[realpatriot71]

I don't think the problem here is with surface proteins, but rather with gene switches in the DNA. Already differentiated cells type nuclei are unable to bring about the complete formation of an organism after insertion into a zygote, at this point. What I think science needs to find is a way to play with the gene switches.

[Nebullis]

I don't think the problem here is with surface proteins, but rather with gene switches in the DNA.

Good point. But it turns out that many of these genetic switches are controlled by surface proteins which are activated by changes in the extracellular environment.

[aruanan]

I don't think the problem here is with surface proteins, but rather with gene switches in the DNA. Already differentiated cells type nuclei are unable to bring about the complete formation of an organism after insertion into a zygote, at this point. What I think science needs to find is a way to play with the gene switches.

One of the reasons a particular cell is a particular cell is because its progenitors have responded to a series of factors that have pushed its differentiation in one direction rather than another. Some of these factors act through surface receptors such as the heterotrimeric G proteins. Some diffuse in. Some apparently work directly through a mechanical connection between cell-surface receptors and the gene in question. These means are the switches that drive gene expression. DNA all on its own is just a passive strip of protein recipes and doesn't control anything. It responds to pre-existing cellular conditions. Finding ways to manipulate certain conditions can control gene expression. Finding ways to control gene expression can change the conditions and, hopefully, alter the identity of the cell. Perhaps a particular disease phenotype is due to a mutation that affects the behavior of a certain splice variant which happens to be the developmental default isoform. Instead of trying to do gene therapy and replace the entire gene, it may be possible to preferentially induce expression of one of the other splice variants that could work just as well and so partially or completely ameliorate the disease phenotype. Of course, if the error was far down within a developmental chain such that D depended on C, and C on B, and B was screwed up by something happening in A, then by the time you get to the point of altered D or non-existing D, it may be too late to fix. However, there could be certain embryonically-lethal or perinatally lethal mutations that could be sidestepped by this approach at an early enough point of intervention. That would be really cool. Something like this is a standard technique anyway with a rescue transgene (this is a way of finding out just what role a particular gene does or doesn't have in development). In Drosophila you can have a lethal mutation for a protein that is absolutely critical for development. There may be a larva which simply will die without pupating. However, if you transfect the egg with a transgene with an inducible promoter for that gene, you can turn it on and get the larva over the developmental hump, after which it may be fine. Unfortunately, the way things have been going with humans, even if it were possible to avoid such a developmental blind end, genetic tests that revealed the condition early enough on would just be used to counsel the mother to abort the kid, "for his own good".