From: Robert J. Bradbury (bradbury@aeiveos.com)
Date: Sun Mar 09 2003 - 11:46:00 MST
On Thu, 6 Mar 2003, Joao Magalhaes wrote:
> Robert, I understand how we can apply whole genome engineering to
> microbes--particularly since I'm a microbiologist. Yet I imagine that you
> want to do the same for humans. May I ask how? Or is it some sort of
> industrial secret.
I too would like to know how... :-) And I think I do. There may be some
further problems down the road, but if look at the costs for DNA synthesis
right now it would cost you $1 million to synthesize a complete genome.
Nobody is going to even try that when they expect the first one will
not function correctly... Sure you could build one just like an existing
genome but what have you learned then? You would have to build a dozen
more changing them little by little to see if you understood things well
enough. You have to lower that cost. But I think I know how to solve
that problem. How I want to try to do that is an industrial secret.
> My reasoning is that for microbes we have dozens of
> genomes from which not only we can analyze what are the important parts,
> but also how to engineer a microbe's genome to suit our tasks--or even
> create a whole genome from scratch as I believe some scientists are trying
> to.
Some industrial engineering takes place, but I believe it is only in Japan
where they want to produce large quantities of vitamins (I'm reasonably
sure most of the vitamins sold in the world come from there) or in the
big Pharma companies where they want to produce antibiotics. In those
cases they may add multiple genes for the enzymes they need. An example
I think in eukaryotes would be "golden rice". But it is moderately
difficult work (to completely optimize the genome to maximize
production) and so people are only going to try it if there is
a pile of money to be made.
Venter's crew at the Institute for Biological Energy Alternatives
or perhaps the people from TIGR are trying to create a whole
genome from scratch but I think they are first planning to spend
a great deal of time to confirm what a minimal genome is.
That makes a certain amount of sense because given the cost
of assembling a genome one would want to make it as small as
possible.
> My problem with engineering the human genome is that we need a lot more
> mammalian genomes. You mention we can "patch our genomes using
> intracellular bacteria". The idea is certainly exciting, and I sure hope
> you can make it real in a few years. But I can think of so many technical
> problems in the idea that I wonder how you will pull it off. Even so I wish
> you the best of luck.
Oh, I don't intend to *start* there -- there is a lot of experience that
will be required before one attempts that. But I think we may be able
to get there in a decade. One of your problems will probably be solved
by US Genomics they are working on some methods that will allow very
fast and inexpensive sequencing. We will have a lot more genomes when
they finally produce results. And we will not be able to just throw
bacteria with whole genomes into your blood -- your immune system will
go crazy. But I think I've got at least two solutions for this problem
as well.
> As for HIV, I'm sure we will eventually find a cure. My argument is that
> the pace at which developments at a molecular level have been made in the
> fight against HIV will not suffice if we are to find ways to cure aging
> within a reasonable future.
RNAi for the transcribed genes, receptor blockers combined that allow
the virus to enter the cells with the current enzyme blocking drugs
are going to make it *pretty* tough for the virus to reproduce or spread.
I would tend to agree that up until now it has seemed "impossible".
But now we know all of the genes that cause the accelerated aging
diseases, except one. We know Xeroderma Pigmentosum, Trichothiodystrophy,
Fanconi's Anemia, Werner's Syndrome, Ataxia Telangectasia. The only
one we don't know is Progeria. Given our existing methods, it should
only be a few years to know who the proteins for these genes interact
with and what precisely they do. That is going to help alot (and helps
explain why progress has been so slow to date). Before we had a
fundamental lack of understanding as to what aging was.
> They won't even suffice to cure cancer.
Here you are partially correct. But very precise drugs like Gleevec
are performing *very* *very* well. There is also some good promise
for the angiogenesis inhibitors or similar drugs once they figure
out which combination(s) to use. I strongly suspect once you block
one angiogenesis protein the cancer just pulls another one out of the hat
since there are several of them.
> So we
> need faster ways to gather data, interpret results, and develop effective
> interventions in medical science. You give a few examples of how things are
> shaping up and I certainly hope they do because the current pace of medical
> progress won't be enough.
I gave a few above. Others would include the current rapid drug
screening methods.
> Finally, I'm totally in favor of a comparative genomics approach to study
> aging. As I mentioned in previous e-mails, I've been trying to publish an
> article on the subject for months now. Yet my approach is based on
> mammalian genomes, since I estimate several of these will be available long
> before we have, say, a turtle's genome.
Not if US Genomics pulls the rabbit out of the hat.
> All the best.
Same to you. Sorry it took so long for me to respond, I didn't previously
notice the note.
Robert
This archive was generated by hypermail 2.1.5 : Sun Mar 09 2003 - 11:50:59 MST