вход по аккаунту


How Unique Is the Genetic Code.

код для вставкиСкачать
Coded Amino Acid No. 22
How Unique Is the Genetic Code?**
Christiane Fenske, Gottfried J. Palm, and Winfried Hinrichs*
amino acids ¥ enzymes ¥ genetic code ¥
pyrrolysine ¥ RNA
In two recent publications in Science,
a new type of gene-encoded amino acid
has been described by the groups of
Krzycki and Chan, extending the genetic code to 22 amino acids. This modified
lysine side chain was identified by
genetic[1] and crystallographic[2] investigations and is clearly the active site in
enzymes that play an essential role in
catabolizing methylamines in methanogenic Archaea. Carbon assimilation and
methanogenesis in Methanosarcina species rely on specific methyl transferases,
which use mono-, di-, or trimethylamines as substrates.
In the first paper it is shown that this
new amino acid is added to the growing
polypeptide chain on the ribosome at
the position of an UAG stop codon,
which usually interrupts translation during protein synthesis. A specific tRNA is
used which is charged by its own tRNA
synthetase. The second paper describes
the characterization of the monomethylamine methyl transferase (458 amino
acid residues) from Methanosarcina barkeri by X-ray crystallography and identifies residue 201 as a 4-methyl-pyrroline-5-carboxylate-modified
called pyrrolysine.
The discovery that a stop codon does
not always mean ™stop∫, but may en-
[*] Prof. Dr. W. Hinrichs, Dipl.-Biol. C. Fenske,
Dr. G. J. Palm
Institut f¸r Chemie und Biochemie
Ernst-Moritz-Arndt-Universit‰t Greifswald
Soldmannstra˚e 16
17 489 Greifswald (Germany)
Fax: (þ 49) 3834-86-4373
[**] Financial support to W.H. by Fonds der
Chemischen Industrie is gratefully acknowledged.
In memory of Max Perutz
code a nonstandard amino acid in some
organisms urged us to reconsider the
universality of the genetic code. Could it
be that in former times every codon
defined a different amino acid and after
millions of years of evolution some rare
amino acids were replaced by common
ones that now have up to six different
codons? If this holds true, it could be
compared with the loss of languages in
the world: Chinese, English, and Arabic
are spoken by more and more people,
whereas each year many languages become extinct because indigenous people
are no longer able to uphold their
Structure and Chemistry of Pyrrolysine
X-ray crystallography reveals the
three-dimensional folding pattern of
(polypeptides, polynucleotides, etc.) and the
interaction of these macromolecules
with other biologically relevant molecules. This method opens the door to
understanding of the chemistry of biological processes at the atomic level, for
example, regulation mechanisms, chemical reactions, or transport phenomena.
The chemical composition of the crystallized compounds is usually known,
but not the conformation of the building
blocks and the specific type of interactions within the protein or polynucleotide and their recognition of ligands.
This information can still not be predicted very well from the sequence by
ab initio methods, and thus high-resolution structures are only provided by
crystallography or NMR spectroscopy.
Monomethylamine methyl transferase (MtmB) transfers the methyl group
of monomethylamine to the corrinoid
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4206-0606 $ 20.00+.50/0
cofactor of another associated methyl
transferase, MtmC. The crystal-structure analysis of MtmB from the archaea
Methanosarcina barkeri at 1.55 ä resolution revealed the chemical structure of
a previously unknown amino acid residue in the polypeptide.[2] The electron
density observed for the new residue
was interpreted as a lysine side chain
connected to 4-methyl-pyrroline-5-carboxylate through an amide linkage (Figure 1). Therefore the authors suggested
the name l-pyrrolysine for this novel
amino acid (three-letter code: Pyl; our
suggested one-letter code: O). The direct precursor of l-proline, the D1-pyrroline-5-carboxylate, does not fit to the
diffraction data because of the d chirality and the substituent at C4.
Electrospray mass spectrometry of
the MtmB/MtmC complex supports the
assignment as pyrroline-5-carboxylate
with an additional methyl substituent
at C4, and excludes the alternative
interpretation of the corresponding
electron density as an amino or hydroxy
substituent. The enzyme adopts an a/b
TIM barrel fold.[3, 4] Although there is no
significant sequence identity of MtmB
to other corrinoid-cofactor-associated
enzymes, the authors point out that the
topology is reminiscent of tetrahydrofolate:corrinoid/iron±sulfur methyl transferases, diol hydratase, and racemases/
As in the case of other methyl
transferases, the barrel of MtmB features a deep, negatively charged cavity,
which is clearly required to bind the
methylammonium cation substrate. The
position of the newly identified residue
at the bottom of this typical cavity
suggests its function in the catalytic
mechanism. Pyrrolysine positions and
displays the methyl group of the substrate for attack by the corrinoid cofacAngew. Chem. Int. Ed. 2003, 42, No. 6
tor. The proposed methylamine-activation mechanism is based on the two
different conformations of the active
site seen in the crystal forms obtained
with NaCl and (NH4)2SO4 as precipitating agents, diffracting to 1.55 ä and
1.70 ä, respectively. As a reaction intermediate an amine substituent at C2 of
after methyl transfer from methylamine
to the corresponding corrinoid protein,
The challenge in the interpretation
of electron-density maps at the borderline of atomic resolution is the disorder
of the pyrroline moiety in both structures. The pyrroline head group is found
in two orientations rotated approximately 908 relative to each other. The
relative occupancies of the alternate
orientations differ in the two crystal
forms, because different sets of hydrogen-bonding interactions with the protein determine the dominant conformation of the pyrrolysine side chain. However, the observed disorder in the two
structures also provides additional information on possible active-site conformations, which hints at the catalytic
mechanism shown in Scheme 1 and
supports the assigned imine double
bond (N1¼C2).
Genetic Background for Pyrrolysine Coding
Figure 1. a) Modeling of (4R,5R)-4-substituted-pyrroline-5-carboxylate to the 2 FOFC electron density (3s) of the NaCl crystal form (orientation 1). The substituent at C4 is shown as
a methyl, but it could also be an ammonium
or an hydroxy group. b) Structure of the proposed l-pyrrolysine amino acid. c) Residual
FOFC electron-density-difference map of the
(NH4)2SO4 crystal form after incorporation of
a 40 %-occupancy model consisting of l-pyrrolysine in orientation 1 and an exogenous ammonium ion. This remaining difference density suggests that l-pyrrolysine adopts a different orientation (orientation 2) at 60 % occupancy in (NH4)2SO4 with an amine at C2 of
the pyrroline ring. Reprinted with permission
from reference [2]. Copyright 2002 American
Association for the Advancement of Science.
the pyrroline ring was found in the
crystal obtained by precipitation with
(NH4)2SO4, but not in the other crystal
structure. This substituent undergoes
hydrogen bonding to Glu and Gln side
chains and most probably represents the
ammonium group, which is the product
Angew. Chem. Int. Ed. 2003, 42, 606 ± 610
All known organisms synthesize the
polypeptide chains of their proteins out
of a subset of 20 different l-amino acids.
The sequence of each protein is stored in
the DNA sequences and is copied to
mRNA (transcription) for polypeptide
syntheses, which determines the sequence-specific peptide bond formation
in the ribosomal machinery (translation). The fundamental textbook coding
system is based on 20 proteinogenic
amino acids, which are determined by
the variation of four different RNA
nucleotides. For almost 50 years it has
been known that each amino acid residue is encoded by a triplet of nucleotides in the polynucleotide chain of the
DNA and RNA molecules.
Hundreds of nonstandard amino
acids in proteins are known,[9] but these
are all formed by chemical modification
after ribosomal polypeptide syntheses
or in nonribosomal processes. The first
observed exception was selenocysteine,
which is found in proteins of various
organisms, from Archaea to mammals.[10±13] Selenocysteine is not formed
by posttranslational modification as is
the case for all nonstandard amino acids,
but is encoded by UGA, the so-called
opal stop codon.[14, 15] Now, more than
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
one decade later, pyrrolysine has been
found as another example.[1, 2] Krzycki
and co-workers identified an uncommon
tRNAPyl and a corresponding lysyltRNAPyl synthetase (PylS), which enable
the translation of the stop codon as
pyrrolysine in various methane-producing Archaea. The genes that encode
these methyl transferases have an inframe stop codon (UAG, so-called amber stop codon), which does not stop
ribosomal polypeptide elongation. Dimethyl and monomethyl methyl transferases have very similar genes in the
same genome, and the read-through is
very efficient, unlike other stop-codon
overrunning events.[16] The function of
UAG as a sense codon for this specific
type of protein was unknown in archaeal
genes. The translation of the stop codon
as pyrrolysine is made possible by a
specific tRNAPyl that can be charged
with lysine by a corresponding lysyltRNA synthetase (LysRS).
The 20 standard aminoacyl synthetases can be divided into two structurally
conserved classes. Interestingly, LysRS
enzymes are the only examples that
feature nonhomologous members in
both class I and class II.[17, 18] In M.
barkeri, two other LysRS enzymes were
identified: one belongs to class I and the
other to class II. Sequence alignment of
the newly identified synthetase suggests
that PylS represents a third type of
LysRS that probably belongs to a new
subclass of class II. There is no evidence
for modification of the mRNA (transcript editing) prior to translation.
Based on all the genetic and biochemical results, Krzycki and co-workers concluded that pyrrolysine is inserted into the polypeptide chain of MtmB
during translation. Unfortunately, pyrrolysine itself has not yet been synthesized chemically and could not be tested
as a substrate for PylS, which shows a
high affinity but a low reaction rate for
lysine. Pyrrolysyl-tRNAPyl is probably
synthesized by enzymatic condensation
of pyrroline-5-carboxylate to the amino
group of the lysine side chain of the
charged tRNAPyl.
Alternatively, tRNAPyl might be
charged with preformed pyrrolysine.
Both routes would allow direct translation of the UAG codon as pyrrolysine
in ribosomal polypeptide elongation.
However, an enzyme for either pathway
1433-7851/03/4206-0607 $ 20.00+.50/0
Scheme 1. Hypothetical model for the role of the amber-encoded residue in the catalytic cycle. The proposed intermediates for a), b), c), and f)
are based on the structures of l-pyrrolysine (X ¼ CH3, NH2, or OH) in orientation 1 (85 % occupancy) in the NaCl crystal form and on both orientations of l-pyrrolysine in the (NH4)2SO4 crystal form. Intermediates d) and e) are based on a preliminary docking model of MtmB with the corresponding corrinoid protein, MtmC. (Reprinted with permission from reference [2]. Copyright 2002 American Association for the Advancement of
remains to be identified. Since sustaining both pathways would require two
enzymes and pyrrolysine would compete with lysine as substrate for PylS,
Krzycki et al. favor the route in which
Pyl synthesis occurs on the tRNA molecule.
A strong analogy between pyrrolysine and selenocysteine is observed
because both are identified at positions
encoded by canonical stop codons.
Moreover, for both selenocysteine and
pyrrolysine, an uncommon tRNA
(which deviates structurally from the
predicted secondary structure) is involved.[19]
It has been noted that the readthrough of the selenocysteine ™stop∫
codon is only possible with the special
sequence and the structural organization of the corresponding mRNA, which
permits continuous decoding of two
separated open reading frames and, of
course, only this very subset of stop
codons will not terminate polypeptide
elongation.[20] The same mechanism can
be expected for pyrrolysine.
Deviation from the ™Universal∫
Variations in the reading of the
genetic code (as a result of a redefinition
of codons) can lead to the formation of
different products. This is possible by
the interaction of a specific tRNA with
the corresponding aminoacyl-tRNA
synthetase supported by specific features of the mRNA. There are two
possibilities to introduce amino acid
modifications prior to translation:
1. Instead of aminoacylating a tRNA
directly with a new amino acid, a
standard amino acid is attached to a
specific tRNA and modified enzymatically. This procedure is described
which is loaded with serine and
enzymatically converted into selenocysteine.[21]By the criteria discussed for Sec and Pyl we have at
least 23 encoded amino acids, because the long known formylmethionine (fMet) in bacteria can also be
considered as an encoded amino
acid: its AUG codon is recognized
as a special codon with help of the
sequence; tRNAfMet is exclusively used
for the start codon, charged with
methionine and enzymatically formylated to fMet-tRNAfMet prior to
translation. For methionines in other
than start positions a different tRNA
(Met-tRNAMet) is used, which cannot be formylated. Modifications of
standard amino acids attached to
specific tRNAs are known from
various examples. The use of tRNA
in the synthesis of amino acids is not
restricted to Sec, Pyl, and fMet. In
some instances, the amide side
chains of Asn and Gln are formed
only after their corresponding acids
(Asp, Glu) have been loaded onto
their tRNAs.[21]
2. In certain organisms or organelles,
the meaning of a subset of codons is
reassigned–that is, it differs from
the ™universal∫ code–and holds for
all mRNAs in which these codons
occur. It is well-known that the
Angew. Chem. Int. Ed. 2003, 42, 606 ± 610
mitochondrial and nuclear reading
of mRNA is different. For example,
in humans and other mammals,
UGA is a stop codon in nuclear
mRNA, but encodes Trp in mitochondrial mRNA. Other examples
and differences within various species are reported[22] and can be found
in databases.[23]
RNA and Convergent Evolution
The involvement of RNA not only in
translation but also in amino acid synthesis might be a relict of an RNA world.
The distribution of Sec in all kingdoms
and of fMet in bacteria points to an
ancient origin. It is likely that the
substrate of the methylamine methyl
transferases was already available in the
primordial soup. The corrinoid methyl
acceptor has also been hypothesized to
be accessible for prebiotic chemical
Most of the 20 standard amino acids
have degenerated codons. For example,
the four triplets for alanine could be the
simplified pool of several different aliphatic amino acids that behave similarly.
Of the small, aliphatic l-amino acids (up
to C5), only the linear 2-aminobutanoic
and -pentanoic acids are missing today;
they may have been replaced by alanine,
which shows similar helix-forming propensity.
The reason to keep an amino acid
during evolution should be either structural or functional, and indeed Sec and
Pyl are catalytic residues. Apparently,
organisms with a reduced set of coded
amino acids were able to survive evolution because of more efficient metabolic
pathways. Based on these statements,
the approximately 20 amino acids today
seem to be a reduced set rather than an
expanded one; as genome sequencing
and proteome analysis progresses further, more relicts may be found.
Entry to an Expanded Set of Encoded Amino Acids
Special tRNA and aminoacyl-tRNA
synthetases are the key to modify the
transcription signal. This might be a
general approach to design proteins for
biotechnological applications by inAngew. Chem. Int. Ed. 2003, 42, 606 ± 610
creasing the actual genetic repertoire
with a variety of amino acids with novel
physical, structural, and chemical properties, which are not found in the
standard set of 20±23 amino acids.
Novel amino acids have been introduced into proteins by using the biosynthetic machinery of the Escherichia
coli ribosome. In E. coli mutants that
charge tRNAVal incorrectly with cysteine, 20 % of the valine in the cellular
protein could be replaced by the noncanonical aminobutyrate, sterically similar to cysteine.[25] Modified pairs of
were used to incorporate in vivo synthetic O-methyl- or O-allyl-l-tyrosine
into enzymes in response to an amber
stop codon.[26]
Moreover, it happens that a gene
that is transferred into another organism
is not read with the expected efficiency
by the new host, because a certain codon
for an amino acid is not as well-recognized. These codons, which are not used
often, support the hypothesis that in
former times more than 23 amino acids
were encoded by distinct base triplets.
Genetic Variability on All Levels
Whereas in humans the combination
of XY or XX defines male or female,
respectively, in some bugs (Insecta:
Heteroptera) only the X chromosome
is required: the females have two (XX)
and the males only one (X0). In butterflies, some fish, amphibians, reptiles, and
birds, the females are heterogametic
(WZ) and the males homogametic
Thus, not only words (codons) of the
genetic code can be read with different
meanings. Even whole books (chromosomes) of the library (genome) can be
interpreted with surprising variations. It
is not only the written genetic information (DNA sequence), but also the
biochemistry of the reading system
(mRNA, tRNA, synthetases, ribosomes), that determines the sequence
and the biological function of the gene
[1] G. Srinivasan, C. M. James, J. A.
Krzycki, Science 2002, 296, 1459 ± 1462.
[2] B. Hao, W. Gong, T. K. Ferguson, C. M.
James, J. A. Krzycki, M. K. Chan, Science 2002, 296, 1462 ± 1466.
[3] D. W. Banner, A. C. Bloomer, G. A.
Petsko, D. C. Phillips, C. I. Pogson,
I. A. Wilson, P. H. Corran, A. J. Furth,
J. D. Milman, R. E. Offord, J. D. Priddle,
S. G. Waley, Nature 1975, 255, 609 ± 614.
[4] C. A. Orengo, A. D. Michie, S. Jones,
D. T. Jones, M. B. Swindells, J. M.
Thornton, Structure 1997, 5, 1093 ± 1108.
[5] T. Doukov, J. Seravalli, J. J. Stezowski,
S. W. Ragsdale, Structure 2000, 8, 817 ±
[6] J. Masuda, N. Shibata, Y. Morimoto, T.
Toraya, N. Yasuoka, Structure 2000, 8,
775 ± 788.
[7] F. Mancia, N. H. Keep, A. Nakagawa,
P. F. Leadlay, S. McSweeney, B. Rasmussen, P. Bosecke, O. Diat, P. R. Evans,
Structure 1996, 4, 339 ± 350.
[8] R. Reitzer, K. Gruber, G. Jogl, U. G.
Wagner, H. Bothe, W. Buckel, C. Kratky,
Structure 1999, 7, 891 ± 902.
[9] G. C. Barrett in Amino acid derivatives:
a practical approach (Ed.: G. C. Barrett), Oxford University Press, New
York, NY, 1999, pp. 15 ± 30.
[10] J. F. Atkins, A. Bˆck, S. Matsufuji, R. F.
Gesteland in The RNA World (Eds.:
R. F. Gesteland, T. R. Cech, J. F. Atkins), Cold Spring Harbor Laboratory
Press, Cold Spring Harbor, NY, 1999,
pp. 637 ± 673.
[11] A. Bˆck, K. Forchhammer, J. Heider, W.
Leinfelder, G. Sawers, B. Veprek, F.
Zinoni, Mol. Microbiol. 1991, 5, 515 ±
[12] M. Rother, R. Wilting, S. Commans, A.
Bˆck, J. Mol. Biol. 2000, 299, 351 ± 358.
[13] M. J. Berry, R. M. Tujebajeva, P. R.
Copeland, X. M. Xu, B. A. Carlson,
G. W. Martin 3rd, S. C. Low, J. B. Mansell, E. Grundner-Culemann, J. W. Harney, D. M. Driscoll, D. L. Hatfield, Biofactors 2001, 14, 17 ± 24.
[14] I. Chambers, J. Frampton, P. Goldfarb,
N. Affara, W. McBain, P. R. Harrison,
EMBO J. 1986, 5, 1221 ± 1227.
[15] F. Zinoni, J. Heider, A Bˆck, Proc. Natl.
Acad. Sci. USA 1990, 87, 4660 ± 4664.
[16] L. Paul, D. J. Ferguson, J. A. Krzycki, J.
Bacteriol. 2000, 182, 2520 ± 2529.
[17] D. Moras, Trends Biochem. Sci. 1992, 17,
159 ± 164.
[18] M. Ibba, S. Morgan, A. W. Curnow,
D. R. Pridmore, U. C. Vothknecht, W.
Gardner, W. Lin, C. R. Woese, D. Sˆll,
Science 1997, 278, 1119 ± 1122.
[19] S. Commans, A. Bˆck, FEMS Microbiol.
Rev. 1999, 23, 335 ± 351.
[20] R. F. Gesteland, J. F. Atkins, Annu. Rev.
Biochem. 1996, 65, 741 ± 768.
[21] M. Ibba, D. Sˆll, EMBO Rep. 2001, 2,
382 ± 387.
[22] S. Osawa in Evolution of the Genetic
Code, Oxford University Press, Oxford,
UK, 1995.
Utils/wprintgc.cgi?mode ¼ c.
[24] G. Ksander, G. Bold, R. Lattmann, C.
Lehmann, T. Fr¸h, Y.-B. Xiang, K.
Inomata, H.-P. Buser, J. Schreiber, E.
Zass, A. Eschenmoser, Helv. Chim. Acta
1987, 70, 1115 ± 1172.
[25] V. Dˆring, H. D. Mootz, L. A. Nangle,
T. L. Hendrickson, V. de Crÿcy-Lagard,
P. Schimmel, P. Marliõre, Science 2001,
292, 501 ± 504.
[26] a) L. Wang, A. Brock, B. Herberich,
P. G. Schultz, Science 2001, 292, 498 ±
500; b) Z. Zhang, L. Wang, A. Brock,
P. G. Schultz, Angew. Chem. 2002, 114,
2964 ± 2966; Angew. Chem. Int. Ed.
2002, 41, 2840 ± 2842.
[27] Rolf Siewing, Lehrbuch der Zoologie,
Vol. 1, Allgemeine Zoologie, Gustav
Fischer, Stuttgart, 1980, p. 297.
Angew. Chem. Int. Ed. 2003, 42, 606 ± 610
Без категории
Размер файла
122 Кб
unique, genetics, code
Пожаловаться на содержимое документа