Harvard University, Cambridge, Massachusetts, USA

Thanks to the discovery of a new catalytic RNA, a chemical biologist can satisfy his student's curiosity.

An first-year undergraduate recently asked me a remarkable question: are all natural ribozymes, RNA molecules with catalytic activity, simply leftovers from the 'RNA world'? The RNA-world hypothesis supposes that RNA molecules were precursors to the first primitive forms of life, before the evolution of DNA and proteins.

Unanswered, my student's question preoccupied me until I encountered a recent paper by Jack Szostak of Massachusetts General Hospital, Boston, and his co-workers (K. Salehi-Ashtiani et al. Science 313, 1788–1792; 2006).

Prior to this work, only two of the known natural ribozymes were associated with mammals. The rarity of catalytic RNAs in more recently evolved, higher-order cells could reflect their attrition on the evolutionary battlefield during the rise of more highly functional protein enzymes.

This paper, however, supports a different conclusion. Szostak's group designed an ingenious system to isolate self-cleaving RNA molecules from RNA encoded in the human genome. Using this system, they discovered several new ribozymes.

One of these ribozymes, associated with a gene known as CPEB3, is highly conserved among placental mammals and marsupials, but is absent from non-mammalian vertebrates. This observation suggests that it arose relatively recently, around 200 million years ago.

We can therefore infer that some ribozymes have evolved in modern organisms, long after the era of the RNA world. The work elegantly demonstrates a new approach to the study of ancient molecules — and also reminds me that our youngest students can ask some of the best questions.

Posted by Jenny Hogan at 06:00 PM | Permalink | Comments (0) | TrackBacks (0)

Max Planck Institute of Neurobiology, Martinsried, Germany

A neuroscientist asks whether neurons are enough when it comes to learning.

One of my main scientific interests is understanding how the brain adapts to and learns from experience. By 'brain' I normally mean networks of neurons, because the electrical impulses they produce are the common currency of sensation and perception; learning involves changes in the synaptic connections between these cells driven by sensory experience. But it's beginning to seem that the brain's plasticity depends on more than neurons alone.

A few months ago, I found myself debating with a colleague what role glial cells might play. I was sceptical. Some years ago, the various sorts of non-neuronal cells that go by the name of glia were thought of simply as a glue (which is what glia means in greek) that holds the brain together. More recently, though, these cells have been shown to form extensive networks important for regulating the local brain environment and to communicate with neurons.

Despite knowing all of this, I had not considered glia as serious players in neuronal plasticity. My thinking changed after reading two recent studies showing that glia not only change their activity after sensory stimulation (X. Wang et al. Nature Neurosci. 9, 816–823; 2006) but also influence the strength of synaptic connections on neurons via a secreted soluble protein (D. Stellwagen and R. Malenka, Nature 440, 1054–1059; 2006). Because the amount of secreted factor depended on the level of surrounding neuronal activity, these results may provide a glial link between sensory experience and synaptic plasticity. The challenge is now to show this directly in the intact brain. I am willing to give it a try — the glia may have changed my mind!

Posted by Jenny Hogan at 06:00 PM | Permalink | Comments (0) | TrackBacks (0)

University of Toronto, Canada

The molecular dance of a protein allows a chemist's secret wish to come true.

One fascinating aspect of molecular function is the way information propagates between parts of a molecule that can be many tens of angstroms apart.

Our understanding of how proteins do this, a process termed allostery, emerged from Max Perutz's pioneering studies of oxygen-carrying haemoglobin. Three-dimensional images show that when a ligand binds to part of the molecule, a discrete set of structural changes take place at distinct sites. This, in turn, influences the ease with which subsequent ligands bind.

Nature has chosen this model in designing many allosteric proteins. However, as a practising nuclear magnetic resonance (NMR) spectroscopist with a strong interest in protein dynamics, I was secretly hoping she might design proteins in which information is communicated through changes in the dynamics between distal sites, with little or no change in overall structure. Moreover, I was rooting for NMR to play a major role in characterizing such a system.

How exciting it was, therefore, to read that Charalampos Kalodimos and his co-workers recently found such a case by studying the motional properties of a protein in different ligated states (N. Popovych et al. Nature Struct. Mol. Biol. 13, 831; 2006). Using NMR spectroscopy, the team quantified protein dynamics for a wide range of timescales. Remarkably, ligand binding at one site is linked to changes in motion far removed, over the complete set of timescales, while a corresponding propagation of structural changes does not occur.

The work of Popovych et al. provides a striking example of the importance of protein dynamics to information transfer. I eagerly await the discovery of more molecular dances and of how they, too, will relate to biological function.

Posted by Jenny Hogan at 06:00 PM | Permalink | Comments (0) | TrackBacks (0)