This is another archival repost, this time from oct 2007. The post is part two in a series. The series so far can be found here.
Knowing how and why things go wrong tells us a lot about how and why things work when they go right. Indeed, this is such an important principle in basic and medical research that the Nobel prize for physiology or medicine was awarded for discoveries of mechanisms which can be used to artificially switch off genes, in both 2006 (RNA interference) and 2007 (gene targeting). Similarly, developmental disorders give us clues about how normal development occurs. This week’s syndrome is Cri du chat (OMIM:#123450). Cri du chat is caused by deletions in the short arm of chromosome 5. The deletion may be as small as a single gene, though larger deletions, involving more genes, have more severe effects. Symptoms include low birth weights, poor muscle tone at birth, and slow growth; unusual facial characteristics, such as a round face, widely spaced eyes, small chin, and low set ears; severe psychomotor and mental retardation, due to abnormal brain architecture; and the characteristic “cat cry” sound of the newborn, caused by abnormal larynx development.
How can such a diverse range of symptoms arise from such a small loss of data? Surely we should expect a change in one or two genes to correspond to a change in one or two features? This is the impression one might get from media coverage of genetics. The coverage seems to be improving, but we are still regularly treated to stories about the discovery of the gene “for” characteristic X. One of the genes lost in cri du chat could be described as “for” high ear position, since its absence causes low set ears. Another[1] could be described as “for” normal speech without cat-like cries. The reality for the majority of genes is that it is impossible to connect them directly to any physical or mental characteristic. Take the three genes that are most commonly involved in cri du chat. One of the genes is called TERT, short for telomerase reverse transcriptase, and it produces part of an enzyme (a piece of molecular machinery) called telomerase. During cell division, a duplicate of the genome is produced: for each of the chromosomes, a set of machinery clamps on, and moves along producing a copy. However, this machinery can not copy the first few letters at the very tip of the chromosome — it has to clamp on there, and so it gets in the way of itself! — so, over time, the chromosomes reduce in length, and eventually the genes become at risk. The most immediate purpose of telomerase is to produce some gibberish DNA with which to cap the chromosome, and thus protect the genes at the ends of chromosomes. Expression of TERT is activated at several stages in development, and in response to certain events. Notably, from the point of view of cri du chat, it appears to be used early in the development of the nervous system, and to a lesser extent in the survival of neurons. However, those aren’t its only roles: it is also known to be involved in maintaining the immortality of stem cells, and in cells’ responses to external signals, such as the hormone estrogen and cytokines, both of which are in turn involved in a range of functions, including regulating cell division and programmed cell death. Indeed, you may already have heard of telomerase, because it is the subject of research in aging and cancer.[2] Another gene involved in cri du chat is Semaphorin F (also known as Sema4C). The product of this gene is a protein which sits on the surface of the growing neuron, helping to guide it to the cells with which it should interact. However, it may also moonlight in the immune system, and bone development. The third gene is delta-catenin, whose product sits on the surface of dendrites — the branches of the neuron — and interact with another protein on neighbouring cells to keep the neurons together and communicating.[3]
These genes, then, can not be described as directly responsible for anything, except the proteins that they produce. They may directly map to a molecular phenotype, but they only indirectly map to visible anatomical, physiological and mental phenotypes, and may have a hand in many diverse systems throughout the body. At the level of anatomical structures and psychological circuits, so many genes must be involved to produce and fine tune even the smallest components that it often becomes difficult to determine their specific roles, or to visualise how the complexity of the macro emerges from the micro-world. Genes, therefore, can not be described as being “for” anything other than the molecule that they encode. What we have learnt about development from what happens when things go wrong therefore also shows us that there are limitations to what we can learn from these situations. Those who use RNA interference and gene targeting to “switch off” genes must treat their findings with caution: there are so many variables confusing the situation that it is difficult to draw concrete conclusions about anything above the molecular level. And we must bear it in mind throughout this series: when the situation appears to be simple, it probably is not. Next week, we’ll invite in environmental variables, and make the situation an order of magnitude more complicated again.
References
- ^ Or perhaps the same one
- ^ Mattson MP, Fu W, Zhang P (2001) Emerging roles for telomerase in regulating cell differentiation and survival: a neuroscientist’s perspective. Mech. Ageing Dev. 122 (7): 659-71.
- ^ I. Israely, R. M. Costa, C. w. Xie, A. J. Silva, K. S. Kosik and X. Liu. (2004) Deletion of the neuron-specific protein delta-catenin leads to severe cognitive and synaptic dysfunction. Current Biology 14:1657-63.