Nature Genetics pp 131 - 138 and pp 107 - 108

Amyotrophic lateral sclerosis (ALS; also known as Lou Gehrig's disease) is a common neurodegenerative disease, affecting only motor neurons and killing about 1 in every 1,000 people. Awareness of the disorder has been raised by public figures who have had (or have) it—Lou Gehrig, the legendary baseball player, died shortly after first showing symptoms, and Stephen Hawkins, the astrophysicist, suffers from a slower form of the disease, losing speech and movement while conserving his legendary brain power. In less than 2% of people with ALS, the disease is caused by mutation of the gene encoding the superoxide dismutase SOD1, an enzyme involved in the control of oxidative stress. Over 98% of ALS cases are of unknown cause.

On page 131, Peter Carmeliet (of University of Leuven, Belgium) and colleagues implicate the vascular endothelial cell growth factor (VEGF) in adult-onset degeneration of motor neurons in mice. They originally set out to modify the 'control' region upstream of the VEGF gene, so as to repress the increased rate of production of VEGF in response to oxygen deprivation (hypoxia). Surprisingly, the mutant mice developed an age-dependent degeneration of motor neurons, reminiscent of ALS in humans. This indicates that a primary cause of ALS may be a chronic deficit in VEGF-dependent supply of blood (and its bio-available oxygen) to motor neurons. The authors also demonstrate that VEGF directly affects the survival of motor neurons, and thus indicate another potential means by which aberrations in VEGF may give rise to ALS.

Pate Skene (of Duke University, USA) and Don Cleveland (of University of California in San Diego, USA) suggest, in an accompanying News & Views article that, of all the neurons in the body, motor neurons are the most sensitive to deficits in the blood supply because they are among the largest cells in the body and thus have extremely high metabolic needs.

The mice obtained by Carmeliet and collaborators provide an interesting animal model with which to explore possible interactions between SOD1, VEGF, and other cellular components in the pathogenesis of motor neuron disorders. Whether mutations affecting VEGF levels result in ALS will now be the focus of intense investigation. And, as VEGF and its receptors are the targets of current anti-cancer drugs—designed to deprive tumors of a blood supply—the possible side effects of such drugs on motor function should draw the attention of oncologists.

Nature Genetics pp 169 - 172 and pp 101 - 102

Tumor suppressor genes, which encode proteins that normally inhibit aberrant cell proliferation, are frequently mutated in human cancer. In most cases, cells with one normal copy of a particular tumor suppressor gene (heterozigosity) will not give rise to a tumor. When the normal copy is lost, however, a cell is predisposed to abnormal proliferation and, if enough mutations accumulate, to tumor formation. This 'loss of heterozygosity,' or LOH, is a rate-limiting step in tumorigenesis, but its causes are not fully understood.

One way in which LOH can occur is through mitotic recombination, in which homologous chromosomes exchange DNA when they are lined up during the process of cell division. A group led by Jay Tischfield (of Rutgers University) has now shown that the level of mitotic recombination in mice is affected by the degree of relatedness between the two sets of chromosomes, one set being inherited from each parent. Tischfield and colleagues studied mice that carry only one copy of the gene encoding the enzyme adenine phosphoribosyltransferase (Aprt). Mitotic recombination events in which Aprt is lost, resulting in LOH, can be identified because those cells completely lacking Aprt survive when cultured in the presence of diaminopurine (DAP), whereas cells containing Aprt cannot.

As recombination depends on a high level of DNA sequence identity, the authors hypothesized that mitotic recombination would be suppressed in the hybrid offspring of two distantly related strains of mice whose chromosomes have diverged in sequence. Indeed, fibroblasts isolated from such mice are found to have a dramatically lower frequency of mitotic recombination events when compared to fibroblasts taken from the offspring of two closely related strains. Interestingly, however, mitotic recombination is not suppressed in T-lymphocytes, suggesting that there are tissue-specific mechanisms involved.

In an accompanying News & Views article, Harry Vrieling (of Leiden University Medical Center, The Netherlands) expands on the relevance of these findings for human disease. For example, it may be that small genetic differences between individuals are sufficient to suppress mitotic recombination in certain tissues, thereby reducing the risk of cancer. And, in a provocative conclusion to their paper, the authors suggest that the degree of DNA sequence similarity between parents may be a contributing factor in determining the susceptibility of their offspring to diseases, like cancer, that are associated with mitotic recombination and LOH.

Nature Genetics pp 173 - 177

Four years after the announcement of Dolly the cloned sheep, scientists struggle to produce healthy cloned animals. Cloning attempts often result in spontaneous abortions and animals that survive to birth are often born obese and develop heart abnormalities, lung defects and other problems. A new study from Yong-Mahn Han (of the Korea Research Institute of Bioscience and Biotechnology, Taejon, South Korea) indicates that the developmental abnormalities of cloned embryos may be due to defects in the genetic imprint of the donor DNA.

Genetic imprinting is a process in which genes from the mother and father are chemically marked, so as to balance the expression of the two sets of genes. Some genes carry a mark (a methylation) on the maternal copy that keeps it turned off, so only the paternal gene is expressed. Other genes have a mark on the paternal gene, leaving only the maternal gene expressed. When embryos develop normally, a genome-wide erasure of the marks occurs, starting very early in development.

Han and colleagues studied a specific region of the genome in cells from cloned bovine embryos, and compared its methylation status to that of normal embryos. Unexpectedly, the cells of the cloned blastocysts have genomic methylation patterns that resemble those of the donor cells, but also show wide variation. This indicates that the early genome-wide reprogramming—the erasure of the methylation—has gone awry.

Aberrations in imprinting are already known to result in some human disorders; for instance, the chromosomes of people affected with Beckwith-Wiedemann syndrome have physical characteristics similar to those seen in clones. So, despite recent advances, significant hurdles—such as overcoming deficiencies in genetic reprogramming—must be cleared before animals can be routinely cloned.