Slashdot Mirror
New Amino Acid Discovered
EricMargel writes: "As published in Science, researchers at the Ohio State University claim to have discovered the 22nd known amino acid, pyrrolysine, the first discovered since 1986." I hope rice and beans are still sufficient to get all the needed amino acids.
There are lots of information sourced from documents at this page.
Maybe the discover will revolutionize the way humans feed - should help French vine to be even more flavorful too? :)
The amino acid they discovered in 1986 is selenocysteine, which is also encoded for by a STOP codon (UGA in this case). Maybe there is an entire class of amino acids that are encoded in this manner, in between the 20 directly encoded amino acids and the multifarious post-translationally modified amino acids (e.g., hydroxyproline and hydroxylysine in collagen; gamma-carboxyglutamate in various clotting factors)
And you probably need more than just a STOP codon to incorporate pyrrolysine. With selenocysteine, you need enzymes to convert the serine residue on the tRNA to selenocysteine, an enzyme to activate the inorganic selenium, and a modified translation factor that recognizes this special case.
*Most* but not all enzymes are proteins. Many pieces of RNA, and a few pieces of DNA are also enzymes.
Ah! Here's the original article: Code Breakers. It's definitely worth a read.
Steven N. Severinghaus
Looks like you're right about it being a stop codon. Actually, the article does tell you which one (in jargon):
"Then in 1998, they published a paper showing that the gene had a component called an in-frame amber codon that behaved unusually."
"Amber" = UAG
Nor is it obvious why certain radicals are vital, and most are not. Some of the common radicals are missing in the vital amino acids. Hydrogen and methyl are there, but ethyl, propyl and higher n-alkanes are not. Yet isopropyl, and both 1 & 2-methylpropyl are. Wierd. Perhaps it has something to do with the way exclusionary mechanisms to keep undesirably amino acids out of the protein building machinery.
From an information-theory viewpoint, why are the DNA sequences largely incompressible? Are the three-base pair codons (6 binary digits each) equally probable? Those codons could be decoded into 64 possibilities, yet we have only 22 amino acids. Are some of the codons used for amino acid pairs? Or else we've got alot of missing acids. Untils those codons are themselves decoded (and any bigrams, tridgrams, etc), we should expect surprises. And what of the great expanse of alleged junk? Does nature have a signal-to-noise ratio approaching that of USENET? :)
Yes, yes there are 64 possible amino acid encodings. However, by only yielding 22 possible amino acids, the system provides a level of redundancy. The redundancy usually occurs around the third nucleotide, for instance, Ser can be coded AGU and AGC. This redundancy compensates for "point mutations", mutations that effect only one nucleotide. Because of the redundancy of the genetic code, the effect of point mutations is reduced by about 1/4. From looking at the old Codon Table, it is clear that the new amino acid was coded by UAG (the "amber" discussed in the article). However, UAG obviously stops for some organisms (Scientists dont make things up), so perhaps UAG stops unless the complementary tRNA can be found? If this is the case, then it is likely that all three if the "stop" codons code for certain new amino acids in some organisms. The trick, of course, is finding them.
It's not surprising that there are tRNA's in rare organisms that encode for "non-standard" amino acids -- evolution just selected against them, since the common 20 are so prevalent and easy to produce or obtain from food. Humans actually use 22 amino acids, but two of them are not genetically encoded, but produced by modifying the finished protein (hydroxylation of proline and lysine during collagen biosynthesis. Rice and beans are not sufficient, you need vitamin C to make collagen) Some bugs live in places where "non-standard" amino acids are probably preferred to make proteins more suited to the enivronment -- extreme conditions like Antarctic ice, or thermal vents.
It's important to remember that amino acids aren't the only building blocks -- cell membranes are made of lipids, cholesterol, and polysaccharides (sugars). There are many possible modifications beyond the amino acid sequence. For instance, immune markers (blood type, etc.) are sugar chains which are tacked onto proteins. Sugars on the surface of viruses help them bind to cells. Another common modification is phosphorylation: addition of phosphate to a protein, which is a common method of activating (or deactivating) proteins.
The body also uses lipid derivatives, steroids, and most importantly vitamins to obtain chemical functions not provided by amino acids (catalysis, cell signaling, etc.)
Incidentally, while the genetic code is pretty much universal, there are some variations. For example, in mitochondria, instead of functioning as a STOP codon, UGA encodes for tryptophan; instead of coding for isoleucine, AUA encodes for methionine; instead of coding for arginine, AGA and AGG function as STOP codons.
This case is special not because of the use of a non-standard amino acid, but because it is an *additional* amino acid rather than a replacement. This means that the machinery of translation of an RNA codon to an amino acid (via tRNA) and the construction of the amino acid (via an enzyme) exists in parallel with the machinery for all the other existing amino acids. This is remarkably interesting because it represents a much larger genetic difference in the amino acid translating machinery, and a difference which we have never seen before.