On Mon, 6 Sep 1999, Technotranscendence wrote:
> On Sunday, September 05, 1999 11:20 PM "Hal" hal@finney.org wrote:
Hal, I think you are making a stretch here unless you have some
source I'm unaware of. I've briefly glanced at Dowdy's papers
on the TAT fusion proteins and he claims the mechanism isn't
currently known. He does cite one paper that "claims" the
uptake is not receptor mediated. But Not-receptor-mediated
does not equal protein-unfolding.
> > It is interesting to learn that the reason proteins are able to get
> > into cells is because they are capable of unfolding.
I've done a little further research and there are processes such as endocytosis, pinocytosis and very odd structures called caveolae that are lipid "rafts" with a very high protein content (esp. of caveolin proteins). I think I've seen electron micrographs that show unusual cave-like indentations in the cell membrane.
My suspicion is that the TAT-protein-conjugate may be associated with these processes and is brought into the cells in ways not dissimilar from that of lipoprotein particles or the cholera toxin. You may want to think about how the cells do the normal "cleanup" when nearby cells undergo apoptosis. I do not believe that this involves macrophages, so all cells must have large protein uptake mechanisms. There is a fair amount of evidence that whatever the mechanism is, it is also involved in the uptake of naked DNA (since the above processes are believed to be involved in the delivery of DNA-vaccines and antisense therapies.
[An interesting side-note here is that we are discovering interesting therapeutic uses for biochemical processes developed by the HIV virus... Dowdy's work is using the HIV TAT protein fragment to introduce a carefully engineered protein that causes cell death if and when the cell becomes infected with HIV. This new protein effectively functions as a "trojan horse". Very very slick work.]
> > But apparently it is necessary that they unfold
> > back into their linear form (or something close to it) in order to pass
> > through the narrow channels through cell walls. Then the refold once
> > the get to the other side.
> >
> > This might pose a problem for a nanotech device which wants to make its
> > way into the cell. It will not be able to unfold and so will need to
> > hack its way through the membrane somehow. Perhaps the upcoming book
> > Nanomedicine will include a chapter on breaking and entering.
The process is discussed. Getting into the cell is relatively easy compared with navigating through the cytoskeleton. You shouldn't view the cell wall like a house wall. A lipid bilayer is a relatively flimsy barrier and a nanobot easily has enough energy to poke through it. One difficult problem may be getting in in "stealth" mode. Probably one of the reasons asbestos is so toxic is because the fibers are small enough and sharp enough that tissue movement caused by breathing causes them to poke through lipid bilayers causing the activation of defense mechanisms. Nanobots will have to enter slowly and carefully. But this is doable, since there are a number of intracellular bacteria that have figured out how.
>
> I see three potential solutions. One would be to have the nanotech device
> do some sort of contortion such as proteins to, changing shape to fit
> through, then changing back. (Correct me if I'm wrong, but inside most
> cells the contents are not just in suspension, but there are also structures
> and channels which guide molecules about. Inside it's not purely a
> diffusion-reaction area, right?)
Yep. Robert discusses shape-changing a little, but it is probably simply easier to adopt a javelin like shape. Or you could be a round, as I believe mycobacteria are and they probably use the normal cellular uptake mechanisms (discussed above).
Most small molecules operate on diffusion, larger molecules may be actively transported. Probably one of the major developments that had to occur in going from bacteria to eukaryotes was the development of the active transport processes for larger molecules. If you had to wait for diffusion your growth rate would be really slow.
>
> Another is to have the nanotech device, as Hal puts is, hack it's way
> through, most like repairing what damage its done (or having another device
> do the repairs) after it's inside. Of course, there might be areas to break
> into which are easier to pass through or that repair themselves...
Yep. Nanobots can a have lipid membrane coat and actually fuse with the membrane (the same way soap bubbles fuse). Lipid mebranes are by definition self healing.
>
> Still another solution might be to create a nanotech "gate" ("nanogate"?)
> which will allow nanotech devices to pass into and out of cells. (Of
> course, then we have the problem of how to create such and for some
> applications such gates would be overdoing it.)
You simply attach hydrophobic lipid molecules to the outside of your diamond/saphire shell. It essentially then works the same way protein channels work now only they can be much larger and can open and close much faster. The image that comes to mind is the entrance "door" in the Dyson sphere in the Star Trek NG episode/movie.
A nanogate for nanobots would be a fairly large structure and would seem to be tying up a lot of material/intelligence for what is probably a rarely done function. Of course if it is doing other jobs like functioning in the transponder/positioning network it might make sense.
One application would be a nano-transport-bot, functioning as a "fork-lift" for the cell. It would contain a number of active gates/channels combined with high speed pumps and molecular sorting rotors to selectively control ion flow in/out of the cell. This can be used for lots of interesting things like triggering nerve impulses, absorbing calcium release during strokes to prevent apoptosis, sequestering metal ions to reduce the potential for DNA damage, etc.
Robert