DNA Construction Set
So what else is DNA good for besides evolving meddlesome creatures that cover the planet with reactive oxygen, then others that breathe the oxygen and start looking for other planets? DNA–and its predecessor, RNA–are the most amazing building sets ever known.
A recent article in Science shows how DNA can build self-programmed 3D structures. The most amazing part is how the linear 1D sequence of DNA bases actually determines the 3D structure. The way this works is that short stretches of DNA have bases complementary to other short stretches elsewhere in the long strand. Wherever the short stretches line up, they will twist around each other to form a short double helix. But then, the strands come apart to form double helix with other stretches elsewhere on the molecule.
Why would life’s information molecule have such intriguing properties? If we look at DNA’s precursor, RNA, we see that life is full of 3D RNA molecules composed of “hairpin” turns. The most dramatic example is the RNA skeleton of the ribosome, life’s ubiquitous protein factory. Here is what that looks like; the RNA is the winding turquoise-green part.
What else do you think are DNA and RNA are good for?
Ultraphyte 
There is the area of DNA computing. While Adelman’s early use was interesting in solving smallish problems, I think that DNA might have a more general role as a a structural component for self assembly of more traditional computers. Not as small scale as graphene perhaps, but self assembly would make for low cost manufacture.
I’m also impressed with the work going on at MIT to program cells. DNA resumes its traditional role, but the code programs a very different biological function.
I would still bet on DNA as being most interesting in this role – allowing us to reprogram life functions and new materials. Cells that compute, and new protein fibers for a host of applications
DNA computers have taken off into a whole field, tough for me to grasp–there are certain problems it can do that silicon can’t.
Also, DNA is being used for masks or scaffolds to fabricate things.
Depending on bond strengths, this could be heading towards your idea of making “beanstalk” cables out of anthrax (The Highest Frontier).
There’s a thought. I think though that the aromatic bonds of nanotubes are stronger than the phosphate backbone, in terms of tensile strength.
I’m waiting until they can get proteins to fold reliably. Then we’ll really see some 3-D building.
I find it interesting that life’s mostly gone for proteins and cellulose as the structural component (with all the nitrogen that demands) and limited DNA to the information storage role. This suggests that, fun as it is to work with DNA 3-D shapes, there are better options out there.
In the shorter run, DNA sequence generation is a fairly mature technology, so it will certainly be used for nanomechanical applications.
The “G” diagram resembles a viral case, but I’m concerned about durability, compared to lipoprotein membranes. If there was life consisting only of RNA (more likely than DNA), it must have had a pretty moderate environment in terms of pH, temp, electrical charge…
The ribosome fascinates me endlessly: It’s a highly complex mechanism, a biological computer with few errors or changes among all the life forms on this planet. If there’s ever a proof of creationism (sarcasm), the ribosome is where they ought to be looking.
The reasons living cells use protein more than DNA are:
(1) Diverse chemical function of amino-acid R groups. Proteins are more than structural; they have to do chemistry. If we make functional things out of DNA, we’ll have to modify the bases to have chemical functional groups.
(2) Double-stranded DNA is less flexible than protein, so it cannot make smaller things. The objects pictured above are actually quite large as cell parts go.
As for RNA, we think that RNA originally performed all the functions that protein and DNA provide today (in the first cells of the RNA world). Some cell parts still do use RNA as if it were protein, for catalysis. Back in the RNA world, RNA probably had a much wider range of bases–as transfer RNA does today.
Nevertheless, structural materials are mostly made of sugars – cellulose and chitin. Where CaCO3 is available, layers of CaCO3 and proteins. Proteins are of course also used in silks.
Given that RNA was possibly more functional in early life, it should be asked why it largely gave up its potential role as a structural material, as well as data storage and enzymatic.
While biologists are starting to add new amino acids to the protein repertoire, as well as changing the genetic code, it might well be worth asking what could be achieved if RNA and DNA was modified to handle more kinds of bases? Perhaps little for organisms, but potentially something for manufacturing?
I see the base pair binding as the key feature that one would want to exploit. We see it in some forms of DNA computing and in the images presented in the OP. The structural strength properties are likely inferior to sugars and proteins and therefore not the area of suitable application.
Why RNA “gave up its structural role” — Actually, RNA still has the dominant structural/functional role in the ribosome and in other “small nucleoprotein particles” such as intron splice machines. However, some roles of RNA have diversified and specialized: DNA was better at long-term storage, while protein had more diverse chemistry.
“What could be achieved if RNA and DNA were modified” — Instead of “more kinds of bases” I think secondary modification of bases would be useful (as in living chromosomes).
DNA can be stronger when the phosphodiester bonds are replaced with protein-like peptide bonds (this is called PNA). Even so, DNA molecules have lasted in fossils for 30,000 years or more (enough to sequence much of Neanderthal genome).
One thing that’s bad about using DNA or RNA for structure is the amount of phosphorus embedded in the structure. The advantage of things like carbohydrates (especially in plants) or calcium composites (in animals) is that the elements are mostly surplus. Cells run not just on elements, but on a ratio of elements (such as the Redfield ratios). Having a surplus of an element around isn’t necessarily a good thing. Plants, for example, are very good at fixing carbon. Things like wood, keratin, and chitin are great because they are carbon sinks. The carbon in these compounds serves a useful, structural role, without taking much metabolic energy for maintenance, and it takes surplus carbon out of the cytoplasm. I suspect calcium in shells and bones is performing much the same role, taking an abundant element out of circulation while providing support and protection.
Thing is, phosphorus supplies are generally very limited (as P is insoluble), so DNA and RNA are always going to be a bit limited. Proteins face a similar issue with nitrogen shortages.