Duke Scientists Map 'Silenced Genes'

palegray.net writes "Wired reports on new research into the phenomenon of 'silenced genes', genetic constructs that have no 'partner' in case one goes wrong over the course of your lifetime. Scientists at Duke University have mapped some 200 genes that may 'play a profound role' in the health of the average human. 'Many of the newly found imprinted genes are in regions of chromosomes already linked to the development of obesity, diabetes, cancer and some other major diseases, the researchers reported ... Scientists had thought imprinted genes would account for about 1 percent of the human genome. While scientists must double-check that the newly identified ones are truly silenced, the new map matches that tally.'"

Comments

Uh

[goldaryn]

From TFA:

Sometimes imprinting goes awry before birth, leaving a normally silenced gene "on" or silencing one that should not be.
...
Now a question is how imprinting may be changed to reactivate an imprinted gene after birth.

Am I the only one concerned by this statement?

Re:With

[wizardforce]

you're missing the point entirely. This isn't just a few cells that are affected, this is your entire population of cells. if you inherit a gene from your father that is "switched off" and that gene is the only one that can be inherited [only from Y chromosome for example] then you're kind of screwed. There isn't a second copy that is switched on and functional to prevent the associated disease caused by the first gene being switched off. There are a lot of these kind of genes that are regulated in the levels of expression and on/off states of the gene. Hardly mundane.

Re:With

[LooTze]

Knowing that one of the copies is imprinted helps in three ways -

(a) It helps provide people with genetic counseling e.g. helps in deciding if you want to continue with a pregnancy if you know that your fetus has a genetic defect on the paternal copy (and the maternal copy is silenced) by sequencing an amniocentesis sample.

(b) More fundamental to this is that, is that this might help pin down a gene defect as the cause of a disease. For example you might find some locus often associated with a disease but in the patients you sequence the genes, it turns out one of them has got a perfect copy and the other has a mutation. Since it is difficult to say for a majority of mutations if they would affect function or are simple polymorphisms in the population, you continue searching other genes. OTOH, if you know one of the copies is shut down, and you see one copy has a mutation, you promptly analyze this candidate gene a lot more.

(c) Finally, of course for proper cure, it helps to know what the defect is. e.g. if you know it is a defect in an ion transporter, you might try some types of drugs and if it is an inflammatory defect you will try something else - so (b) is useful in guessing plus making animal models to test them.

(d) there is the hope that one day we will be able to fix things gene therapy which again is dependent on figuring out the molecular defect by (b).

Re:With

[wizardforce]

Completely irrelevant. Unless you are in a position to DO something about it.

just a note, the university I am doing my studies was one of the first to work in this area so yes I could probably end up doing something about this.

We already know there are many lethal gene combinations, that produce in utero abortion or neonatal death. Your point is?

just a note that not all changes in gene expression cause in utero abortions.

Now if you could point out a case of a single gene being altered AFTER embryogenesis (by environmental factors or whatever) that produces disease, then we're talking about eventually being able to work on a way of preventing this.

even better a whole mechanism for many genes being altered- anything that methylates cytosine residues will alter gene expression although they also increase the incidence of point mutations because of hyrolysis of 5'cytosine to thymine.

However as far as I know, altering the gene in a single cell will damage THAT cell

sigh... you over simplify this far too much. You never considered the fact that these kind of changes in expression become very important in egg and sperm cells which combine to produce the next generation passing on the change in expression that occured in one single cell. if the same thing happens in the zygote you can see that that single change in expression for that particular cell can have great consequences.

While perhaps you might also be able to identify non-desirable traits that are more likely to be passed on to offspring, this won't be much use until you start obliging probable carriers to be sterilized.

gene expression can be altered possibly alleviating illness, not only that but eugenics is out of the question.

Is this how you plan to "fight" disease? I am not sure I want to live in that world.

the only one that implied such a future is *you* for I certainly do not want to see such a future.

Re:With

[Anonymous Coward]

Now if you could point out a case of a single gene being altered AFTER embryogenesis (by environmental factors or whatever) that produces disease, then we're talking about eventually being able to work on a way of preventing this. However as far as I know, altering the gene in a single cell will damage THAT cell only. Now with the exception of the cancers (which I made in my previous post) please point out a disease resulting from a mutation of a single cell?

It's not a mutation! You inherit two copies. One of these copies is disabled EPIGENETICALLY. It's not mutated, it's methylated. This is a reversible process. The choice of which of these two copies is methylated is very important, because some tissues/cells work with the paternal and others with the maternal copy. This is why some times you get diseases with variable penetrance (autosomal dominant, but not always present or present in variable degrees, for example).

Concerning the treatment, if you can reverse or control epigenetic silencing, you can alter a persons genetic profile by choosing which copy (paternal or maternal) is active. This is currently impossible.