World's first plastic antibody works in mice

http://www.newscientist.com/article/dn19031-worlds-first-plastic-antibody-works-in-mice.html

by bshep • 16 years ago

42 8 16 years ago

phren0logy• 16 years ago

Man, that is amazingly cool. Does this have implications for other problems that have to this point been limited by the difficulties of predicting the outcome of protein folding?

kiba• 16 years ago

Yes, it's nice and all that it works in mice.

But does it work in the human body?

ahk• 16 years ago

Nice! Hoping to see the "no known antidote exists" tag disappear by the next decade or so.

jballanc2• 16 years ago

Well, not to be a downer, but there are a number of caveats to be attached to this discovery. First, melittin is over 7.5 kDa which, while small for a protein, is still much, much larger than your typical household toxin. It is also larger than the smallest molecules that antibodies can be trained against.

Second, the interactions here appear to be entirely steric (that means size/shape related). One of the advantages of natural antibodies is that they can bind their targets via H-bonding, ionic interactions, van der Waals interactions, and hydrophobic interactions in addition to steric interactions.

Third, clearance by the liver runs the risk of just having all of the toxin concentrated in the liver, and it's not clear that the toxins will remain bound to the bead until they are inactivated. Certainly, if we're talking about a snake-venom which is a neurotoxin, and the "down-side" to taking this and not dying is that you need a liver transplant, I think it would still be useful, but this is not any sort of panacea I would think.

Edit: Also, it seems like the "imprinting" would require a fairly large interaction surface with the target molecule. This means that its probably not even very useful for molecular biology work, since natural antibodies can distinguish between proteins with very small differences and only require half-a-dozen or so amino acids for a match. Really, I think the lesson of this article should be just how amazingly cool natural antibodies are, and that it's taken until now to maybe come up with something that could possibly replace one of these uses.

ars• 16 years ago

> clearance by the liver runs the risk of just having all of the toxin concentrated in the liver

Could you bind the toxins, then do a full body blood transfusion?

nooneelse• 16 years ago

I believe there has been research taking these sorts of molecule imprinted polymers beyond strictly steric interactions. I'm fairly certain it was Nicholas Peppas that I saw talking about it at a conference; sorry I don't recall a more specific name closely connected to it. Iirc, they were first binding several other molecules to the target molecule, then enveloping the target plus extras in the polymer, and then getting rid of the target molecule but not the others. So not only is the shape of the target imprinted in the polymer, but also, around the target-shaped hole are several molecules that would bind to the target were it to get in the hole again.

Now that I think about it, I don't recall if it helped with the kind of selectivity wanted for this "antibody". They were more concerned, I think, with using the binding of the target to the extra molecules around the hole as a trigger for degrading the polymer.

World's first plastic antibody works in mice

Related Stories