 The mycobacteriophage Bxb1 uses a novel enzyme to remove itself from its bacterial host's genome before launching an infection.
resistance and another conferring sucrose sensitivity (thus, stunting growth) inserted between the attL and attR junctions. Successful excisive recombination would remove the intervening sequence and confer sucrose resistance, allowing the strain to grow in sucrose.
To identify excision-related sequences, Ghosh et al. created a Bxb1 DNA library and incorporated it into the tester strain, producing colonies of genetically transformed strains. The resulting colonies produced sucroseresistant strains, which had undergone site-specifi c excision recombination between attL and attR, the researchers confi rmed, which had been mediated by a fragment of Bxb1 DNA and a putative RDF. And by creating genetic variants of the tester strains, they showed that excisive recombination requires both the Bxb1 integrase gpInt and a candidate Bxb1 RDF.
Only one region in this putative RDF, gp47, had been previously described. And though it is much larger and unlike any other known RDF, the researchers show that gp47 does indeed catalyze excisive recombination and can inhibit integration, like other RDFs. Bxb1 gp47 distinguishes itself, however, by regulating site-specifi c recombination through novel interactive mechanisms that don't require complex macromolecular structures or direct DNA binding. Instead, the researchers propose, it instigates excisive recombination through transient interactions with gpInt-attL/R junctions-and blocks integrative recombination through more stable interactions that prevent gpInt-attB and gtInt-attP complexes from forming productive associations.
Altogether, these results point to a novel RDF with an unusual strategy for regulating integration and excision. Though the gp47 RDF acts through protein-protein interactions rather than through direct DNA-binding interactions, its effects depend on the specifi c DNA site that gpInt binds. With evidence that gp47 also plays a role in DNA replication, the researchers plan to investigate whether this novel enzyme uses the same strategy to interact with other phage proteinsand maybe even to commandeer host machinery and unleash infection. As it rolls off the DNA assembly line, RNA faces an uncertain fate. Down one path lies further processing and, ultimately, translation into protein.
Down the other, though, lies obliviondegradation by RNA-digesting enzymes, and recycling back into the nucleotides from which it came. A new study by Martin Hicks, Klemens Hertel, and colleagues shows that effi cient processing is promoted, and oblivion avoided, by a close linkage between the machinery of RNA synthesis and that of RNA processing.
Making a protein from a DNA gene requires fi rst making an RNA copy, a process handled by the enzyme complex RNA polymerase II (pol II). This product, called pre-messenger RNA (pre-mRNA) gets a cap and tail, which protect it from exonucleases, enzymes that could otherwise degrade the RNA starting from its ends. They do not protect it from endonucleases, however, which can attack the middle of the RNA strand and render it useless. Endonucleases are prevented from attacking the pre-mRNA when it is bound by the splicing machinery, another enzyme complex that removes noncoding bits, called introns, and splices together the remaining exons.
To investigate whether pol II helps promote binding of the splicing machinery and thus prevent pre-mRNA destruction, the authors compared pol II, found in all animal cells, with a viral RNA polymerase, T7, in otherwise identical cell-free systems. Both generated similar amounts of pre-mRNAs, but spliced mRNAs were created much more effi ciently in the pol II system. This might be due either to faster binding of the splicing machinery to the pre-mRNA or to faster processing of the RNA once it is bound. By analyzing the reaction mixture at different time points, the authors found that binding, not splicing, is the key step affected by the presence of pol II.
This suggests that pol II, while it is not part of the splicing machinery

RNA Polymerase Promotes Splicing, Prevents Degradation
Richard Robinson | DOI: 10.1371/journal.pbio.0040183 DOI: 10.1371/journal.pbio.0040183.g001 RNA synthesis by Pol II promotes the recruitment of the spliceosome to exon/intron junctions. The rapid association of RNA processing factors with the pre-mRNA prevents its premature degradation.
per se, somehow increased the ability of the splicing machinery to fi nd and latch on to the pre-mRNA, thereby preventing RNA degradation and ultimately promoting splicing. They found further support for this conclusion by showing that functional splice sites on the pre-mRNA were needed for this effect to occur. Finally, they built a mathematical model of the entire range of RNA interactions, from synthesis to splicing or degradation, and showed that the key variable in determining the relative concentrations of the various RNA species was the affi nity of the splice machinery for the pre-mRNA, highlighting the importance of this step in controlling the fate of the RNA.
This study did not address whether pol II and splicing are functionally linked-that is, whether some structural property or dynamic activity of pol II is coupled with the splicing machinery, or, instead, whether the mere proximity of the two ensures effi cient transfer of RNA from one to the other. But other studies have highlighted the importance of pol II's C-terminal domain. Located at one end of this molecular behemoth, this region appears to promote other RNA processing events, including addition of the protective cap, and some splicing factors are known to associate with this domain. Further investigation of these interactions may reveal just how intimately linked transcription and splicing are. Though scientists recently downsized estimates of insect species from over 30 million to around 4 million to 6 million, it's safe to say that insects are among Earth's most numerous and diverse group of organisms. At least some of this success derives from mutually benefi cial interactions with bacteria, established as long as 270 million years ago, that allow insects to thrive in otherwise unsuitable niches. These symbiotic bacteria-called endosymbionts because they live inside cells-synthesize essential nutrients for their hosts, while the insects provide an ecological niche for the bacteria.
In a new study, Dongying Wu, Jonathan Eisen, Nancy Moran, and colleagues used comparative genome analysis to investigate the inner life of a widespread agricultural pest, the glassy-winged sharpshooter (Homalodisca coagulata). As the vector of Pierce disease (caused by the Xylella fastidiosa bacterium), the sharpshooter poses a major threat to California crops, having already destroyed an estimated $14 million of Southern California grapevines. It lives on xylem sap-which distributes salts and water throughout the plant but provides very few organic nutrients to the insect-and houses two endosymbionts, Baumannia cicadellinicola and Sulcia muelleri. These two unrelated endosymbionts, they show, play distinct, complementary roles that supplement their host's diet.
Endosymbionts were fi rst described through microscopy in the early 20th century, but their metabolic secrets The genomes of two bacterial species that supply essential nutrients for the glassywinged sharpshooter provide potential targets for controlling infestations of this greatly feared agricultural pest have only recently come to light by making inferences about gene function based on genome analysis. After sequencing the B. cicadellinicola genome, Wu et al. built a genomebased evolutionary tree (called a phylogenetic tree) for B. cicadellinicola and related endosymbiont species. B. cicadellinicola was grouped together with endosymbionts of aphids (Buchnera), tsetse fl ies (Wigglesworthia), and ants (Blochmania), in keeping with other genomic studies, but was "the deepest branching symbiont," suggesting it had separated from the group earlier.

Bacterial Symbionts May Prove a Double-Edged Sword for the Sharpshooter
The endosymbiont genomes shared a number of features, including reduced genome size, fewer guaninecytosine (G-C) than adenine-thymine (A-T) base pairs, and rapidly evolving proteins. These trends can be found in endosymbionts arising across all the major bacterial groups, challenging biologists to fi gure out the mechanisms driving them. One explanation centers on changes that arise by chance in small populations (called drift); the other centers on a higher rate of mutations stemming from the loss of DNA repair systems. Interestingly, though some researchers favor one hypothesis over the other, analysis of the B. cicadellinicola genome suggests both are important, but act at different levels. The differences between endosymbionts and free-living species appear to be due to genetic drift. Differences among symbionts, however, appear to be due to differential loss of DNA repair genes.
The researchers argue that the nature of some of B. cicadellinicola's other genome features (such as G-C content) and position on the evolutionary tree can help fi ll in missing gaps between free-living and intracellular species. And because its proteins are evolving more slowly than those of other endosymbionts-making it more likely that similar sequences really do refl ect evolutionary relatedness rather than artifacts that occur when only fastevolving sequences are used-inferences about the evolutionary events promoting intracelluarity can be made with more confi dence.
As for B. cicadellinicola's metabolic capabilities, the researchers found a "relatively limited repertoire." It lacks the necessary genes to support sugar metabolism, and probably does not use sugar as energy source since The human genome lies deep within the cell interior, sequestered behind the double-walled membrane of the nucleus, yet still remains vulnerable to a wide variety of hazardous materials. Two of DNA's worst enemies, ultraviolet light and chemical carcinogens, can wreak havoc on the molecule by mutating individual nucleotides or changing its physical structure. In most cases, genomic integrity is restored by specialized suites of proteins dedicated to repairing specifi c types of injuries.
One mending mechanism, called nucleotide excision repair (NER), recruits and coordinates the services of roughly 25 proteins to recognize and remove structure-impairing lesions, including those induced by ultraviolet light. At the center of this effort is the ten-subunit transcription/repair factor IIH (TFIIH) complex. As its name suggests, in addition to its role in DNA repair, it also regulates transcription; how TFIIH coordinates these very different activities is still under investigation. Regardless, three TFIIH genes-XPB, XPD, and TTDA-have been implicated in the photosensitive form of a rare inherited premature aging syndrome called trichothiodystrophy (TTD), which is characterized by brittle hair and nails, scaly skin, and neurological degeneration.
In a new study, Giuseppina Giglia-Mari, Jan Hoeijmakers, Catherine Miquel, Wim Vermeulen, and colleagues created a fl uorescently tagged version of trichothiodystrophy group A (TTDA) to investigate its role in repair and transcription. By experimentally modifying the transcription function and by triggering DNA repair in human cell lines expressing the fl uorescent TTDA protein, the researchers show that TTDA fi rst of all dynamically associates with TFIIH, and that TTDAs become stably incorporated only while the complex is engaged in NER.
NER steps in when damage interferes with the elongation of a newly transcribed gene, promoting cell survival. It also plays an anti-cancer role by surveying the genome for damage. In ultraviolet-induced NER, TFIIH unzips the

A Tiny Protein Plays a Big Role in DNA Repair
Liza Gross | DOI: 10.1371/journal.pbio.0040184 DOI: 10.1371/journal.pbio.0040184.g001 Fluorescence accumulation in the nucleus indicates TTDA retention during the repair reaction at a locally ultraviolet-irradiated area in a living human fi broblast in this three-dimensional reconstruction. (Image: Pierre Olivier Mari and Giuseppina Giglia-Mari) almost none is present in the xylem sap diet. Energy likely comes from using the more abundant amino acids, one of the main types of organic compounds present in xylem. In return, it provides the insect with a range of vitamins and cofactors (required for enzyme activity). But surprisingly, B. cicadellinicola lacks key enzymes needed to synthesize essential amino acids. Obviously, the insect must get the missing essential nutrients from somewhere, and the most likely candidate, they reasoned, was the other bacterium known to reside in its cells, S. muelleri.
They went back to the sequences that didn't assemble with B. cicadellinicola, realizing they might contain part of the S. muelleri genome, and found genes required for the synthesis of several essential amino acids. Though some of the sequences belonged to other bacterial species, the vast majority required for amino acid synthesis belonged to S. muelleri. Since the sample of this bacterium's genome contained very few genes related to vitamin or cofactor synthesis, Wu et al. concluded that the two endosymbionts perform nonredundant, complementary services for their host. B. cicadellinicola synthesizes most of the sharpshooter's vitamins and cofactors, while S. muelleri appears to provide essential amino acids.
How three unrelated organisms manage to integrate such complex operations as metabolism and gene expression is a question for future study. But targeting any number of these essential bacterial pathways could prove a promising strategy for containing the spread of a potentially devastating pest-a risk perceived as so great that the sharpshooter is the only insect listed as a potential bioterrorism agent. And the researchers make a strong case that endosymbiont systems can help shed light on the molecular and evolutionary changes that allowed free-living organisms to take up residence in the cells of others-a path echoing the evolution of eukaryotes, whose cells contain mitochondria and chloroplasts, the descendants of ancient free-living bacteria. Unscrewing the lid of a honey jar stuck tight takes more than simply muscle power and a vice-like grip. It also takes coordination between your hands-one twisting the lid and the other holding the jar-to open the jar without it spinning out of your grasp.
But you might not suspect that in tasks like this, you unconsciously appoint one of your hands the leader and assign the other a supporting role-and the leader is not necessarily your writing hand. A new study reveals that people are surprisingly fl exible in assigning these roles, and that they can quickly switch dominance between hands depending on the nature of the task. This dominance turns out to be thoroughgoing, from the brain's cortex down to specifi c muscles in each hand.
In the new experiments, Roland S. Johansson and colleagues had 37 people, all right-handers, play a video game similar to the arcade classic Whac-A-Mole: when a target appeared in a random spot on a video screen, the players moved a cursor to touch the target as quickly as possible. Key to the experiments was a custom-made controller, a rectangular box with knobs on two ends, that the players used. To move the cursor horizontally on the screen, the players pushed the controller's knobs inward or pulled them out; twisting the knobs, like screwing or unscrewing a jar's lid, sent the cursor up or down.
With this setup, the researchers could see whether a person's hands were acting perfectly in concert or whether one hand took a dominant role, acting a bit before the other or pushing or twisting harder than the other. The researchers found that one hand did consistently dominate, so the body of the controller would shift or twist slightly at fi rst, before the supporting hand had a chance to compensate.
But which hand was dominant depended on the rules of the game, and the players were fl exible, quickly switching dominance between their hands when the rules of the game

Making Hands Jive: How the Body Manages Hand Coordination
Mason Inman | DOI: 10.1371/journal.pbio.0040196 DOI: 10.1371/journal.pbio.0040196.g001 When the rules of a bimanual task changed, the center of activity switched between hemispheres in motor-control areas in the brain's cortex and cerebellum. double helix to access the injury, and then recruits four proteins to stabilize the open strand, evaluate the damage, and cleave the DNA on either side of the lesion. Other proteins produce new nucleotides to fi ll the gap. Cells collected from patients with TTD-A have TTDA mutations that produce either truncated, nonfunctional versions of the protein or no protein at all, resulting in reduced TFIIH levels. TTD-A cells also have NER defects and are hypersensitive to ultraviolet light.
To study the role of the TTDA subunit in the TFIIH complex's functions, the researchers tagged TTDA and the XPD subunit with green fl uorescent protein (GFP) to monitor and compare their movement and behavior using high-resolution confocal microscopy. Both TTDA-GFP and XPD-GFP could stably incorporate into TFIIH and function in DNA repair, proving reliable tools for studying TFIIH's spatial and temporal distribution and activity during transcription and DNA repair.
Both TTDA-GFP and XPD-GFP were observed in the cytoplasm and nucleus, in contrast to the XPB subunit, which is known to localize only in the nucleus. To determine if TTDA and XPD assemble with TFIIH in the cytoplasm, the researchers used a customized version of a motilitymonitoring technique called fl uorescence recovery after photobleaching (FRAP). Applying a high-intensity laser to a specifi c region in a cell destroys fl uorescence in that area; as unaffected GFP-tagged molecules from other regions replace the zapped molecules, fl uorescence is recovered. Recovery is much faster for free-moving GFP-tagged proteins than for those caught up in interactions with other cellular proteins.
In the cytoplasm, TTDA-GFP moved much faster than XPD-GFP, but did not interact with each other. Both proteins, however, were less motile in the nucleus, where they interacted with TFIIH. These results indicated that TTDA exists in two pools in non-irradiated cells, one free and one bound to TFIIH. After irradiation, however, fl uorescence was not completely recovered for either TTDA-GFP or XPD-GFP, suggesting a lack of motility one would expect by participating in DNA repair. When the researchers inhibited a protein essential for NER initiation in the irradiated cells, they observed a reduced fraction of immobilized TTDA-GFP, confi rming that TFIIH must incorporate TTDA-GFP (thus, reducing its mobility) for NER. By interfering with the transcription machinery, they go on to show that TFIIH does not stably incorporate TTDA when TFIIH is simply bound to DNA; TFIIH only enhances association with TTDA during NER.
Altogether, these results show that NER-specifi c lesions induce TTDA to form a stable association with the TFIIH in living cells. Giglia-Mari et al. propose that once TTDA nestles into place after the core TFIIH complex attaches to a lesion, it triggers a conformational change that recruits the other subunits required for repair. TTDA may also help TFIIH fold properly, preventing it from being degraded and allowing it to accumulate to the levels necessary for NER function. TFIIH molecules spend far longer at NER sites (four to fi ve minutes) than they do at transcription sites (two to ten seconds), suggesting why TFIIH concentrations are so important to NER-and why people that can't produce TTDA experience such debilitating symptoms.