Let's try to avoid creating something with "molecular acid for blood," shall we?

December 14, 2010 9:36 AM

Dmitar Sasselov is an astrophysicist, Director of the Origins of Life Initiative at Harvard and a co-investigator of the Kepler space telescope project to find Earth-like planets around the Cygnus constellation and discover extraterrestrial life. But no matter how successful the Kepler project may be, it still won't answer the most fundamental questions of astrobiology: How diverse is life in the universe? If alien life exists, will it have Earthly DNA and proteins? Or will it run on something else? So Dr. Sasselov has decided to collaborate with two synthetic biologists, asking them to create a life form based on mirror-image versions of what we know as the essential building blocks of living things on Earth.

From the Wired Article:

...go to the synthetic biologists Jack Szostak and George Church. Ask them to create a life-form that runs on an operating system different from our own, based on mirror-image versions of earthly proteins and DNA. Let these alien cells grow and mutate, and see how they survive. If it worked, those new cells—Church called them “mirror life”—could answer one of the deepest questions about the origin of life, not just here on Earth but everywhere in the universe. They might also open up new avenues of discovery in materials science, fuel synthesis, and pharmaceutical research. On the down side, though, mirror life wouldn’t have any predators or diseases to limit its reproduction. They would have to keep an eye on that.

In July, Dr. Sasselov's TED talk (also linked above) was widely misinterpreted by the press.

Kepler space telescope: previously on Metafilter. Also, this post is tangentially related.

posted by zarq (13 comments total)

....so I'm gonna guess these people have never cracked open a single sci-fi thriller, eh?
posted by The Whelk at 9:47 AM on December 14, 2010 [1 favorite]

The Whelk: "....so I'm gonna guess these people have never cracked open a single sci-fi thriller, eh?"

Well, they do say that anything created would have no defenses against our bacteria, viruses etc. So maybe they read War of the Worlds? :D
posted by zarq at 10:11 AM on December 14, 2010

As I understand it, a protein made from D amino acids will not be a mirror image of the same protein (by sequence) made from L amino acids. The primary structure will translate into two very different tertiary structures, which in turn will drastically affect function. That is one of several dozen things wrong with this idea.
posted by dephlogisticated at 10:13 AM on December 14, 2010 [1 favorite]

....so I'm gonna guess these people have never cracked open a single sci-fi thriller, eh?

Meh. It's all rehashes of Mary Shelley - either Frankenstein or The Last Man.

It's all Regency-era guilt tripping, and requires drama and pathos. "man plays god, creates monster" is more interesting to read than "man plays god, creates more cuddly puppy."
posted by Slap*Happy at 10:14 AM on December 14, 2010

"The Technical Error" by Arthur C. Clarke, first published 1954.
posted by DU at 10:16 AM on December 14, 2010

That is one of several dozen things wrong with this idea.

Too bad for the bioindustrial applications ruled out, but I'm kind of relieved. Given the option, I'd rather not be smothered under a carpet of ineradicable levobioforms.
posted by Iridic at 10:42 AM on December 14, 2010

I for one welcome our mirror-life overlords.
posted by DetonatedManiac at 11:19 AM on December 14, 2010

As I understand it, a protein made from D amino acids will not be a mirror image of the same protein (by sequence) made from L amino acids.

I really don't see why it shouldn't be. As far as I know high energy fundamental particle collisions and nuclear decay are the only things significantly different under a mirror symmetry.
posted by topynate at 11:28 AM on December 14, 2010

As I understand it, a protein made from D amino acids will not be a mirror image of the same protein (by sequence) made from L amino acids. The primary structure will translate into two very different tertiary structures, which in turn will drastically affect function. That is one of several dozen things wrong with this idea.

A relevant paper.
posted by kickingtheground at 12:06 PM on December 14, 2010

Even if it is a mirror image of the protein (which I don't think it is because the order of amino acids need to be flipped about somehow too, right?), I don't think it would interact with substrates the same way. I am testing this by trying to cover objects with my right and left hands and I can't make it work in all three dimensions. So I'm not using a terribly rigorous method, but everything the protein interacts with would also have to be different, I think.
posted by maryr at 12:11 PM on December 14, 2010

That every other molecule the organism interacts with would have to be a mirror image is a given; for example, L-glucose has no nutritive value for us, so D-glucose will surely have no nutritive value for a mirror organism. I was questioning that the protein has a 'very different' tertiary structure. It would be a mirror image, which is different, but not different in its participation in mirror image reactions.
posted by topynate at 12:57 PM on December 14, 2010

I'm going to withdrawal my initial objection. It was based on a conversation I had a few years ago with a structural biologist, wherein I inquired about this specific possibility; mirror proteins formed from D-amino acids. He claimed it wouldn't work, though in retrospect he did not bother to explain why in any detail. After doing some superficial lit review, I'm now thinking he may have been mistaken, and in any case I'm not sufficiently versed in structural biology to defend his position (I'm also reminded of how much stereochemistry makes my head hurt).

To add to the confusion, there are multiple ways in which a protein can be mirrored. In addition to the method referred to here, using D-amino acids in place of L-amino acids, a protein's structure can be mirrored by reversing the direction of the peptide backbone. Combining the two types of inversion gives four different permutations, with potentially different effects on tertiary structure. Then there's the issue of side-chain interactions; threonine and isoleucine, for instance, have two stereocenters, and presumably these would have to be reversed as well make the cross-links work. I think it could be possible in principle to create a functional mirror protein, but I'm not going to take a strong stance either way.

But putting aside the issue of protein structure, I still take a very skeptical view of the proposition mentioned in this article. It's not just one chicken-and-the-egg problem, but dozens of them. A mirror ribosome, for example, won't do you much good without mirror chaperones to assist in protein folding and mirror enzymes to perform post-translational modifications. Even if you're aiming for a stripped-down cell, you'll still need to build a nontrivial number of mirror components from scratch and somehow assemble them all. I just don't see it happening.
posted by dephlogisticated at 2:44 PM on December 14, 2010

Even if you're aiming for a stripped-down cell, you'll still need to build a nontrivial number of mirror components from scratch and somehow assemble them all. I just don't see it happening.

And for what it's worth, Dr. Szostak himself says that he thinks it's unlikely Dr. Church will succeed. But it's interesting to speculate about.

(I'm also reminded of how much stereochemistry makes my head hurt).

You and me both.
posted by zarq at 2:56 PM on December 14, 2010