Developmental biology: Embryos gain foothold in the womb

Helen Pearson

Science 299, 405–408 (2003)

Many human embryos that fail to develop fall at the first hurdle — when burrowing into their mother's uterus. So the discovery of a molecule that is potentially vital for implantation might help in the diagnosis and treatment of infertility.

Using spare tissue from a private in vitro fertilization clinic, Olga D. Genbacev et al. found that embryos increase production of a protein called l-selectin on their outer coat of cells six days after fertilization, around the time of implantation. Biopsies showed that at the same time — and indeed every month — the uterine surface displays a set of unique carbohydrates to which l-selectin binds. Polystyrene beads coated in l-selectin stuck to receptive uterine cells.

Genbacev et al. suggest that embryos afloat in the uterus may become anchored by the interaction between l-selectin and its targets, triggering the embryos' invasion of the uterine lining. Intriguingly, l-selectin also helps circulating immune cells called leukocytes stick to the side of blood vessels, so they can invade tissue and fight infection.

If this model stands up, measuring the levels of uterine carbohydrates might help to diagnose women who suffer infertility through faulty implantation. Similarly, the efficiency of in vitro fertilization might be improved if embryos were screened for a healthy dose of sticky l-selectin.

Granular materials: Nut theory gets a shakedown

Philip Ball

Phys. Rev. Lett. 90, 014302 (2003)

Ups and downs: the brazil-nut effect (top panel) and its reverse.

There are various explanations for the 'brazil-nut effect' — the rising of the largest grains to the top in a shaken granular mixture. These invoke mechanisms such as sieving and convection. But it was pointed out several years ago that the reverse can happen, too: the larger grains may sink.

A theory purporting to explain the crossover between the rising and sinking of the large grains has now been subjected to an experimental test by Andreas Breu and colleagues. They placed various mixtures of two different kinds of spherical bead, made from glass, metal, plastic and wood, in stratified layers and shook them vertically in a clear plastic container (see picture). The theory, which claims that the crossover happens when the ratio of particle sizes in a two-component mixture is equal to the inverse ratio of the particle densities, predicts correctly in about 82% of the cases studied.

But the two types of segregation also depended on other factors. A reverse brazil-nut effect could never be produced if the initial lower layer of small grains was too deep. And the same mixture could separate either way, depending on the acceleration of the grains — a function of both the frequency and the amplitude of vibration.

Geochemistry: Zircons and the early Earth

Tim Lincoln

Earth Planet. Sci. Lett. 204, 333–346 (2002)

Grains of zircon can be dated to more than 4 billion years ago — very early in the Earth's history. The chemical composition of the parent magma from which zircon crystallized is an indicator of past geological conditions, but that parent rock may long since have been destroyed. Can the durable zircons be used to infer the existence, at that early time, of continental crust, and of subduction processes in which crust is dragged down into the mantle below it? Martin J. Whitehouse and Balz S. Kamber have doubts about attempts to do so.

The crucial question is whether the composition of zircon is a reliable reflection of the original magma chemistry. From their analysis of the light rare-earth elements present in 3.8-billion-year-old zircon and its existing parent rock, Whitehouse and Kamber say that it isn't. In a second line of argument, they point out that the light rare-earth-element content of zircon samples from the Moon is akin to that of terrestrial zircons of similar age. There is no evidence that subduction processes have ever occurred on the Moon.

But that won't be the end of the story — zircons are just too precious as a potential information source about the early Earth for matters to rest there.

Cardiovascular biology: Chilled-out platelets

Helen Pearson

Cell 112, 87–97 (2003)

Transfusions of blood-clotting cells called platelets can be a life-saver, but there's a problem: platelets have a limited shelf-life because they cannot be stored in the fridge. Karin M. Hoffmeister et al. have now discovered why cooling is bad for these cells.

Mysteriously, when chilled platelets are used in transfusions, they are rapidly engulfed and removed by white blood cells. Hoffmeister et al. show that a molecule called complement type 3 receptor (CR3) is responsible for the platelets' demise. Mice lacking CR3 on their white blood cells sustain refrigerated platelets for longer. The team suspect that cooling enhances CR3's recognition of a molecule on the platelet surface — a receptor for von Willebrand factor, which is an essential component of the clotting process. They found that chilling platelets makes these receptors cluster together into a more recognizable form.

Hoffmeister suggests that it might be possible to protect refrigerated platelets from white blood cells by blocking the interaction with CR3. Adding enzymes that chemically modify the receptor for von Willebrand factor could do the trick, she says, and allow stored blood to be refrigerated.

The team also speculates that purging chilled platelets is a natural physiological response. Platelets may become 'primed' for activation in the cool outer skin — the location of wounds — and are then removed to avoid potentially harmful clotting throughout the circulation. When whole mice were chilled in the fridge, transfused platelets left their bloodstream faster, the authors found.