Frontiers : The month in research

Molecular biology
Unfolding clues
Cell 2006;127:803-15

Misfolded proteins lie at the centre of many debilitating conditions, the most notable being Alzheimer's disease and cystic fibrosis. In the latter, mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) result in a misfolded membrane protein that prevents normal movement of important ions. Researchers have found that tweaking a component of the protein folding machinery can help.

The set of protein-protein interactions that enable other proteins to fold normally is called an interactome. Researchers compared CFTR interactomes, using proteomic analyses, of the normal CFTR and the most common CFTR mutant. They found that partially blocking a helper protein involved in stalling folding, known as a co-chaperone (Hsp90), helped the mutant protein to fold properly. How this happens is not clear-but researchers now wonder if this mechanism can be used to target b amyloids in Alzheimer's disease.

Neuroscience
Demystifying memory
Neuron 2006;52:437-44,445-59, 461-74, 475-84

Memory, perhaps the most important of our mental faculties, has four important components-encoding, con­solidation, storage, and retrieval. Four papers have just been published that have considerably improved our understanding of how memory consolidation works at the genetic level.

The gene, known as Arc/Arg3.1, is expressed in the brain during learning. It has long been used as a marker of neuronal activity, even though its exact role in memory formation remained a mystery. The best way to identify the function of a specific gene is to do a gene knockout study, wherein a specific gene is deleted before the development of an animal, so that its genome does not contain the gene. This is exactly what was done with Arc/Arg3.1.

The animals failed to form long lasting memories, not just for conscious memory involving explicit tasks, but also for memories that do not require consciousness for retrieval. The short term memory, though, was left intact. As expected from these findings, the underlying mechanisms of memory consolidation, long term potentiation, and long term depression in the relevant neurones were significantly impaired.

Studies in vitro suggest that the gene controls the turnover of the AMPA receptors for glutaminergic neurotransmitters on neuronal surfaces. This is consistent with previous findings, which showed close links between glutamate receptors and plasticity of the neurones (that is, the strength of connection between them), the latter being fundamental to learning and memory.

Prof p Motta/dept of anatomy/la sapienza/spl
Yes, I remember

Cardiology
Pressured under failure
JAMA 2006;296:2217-26

A simple systolic blood pressure reading on admission can provide important prognostic information for people presenting to hospital with acute heart failure, say researchers  who found a clear link between higher systolic blood pressure and lower mortality among more than 48 000 adults. The trend was significant, with no apparent ceiling to the effect. Death rates up to 90 days followed a similar pattern.

Low systolic blood pressure was a poor indicator of prognosis, regardless of the patient's ejection fraction. Overall, the odds of dying in hospital increased by 21%  for every 10 mm decrease in systolic blood pressure below 160 mm Hg.

It's likely that systolic blood pressure on admission divides patients with acute heart failure into two distinct groups. Women, African Americans, and people with a well preserved ejection fraction seem over-represented at the top end of the blood pressure range. Men and people with ischaemic heart disease and poor ejection fraction are over-represented at the bottom. Clinical trials of treatments should probably consider these groups separately.

Cancer medicine
A design for efficacy
Science 2006;314: 1308-11

The anaerobic bacterium Clostridium novyi-NT can lyse red blood cells, and researchers have wondered if this property could be used to improve the release of certain drugs within tumours that are surrounded by a membrane called liposome. Liposomes carry the drug to the inside of the target cells by virtue of their lipophilic property. To test this hypothesis, researchers used a single dose of liposomal doxorubicin and simultaneous infection with C novyi-NT to treat large, established tumours in mice. Sure enough, the cancers disappeared.

It seems that release of doxorubicin was facilitated by a previously unidentified protein called liposomase. If incorporated into other liposomal drugs, this protein might be able to improve the therapeutic effects of chemotherapy in human beings.

Internal medicine
Questionable safety
Lancet 2006;368:1771-81

It is well established that non-steroidal anti-inflammatory drugs (NSAIDs) that selectively inhibit cyclo-oxygenase-2 (COX 2) are linked to thrombotic events such as heart attack. But do such adverse events extend to more traditional NSAIDs? A pooled analysis from three large clinical trials says yes. Researchers have found that diclofenac (a traditional drug) had almost identical effects on cardiovascular risk as etoricoxib, a selective COX-2 inhibitor.

The research, sponsored by Merck, concludes that COX-2 inhibitors are no worse in this respect than diclofenac. A linked commentary (pp 1745-7) largely agrees. The similarity arises because both inhibit prostacyclin, a prostanoid that restrains platelet activation, hypertension, and atherogenesis. Patients and doctors now need prospective data on other drugs, such as ibuprofen and naproxen, both of which looked "safer" than diclo­fenac in observational studies .

The HapMap project is so yesterday-it only mapped single nucleotide changes in the human genome. Now an international consortium has mapped even larger areas of differences between individuals, called copy number variants.

This new map is based on data collected from analyses of DNA from 270 different individuals. Each copy number variant involves, by definition, at least 1000 base-pair differences between individuals, and both benign changes and those causing disease have been attributed to such large chunks of DNA. The surprise was that these copy number variants covered almost 12% of the human genome-researchers documented more than 1400 copy number variants.

This discovery revamps our view of interindividual genomic variation, which now seems far more widespread. Unless directly analysed, these differences could be easily missed by the present strategies used to identify genetic mutations in human diseases.