Faculty of 1000

Post-publication peer review

Archive for the ‘Weekly roundup’ Category

Should I stay or should I go

Posted by rpg on 21 April, 2010

Epigenetic regulation is the new black, or at least charcoal. Modifying chromatin—acetylation or methylation of histone tails, or directly on DNA, in CpG islands for example—can change a number of things in addition to imprinting (turning gene expression on or off depending on parental origin). I’ve just finished one write-up of a F1000 Report discussing histone methylation in meiotic recombination hotspot specification and hybrid sterility, and another one of a paper showing that histone marks regulate alternative splicing patterns (look out for them in June’s The Scientist).

Viruses are getting in on the act too. When Epstein-Barr virus infects a cell it can either hide–latency–or replicate and continue the infection. It uses a single transcription factor to drive these states, depending on the epigenetic status of the host cell; in other words it uses the cellular machinery to wake up when conditions are right. Kind of like a molecular clock, actually. (And you’ve got to love an evaluation that quotes Mick Jones.) We’re also seeing strange but pretty strong effects in certain tumors: hypermethylation of two tumor suppressors in the parathyroid is associated with (and probably leads to) adenoma formation.

Fat’s back in town. Lab rats are obese—morbidly so. But the problem is if you don’t let them feed at will then they get stressed, apparently, which is bad if you’re trying to do experiments on a normal, healthy ‘control’ animal model. Lots to think about there. I’m not going to repeat a comment a colleague made about them being a normal model for certain nationalities, but maybe we could look at knocking out hedgehog, which seems to suppress the generation of white but not brown fat.

Finally (sorry, no room for cytoskeletal porn this week), Faculty of 1000 published its 90,000th evaluation yesterday! David McManus at Weill Cornell Medical College wrote about a paper in the Journal of Clinical Oncology that looked at combination therapy in the treatment of advanced clear cell renal carcinoma.

We’re looking forward to 100k.


Posted in Weekly roundup | Comments Off on Should I stay or should I go

At the movies

Posted by rpg on 7 April, 2010

Busman’s Holiday

Faculty of 1000 published 1472 evaluations last month. This is a world record! And it should help you predict when we’re going to hit 90,000. Remember, we’re running a competition: simply twitter the date and time you think we’ll make 90,000 evaluations with the hashtag #F90K for a chance to win some F1000 goodies.

Easter Hollidays

Image: Richard Wheeler, Wikimedia Commons

Faculty member Fyodor Urnov wants to know if you remember the lecture on homologous recombination (HR) from your genetics class in college. I certainly remember tutorials, and seeing micrographs of Holliday Junctions for the first time. I was fascinated and not a little excited at actually seeing a physical representation of an incredibly important biological process.

Fyodor answers his own question,

For many people, sadly, the answer is “no, and not regretting it.” This is a shame — not only are we the products of HR that took place during gametogenesis in our parents but the repair of double-strand breaks (e.g. after a dental X-ray) keeps our genomes intact.

and recommends you read a recent paper in Nature showing that double Holliday junctions are indeed involved in repair of double-strand DNA breaks. He goes on to bang the drum for traditional biochemistry:

single-locus analysis continues to offer remarkable insight into biology, despite the ubiquity of massively parallel omics. A proper Southern blot — of which this paper has many — is a very, very powerful tool.

Roman Holiday

You probably saw that Nature Reviews suffered its first retraction across its stable recently. In an interesting case of intellectual plagiarism, Mariam Sticklen was accused of writing a paragraph that was “paraphrased without attribution”. This becomes interesting in that the principle of anonymous peer review has been challenged: Sticklen reviewed a paper and allegedly lifted the offending ‘thought’ directly from it. The editor of Plant Science, Jonathan Gressel, said
“When you have done something that’s way beyond the pale, you forfeit your anonymity as a reviewer,” and “I think Nature Reviews Genetics was nice to her in allowing her to say ‘paraphrase’.”

The full story by Bob Grant is available at The Scientist. I only mention it here, really, for the comment thread, in which we see both editors getting involved, as well as Sticklen herself and her ex-husband, who gives her a glowing character reference. If there’s any film makers out there who want to make a blockbuster about science, this has all the ingredients.

And finally

Trees are good for you. At least, if you’re an arboreal—i.e. tree-dwelling—mammal. But not a primate or marsupial. If you’re one of those (what are you doing reading this?), you

should have longer lifespan than terrestrial species of similar body mass, the rationale being that arboreality reduces the risk of predation by terrestrial predators.

As Douglas Adams once said, coming down from the trees was a bad idea.

Posted in Weekly roundup | Tagged: , | Comments Off on At the movies

We will rock you

Posted by rpg on 31 March, 2010

Three’s a crowd

Who would have thought that plants could teach us about deafness? Or single-celled yeast about blood vessel development? Orthologous genes in different species can have totally different effects, and a statistical data-mining technique has thrown up not a few surprising models for human disease. The paper is free at PNAS and reviewed at Faculty of 1000. You can read more at The Scientist and Nature.

In another bizarre turn of events, it turns out that dilute biochemistry is all it’s cracked up to be. People have, over the years, ragged on biochemistry for dealing with dilute proteins and ignoring the ‘crowding’ effects of the cytoplasm (despite biochemical predictions often being borne out by in vivo work, but anyway). An outstanding computational paper looks at models of crowding in the bacterial cytoplasm, successfully simulating the relative thermodynamic stabilities of individual proteins:

But the overall take-home message seems to be that the effect of crowding by steric exclusion is largely cancelled by hydrophobic interactions with the crowders. Protein biochemistry in dilute solution has gained new respect.

The full paper, Diffusion, Crowding & Protein Stability in a Dynamic Molecular Model of the Bacterial Cytoplasm, is available from PLoS Computational Biology.

Got rhythm?

Another strange one, lurking in PNAS (Social Sciences/Psychological and Cognitive Sciences), suggests that humans are born to rock and roll:

One of the most curious effects of music is that it compels us to move in synchrony with its beat. This behavior, also referred to as entrainment, includes spontaneous or deliberate finger and foot tapping, head nodding, and body swaying.

Children under the age of two (and pre-verbal) spontaneously to music, but not speech.As Katrin Schulze, down the road at UCL says,

This suggests a predisposition in humans towards engaging rhythmically to a musical beat.


And finally

It’s competition time!

Faculty of 1000 is approaching the publication of 90,000 evaluations. This morning we had 89210 on the two sites, Biology and Medicine. We’ll be running a little internal compo for the people in the office, but we’d like to throw this open to all our Twitter followers and readers too. Use the hashtag #F90K to tell us the day you think we’ll make the 90 thousand. For a tie-breaker, feel free to put in the time, too (best use UTC. And here’s a clue: our editors work London office hours). Visit http://f1000biology.com/ and http://f1000medicine.com/ to help you with your guesswork.

As usual, a bag of F1000 swag for the winner. Good luck!

Posted in Competition, Weekly roundup | 1 Comment »

Cells and drugs and roll-on/roll-off

Posted by rpg on 24 March, 2010

Everybody loves a good structure. Here’s one of the prototype foamy virus (PFV) integrase in complex with DNA ends. The integrase is the complex that binds the termini of viral DNA and catalyses its insertion into the host genome. Nasty piece of work, and incidentally one that has ‘nasty biophysical properties‘—at least the one from HIV and related retroviruses. Fortunately for the future development of retroviral drugs, Stephen Hare from Imperial College and his colleagues managed to persuade the integrase from PFV to not only bind to DNA, but form a complex that was active in solution and amenable to crystallization. Fred Dyda of the NIH says

This outstanding achievement also speaks volumes about the value of high resolution experimental structures of assembled complexes, as it appears that previous models of intasomes based on fragmentary structural information and a lot of imagination have not been reliable.

C’est n’est pas un stem cell

Elsewhere this week, surprising news from the world of plant biology. A favourite tool of plant biologists is the callus, a mass of seemingly totipotent cells. Callus can be induced starting from virtually any differentiated plant tissue and is the basis for the remarkable ability of plants to regenerate the whole zucchini from just about any tissue, as any gardener is well aware when it comes to rooting out dandelions. The surprise is that the callus resembles the growing tip of roots, and is not the result of a simple reprogramming process to an undifferentiated state.

Vorsprung duck technik

In an amazing piece of German engineering scholarship, the itineraries of 16,363 cargo ships in 2007 were used to construct a network map. And if I could get access to the paper, I’d post a pretty picture here. Sorry. Anyway, the point is that this map could tell us lots about the movement of bioinvasive marine species, seeing as bulk dry carriers and oil tankers are the primary vectors for such beasties, as they hitch rides in the ballast. Jonathan Belmaker and Lewi Stone of Tel Aviv University add,

the network is almost scale-free since there are several large highly connected ports through which all smaller ports transact their trade. This property makes the ship network more prone to the spreading and persistence of bioinvasive organisms.

L’amour est comme un papillon

This is rather sweet, as well as pretty mind-blowing from an evolutionary point of view. A duplicated opsin—a visual pigment—in the Heliconius spp of butterfly evolved by positive selection, giving the insect the ability to discriminate between pigments. Wing pigments in the same butterfly underwent coincident evolution, which might help explain why the yellow wing pigments of Heliconius are so varied in the UV range, compared, with its less colourful relatives.

This is a beautiful example of how evolutionary and functional information together can provide strong evidence on the nature of an evolutionary adaptation.

And finally, a quick cytoskeleton fix. We’ve known for a while that short actin oligomers can anneal in vitro to form fibres. Turns out that this also happens in yeast cells at least; which means that if you sequester all the monomers, you can still build filaments close to the plasma membrane at endocytic sites. Could be important if you’re a budding yeast faced with a shed load of latrunculin A.

Posted in Weekly roundup | Comments Off on Cells and drugs and roll-on/roll-off

Waltzing Mathilda

Posted by rpg on 17 March, 2010

You turn me right round

As an undergraduate, I remember being fascinated by the family of rotary motors that is the proton-translocating ATPase. This is the protein complex, resembling something from the imagination of E. E. ‘Doc’ Smith rather than anything merely biological, that either pumps protons from one side of a membrane to another, or uses a proton gradient to drive its rotation and make ATP. One of the less obvious difficulties with this amazing piece of machinery is that because the thing rotates, you need something to counteract the torque generated. So the ATPase has a additional stator component, the peripheral stalk, that links the nucleotide-binding domain to the membrane-embedded domain, which stops the entire thing waltzing away.

Interestingly, despite its essential role, the peripheral stalk is the least conserved element of the structure, varying in composition and number of subunits across the family. Daniela Stock’s lab at the Victor Chang in Sydney has recently published the structure of the peripheral stalk of the Thermus thermophilus H+-ATPase/synthase, in the process discovering a new protein structural motif—a right-handed hetereodimeric coiled-coil.

peripheral stalk

And that’s fascinating because the central stalk contains a left-handed coiled-coil, and it’s tempting to speculate that the opposite coiling of the two stalks is important, either in stabilizing the whirligig and/or for proton pumping. Read all about it (free link):

This paper reports, for the first time, a crystal structure of the bacterial A/V-ATP synthase peripheral stalk complex, which shows the N-terminal domains of the two peripheral-stalk-forming subunits E and G to be folded in a heterodimeric right-handed coiled coil, a protein fold not seen before.

Stephan Wilkens

Some shall pass

Peroxisomes can import fully-folded, native proteins across their lipid bilayer. How they achieve this wasn’t known—hypotheses included the transient formation of pores or a process similar to endocytosis (compare with the nucleus, for example, which transports macromolecules in and out via stable pores existing within the double membrane). It turns out that the membrane-associated receptor Pex5p, which recognizes the peroxisome signal motif Ser-Lys-Leu at the C-terminus of potential cargo, forms a highly dynamic, ion-conducting channel in conjunction with its docking partner Pex14p and cargo. These transient pores can be opened to a diameter of about 9 nm, allowing quite large protein complexes to pass. (Meinecke et al., Nat Cell Biol)

Who are you?

Whole genome sequencing is getting a fair bit of press recently. The ‘$1000 genome’ is still a way off; but a case can be made that even the current cost (~$50,000 and dropping) compares favourably with current diagnostic procedures for genetically heterogenous conditions. A paper in NEJM shows the feasibility of using whole genome sequencing to identify clinically relevant variants in the case of a rare, inherited disease, and provide diagnostic information to inform the care of patients:

the patient described is a compound heterozygote for two different missense mutations, which are easily missed in standard sequencing-based strategies for mutation identification.

(Free F1000 evaluation)

And finally, don’t forget to check out the Faculty Faves at The Scientist homepage. These are selected from the previous day’s most interesting papers at F1000, and available for free.

Posted in Weekly roundup | Tagged: , , , , | Comments Off on Waltzing Mathilda

Richard of York…

Posted by rpg on 10 March, 2010

It’s all in the mind

Neuroscience is a fascinating subject, and not just because our friend Robert Sapolsky is a Faculty Member. Neuroscience (at least at F1000) covers a spectrum of subjects and disciplines, from the molecular basis of odour discrimination, through neural processing in the retina all the way up to anticipation of rewards and what chemical signals drive spontaneous learning and the pursuit of information.

In that vein, then, it’s rather fitting that

the last two decades of neuroscience and biological psychiatry research have revealed [that] there are no meaningful categorical differences in brain structure or function that map onto categorical clinical disorders

Ot to put it another way, it’s being on the ends of the spectrum, not having a discrete genotype, that makes you ‘sick’. Ahmad Hariri of Duke University evaluated a paper from Robert Plomin (just down the road at King’s College) which synthesizes current thinking about mental illness into a conceptual framework summarized by the single statement

there are no categorical psychiatric disorders but rather a continuous range of behaviors which, at the extremes, may interfere with daily functioning.

The critical thing is that mental illness appears to be polygenic; that is there are multiple genes, each which have a small effect. Thus qualitative disorders, the ones that occupy pscychologists and psychiatrists, are simply the extremes of the bell curve of quantitative effects—there is a continuous range of brain structures and functions mapping onto a continuous range of behaviours. (‘Simply’, he says.) Now that’s all very well, but for the most part we don’t actually know what those quantitative traits, those individual genes, are. That shouldn’t be a major problem: Hariri recommends replacing categorical endpoints (exclusive) with quantitative biological and behavioural phenotypes (exclusive) in ongoing research. Extremes of the spectrum are important—medically and socially—and accepting this framework has enormous implications for the diagnosis, treatment and prevention of mental illness.

Which way is ‘up’?

I wrote a piece for an upcoming new section in The Scientist last week (actually, I wrote three but only one is of concern now). It turns out that a long-held belief (there’s no other word) about segregation of different cell types in the developing embryo might well be wrong. The proposed mechanism is really intriguing but you’ll have to wait for me to tell you about it. In a almost-parallel development (pun not intended) another paper on F1000 looks at a similar problem in plants—in this case the generation of apical-basal patterning in Arabidopsis. This is important for establishment of root and shoot identities at opposite ends of the plant. Turns out that the previously-characterized PLETHORA genes and the CLASS III HOMEODOMAIN-LEUCIN ZIPPER transcription factor(don’t ask me why plant biologists have to shout so much) act antagonistically to define shoots versus roots.

While we’re talking about plants, Swiss researchers have found that the membrane at the Casparian strip is a diffusional barrier, and much like tight junctions in animal cells, generates polarity within cells. Two boron transporters (BOR1 and NIP5;1) are markers for polarization and allowed the researchers to trace polarity’s back to the embryo.

Changing your mind

An NMR structure

Anyone who has spent any time at all looking at protein structures knows that NMR and x-ray crystallography tell you different things. Crystal structures give you the impression of something fixed in space, and NMR structures usually look like something constructed from over-cooked spaghetti. They’re both representations of course, but someone has finally pointed out that native structures of proteins should be described in terms of ensembles rather than some ‘average’ structure. I’m not convinced this counts as a ‘paradigm shift‘ but it’s good to see these thoughts in print. Oh, and I get to use my new ‘Hidden Jewel’ icon! Tell me what you think. Hidden Jewel

Another challenging paper was published in Cell last month, identifying the tumour suppressor NF2 as a partner of (yet another) E3 ubiquitin ligase. The protein of NF2 is called ‘Merlin’, for ‘Moesin-Ezrin-Radixin-related protein’, and has signficant similarity to one domain of the cytoskeletal protein I did my thesis on. So the surprise is that Merlin appears to be active in the nucleus, rather than (or in addition to?) at the plasma membrane.

This paper is going to alter the way we think about the tumor suppressor NF2 […] These findings are surprising and will make us reassess how Merlin works.

says Jonathan Chernoff of the Fox Chase Cancer Center. This has shades of the whole ‘nuclear actin’ controversy.

Men of Harlech

Worms Armageddon Finally, all is not what it seems in the Welsh valleys. Mutant monster worms are using calcium binding and sensing pathways to deal with environmental lead, turning into mutant killing compost eaters of DOOM! Well, not quite, but it does appear that a population of the little wrigglers in the unpronounceable Cmystwhyth Valley diverged from the rest of the species 17 thousand years ago. One lineage is pretty normal, the other eats lead for breakfast. The authors of the paper recommend mandatory genotyping of all individuals prior to field-based ecotoxicological assays, particularly those using discriminating genomic technologies. I say we take off and nuke the site from orbit, etc.

Posted in Weekly roundup | Tagged: , , , , | 4 Comments »

Separation Anxiety

Posted by rpg on 4 March, 2010

Stem cells for dummies

The ability to maintain embryonic stem cell lines—more importantly, to preserve their pluripotence—in culture held out great promise for the treatment of a range of conditions from cancer to Parkinson’s disease. Unfortunately the technology ran into trouble when in 2001 the US Government restricted Federal funding to work done with the 21 lines that existed then, effectively kiboshing meaningful stem cell research in the US. Although the ability to induce pluripotency in adult cells (‘induced pluripotent stem’ or ‘iPS’ cells) avoided much of the controversy associated with deriving cells from human embryos, iPS cells are not as reliable as the real thing. All this changed last year when President Obama revoked the bill and enabled the development of new guidelines for Federal funding. And last month, the federal contract with the National Stem Cell Bank—the only one in the US—expired, throwing the stem cell game wide open.

All this means that we’re likely to see a lot more requests for stem cell line approval, like this:

Request for Human Embryonic Stem Cell
Line to be Approved for Use in NIH Funded Research. Type of Information
Collection Request: Revision, OMB 0925-0601, Expiration Date 02/28/
2010, Form Number: NIH 2890.

—and there’s even a Stem Cells for Dummies! (H/T)

Naturally, F1000 has its fair share of stem cell excitement. A study from Osaka University shows that embryonic stem cells do not have circadian rhythms; in fact, they do not express the transcriptional-translational feedback loops that generate rhythm in somatic cells. And although differentiation appears to induce circadian rhythm in these cells, subscequent treatment with the factors used to create iPS cells turns it off again. It’s not that straightforward though: an older paper, from Kyoto University, evaluated last week, finds a role for a cyclic gene that is expressed in stem cells, and in fact its expression level at the point of differentiation determines cell fate. So it would appear that circadian rhythm and commitment to cell differentiation are pretty much inseparable in development.

Six degrees of separation

It’s been an intriguing week for Structural Biology, too. The apparent similarity of many protein folds looks like an accident of physics rather than of evolution: there are only so many folds available to the 20 natural amino acids. Almost any two protein domains are separated by seven or fewer intermediate structurally similar domains—and this holds even for artificially-created polypeptide sequences.

The fact that evolutionary divergence need not be invoked to explain the continuous nature of protein structure space has implications for how the universe of protein structures might have originated, and how function should be transferred between proteins of similar structure.

Intrinsically unstructured proteins can lead to pathological conditions such as cancer and amylopathies. A paper from Madan Babu’s lab at the MRC-LMB (where I spent six happy years) shows that these proteins are actually very tightly and differentially regulated. Solving the three-dimensional shape of structured proteins, on the other hand, is fraught with difficulties. It’s exciting then to see a method that nearly doubles the size of proteins that potentially could be solved by NMR, by only looking at the peptide backbone assignments. Expect to see more NMR structures, and faster, then.

At the movies

I don’t think there’s any denying that studying the cytoskeleton gives the most opportunity for biology eye candy. A group at Southwestern Medical Center in Texas looked at how semaphorin receptors talk to the actin cytoskeleton, and show that fly cell ‘bristles’ reflect what’s going on inside the cell. They also show that a protein called Mical unbundles and depolymerizes actin fibres.

Finally, there’s been a lot of interest in how sticking cells down to a surface helps them stick mnore tightly to each other. A paper in the Journal of Cell Science, evaluated twice on F1000, describes an intriguing technique to test this: the researchers used two micropipettes to pull cells apart, measuring the force required as they do so. Reviewers Ekaterina Papusheva and Carl-Philipp Heisenberg say

In future studies, this could be used to visualize the spatial transmission of the signals between integrin and cadherin adhesion sites e.g. by implementing fluorescence resonance energy transfer biosensors

which should make for some very pretty data indeed. There is a movie of the technique, that I found fascinating. Can you predict when they’re finally going to separate?

Posted in Weekly roundup | Tagged: , , , , , , , , | Comments Off on Separation Anxiety