Structural basis for the exceptional stability of bisaminoacylated nucleotides and transfer RNAs

Article abstract of DOI:10.1021/ja203994e

At least one bisaminoacyl-tRNA is synthesized in nature (by Thermus thermophilus phenylalanyl-tRNA synthetase), and many disubstituted tRNAs have been prepared in vitro. Such misacylated tRNAs are able to participate in protein synthesis, even though they lack the free 2-OH group of the 3-terminal adenosine moiety. Their ready participation in protein synthesis implies significant chemical reactivity. The basis for this reactivity has been documented previously. Surprisingly, the aminoacyl moieties of these tRNAs also exhibit exceptional chemical stability. In the present report, bisaminoacylated nucleotides are investigated computationally and experimentally to define the basis for the stability of such species. Molecular modeling of bisalanyl-AMP in the absence of solvent and in the presence of a limited number of water molecules revealed two common features among the low-energy structures. The first was the presence of H-bonding interactions between the two aminoacyl moieties. The second was the presence of a H-bonding interaction between the 2-O-alanyl moiety and the N-3 atom of the adenine nucleobase, typically mediated through a water molecule. The prediction of an interaction between an aminoacyl moiety and the adenine nucleobase was confirmed experimentally by comparing the behavior of bisalanyl-AMP and bisalanyl-UMP in the presence of model nucleophiles. This study suggests a possible role for the adenosine moiety at the 3-end of aminoanyl-tRNAs in controlling the stability and reactivity of the aminoacyl moiety and has important implications for the reactivity and stability of normal aminoacyl-tRNAs.

Full text of DOI:10.1021/ja203994e

ARTICLE
pubs.acs.org/JACS
Structural Basis for the Exceptional Stability of Bisaminoacylated
Nucleotides and Transfer RNAs
Maria Duca,† Carl O. Trindle, and Sidney M. Hecht*,‡
Departments of Chemistry and Biology, University of Virginia, Charlottesville, Virginia 22904, United States
S Supporting Information
b
ABSTRACT: At least one bisaminoacyl-tRNA is synthesized in nature
(by Thermus thermophilus phenylalanyl-tRNA synthetase), and many
disubstituted tRNAs have been prepared in vitro. Such misacylated
tRNAs are able to participate in protein synthesis, even though they
lack the free 20-OH group of the 30-terminal adenosine moiety. Their
ready participation in protein synthesis implies significant chemical
reactivity. The basis for this reactivity has been documented pre-
viously. Surprisingly, the aminoacyl moieties of these tRNAs also
exhibit exceptional chemical stability. In the present report, bisami-
noacylated nucleotides are investigated computationally and experi-
mentally to define the basis for the stability of such species. Molecular
modeling of bisalanyl-AMP in the absence of solvent and in the
presence of a limited number of water molecules revealed two
common features among the low-energy structures. The first was the presence of H-bonding interactions between the two
aminoacyl moieties. The second was the presence of a H-bonding interaction between the 20-O-alanyl moiety and the N-3 atom of
the adenine nucleobase, typically mediated through a water molecule. The prediction of an interaction between an aminoacyl moiety
and the adenine nucleobase was confirmed experimentally by comparing the behavior of bisalanyl-AMP and bisalanyl-UMP in the
presence of model nucleophiles. This study suggests a possible role for the adenosine moiety at the 30-end of aminoanyl-tRNAs in
controlling the stability and reactivity of the aminoacyl moiety and has important implications for the reactivity and stability of
normal aminoacyl-tRNAs.
’ INTRODUCTION
While bisphenylalanyl-tRNA is presently the only tandemly
activated tRNA reported as a constituent of a biochemical system,
it has been possible to prepare a number of bisaminoacylated
tRNA transcripts.6,7 This has been achieved by modification of a
scheme used to produce misacylated tRNAs, in which a tRNA
transcript lacking the 30-terminal CpA sequence8 is ligated enzy-
matically to an aminoacylated pdCpA that has been prepared by
chemical synthesis.9 By the use of chemically synthesized bisa-
minoacylated pdCpA derivatives, it is also possible to prepare
bisaminoacylated tRNAs containing the same6,7 or different6a,10
amino acids on the 20-and 30-positions of adenosine (Scheme 1).
The bisaminoacylated tRNA transcripts prepared by the
strategy outlined in Scheme 1 have been shown to participate
quite efficiently in protein biosynthesis in in vitro systems.6,7 By
defining the structural elements in the bisaminoacylated tRNAs
required to support their participation in protein synthesis and by
studying the reactivity of modified bisaminoacylated pdCpA
derivatives with model N-nucleophiles, we have been able to
formulate a chemically plausible mechanism consistent with the
observed participation of the bisaminoacylated tRNAs in ribo-
somally mediated protein synthesis.7
Aminoacyl-tRNAs are essential to the process of translation
of mRNA to form proteins, being responsible for the introduc-
tion of the appropriate amino acids into the growing peptide
chain. This in turn requires activation of the aminoacyl moiety for
the peptidyltransferase reaction.1 Each tRNA usually bears an
amino acid linked to the 20- or 30-hydroxyl groups on the 30-
terminal adenosine moiety; the 30-regioisomer group participates
in protein synthesis.2 In the generally accepted mechanism,
the 20-OH group plays an essential role catalyzing peptide bond
formation, probably by acting as a “proton shuttle” between
the amino acid in the 30-position and the growing peptide chain.3
However, it has recently been shown that the 20-OH group
of adenosine is not essential for protein synthesis and it was
suggested that a nucleophile located on 23S RNA close to the
ribosomally bound tRNAs acts as the proton shuttle.4
It has been shown that phenylalanyl-tRNA synthetase from
Thermus thermophilus is able to form a tandemly activated tRNA,
leading to the formation of bis-20,30-(O-phenylalanyl)-tRNAs
using T. thermophilus tRNAPhe as a substrate.5 Escherichia coli
tRNAPhe and an RNA transcript identical in sequence with E. coli
tRNAPhe5 were also shown to undergo bisaminoacylation with
phenylalanine.
Received: April 30, 2011
Published: June 07, 2011
r
2011 American Chemical Society
11368
dx.doi.org/10.1021/ja203994e J. Am. Chem. Soc. 2011, 133, 11368–11377
|

Journal of the American Chemical Society
ARTICLE
Scheme 1. Synthesis of Mono-20(30)-O-alanyl-AMP (4a), Mono-20(30)-O-alanyl-UMP (5a), Bis-20,30-O-alanyl-AMP (4b), and Bis-20,
30-O-alanyl-UMP (5b)
Figure 1. General structures of the bisaminoacylated derivatives studied: (A) tandemly activated tRNA, (B) bisaminoacyl-AMP, and (C) bisaminoacyl-UMP.
In addition to providing new insights into the mechanism of
aminoacyl-tRNA reactivity in protein synthesis, the bisaminoa-
cylated tRNAs hold the promise of permitting significantly
increased amounts of proteins to be elaborated in vitro. This
is a consequence of the findings that both of the amino acid
moieties of bisaminoacylated tRNAs are incorporated into
protein and, especially, that the bisacylated species have much
greater chemical and biochemical stability than normal mono-
aminoacylated tRNAs. The latter feature permits in vitro protein
synthesis to proceed over an extended time period.
One facet of the behavior of bisaminoacylated tRNAs that
is not well understood is their exceptional stability. Presently,
we report a computational and experimental study of bisami-
noacylated nucleotides that is focused on the stability issue. The
results of this study identify H-bonding between the aminoacyl
moieties as one likely source of the observed stability of the
bisaminoacylated nucleotides and tRNAs. A second, and more
surprising, source of stability is a H-bonding interaction between
the 20-O-aminoacyl moiety and N-3 of the adenine nucleobase at
the 30-end of the tRNA. The present findings have important
implications for the sources of reactivity and stability of normal
aminoacyl-tRNAs during protein synthesis.
’ RESULTS
The chemical and biochemical data reported previously6,7
permitted the formulation of a molecular mechanism for protein
synthesis when bisaminoacyl-tRNAs are employed.6a,10 How-
ever, these data provided no insight concerning the molecular
basis for the chemical stability of bisaminoacyl-tRNAs and the
corresponding pdCpAs. Thus, a molecular modeling study was
undertaken initially in order to identify the possible interactions
responsible for the chemical stability of these species.
Computational Study of the Stability of Bisacyl-AMP
and Bisacyl-UMP Derivatives. Bisaminoacyl-AMP (Figure 1B)
11369
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
Figure 2. Computationally determined energetically favorable structures of a (phosphate-methylated) 50-AMP derivative containing alanyl esters at the
20 and 30-positions. (A) Representative structure illustrating the strong interaction of a hydronium ion with N-3 of the adenine nucleobase as well as both
alanyl moieties. Hydronium Oꢀnucleobase N-3 distance, 2.77 Å; hydronium OꢀNR (20-alanyl) distance, 2.60 Å; hydronium Oꢀcarboxyl O (30-alanyl)
distance, 2.63 Å. (B) Representative structure illustrating a weaker interaction of water with N-3 of the adenine nucleobase as well as both alanyl moieties.
Water Oꢀnucleobase N-3 distance, 3.14 Å; water OꢀNR (20-alanyl) distance, 3.06 Å; water Oꢀcarboxyl O (30-alanyl) distance, 3.24 Å. (C)
Representative structure illustrating the interaction of water with both alanyl esters. Water Oꢀcarboxyl O (20-alanyl) distance, 3.04 Å; water Oꢀcarboxyl
O (30-alanyl) distance, 2.90 Å. The diagrams to the extreme right illustrate potential H-bonding interactions.
was chosen as a simplified model to study possible three-dimen-
sional structures of bisaminoacylated pdCpA derivatives and the
interactions that could be realized to confer the observed chemical
stability to the molecule. Using the software Spartan, a series of
torsional isomer scans was conducted beginning with molecular
mechanics and PM3 models, followed by HF/STO-3G and finally
B3LYP/6-31G(d) calculations. This procedure has been applied
to (i) the neutral structure, (ii) structures with one of the amino
groups of the aminoacyl chains protonated, (iii) neutral structures
including one water, as well as (iv) structures containing a
protonated water molecule. In each case, about 20 structures
survived the screeningand werejudged by B3LYP/6-31G(d) tolie
within 5 kcal/mol of the most stable form. The most stable
molecules contain certain common features: hydrogen bonds link
11370
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
In analogy with the method used previously for the preparation
of bisaminoacyl-pdCpAs,6,7 the tetra-n-butylammonium (TBA)
salts of AMP and UMP were prepared in order to render these
compounds soluble in organic solvents. As outlined in Scheme 1,
the TBA salt of AMP was treated with N-(4-pentenoyl)-S-alanine
cyanomethyl ester (1) in anhydrous DMF containing Et3N. After
3 days of stirring at room temperature, the reaction mixture was
first analyzed and then purified by HPLC using a C18 reversed-
phase column with a gradient of 0% f 63% CH3CN in 50 mM
NH4OAc, pH 4.5, over a period of 35 min. Mono-20(30)-O-
[N-(4-pentenoyl)-S-alanyl]-AMP (2a) and bis-20,30-O-[N-(4-
pentenoyl)-S-alanyl]-AMP (2b) were obtained in 40% and 60%
yields, respectively. Each of the purified compounds (2a and 2b)
was separately treated with iodine in a mixture 1:1 THFꢀH2O
for 50 min. HPLC purification using the same conditions
described above afforded 20(30)-S-alanyl-AMP (4a) and bis-
20,30-O-alanyl-AMP (4b) in 80% and 75% yields, respectively.
Analogously, the TBA salt of UMP was treated with N-(4-
pentenoyl)-S-alanine cyanomethyl ester (1)11 in anhydrous DMF
containing Et3N. After 3 days of stirring at room temperature, the
reaction mixture was analyzed and purified by HPLC using a C18
reversed-phase column and a gradient of 0% f 63% CH3CN in
50 mM NH4OAc, pH 4.5, over a period of 35 min. Mono-20(30)-
O-[N-(4-pentenoyl)-S-alanyl]-UMP (3a) and bis-20,30-O-[N-(4-
pentenoyl)-S-alanyl]-UMP (3b) were obtained in 25% and 75%
yields, respectively. Each compound was then treated with iodine
in a mixture 1:1 THFꢀH2O for 50 min. HPLC purification using
the same conditions described above afforded mono-20(30)-O-
alanyl-UMP (5a) and bis-20,30-O-alanyl-UMP (5b) in 98% and
57% yields, respectively.
Chemical Study of the Stability of Bisacylated AMP Com-
pared to pdCpA Derivatives. The treatment of the synthesized
molecules in the presence of nucleophiles was performed in
order to model the chemical conditions during acyl transfer and
peptide bond formation and, more generally, to understand the
stability of the bisaminoacylated nucleotides. Compounds 4a, 4b,
5a, and 5b were treated with aqueous n-butylamine or imidazole
in order to characterize their chemical behavior and permit
comparison to the analogous pdCpA derivatives 6a and 6b
studied previously (Figure 3).7 Each compound was incubated
for 2 h in the presence of 2 mol equiv of aqueous n-butylamine or
imidazole, and aliquots were taken periodically. The reaction
mixtures were then analyzed by C18 reversed-phase HPLC after
15, 30, 60, and 120 min. AMP derivatives 4a and 4b were first
studied and compared to the respective pdCpA analogues. The
use of AMP in comparison to pdCpA was intended to improve
our understanding of the possible role of the dC moiety of
pdCpA in conferring stability to the aminoacylated nucleotides.
Concerning the monoacylated derivatives, it was found that 4a
and 6a were completely deacylated within 2 h in the presence
both of n-butylamine and imidazole (Figure 4A). There was no
difference in the kinetics of deacylation between the pdCpA and
AMP derivative. The bisacylated derivatives, 4b and 6b, were
much more stable than the corresponding monoalanine deriva-
tives, as expected (Figure 4B).6,7 However, the kinetic profile
indicated that deacylation of the AMP derivative 4b was slightly
faster than that of the pdCpA derivative 6b in the presence both
of n-butylamine and imidazole.
Figure 3. Compounds synthesized and studied for their reactivity toward
nucleophiles. Mono-20(30)-O-alanyl-AMP (4a), bis-20,30-O-alanyl-AMP
(4b), mono-20(30)-O-alanyl-UMP (5a), bis-20,30-O-alanyl-UMP (5b),
mono-20(30)-O-alanyl-pdCpA (6a), and bis-20,30-O-alanyl-pdCpA (6b).
the aminoacyl moieties, and also link the 20-aminoacyl moiety with
the N-3 ring atom of the adenine nucleobase. Figure 2 provides
examples of structures in which a hydronium ion exhibits a strong
interaction with N-3 of the adenine nucleobase and both aminoa-
cyl moieties (A), water interacts with the nucleobase and two
aminoacyl moieties (B), and water interacts with the two aminoa-
cyl moieties (C). Parallel studies of models of bis-acyl UMP
showed no H-bonding of types A or B, as is to be expected from
the structural differences between bisacyl UMP and bisacyl AMP.
Synthesis of New Monoaminoacylated and Bisaminoacy-
lated Derivatives of AMP and UMP. The computational study
of AMP derivatives strongly suggested that the two NH2 groups
of amino acids in the 20- and 30-positions and the N-3 atom of
adenosine could interact via the formation of hydrogen bonds
and were thus involved in the observed chemical stabilization6,7
of bisaminoacyl-pdCpAs and tRNAs. The chemical synthesis of
appropriate bisaminoacylated derivatives was carried out in order
to permit experimental study of the interactions suggested by the
computational data. Thus, bis-20,30-O-alanyl-AMP (4b) and bis-
20,30-O-alanyl-UMP (5b) (Figure 3) were synthesized as well as
their corresponding monoaminoacylated analogues, 4a and 5a.
Although a slight difference in the reaction rates between AMP
and pdCpA derivatives was observed for the bisaminoacylated
species, the data lead to the conclusion that the chemical behaviors
of the AMP and pdCpA derivatives are quite similar and that the
11371
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
Figure 4. (A) Kinetic analysis of the deacylation of mono-20(30)-O-alanyl-AMP (4a) and mono-20(30)-O-alanyl-pdCpA (6a) in the presence of 2 equiv
of n-butylamine (red and orange, respectively) and imidazole (dark blue and light blue, respectively). (B) Kinetic analysis of the deacylation of bis-20,30-
O-alanyl-AMP (4b) and bis-20,30-O-alanyl-pdCpA (6b) in the presence of 2 equiv of n-butylamine (red and orange, respectively) and imidazole (dark
blue and light blue, respectively). The percent formation of AMP and pdCpA after treatment is shown over a period of 120 min as analyzed by reversed-
phase HPLC (250 ꢁ 4.6 mm, 5 μm column). The column was washed with 0% f 63% CH3CN in 50 mM NH4OAc, pH 4.5, over a period of 35 min at a
flow rate of 1 mL/min (monitoring at 260 nm). Aliquots were analyzed after 15, 30, 60, and 120 min.
aminoacylated AMP can be used experimentally as a simplified
model for pdCpA.
of ribose was probably responsible for the greatly increased
reactivity of the bisalanyl-UMP derivative. The data obtained for
compounds 5a and 5b in panels A and B of Figure 5 are replotted
as Figure 6. As is clear from the latter figure, in complete con-
trast to the mono- and bisaminoacylated derivatives of AMP and
pdCpA, bisalanyl-UMP (5b) was actually deacylated at a rate
comparable to that of monoalanyl-UMP (5a) in the presence of
n-butylamine and imidazole.
Chemical Study of the Stability of Bisacylated UMP Com-
pared to AMP and pdCpA Derivatives. Both the molecular
modeling study described above and previously published data
on the reactivity of mono- and bis-aminoacyl-pdCpAs suggested
the possible participation of the adenine nucleobase in the stabili-
zation of the bisaminoacylated nucleotides. The use of UMP
analogues precludes possible hydrogen-bond formation in direct
analogy with AMP and could potentially confirm the hypothesis
regarding the stabilizing participation of the adenosine N-3 atom.
At first, mono-20(30)-O-alanyl-AMP (4a) and mono-20(30)-O-
alanyl-UMP (5a) were treated in the presence of n-butylamine
and imidazole. As shown in Figure 5A, both compounds were
sensitive to nucleophilic deacylation, leading to 90%ꢀ100%
deacylated product after 2 h in the presence of n-butylamine or
imidazole. The AMP derivative was slightly more reactive than
the UMP derivative in both cases. Figure 5B shows the compar-
able kinetic profiles for bis-20,30-O-alanyl-AMP (4b) and bis-20,
30-O-alanyl-UMP (5b). While 4b was deacylated to the extent
of ∼40% after 2 h, 5b was completely deacylated at the same
reaction time in the presence of n-butylamine and >80% deacy-
lated when treated with imidazole.
’ DISCUSSION
The 20,30 cis-diol moiety at the 30-terminus (A76) of tRNA has
important functions at a number of steps in the overall process of
protein translation, in addition to providing the attachment site
for the activated amino acid or peptide.12 In solution, the amino
acid or peptide ester attached to the A76 ribose is in rapid
equilibrium between the 20- and the 30-hydroxyl groups,13 but
the 30-linked ester is the regioisomeric form of the substrate that
participates in the peptidyltransferase reaction at both the A- and
P-sites.2 Thus, the 20-aminoacyl (peptidyl) ester is not required
to form any essential covalent linkage. While the detailed molec-
ular mechanism of protein synthesis catalyzed by the ribosome is
not yet completely elucidated, a “proton shuttle” model has been
suggested, which would involve the 20-OH group in the pepti-
dyltransferase reaction.12,14 In fact, this hydroxyl group may
participate as a H-bond acceptor from the amino group of the
aminoacyl moiety and also donate a H-bond to the 30-oxygen of
This result confirmed what had been suggested by the
computational study. The absence of hydrogen bonding between
the N-3 of adenosine and the aminoacyl moiety at the 20-position
11372
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
Figure 5. (A) Kinetic analysis of the deacylation of mono-20(30)-O-alanyl-AMP (4a) and mono-20(30)-O-alanyl-UMP (5a) in the presence of 2 equiv of
n-butylamine (red and green, respectively) and imidazole (blue and yellow, respectively). (B) Kinetic analysis of the deacylation of bis-20,30-O-alanyl-
AMP (4b) and bis-20,30-O-alanyl-UMP (5b) in the presence of 2 equiv of n-butylamine (red and green, respectively) and imidazole (blue and yellow,
respectively). The percent formation of AMP and UMP after treatment is shown over a period of 120 min as analyzed by reversed-phase HPLC (250 ꢁ
4.6 mm, 5 μm column). The column was washed with 0% f 63% CH3CN in 50 mM NH4OAc, pH 4.5, over a period of 35 min at a flow rate of 1 mL/min
(monitoring at 260 nm). Aliquots were analyzed after 15, 30, 60, and 120 min.
ribose, leading to the activation of both groups involved in the
peptidyltransferase reaction. However, a recent study suggested
that the proton shuttling from the incoming aminoacyl-tRNA to
the peptidyl-tRNA during peptide bond formation need not
involve the 20-OH group of the peptidyl-tRNA; it seems possible
that some other nucleophile spatially close to this tRNA acts as
the proton shuttle.4
Despite the importance of the discovery that bisaminoacy-
lated tRNAs are synthesized in nature by Thermus thermophilus
phenyalanyl-tRNA synthetase, such tandemly activated tRNAs
have not been studied in detail, probably because of the difficulty
of obtaining them. The techniques developed for synthesizing
misacylated tRNAs in vitro has facilitated the preparation and
the detailed study of tandemly activated tRNAs.6,7,10 While the
substitution of the 20-hydroxyl group with different functional
groups (such as 20-OCH3 or 20-H) led to tRNAs with diminished
facility of function in protein synthesis,15ꢀ17 its substitution with
a second aminoacyl moiety results in complete maintenance of
activity in protein synthesis and also confers a surprising chemical
stability to the molecule.6,7
group of corresponding monoaminoacyl-pdCpAs in activating
the aminoacylmoiety for participationinpeptidebondformation.7
However, the molecular basis for the distinctive chemical stability
of the bisaminoacylated pdCpA and tRNA derivatives remains
to be explored and superficially seems at odds with the ability
of these species to participate in protein synthesis. In order to
understand the molecular basis for the observed stability of bisa-
minoacylated pdCpA derivatives and their corresponding tRNAs,
a molecular modeling study of simplified bisaminoacylated analo-
gues was undertaken. As described above, low-energy conforma-
tions of bisaminoacyl-AMP (Figure 1B) were investigated. Water
and hydronium establish a variety of types of hydrogen-bonding
interactions between aminoacyl chains and between the aminoacyl
moieties and the adenine nucleobase. The latter finding is
unprecedented and prompted us to investigate the structure of
bisaminoacyl-UMP (Figure 1C), which should be incapable of
analogous stabilization.
In order to confirm experimentally the role of the adenine
nucleobase in the chemical stability of bisaminoacyl-pdCpAs,
bisalanyl-AMP and UMP have been prepared. The synthesis of
these compounds was carried out in analogy to the synthetic pro-
cedure reported for the preparation of bisaminoacyl-pdCpAs.6
The use of AMP derivatives allowed us to determine whether the
cytidine residue adjacent to the adenosine bearing the aminoacyl
moieties could influence the reactivity of the aminoacyl moiety,
The ability of bisaminoacylated tRNAs to participate in protein
synthesis has been analyzed by biochemical investigation of
suitably modified bisaminoacyl-pdCpAs and tRNAs bearing free
or protected NH2 groups on the aminoacyl moieties.6a,7 It has
been demonstrated that the NH2 functional group of the 20-
aminoacyl residue can serve a role analogous to that of the 20-OH
11373
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
play a key role in the chemical stability of such compounds.
Monoaminoacyl- and bisaminoacyl-UMP have essentially the
same sensitivity to nucleophiles (Figure 6), while the AMP deriva-
tives clearly do not (cf. Figures 4 and 5).
It has recently been shown that the transesterification rate of
aminoacylated adenosine derivatives depends strongly on the
presence both of free vicinal 20,30-OH groups and the nature of
the attached nucleobase.18 This supports the hypothesis that the
nucleobase in tRNA position A76 can help to define the reactivity
of the aminoacyl moiety of an activated tRNA.
On the basis of these theoretical and experimental results, we
suggest that in the presence of water, three hydrogen-bonding
interactions are present in bisaminoacylated adenosines, as shown
in Figure 2. These interactions must contribute to the chemical
stability observed for this class of molecules. Further, the possibility
of hydrogen bonding between the NH2 group of 20-aminoacyl
moiety and the carbonyl oxygen of the 30-aminoacyl moiety in
bisaminoacyl-tRNAs likely activates the aminoacyl moiety for the
peptidyltransferase reaction during protein synthesis.7
’ CONCLUSIONS
The present findings provide strong computational and experi-
mental support for the participation of the N-3 atom in the
adenine nucleobase as a determinant of the stability and reactivity
of the activated amino acid added to the growing polypeptide
chain from aminoacyl-tRNAs during protein synthesis. It seems
unlikely that the presence of adenosine as the 30-terminal
nucleotide in every tRNA in all organisms has occurred by
chance, but rather helps to define the stability and reactivity of
(mono)aminoacylated tRNAs.19 There is also direct evidence
for the involvement of the 30-terminal adenosine moiety of
T. thermophilus tRNAPhe in activation by its cognate aminoa-
cyl-tRNA synthtease.20 Bisaminoacylated tRNAs have also been
shown to bind to elongation factor Tu,6a and it would be of great
interest to assess the roles of both amino acids, as well as the 30-
terminal adenosine moiety, in modulating interaction with this
protein factor.21 These findings contribute to our understanding
of the chemical behavior of bisaminoacylated tRNA, provide
novel insights into the protein synthesis mechanism involving
such bisaminoacylated tRNAs, and suggest a more general role
for the 30-terminal adenosine moiety of tRNA in protein synthesis.
Figure 6. Kinetic analysis of the deacylation of mono-20(30)-O-alanyl-
UMP (5a) and bis-20,30-O-alanyl-UMP (5b) in the presence of 2 equiv
of n-butylamine (red and yellow, respectively) and imidazole (dark blue
and light blue, respectively). The percent formation of UMP after
treatment is shown over a period of 120 min as analyzed by reversed-
phase HPLC (250 ꢁ 4.6 mm, 5 μm column). The column was washed
with 0% f 63% CH3CN in 50 mM NH4OAc, pH 4.5, over a period of 35
min at a flow rate of 1 mL/min (monitoring at 260 nm). Aliquots were
analyzed after 15, 30, 60, and 120 min.
while the use of activated UMP derivatives led to important insights
concerning the role of the adenine nucleobase.
Mono-20(30)-O-alanyl-AMP (4a) proved to be very unstable
in the presence of the nucleophiles n-butylamine and imidazole;
its behavior was thus comparable to that of 20(30)-O-alanyl-
pdCpA (6a) (Figure 4A) and the corresponding activated tRNA.
In contrast, bis-20,30-O-alanyl-AMP (4b) was quite stable in the
presence of n-butylamine and imidazole, although a slightly faster
deacylation rate was observed compared to that of bis-20,30-O-
alanyl-pdCpA (6b) (Figure 4B). These results indicated that the
cytidine residue adjacent to the bisaminoacylated adenosine
moiety was only marginally involved in the chemical stability
of the aminoacyl moiety in the bisaminoacyl-pdCpAs and con-
firmed that mononucleotides can be used as a simplified model
system to study bisaminoacylated pdCpA and tRNA chemistry.
Regarding the UMP derivatives, it has been shown that both
20(30)-O-alanyl-UMP (5a) and bis-20,30-O-alanyl-UMP (5b) were
very sensitive to nucleophilic attack, since after 2 h these
compounds were converted to free UMP (Figure 6). Compar-
ison with the corresponding AMP analogs 4a and 4b (Figure 5)
underscores the ready deacylation of all of the monoaminoacy-
lated nucleotide derivatives by nucleophiles. In the case of UMP,
the bisaminoacylated derivative also proved to be sensitive to
nucleophilic attack, in complete contrast to the stability of
bisacylated AMP (4b). Thus, the adenine nucleobase of 4b must
’ MATERIALS AND METHODS
General Methods. All chemicals reagents were purchased from
Aldrich Chemical Co. (Milwaukee, WI) or Sigma Chemical Co.
(St. Louis, MO) and used without further purification. Anhydrous grade
DMF and acetonitrile were purchased from VWR scientific. Moisture
sensitive reactions were conducted under argon in oven-dried glassware.
Analytical thin-layer chromatography (TLC) was performed on 60 F254
(E. Merck) plates that were visualized by irradiation (254 nm) or by
staining with Hanessian’s stain (cerium molybdate). Flash chromatog-
raphy was performed using Silicycle 40ꢀ60 mesh silica gel. High-
resolution mass spectra were recorded at the Mitchigan State Universityꢀ
NIH Mass Spectrometry Facility. 1H and 13C NMR spectra were
recorded on a 300 MHz Varian instrument. Chemical shifts are reported
in parts per million (ppm, δ) referenced to the residual 1H resonance of
the solvent (CDCl3, δ 7.26; CD3OD, δ 3.31). 13C spectra were
referenced to the residual 13C resonance of the solvent (CDCl3,
δ 77.3; DMSO-d6, δ 39.5). Splitting patterns are designated as follows: s,
singlet; br, broad; d, doublet; dd, doublet of doublets; t, triplet; q,
quartet; m, multiplet. HPLC was performed using a Varian 9012 pump
11374
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
coupled with a Varian 2050 UV detector. Analytical HPLC was per-
formed using an Alltech Alltima RPC18 column (250 ꢁ 4.6 mm, 5 μm),
while semipreparative HPLC was performed using an Alltech Alltima
RPC18 column (250 ꢁ 10 mm, 5 μm).
8.22 (s, 1H), and 8.49 (s, 1H); 13C NMR (D2O) δ 15.6, 16.4, 29.4, 29.5,
34.5, 34.8, 48.3, 48.8, 73.0, 75.0, 82.7, 84.9, 115.6, 115.8, 118.7, 136.8,
137.1, 141.0, 149.1, 156.0, 161.0, 171.3, 173.1, 173.3, 175.8, and 176.3;
mass spectrum (ESI), m/z 654.2284 (M + H)+ (C26H37N7O11P
requires 654.2283).
Molecular Modeling. All modeling results were obtained with
Spartan 04 and 08 software; complementary results were obtained for
simpler systems in Gaussian 03. Spartan’s automated conformational
energy search feature was central to the systematic generation of candi-
date structures. The conformation generator was applied to (a) the
neutral model of bis-acyl AMP we call model A, (b) model A with amino-
protonation, (c) model A with a single water variously sited in the initial
structure, and (d) model A with a hydronium ion. To begin this process,
torsional angles in the aminoacyl chains, including the angle defined by
their attachment to the furan ring, were stepped systematically in 30°
increments through full 360° rotations. More than 20 000 structures
were produced. Each unique conformation was then optimized by
molecular mechanics (MM2). This produced many replicants, which
were excluded from the candidate list. Of unique structures, those within
20 kcal/mol of the minimum energy structures were retained for the next
screening step. Roughly 1000 retained structures were reoptimized with
a semiempirical method, PM3. The 100 lowest-energy forms were
retained for further study. Those predicted by the simplest ab initio
method HF/STO-3G to lie within 10 kcal/mol of the lowest energy
form were reoptimized with the well-tested and reliable density func-
tional method B3LYP/6-31G(d). About 20 isomers of each type (aꢀd)
survived this screening. These low-energy forms all displayed multiple
hydrogen bonding established by water or hydronium ion and bridging
aminoacyl chain to chain and often chain to base. Details of the low-
energy structures are available in the Supporting Information.
Mono-20(30)-N-(4-pentenoyl)-S-alanyl-UMP (3a) and Bis-20,
30-[N-(4-pentenoyl)-S-alanyl]-UMP (3b). To a conical vial con-
taining 13 mg (62 μmol) of N-4-pentenoyl-S-alanine cyanomethyl ester
(1)11 was added a solution of 5.0 mg (6.2 μmol) of the tetra-n-
butylammonium salt of UMP in 100 μL of DMF, followed by 20 μL
of Et3N. The reaction mixture was stirred at room temperature. After 24
h a 5-μL aliquot of the reaction mixture was diluted with 100 μL of 1:2
CH3CNꢀ50 mM NH4OAc, pH 4.5, and was analyzed by HPLC using a
semipreparative C18 reversed-phase column (250 ꢁ 10 mm). The
column was washed with 0% f 63% acetonitrile in 50 mM NH4OAc,
pH 4.5, over a period of 35 min at a flow rate of 3.5 mL/min (monitoring
at 260 nm). The remaining reaction mixture was diluted to a total
volume of 400 μL with 1:2 CH3CNꢀ50 mM NH4OAc, pH 4.5, and
purified using the same C18 reversed-phase column. After lyophilization
of the appropriate fractions, both compounds were obtained as colorless
solids: mono-20(30)-N-(4-pentenoyl)-S-alanyl-UMP (3a) (retention
times 18.3 and 18.7 min for the two positional (20,30) isomers): yield
0.7 mg (25%); 1H NMR (D2O) δ 1.34 and 1.38 (2 d, J = 7.0 and 7.0 Hz,
together integrating for 3H), 2.20ꢀ2.35 (m, 4H), 3.90ꢀ4.00 (m, 2H),
4.25ꢀ4.35 (m, 1H), 4.38ꢀ4.45 (m, 1H), 4.48 and 4.54 (t, J = 7.0 and 7.5
Hz, together integrating for 1H), 4.95ꢀ5.10 (m, 2H), 5.25ꢀ5.30
(m, 1H), 5.70ꢀ5.80 (m, 1H), 5.85 and 5.90 (d, J = 8.0 and 8.5 Hz,
together integrating for 1H), 6.00 (d, J = 7.0 Hz, 1H) and 7.87 and 7.91
(d, J = 8.0 and 8.0 Hz, together integrating for 1H); 13C NMR (D2O)
δ 16.4, 29.5, 34.8, 49.0, 64.5, 68.6, 72.4, 82.4, 87.6, 103.4, 115.9, 137.1,
141.8, 152.2, 166.4, 173.6 and 176.5; mass spectrum (ESI), m/z
478.1230 (M + H)+ (C17H25N3O11P requires 478.1223). Bis-[N-(4-
pentenoyl)-S-alanyl]-UMP (3b) (retention time 23.6 min): yield 2.9 mg
(75%); 1H NMR (D2O) δ 1.29 (d, 3H, J = 7.5 Hz), 1.37 (d, 3H, J = 7.5
Hz), 2.20ꢀ2.35 (m, 8H), 4.00ꢀ4.10 (m, 2H), 4.25ꢀ4.30 (m, 1H),
4.40ꢀ4.45 (m, 2H), 4.95ꢀ5.05 (m, 4H), 5.42 (t, 1H, J = 6.0 Hz),
5.50ꢀ5.55 (m, 1H), 5.65ꢀ5.80 (m, 2H), 5.88 (d, 1H, J = 8.5 Hz),
6.11 (d, 1H, J = 6.0 Hz), and 7.86 (d, 1H, J = 8.0 Hz); 13C NMR (D2O)
δ 16.2, 16.3, 29.4, 29.5, 34.7, 34.8, 48.7, 48.8, 64.4, 72.2, 74.0, 82.0, 86.3,
103.4, 110.0, 115.8, 115.9, 137.1, 141.6, 151.7, 166.3, 173.1, 173.3,
176.2, and 176.3; mass spectrum (ESI), m/z 631.2009 (M + H)+
(C25H36N4O13P requires 631.2011).
Synthesis of Aminoacylated pdCpA Derivatives. Mono-
20(30)-N-(4-pentenoyl)-S-alanyl-AMP (2a) and Bis-20,30-[N-(4-
pentenoyl)-S-alanyl]-AMP (2b). To a conical vial containing 12.6
mg (60.2 μmol) of N-(4-pentenoyl)-S-alanine cyanomethyl ester (1)11
was added a solution of 5.00 mg (6.02 μmol) of the tetra-n-butylammo-
nium salt of AMP in 100 μL of DMF, followed by 20 μL of Et3N.
The reaction mixture was stirred at room temperature. After 24 h a
5-μL aliquot of the reaction mixture was diluted with 100 μL of 1:2
CH3CNꢀ50 mM NH4OAc, pH 4.5, and was analyzed by HPLC using a
semipreparative C18 reversed-phase column (250 ꢁ 10 mm). The
column was washed with 0% f 63% acetonitrile in 50 mM NH4OAc,
pH 4.5, over a period of 35 min at a flow rate of 3.5 mL/min (monitoring
at 260 nm). The remaining reaction mixture was diluted to a total
volume of 400 μL with 1:2 CH3CNꢀ50 mM NH4OAc, pH 4.5, and
purified using the same C18 reversed-phase column. After lyophilization
of the appropriate fractions both compounds were obtained as colorless
solids: mono-20(30)-N-(4-pentenoyl)-S-alanyl-AMP (2a) (retention times
16.6 and 16.9 min for the two positional (20,30) isomers): yield 1.2 mg
Mono-20(30)-O-alanyl-AMP (4a). To a conical vial containing 1.2
mg (2.4 μmol) of protected N-(4-pentenoyl)-(S)-alanyl-AMP (2a) was
added 150 μL of a solution of I2 in 15:85 THFꢀH2O.11,22 The reaction
mixture was stirred at room temperature for 50 min. A 5-μL aliquot of
the reaction mixture was diluted with 100 μL of 1:2 CH3CNꢀ50 mM
NH4OAc, pH 4.5, and was analyzed by HPLC using a semipreparative
1
(40%); H NMR (D2O) δ 1.30 and 1.43 (2 d, J = 7.5 and 7.0 Hz,
C
18 reversed-phase column (250 ꢁ 10 mm). The column was washed
together integrating for 3H), 2.45ꢀ2.65 (m, 4H), 3.90ꢀ4.10 (m, 2H),
4.25ꢀ4.35 (m, 1H), 4.36 and 4.49 (t, J = 7.0 and 7.5 Hz, together
integrating for 1H), 4.38ꢀ4.42 (m, 1H), 4.95ꢀ5.15 (m, 2H), 5.55 and
5.44 (t, J = 5.5 and 5.5 Hz, together integrating for 1H), 5.59ꢀ5.65 and
5.78ꢀ5.83 (m, together integrating for 1H), 6.07 and 6.21 (d, J = 7.5 and
7 Hz, together integrating for 1H), 8.11 and 8.13 (s, together integrating
for 1H), and 8.42 and 8.45 (s, together integrating for 1H); 13C NMR
(D2O) δ 16.4, 29.4, 34.8, 49.0, 64.5, 69.3, 73.3, 82.8, 86.2, 115.9, 118.7,
137.1, 140.0, 152.9, 155.8, 173.7, 176.5, and 181.6; mass spectrum (ESI),
m/z 501.1492 (M + H)+ (C18H26N6O9P requires 501.1493). Bis-20,
30-[N-(4-pentenoyl)-S-alanyl]-AMP (2b) (retention time 23.6 min):
yield 2.3 mg (60%); 1H NMR (D2O) δ 1.22 (d, 3H, J = 7.0 Hz), 1.43
(d, 3H, J = 7.5 Hz), 2.00ꢀ2.15 (m, 4H), 2.25ꢀ2.40 (m, 4H), 4.10ꢀ4.15
(m, 2H), 4.20ꢀ4.25 (m, 1H), 4.45ꢀ4.50 (m, 1H), 4.55ꢀ4.60 (m, 1H),
4.85ꢀ4.90 (m, 2H), 4.98ꢀ5.10 (m, 2H), 5.50ꢀ5.60 (m, 1H),
5.65ꢀ5.70 (m, 1H), 5.75ꢀ5.85 (m, 2H), 6.30 (d, 1H, J = 7.0 Hz),
with 0% f 63% acetonitrile in 50 mM NH4OAc, pH 4.5, over a period of
35 min at a flow rate of 3.5 mL/min (monitoring at 260 nm). The
remaining reaction mixture was diluted to a total volume of 200 μL with
1:2 CH3CNꢀ50 mM NH4OAc, pH 4.5, and purified using the same C18
reversed-phase column. After lyophilization of the appropriate fractions,
compound 4a [retention times 14.7 and 15.3 min for the two positional
(20,30) isomers] was obtained as a colorless solid: yield 0.8 mg (80%); 1H
NMR (D2O) δ 1.38 (d, 3H, J = 7.5 Hz), 4.00ꢀ4.10 (m, 2H), 4.25ꢀ4.30
(m, 1H), 4.40ꢀ4.45 (m, 1H), 4.60ꢀ5.70 (m, 2H), 6.04 (t, 1H, J = 6.0
Hz), 8.15 (s, 1H), and 8.40 (s, 1H); 13C NMR (D2O) δ 16.4, 22.4, 64.6,
70.7, 74.7, 84.4, 87.3, 118.8, 140.3, 149.1, 152.1, 155.1, and 179.9; mass
spectrum (ESI), m/z 419.1082 (M + H)+ (C13H20N6O8P requires
419.1080).
Bis-20,30-O-alanyl-AMP (4b). To a conical vial containing 2.3 mg
(4.7 μmol) of protected bis-20,30-[N-(4-pentenoyl)-(S)-alanyl]-AMP
11375
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
(2b) was added 150 μL of a solution of I2 in 15:85 THFꢀH2O.11,22 The
reaction mixture was stirred at room temperature for 50 min. A 5-μL
aliquot of the reaction mixture was diluted with 100 μL of 1:2
CH3CNꢀ50 mM NH4OAc, pH 4.5, and was analyzed by HPLC using
a semipreparative C18 reversed-phase column (250 ꢁ 10 mm). The
column was washed with 0% f 63% acetonitrile in 50 mM NH4OAc,
pH 4.5, over a period of 35 min at a flow rate of 3.5 mL/min (monitoring
at 260 nm). The remaining reaction mixture was diluted to a total
volume of 200 μL with 1:2 CH3CNꢀ50 mM NH4OAc, pH 4.5, and
purified using the same C18 reversed-phase column. After lyophilization
of the appropriate fractions, compound 4b (retention time 18.6 min)
was obtained as a colorless solid: yield 1.7 mg (75%); 1H NMR (D2O)
δ 1.39 (d, 3H, J = 7.0 Hz), 1.37 (d, 3H, J = 7.0 Hz), 3.65ꢀ3.75 (m, 1H),
4.00ꢀ4.10 (m, 2H), 4.20ꢀ4.25 (m, 1H), 4.40ꢀ4.45 (m, 1H), 4.60ꢀ
4.65 (m, 2H), 6.03 (d, 1H, J = 5.5 Hz), 8.13 (s, 1H), and 8.41 (s, 1H);
13C NMR (D2O) δ 16.4, 23.3, 49.0, 64.4, 70.7, 74.7, 84.5, 87.2, 110.0,
118.8, 140.1, 149.2, 152.9, 155.7, 176.1, and 181.5; mass spectrum (ESI),
m/z 490.1453 (M + H)+ (C16H25N7O9P requires 490.1451).
mixture was stirred for 2 h. After 15, 30, 60, and 120 min, 15 μL of each
solution was analyzed by HPLC using a reversed-phase column (Alltima
RPC18 4.6 ꢁ 250 mm, 5 μm). The column was washed with 0% f 63%
CH3CN in 50 mM NH4OAc, pH 4.5, over a period of 35 min at a flow
rate of 1 mL/min (monitoring at 260 nm).
’ ASSOCIATED CONTENT
S
Supporting Information. PDB files of the computationally
b
determined energetically favorable structures of a (phosphate-
methylated) 50-AMP derivative containing alanyl esters at the 20-
and 30-positions. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
sid.hecht@asu.edu
Mono-20(30)-O-alanyl-UMP (5a). To a conical vial containing
0.8 mg (1.7 μmol) of protected mono-20(30)-N-(4-pentenoyl)-(S)-alanyl-
UMP (3a) was added 150 μL of a solution of I2 in 15:85 THFꢀH2O.11,22
The reaction mixture was stirred at room temperature for 50 min.
A 5-μL aliquot of the reaction mixture was diluted with 100 μL of 1:2
CH3CNꢀ50 mM NH4OAc, pH 4.5, and was analyzed by HPLC using a
semipreparative C18 reversed-phase column (250 ꢁ 10 mm). The
column was washed with 0% f 63% acetonitrile in 50 mM NH4OAc,
pH 4.5, over a period of 35 min at a flow rate of 3.5 mL/min (monitoring
at 260 nm). The remaining reaction mixture was diluted to a total
volume of 200 μL with 1:2 CH3CNꢀ50 mM NH4OAc, pH 4.5, and
purified using the same C18 reversed-phase column. After lyophilization
of the appropriate fractions, compound 5a [retention times 16.5 and
17.5 min for the two positional (20,30) isomers] was obtained as a
colorless solid: yield 0.65 mg (98%); 1H NMR (D2O) δ 1.31 (d, 3H J =
7.0 Hz), 3.60ꢀ3.65 (m, 1H), 3.85ꢀ4.00 (m, 2H), 4.10ꢀ4.25 (m, 3H),
5.79 (d, 1H, J = 8.0 Hz), 5.84 (d, 1H, J = 5.0 Hz), and 7.88 (d, 1H, J = 8.0
Hz); 13C NMR (D2O) δ 16.3, 50.6, 70.0, 74.1, 88.5, 102.7, 141.9, 152.0,
163.2, 166.5, 176.0, and 185.5; mass spectrum (MALDI-TOF), m/z
396.5 (M + H)+ (theoretical m/z 396.1).
Present Addresses
†CNRS UMR6001 LCMBA, Universitꢀe de Nice-Sophia Anti-
polis, ParcValrose, 06108 Nice Cedex 2, France.
‡Center for BioEnergetics, Biodesign Institute and Department
of Chemistry and Biochemistry, Arizona State University, Tempe,
Arizona 85287, United States.
’ REFERENCES
(1) (a) Crick, F. H. C. Nature 1970, 227, 561. (b) Green, R.; Noller,
H. F. Annu. Rev. Biochem. 1997, 66, 679. (c) Rodnina, M. V.; Wintermeyer,
W. Curr. Opin. Struct. Biol. 2003, 13, 334.
(2) (a) Hecht, S. M.; Kozarich, J. W.; Schmidt, F. J. Proc. Natl. Acad.
Sci. U. S. A. 1974, 71, 4317. (b) Hecht, S. M. Acc. Chem. Res. 1977,
10, 239. (c) Wagner, T.; Cramer, F.; Sprinzl, M. Biochemistry 1982,
21, 1521. (d) Hansen, J. L.; Schmeing, T. M.; Moore, P. B.; Steitz, T. A.
Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 11670. (e) Sprinzl, M. J. Biosci.
2006, 31, 489. (f) Huang, K. S.; Weinger, J. S.; Butler, E. B.; Strobel, S. A.;
Steitz, T. A. J. Am. Chem. Soc. 2006, 128, 3108.
(3) (a) Schmeing, T. M.; Huang, K. S.; Kitchen, D. E.; Strobel, S. A.
Mol. Cell 2005, 20, 437. (b) Trobro, S.; Aqvist, J. Proc. Natl. Acad. Sci.
U. S. A. 2005, 102, 12395.
(4) Koch, M.; Huang, Y.; Sprinzl, M. Angew. Chem., Int. Ed. 2008,
47, 7242.
(5) (a) Stepanov, V. G.; Moor, N. A.; Ankilova, V. N.; Lavrik, O. I.
FEBS Lett. 1992, 311, 192. (b) Stepanov, V. G.; Moor, N. A.; Ankilova,
V. N.; Vasil’eva, I. A.; Sukhanova, M. V.; Lavrik, O. I. Biochim. Biophys.
Acta 1998, 1386, 1.
(6) (a) Wang, B.; Zhou, J.; Lodder, M.; Anderson, R. D.; Hecht,
S. M. J. Biol. Chem. 2006, 281, 13865. (b) Duca, M.; Maloney, D.;
Lodder, M.; Wang, B.; Hecht, S. M. Bioorg. Med. Chem. 2007, 15, 4629.
(c) Duca, M.; Chen, S.; Hecht, S. M. Methods 2008, 44, 87.
(7) Duca, M.; Chen, S.; Hecht, S. M. Org. Biomol. Chem. 2008,
6, 3292.
(8) Noren, C. J.; Anthony-Cahill, S. J.; Suich, D. J.; Noren, K. A.;
Griffith, M. C.; Schultz, P. G. Nucleic Acids Res. 1990, 18, 83.
(9) (a) Noren, C. J.; Anthony-Cahill, S. J.; Griffith, M. C.; Schultz,
P. G. Science 1989, 244, 182. (b) Bain, J. D.; Diala, E. S.; Glabe, C. G.;
Dix, T. A.; Chamberlin, A. R. J. Am. Chem. Soc. 1989, 111, 8013. (c)
Hecht, S. M. Acc. Chem. Res. 1992, 25, 545. (d) Cornish, V. W.; Mendel,
D.; Shultz, P. G. Angew. Chem., Int. Ed. 1995, 34, 621. (e) Mamaev, S. V.;
Laikhter, A. L.; Arslan, T.; Hecht, S. M. J. Am. Chem. Soc. 1996,
118, 7243.
Bis-20,30-O-alanyl-UMP (5b). To a conical vial containing 3.4 mg
(5.4 μmol) of protected 20,30-[N-(4-pentenoyl)-(S)-alanyl]-UMP (3b)
was added 150 μL of a solution of I2 in 15:85 THFꢀH2O.11,22 The
reaction mixture was stirred at room temperature for 50 min. A 5-μL
aliquot of the reaction mixture was diluted with 100 μL of 1:2
CH3CNꢀ50 mM NH4OAc, pH 4.5, and was analyzed by HPLC using
a semipreparative C18 reversed-phase column (250 ꢁ 10 mm). The
column was washed with 0% f 63% acetonitrile in 50 mM NH4OAc,
pH 4.5, over a period of 35 min at a flow rate of 3.5 mL/min (monitoring
at 260 nm). The remaining reaction mixture was diluted to a total
volume of 200 μL with 1:2 CH3CNꢀ50 mM NH4OAc, pH 4.5, and
purified using the same C18 reversed-phase column. After lyophilization
of the appropriate fractions, compound 5b (retention time 20.2 min)
was obtained as a colorless solid: yield 1.43 mg (57%); 1H NMR (D2O)
δ 1.29 (d, 3H, J = 7.5 Hz), 1.37 (d, 3H, J = 7.0 Hz), 3.55ꢀ3.65 (m, 1H),
3.85ꢀ4.00 (m, 2H), 4.10ꢀ4.15 (m, 1H), 4.15ꢀ4.25 (m, 3H), 5.77 (d,
1H, J = 8.0 Hz), 5.82 (d, 1H, J = 5.0 Hz), and 7.83 (d, 1H, J = 8.0 Hz);
13C NMR (D2O) δ 16.3, 22.8, 50.6, 64.1, 70.0, 74.1, 83.6, 88.6, 102.7,
141.8, 152.0, 166.5, 169.8, 176.0, and 180.7; mass spectrum (MALDI-
TOF), m/z 467.2 (M + H)+ (theoretical 467.1).
General Procedure for the Treatment of Aminoacylated
AMP (4a, 4b) and UMP (5a, 5b) Derivatives and Aminoacy-
lated Dinucleotides (6a, 6b) with Nucleophiles. Compounds
4a, 4b, 5a, 5b, 6a, and 6b (0.2 mg) were each dissolved in 30-μL aliquots
of water. To these stirred solutions was added 30 μL of DMF containing
2 equiv of n-butylamine or imidazole, and the corresponding reaction
(10) Maloney, D. J.; Ghanem, N.; Zhou, J.; Hecht, S. M. Org. Biomol.
Chem. 2007, 5, 3135.
(11) Lodder, M.; Golovine, S.; Laikhter, A. L.; Karginov, V. A.;
Hecht, S. M. J. Org. Chem. 1998, 63, 794.
(12) Weinger, J. S.; Strobel, S. A. Biochemistry 2006, 45, 5939.
11376
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377

Journal of the American Chemical Society
ARTICLE
(13) Griffin, B. E.; Jarman, M.; Reese, C. B.; Sulston, J. E.; Trentham,
D. R. Biochemistry 1966, 5, 3838.
(14) (a) Dorner, S.; Polacek, N.; Schulmeister, U.; Panuschka, C.;
Barta, A. Biochem. Soc. Trans. 2002, 30, 1131. (b) Weinger, J. S.; Strobel,
S. A. Blood Cells Mol. Dis. 2007, 38, 110.
(15) (a) Sprinzl, M.; Cramer, F. Proc. Natl. Acad. Sci. U. S. A. 1975,
72, 3049. (b) Chinault, A. C.; Tan, K. H.; Hassur, S. M.; Hecht, S. M.
Biochemistry 1977, 16, 766.
(16) Weinger, J. S.; Parnell, K. M.; Dorner, S.; Green, R.; Strobel,
S. A. Nat. Struct. Mol. Biol. 2004, 11, 1101.
(17) (a) Hecht, S. M.; Hawrelak, S. D.; Kozarich, J. W.; Schmidt,
F. J.; Bock, R. M. Biochem. Biophys. Res. Commun. 1973, 52, 1341. (b)
Hecht, S. M.; Chinault, A. C. Proc. Natl. Acad. Sci. U. S. A. 1976, 73, 405.
(18) Changalov, M. M.; Ivanova, G. D.; Rangelov, M. A.; Acharya, P.;
Acharya, S.; Minakawa, N.; Foldesi, A.; Stoineva, I. B.; Yomtova, V. M.;
Roussev, C. D.; Matsuda, A.; Chattopadhyaya, J.; Petkov, D. D. Chem-
BioChem 2005, 6, 992.
(19) Limmer, S.; Hofmann, H.-P.; Ott, G.; Sprinzl, M. Proc. Natl.
Acad. Sci. U. S. A. 1993, 90, 6199.
(20) (a) Moor, N. A.; Stepanov, V. G.; Ankilova, V. N.; Favre, A.;
Lavrik, O. I. Biochemistry (Moscow) 1998, 63, 1044. (b) Vasil’eva, I. A.;
Ankilova, V. N.; Lavrik, O. I.; Moor, N. A. Biochemistry (Moscow) 2000,
65, 1157. (c) Vasil’eva, I. A.; Favre, A.; Lavrik, O. I.; Moor, N. A.
Biochemistry (Moscow) 2004, 69, 192.
(21) LaRiviere, F. J.; Wolfson, A. D.; Uhlenbeck, O. C. Science 2001,
294, 165.
(22) Lodder, M.; Golovine, S.; Hecht, S. M. J. Org. Chem. 1997,
62, 778.
11377
dx.doi.org/10.1021/ja203994e |J. Am. Chem. Soc. 2011, 133, 11368–11377