lpetrich
29 Mar 2009, 06:51 PM
Evolution of Amino Acid Frequencies in Proteins Over Deep Time: Inferred Order of Introduction of Amino Acids into the Genetic Code (http://www.ncbi.nlm.nih.gov/pubmed/12270892), by Dawn J. Brooks, Jacques R. Fresco, Arthur M. Lesk and Mona Singh.
These authors reconstructed a large number of "Archean Park" proteins, those belonging to the inferred common ancestor of all cellular life. They worked out amino-acid frequencies in these proteins and then compared those frequencies to present-day ones.
The ancestral proteins had more of the amino acids that are abundantly produced in prebiotic-chemistry experiments and less of those that are rare or absent in such experiments.
Which is consistent with amino acids having been appropriated from some "Primordial Soup"; what reason would some intelligent designer have for creating proteins with this pattern of abundances?
Molecular evolution before the origin of species (http://www.ncbi.nlm.nih.gov/pubmed/12225777), By Brian K. Davis.
His paper used a smaller set of proteins than the earlier one I'd mentioned, but examined them in more detail. The oldest one found was low-potential ferredoxin, which has an iron-sulfur cluster in it and one end strongly acidic. This protein uses mainly the simpler amino acids, which suggests that it dates back before the completion of the genetic code. A conclusion supported by the more detailed examination in BD's paper, which also points to evidence of a gene duplication early in that protein's history -- two copies became concatenated to produce one new protein.
That protein's structure fits in well with Wachterhauser's picture of prebiochemistry as taking place in clay-like materials and involving iron/sulfur chemistry; the clay's metal ions would make it positively charged, attracting the negatively-charged ferredoxin end. The rest of the molecule would remove the hydrogen from H2S and transfer it to CO2, helping make organic acids (with COOH groups) and the like. Which easily stick to those metal ions, keeping them easily available for more (pre)biosynthesis.
BD's paper scored amino acids by biosynthesis-step distance from the Krebs Cycle, which it used as a reference point in early metabolism. The best scorers are aspartate and glutamate, whose aspartate -> asparagine and glutamate -> glutamine reactions were inferred to be an early form of fixing of nitrongen from dissolved ammonia. These close ones are also acidic, enabling them to easily stick to metal ions, as are all of the Krebs Cycle members.
After those ones were those simpler amino acids, those easily formed in prebiotic-chemistry experiments, and then the more difficult ones, which include the ones with a benzene ring (phenylalanine, tyrosine) and the alkaline ones (lysine, arginine).
The other proteins examined included two that are intimately involved with cell structure:
Part of ATPase, an enzyme that sits on cell membranes, and that assembles ATP from the energy of membrane-crossing ions.
The protein FtsZ, which is involved with the splitting-in-two aspect (septation) of prokaryotic cell division.
These are younger than ferredoxin, suggesting that ferredoxin is older than distinct cells. Which is logical when one considers its adaptation to sticking to mineral surfaces.
The formation of distinct cells was followed by the biosynthesis of the more difficult amino acids -- and DNA. But although a DNA-to-RNA enzyme dates to before the last universal common ancestor (LCA, LUA, LUCA), the DNA-to-DNA enzymes are inferred to be younger on account of their very different structures between the bacterial and the archaeo-eukaryotic branches of the family tree of life.
Meaning that the LUCA may not have had DNA-to-DNA copying, but instead, DNA-to-RNA-to-DNA copying in its genome. With DNA-to-DNA copying being invented twice in its descendants.
BD's paper had not addressed the question of the origin of RNA, though it or some predecessor had been at least as old as ferredoxin -- and older than cells.
Interestingly, one of the proteins discussed in BD's paper was a reverse transcriptase. Which makes me wonder what "normal" cellular function involves RNA-to-DNA transcription, because there would otherwise be no "incentive" to keep it.
Here are all the proteins discussed, in order of the "code age" of its amino acids:
"Code Age" = 7
Ferredoxin (electron transfer)
Proteolipid h1 (part of ATPase)
"Code Age" = 8 to 11
FtsZ (involved in cell division)
FEN-1 (flap exonuclease; digests RNA)
RNAP-beta (DNA-dependent RNA polymerase: DNA to RNA)
Reverse transcriptase (RNA to DNA)
"Code Age" = 12 to 13
Topoisomerase-I (involved in DNA supercoiling)
RNR (Ribonucleotide reductase: RNA to DNA nucleotides)
An amino acid's "code age" is determined by the number of metabolic steps needed to make it from a Krebs-Cycle precursor:
2: Aspartate, Glutamate, Asparagine, Glutamine
4: Alanine, Proline, Serine, Valine
5: Cysteine, Glycine
6: Threonine
7: Isoleucine, Leucine, Methionine
9: Arginine
10: Lysine
11: Phenylalanine, Tyrosine
13: Histidine
14: Tryptophan
The upper ones are inferred to be used before the lower ones.
And from that Brooks et al. paper on the LUCA's amino-acid content, here are the most abundant prebiotic amino acids:
Alanine, Aspartate, Glutamate, Glycine, Isoleucine, Leucine, Proline, Serine, Threonine, Valine
These authors reconstructed a large number of "Archean Park" proteins, those belonging to the inferred common ancestor of all cellular life. They worked out amino-acid frequencies in these proteins and then compared those frequencies to present-day ones.
The ancestral proteins had more of the amino acids that are abundantly produced in prebiotic-chemistry experiments and less of those that are rare or absent in such experiments.
Which is consistent with amino acids having been appropriated from some "Primordial Soup"; what reason would some intelligent designer have for creating proteins with this pattern of abundances?
Molecular evolution before the origin of species (http://www.ncbi.nlm.nih.gov/pubmed/12225777), By Brian K. Davis.
His paper used a smaller set of proteins than the earlier one I'd mentioned, but examined them in more detail. The oldest one found was low-potential ferredoxin, which has an iron-sulfur cluster in it and one end strongly acidic. This protein uses mainly the simpler amino acids, which suggests that it dates back before the completion of the genetic code. A conclusion supported by the more detailed examination in BD's paper, which also points to evidence of a gene duplication early in that protein's history -- two copies became concatenated to produce one new protein.
That protein's structure fits in well with Wachterhauser's picture of prebiochemistry as taking place in clay-like materials and involving iron/sulfur chemistry; the clay's metal ions would make it positively charged, attracting the negatively-charged ferredoxin end. The rest of the molecule would remove the hydrogen from H2S and transfer it to CO2, helping make organic acids (with COOH groups) and the like. Which easily stick to those metal ions, keeping them easily available for more (pre)biosynthesis.
BD's paper scored amino acids by biosynthesis-step distance from the Krebs Cycle, which it used as a reference point in early metabolism. The best scorers are aspartate and glutamate, whose aspartate -> asparagine and glutamate -> glutamine reactions were inferred to be an early form of fixing of nitrongen from dissolved ammonia. These close ones are also acidic, enabling them to easily stick to metal ions, as are all of the Krebs Cycle members.
After those ones were those simpler amino acids, those easily formed in prebiotic-chemistry experiments, and then the more difficult ones, which include the ones with a benzene ring (phenylalanine, tyrosine) and the alkaline ones (lysine, arginine).
The other proteins examined included two that are intimately involved with cell structure:
Part of ATPase, an enzyme that sits on cell membranes, and that assembles ATP from the energy of membrane-crossing ions.
The protein FtsZ, which is involved with the splitting-in-two aspect (septation) of prokaryotic cell division.
These are younger than ferredoxin, suggesting that ferredoxin is older than distinct cells. Which is logical when one considers its adaptation to sticking to mineral surfaces.
The formation of distinct cells was followed by the biosynthesis of the more difficult amino acids -- and DNA. But although a DNA-to-RNA enzyme dates to before the last universal common ancestor (LCA, LUA, LUCA), the DNA-to-DNA enzymes are inferred to be younger on account of their very different structures between the bacterial and the archaeo-eukaryotic branches of the family tree of life.
Meaning that the LUCA may not have had DNA-to-DNA copying, but instead, DNA-to-RNA-to-DNA copying in its genome. With DNA-to-DNA copying being invented twice in its descendants.
BD's paper had not addressed the question of the origin of RNA, though it or some predecessor had been at least as old as ferredoxin -- and older than cells.
Interestingly, one of the proteins discussed in BD's paper was a reverse transcriptase. Which makes me wonder what "normal" cellular function involves RNA-to-DNA transcription, because there would otherwise be no "incentive" to keep it.
Here are all the proteins discussed, in order of the "code age" of its amino acids:
"Code Age" = 7
Ferredoxin (electron transfer)
Proteolipid h1 (part of ATPase)
"Code Age" = 8 to 11
FtsZ (involved in cell division)
FEN-1 (flap exonuclease; digests RNA)
RNAP-beta (DNA-dependent RNA polymerase: DNA to RNA)
Reverse transcriptase (RNA to DNA)
"Code Age" = 12 to 13
Topoisomerase-I (involved in DNA supercoiling)
RNR (Ribonucleotide reductase: RNA to DNA nucleotides)
An amino acid's "code age" is determined by the number of metabolic steps needed to make it from a Krebs-Cycle precursor:
2: Aspartate, Glutamate, Asparagine, Glutamine
4: Alanine, Proline, Serine, Valine
5: Cysteine, Glycine
6: Threonine
7: Isoleucine, Leucine, Methionine
9: Arginine
10: Lysine
11: Phenylalanine, Tyrosine
13: Histidine
14: Tryptophan
The upper ones are inferred to be used before the lower ones.
And from that Brooks et al. paper on the LUCA's amino-acid content, here are the most abundant prebiotic amino acids:
Alanine, Aspartate, Glutamate, Glycine, Isoleucine, Leucine, Proline, Serine, Threonine, Valine