On the electrophilicity of cyclic acylphosphoramidates (CAPAs) postulated as intermediates

Article abstract of DOI:10.1002/ejoc.200900227

Cyclic acylphosphoramidates (2-hydroxy-2-oxy-l,3,2-oxazaphospholidine-5- ones, CAPAs) are phosphate-activated amino acids possessing both a carboxyl-phosphoryl anhydride and a phosphoramidate bond in a ring. They structurally resemble NCAs (amino acid N-c

Full text of DOI:10.1002/ejoc.200900227

FULL PAPER
DOI: 10.1002/ejoc.200900227
On the Electrophilicity of Cyclic Acylphosphoramidates (CAPAs) Postulated as
Intermediates
Feng Ni,[a] Xiang Gao,[a] Zeng-Xiang Zhao,[a] Chao Huang,[a] and Yu-Fen Zhao*[a,b]
Keywords: Amino acids / Bielectrophilicity / NMR spectroscopy / Phosphorus
Cyclic acylphosphoramidates (2-hydroxy-2-oxy-1,3,2-oxaza-
phospholidine-5-ones, CAPAs) are phosphate-activated
Ala-CAPA) was identified by isotopic analysis (18O, 15N) and
further proved by trapping α-CAPA with nucleophiles such
amino acids possessing both a carboxyl–phosphoryl anhy- as water, amino acids, phosphate and methanol in alkaline
dride and a phosphoramidate bond in a ring. They structur-
ally resemble NCAs (amino acid N-carboxyanhydrides,
Leuchs’ anhydrides), which have been extensively investi-
gated. By contrast, the chemistry of CAPAs has remained al-
most unexplored since it was proposed in a prebiotic reaction
of inorganic polyphosphates (PolyPs) with amino acids. In the
present work, the bielectrophilicity of α-CAPAs (Gly-CAPA,
media, which yielded N-phosphoamino acids and peptide
phosphoanhydride and phosphate ester derivatives. By com-
parison with the reactivity of NCAs, the bielectrophilicity of
CAPAs indicates that CAPAs can provide rich chemistry.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2009)
phosphate,[6] aminoacyl phosphates[7] and aminoacyl ade-
nylates[8] have been extensively explored as model com-
pounds. Meanwhile, the biologically less popular phos-
phoramidate bond has attracted increased attention[9] since
the emergence of histidine kinase research.[10] CAPAs pos-
sess both a carboxyl–phosphoryl anhydride moiety and a
phosphoramidate bond in a ring. Although intact CAPAs
have not been detected in living organisms yet, the study
of such bifunctional model compounds may enhance our
understanding of biological phosphate activation and trans-
fer. Moreover, it has been suggested that inorganic polyp-
hosphate was an ancient energy carrier preceding ATP,[11]
and the earliest organisms may have used PolyPs through
the use of PolyP-utilizing enzymes;[12] therefore, PolyPs and
amino-acid-induced CAPA might serve as a primitive
model and provide clues to the later-developed mechanism
of PolyP-utilizing enzymes.
Introduction
α-Amino acid N-carboxyanhydrides (Leuchs’ anhydrides,
NCAs, 4Ј, Scheme 1) are cyclic and carbonyl-activated
amino acids that were recently shown to be accessible inter-
mediates in plausible prebiotic reactions mediated by car-
bonyl sulfide (COS)[1] or cyanate.[2] NCAs have been inves-
tigated extensively with a 100-year[3] history of success in
aspects of stepwise peptide syntheses, ring-opening polyme-
rizations and molecular evolution. Interestingly, cyclic acyl-
phosphoramidates (CAPAs, 4, Scheme 1) are cyclic and
phosphoryl-activated amino acids that were first proposed
as intermediates in a prebiotic reaction of inorganic poly-
phosphates (PolyPs) and amino acids with peptide yields up
to 36%.[4] Because CAPAs structurally resemble NCAs, it
will be of great significance to compare the intrinsic simi-
larities and differences between the two. Unfortunately, the
chemistry of CAPAs has remained relatively unexplored
since its discovery.
Hence, we initiate the present investigation of CAPAs to
get insight into the merits of amino acid and phosphate
activations.
In addition, because of the prevalence of carboxyl–phos-
phoryl anhydride containing species in phosphate-transfer
reactions catalyzed by related enzymes (e.g., EC 2.7.2 sub-
family), compounds such as formyl phosphate,[5] acetyl
Results and Discussion
[a] Department of Chemistry and Key Laboratory of Chemical Bi-
ology of Fujian Province College of Chemistry and Chemical
Engineering, Xiamen University,
CAPAs could be produced through the reaction of the
prebiotically available[13,14] linear polyphosphates or cyclo-
triphosphate (P3m, 1) with amino acids 2. Notably, P3m is
the most efficient reagent.[4b] The mechanism involves auto-
matic intramolecular ring closure of N-triphospho-α-amino
acids 3 followed by release of pyrophosphate P2 to yield
CAPAs (Scheme 1). Interestingly, the corresponding N-tri-
phospho-β-amino acids and N-triphospho-γ-amino acids
Xiamen 361005, P. R. China
Fax: +86-86-592-218-6292
E-mail: yfzhao@xmu.edu.cn
[b] Key Laboratory of Bioorganic Phosphorus Chemistry and
Chemical Biology (Ministry of Education), Department of
Chemistry Tsinghua University,
Beijing 100084, P. R. China
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900227.
3026
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3026–3035

On the Electrophilicity of CAPAs Postulated as Intermediates
Scheme 1. Bielectrophilicity of CAPAs towards H218O, amino acids, phosphate and methanol.
were fairly stable at room temperature.[15] In the present ex- were determined by comparison with authentic samples
periments, the CAPAs were generated in situ by treating 1 (Figures S1 and S2, Supporting Information).
with α-amino acids 2.
A 5:1 ratio of 5I/5II was observed by 31P NMR spectro-
scopic analysis of the freshly prepared reaction mixture
(Figure 1). The observed upfield isotope shifts (≈0.03 ppm)
from 5II to 5I as a result of 18O bonded to phosphorus are
consistent with literature data.[18] Furthermore, N-phos-
phoglycine 5a and 5Ia/5IIa derived from regular H2O and
H218O, respectively, were analyzed by ESI MS–MS. Mass
spectra showed a 2 Da unit increment for 18O-incorporated
products 5Ia or 5IIa (m/z = 178; Figure 2, a and b). The
tandem MS2 fragmentation of 5Ia and 5IIa gave the meta-
Bielectrophilicity of CAPAs towards H218O
N-Phosphoamino acids 5 were detected and proposed as
hydrolytic products of CAPAs in previous reports.[4b,4c,16]
Such a hydrolysis reaction was recently applied to the syn-
thesis of 13 N-phosphoamino acids;[17] however, the de-
tailed hydrolytic mechanism was not confirmed. As with
acetyl phosphate[6b,6e] and formyl phosphate,[5c] CAPAs
possess a highly reactive carboxyl–phosphoryl mixed anhy-
dride. Depending on the pH, hydrolysis of such anhydrides
may undergo bond cleavage in two ways: C–O cleavage or
P–O cleavage. In this paper, 18O-isotope analysis was used
to determine the cleavage mode in the hydrolysis reaction
of CAPAs generated in situ through the reaction of amino
acids (Gly or Ala 0.1 ) 2 with 1 (0.1 ) in H218O (96%
18O) at pH 11.0 and 40 °C. If H218O attacks the phosphorus
centre followed by P–O cleavage, the 18O label should be on
the phosphoryl group, leading to product 5I. If H218O at-
tacks the carboxyl carbon followed by C–O cleavage, the
18O label should be on the carboxyl group, leading to prod-
–
phosphate anion PO218O– (m/z = 81) and PO3 (m/z = 79),
respectively, in a 5:1 ratio (Figure 2c), consistent with the
31P NMR spectroscopic result. Hence, NMR spectroscopy
and MS show that CAPAs can be attacked by water either
at the phosphorus centre or the carbonyl carbon atom un-
der basic conditions. However, attack at phosphorus is
dominant for CAPAs, in contrast to the predominant attack
at carbonyl centres of acetyl phosphate and formyl phos-
phate under basic conditions.[5c,6b,6e]
Bielectrophilicity of CAPAs towards the Amino Group
To study the ammonolysis reaction of CAPAs, P3m
uct 5II. The structures of N-phosphoamino acids 5I/5II (0.1 ) was treated with an excess amount of glycine (1 )
Eur. J. Org. Chem. 2009, 3026–3035
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3027

F. Ni, X. Gao, Z. X. Zhao, C. Huang, Y. F. Zhao
FULL PAPER
at pH 11.0 in water at room temperature and tracked by 31
P
NMR spectroscopy (Figure 3, a and b). Reactant 1 [δ =
20.70 (s) ppm] was consumed very rapidly. The first inter-
mediate 3a [δ = –0.51 (d), –4.53 (d), –19.47 ppm (t)] ap-
peared immediately, increased to a maximum after 30 min
and then slowly disappeared over 200 min. A second inter-
mediate 7a [δ = 13.40 (s) ppm] was formed with a maximum
concentration of 0.033  after 40 min and then gradually
disappeared after 150 min. Products 5a [δ = 9.11 (s) ppm]
and 6a [δ = 8.75 (s) ppm] were formed at a later stage and
then remained almost constant after 150 min. Byproduct
pyrophosphate P2 [δ = 4.78 (s) ppm] appeared rapidly and
increased to a maximum after 200 min.
Figure 1. (a) 31P NMR of 5Ia/5IIa showing a 5:1 ratio. (b) 31P
NMR of 5Ib/5IIb showing a 5:1 ratio.
Figure 3. (a) Time-dependent 31P NMR stacking spectra of the re-
action of P3m (0.1 ) with an excess amount of glycine (1 ) at
pH 11.0, room temp. (b) Time-dependent concentration curves of
reactant, intermediate and products in (a). The concentrations were
calculated on the basis of the integration of the 31P NMR peaks.
The observed signal of intermediate N-triphosphoglycine
3a was consistent with previous data.[16] N-Phosphoglycyl-
glycine 6a (yield ≈ 70%) derived from glycine nucleophilic
attack at the carbonyl of CAPA was confirmed by compari-
son with a synthetic sample (Figure S3, Supporting Infor-
mation). N-phosphoglycine 5a was also identified by com-
parison with an authentic sample. The remaining significant
signal was a decoupled peak for 7a at δ = 13.40 ppm (sing-
Figure 2. (a) Negative ion ESI MS–MS spectra of authentic N-
phosphoglycine 5a. (b) Negative ion ESI MS–MS spectra of 18O-
incorporated sample of 5Ia/5IIa. (c) Magnified spectrum of (b)
showing a 5:1 ratio of fragmentation ion (18O-incorporated meta-
phosphate ion, m/z = 81) from 5Ia to the fragmentation ion (meta-
phosphate ion, m/z = 79) from 5IIa.
let), which was transformed into a quintuplet (JH,P
=
1
8.0 Hz) in the H-coupled 31P NMR spectrum (Figure 4a),
suggesting that there were four equivalent protons from two
glycine units to split the phosphorus signal.
3028
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3026–3035

On the Electrophilicity of CAPAs Postulated as Intermediates
reacted in Na2HPO4 solution (1 ) at pH 11.0, 40 °C for
12 h, the 1H-decoupled 31P NMR spectrum showed two
new peaks with equal intensity at –1.36 and –5.96 ppm,
each one a doublet with JP,P = 20.7 Hz (Figure 5b), suggest-
ing that there are two adjacent phosphoryl groups in the
new species.
Figure 4. (a) 31P NMR spectrum of 7a in the reaction of P3m
(0.1 ) with glycine (1 ) at pH 11.0, room temp. (b) 31P NMR
spectrum of 7a in the reaction of P3m (0.1 ) with 15N-labelled
glycine (1 ) at pH 11.0, room temp.
1
Indeed, the H-decoupled 31P NMR spectrum of a reac-
tion mixture derived from 15N-labelled glycine and P3m
showed a triplet (JN,P = 22.4 Hz, close to a reported JN,P
=
25.8 Hz[19]) at δ = 13.4 ppm (Figure 4b), indicating that
there were two P–N bonds in 7a. The NMR results support
the formation of N,NЈ-phosphorylated bisglycine 7a, which
is derived from glycine attack at the phosphorus atom of
CAPA. The 31P NMR signal of bisalanine derivative 7b and
N-phosphoalanylalanine 6b were also detected in the reac-
tion of P3m with alanine (Figure S4, Supporting Infor-
mation). During the reaction, products 7 gradually disap-
peared with increasing amounts of products 5 and 6 at the
same time, which indicated that products 7 are able to con-
vert back to the intermediate CAPAs 4 through automatic
ring closure and release of the amino acids. The hypothesis
7Ǟ4 is supported by a very close mechanism found in the
biological degradation of phosphoric and phosphonic di-
amide prodrugs.[20] In summary, CAPAs can be attacked by
amino acids either at the phosphorus atom or at the car-
bonyl carbon atom to produce kinetically favourable prod-
ucts 7 and thermodynamically stable products 6, respec-
tively.
1
Figure 5. (a) H-coupled 31P NMR spectrum of 9a at pH 11.0. (b)
1H-decoupled 31P NMR spectrum of 9a at pH 11.0.
In addition, the peaks at –1.36 ppm turned out to be a
doublet of triplets (JP,P = 20.7 Hz, JH,P = 6.3 Hz) in the 1H-
coupled 31P NMR spectrum (Figure 5a), indicating that this
phosphoryl group was adjacent to glycine and also con-
nected to the other phosphoryl group. Consequently, the
structure of N-pyrophosphoglycine 9a is proposed. To ver-
ify this assignment, 9a was isolated from the reaction mix-
ture (≈60% conversion yield; ≈30% isolated yield) and its
structure was confirmed by H, 13C and 31P NMR spec-
troscopy and HRMS–MS (Figure S5, Supporting Infor-
mation).
1
The 31P NMR signal of the N-pyrophospho derivatives
of other amino acids (such as Ala, Phe, Ser) were also de-
tected in the corresponding reaction mixtures (Figure S6,
Supporting Information) but yields were lower because
these N-pyrophosphoamino acids were less stable and easily
decomposed into N-phosphoamino acids 5 and inorganic
phosphate. Such a situation prevented us from isolating
these N-pyrophospho derivatives. Hence, we turned to a
further reactivity study of stable N-pyrophosphoglycine 9a
in hydrolysis experiments at pH 6.8. Time-dependent con-
centration curves of the hydrolysis products (Figure 6a) re-
vealed that 9a decomposed into N-phosphoglycine 5a and
orthophosphate. Because only trace amounts of P2 were
produced, the mechanism prefers the breaking of the pyro-
It should be pointed out that previous literature[16]
claimed the direct observation of CAPA 4a by 31P NMR
spectroscopy at δ = 13.4 ppm in pH 12 aqueous solution;
however, our work shows that CAPAs are too labile to be
observed in the presence of nucleophiles such as water and
amino acids. In addition, the structural elucidation above
and the reaction profile in Figure 3b support the conclusion
that the signal near 13.4 ppm in the 31P NMR spectrum
comes from the local intermediate N,NЈ-phosphorylated
bisglycine 7a, but not the global intermediate CAPA 4a.
Electrophilicity of CAPAs towards Phosphate and Methanol
Since phosphate activation is important in modern bio- phosphate bond rather than the P–N bond. Together with
chemistry, the CAPA generated in situ was treated with 31P NMR–pH titration profiles of 9a (Figure 6b), the hy-
phosphate to see whether phosphate would be activated or drolytic mechanism could be rationalized that 9a was firstly
not. When a mixture of P3m (0.1 ) and Gly (0.1 ) was protonated on the β-phosphate, which facilitates the car-
Eur. J. Org. Chem. 2009, 3026–3035
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3029

F. Ni, X. Gao, Z. X. Zhao, C. Huang, Y. F. Zhao
FULL PAPER
boxyl group to expel the orthophosphate anion with the
An alkoxide nucleophile was also investigated. When gly-
formation of CAPA 4a, a path similar to 3aǞ4a+P2. cine (0.1 ) was treated with P3m (0.1 ) in methanol/water
Thereby, 5a is derived from hydrolysis of regenerated CAPA (1:4) at pH 11.0, 40 °C, 48 h, the corresponding N-mono-
4a.
methoxyphosphoglycine 11a and its hydrolysis product
methyl phosphate (MeP) were detected and confirmed by
comparing with authentic samples (Figures S7 and S8, Sup-
porting Information). However, product 10 was not de-
tected in the reaction mixture because of the instability of
its ester bond in alkaline solution. Concerning the signifi-
cant role of nucleosides in prebiotic chemistry,[22] a system-
atic survey of the activation of nucleosides by CAPA is
now[23] in progress.
CAPAs vs. NCAs
The formation of and the bielectrophilicity of NCAs to-
ward common nucleophiles such as water, amino acids,
phosphate and methanol, which exist in the literature, are
compiled in Scheme 2 for comparison with CAPAs. In ad-
dition to artificial agents such as phosgene,[24] the forma-
tion of NCAs can be achieved by using carbonyl sulfide
(COS)[1] and hydrogen cyanate (HNCO).[2] Like pyrophos-
phate for the formation of CAPAs, H2S and NH3 act as
good leaving groups to drive the reaction toward NCAs.
Bielectrophilicity
Results of CAPAs suggested that both the phosphorus
and carbonyl carbon atoms are possible electrophilic sites
and attack at phosphorus is more favourable than attack at
the carbonyl carbon. NCAs also possess two electrophilic
centres, namely, the carbamoyl group (C-2) and the car-
bonyl group (C-5). From Scheme 2, C-5 is a preferable reac-
tive site than C-2 towards nucleophiles such as water, amino
acids, phosphate and methanol to yield N-carboxyl prod-
ucts such as N-carboxyl amino acids[25] 5Ј, N-carboxyl di-
peptides[26] 6Ј, N-carboxyl aminoacyl phosphate[1a,27] 8Ј
and N-carboxyl amino acids ester[28] 10Ј. Meanwhile, C-2-
Figure 6. (a) Time-dependent concentration curves of the hydroly-
sis products of 9a at pH 6.8 (triethanolamine buffer) 25 °C, I = 0.
(b) 31P NMR–pH titration profiles of N-pyrophosphoglycine 9a,
25 °C, I = 0.
The lability of 9 may be explained by the concept of near directed products such as N-alkoxycarbonyl amino acids[29]
attack conformers (NACs) from Bruice.[21] For other amino and N,NЈ-carbonyl bisamino acids[1a,2,30] 7Ј were also re-
acids, it could be argued that their side chains provide steric ported. However, the C-2 related process was proved to be a
compression, which positions the carboxyl and phosphoryl two-step mechanism involving the formation of isocyanate
groups closer, consequently enhancing the intramolecular intermediate[28c,30] 12Ј followed by nucleophilic attack of
carboxyl-catalyzed decomposition of 9. It is noteworthy the amino group or methanol at C-2 to obtain the final
that a similar mechanism for intermediate 3 should produce products. The intrinsic electrophilicity difference between
the products even faster, as pyrophosphate is a better leav- CAPAs and NCAs could be interpreted by the electronic
ing group than inorganic phosphate.
nature of the phosphoramidate group and the carbamoyl
Indeed, N-triphosphoglycine 3a (t1/2 Ͻ 40 min, pH 11.0, group. First, Pauling electronegativity of phosphorus is
room temp., Figure 3) was less stable than N-pyrophospho- somewhat lower than that of carbon (2.19 vs. 2.55) when
glycine 9a (t1/2 = several weeks, pH 11.0, room temp.), bonded to atoms with larger electronegativity (O, N), and
which implies that 9 might be a good phosphate donor and phosphorus retains the larger positive charge, making it
energy carrier. In principle, the reaction of CAPAs with more electrophilic. Second, participation of the d orbital
phosphate should also produce products 8 derived from at- may allow the phosphoramidate group to form a possible
tack of phosphate at the carbonyl centre. However, prod- pentacoordinate intermediate[31] during the nucleophilic re-
ucts 8 were not detected in the reaction as a result of the action. Besides the intrinsic structural factors, solvent ef-
instability of their carboxyl–phosphoryl mixed anhydride fects should also be taken into account. In the present
bond in basic aqueous solution.
work, the reactions of CAPAs were carried out in water,
3030
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3026–3035

On the Electrophilicity of CAPAs Postulated as Intermediates
Scheme 2. Formation of NCAs and their bielectrophilicity toward H2O, amino acids, phosphate and alcohol (data collected from the
literature).
which favours systems with large dipole moments like cleosides, leading to peptide and nucleic acid oligomeriza-
CAPAs, and thus a change in electrophilicity preference tion, respectively (Scheme 3Ͻyschr3 pos="x22"Ͼ).
might be expected if the reactions are performed in organic
solvent.
The concept that “phosphorus is a carbon copy” is suc-
cessful in the interpretation of similarity between low-coor-
We previously reported another species,[32] pentacoordi- dinate phosphorus compounds and unsaturated carbon
nated cyclic acylphosphoramidates (P5-CAPAs), which also compounds.[33] Apparently, this concept could not be sim-
resemble NCAs. They can be formed by automatic ring clo- ply extended to high-coordination phosphorus chemistry
sure of N-dialkyloxyphosphoryl α-amino acids. P5-CAPAs and related carbon chemistry with the examples of CAPAs
are bielectrophilically reactive towards amino acids and nu- and NCAs here.
Scheme 3. Formation and bielectrophilicity of pentacoordinated cyclic acylphosphoramidates (P5-CAPA).
Eur. J. Org. Chem. 2009, 3026–3035
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3031

F. Ni, X. Gao, Z. X. Zhao, C. Huang, Y. F. Zhao
FULL PAPER
Peptide Formation, Phosphate Activation and
Phosphorylation
that they are not only phosphate-activated amino acids but
also amino-acid-activated phosphates. Thus, they have the
potential to be used as self-protected and self-activated
blocks in peptide coupling reactions. Furthermore, the for-
mation process and the reactivity of CAPAs in phosphoryl
transfer may have potential biological significance in con-
nection with PolyP-utilizing enzymes, which use PolyPs as
phosphoryl and energy sources. Finally, the corresponding
phosphorylated products were not observed for the analo-
gous β- and γ-amino acid CAPAs at moderate temperature
(data not shown), which thus highlights the unique role of
α-amino acids in potential prebiotic evolutionary events.[35]
However, a limitation to the extensive application of
CAPAs is the lack of an efficient preparation method of
CAPAs. Hence, great effort should be focused on the prepa-
ration of pure CAPAs in the future, even though the prepa-
ration of Gly-CAPA has been described.[36] It is noteworthy
that the reversible conversion between Gly-CAPA 4a and
N-pyrophosphoglycine 9a, demonstrated in this work, may
provide an alternative path to the in situ generation of
CAPAs, which makes investigations of CAPAs feasible in
neutral (Figure 6a) or acidic solutions. Because the develop-
ment of the chemistry of NCAs is more extensive than that
of CAPAs, we believe that successful experiences on NCAs
can serve as a reference for the chemistry of CAPAs.
The carboxyl group of the amino acid in NCAs is highly
activated such that peptide formation can occur at room
temperature. The ease of decarboxylation (releasing CO2
gas) of the initial product N-carboxyl peptide makes NCAs
prominent in peptide oligomerization and polymerization.
When stepwise peptide synthesis is required, N-substituted
NCAs were developed[3b] because N-protecting group for-
bids the amino group in the decarboxylated product to en-
gage in further coupling and such a N-protecting group can
be easily removed after the coupling step. Although reac-
tions of amino acids with CAPAs at two electrophilic
centres produce N,NЈ-phosphorylated bisamino acids 7 be-
sides peptide derivatives 6. Such a bielectrophilicity will not
sacrifice their potential in peptide formation because 7 is
thermodynamically unstable and can spontaneously con-
verted back to CAPAs, which again could react with free
amino acids to form 6 (Figure 3b). Nevertheless, relative
stability of the P–N bond in N-phosphoryl peptides under
basic condition make CAPAs a potential candidate in step-
wise peptide synthesis rather than ring-opening polymeriza-
tion.
Another interesting aspect of NCAs is phosphate acti-
vation related to prebiotic evolution. Reaction of NCAs
with inorganic phosphate (Scheme 2) produces aminoacyl
phosphates via the precursor N-carboxyl aminoacyl phos-
phates 8Ј.[1a,27] In the present work, reaction of CAPAs with
inorganic phosphate leads to pyrophosphate derivatives 9,
which are another form of activated phosphate (Scheme 1).
Finally, phosphorus is a reactive centre toward several nu-
cleophiles, especially alkoxide nucleophiles, and this makes
CAPAs a promising phosphorylating agent having an
amino acid conjugate. This may provide an alternative route
to the synthesis of phosphoramidate prodrugs of nucleoside
analogues.[34] Although alcohol phosphorylation (e.g., nu-
cleosides) by CAPAs in water is not efficient, the problem
could be circumvented by transfer of the reaction into an-
hydrous solvents.
Experimental Section
General: Reagents were purchased from various commercial
sources. 15N-glycine from Cambridge Isotope lab and H218O
Shanghai Research Institute of Chemical Industry; 1H, 13C and
31P NMR spectra were recorded with a Bruker 400 MHz NMR
spectrometer. 1H NMR chemical shifts were measured relative to
D2O (δ = 4.70 ppm). 13C NMR chemical shifts were measured rela-
tive to CD3OD (δ = 49.5 ppm). 31P NMR chemical shifts in D2O
were externally referenced to 85% H3PO4 (δ = 0.0 ppm). Chemical
shifts are given in ppm and signals are reported as s (singlet), d
(doublet), t (triplet), q (quartet), quint. (quintuplet) and m (mul-
tiplet). MS (ESI) spectra were acquired with a Bruker Dalton Es-
quire 3000 plus ion-trap mass spectrometer (Bruker-Daltonik Co.,
Bremen, Germany). High-resolution mass spectra were performed
with an ESI-Q-TOF-MS spectrometer (Micromass, England) and
LC–MS-IT-TOF (Shimadzu, Japan).
On the basis of the above discussion, CAPAs are not only
phosphate-activated amino acids but also amino-acid-acti-
vated phosphates as unique phosphoryl donors. Conse-
quently, CAPAs might be considered as potential models
that can be used to understand enzymatic phosphoryl trans-
fer mechanisms and aminoacyl transfer processes.
Reaction of Gly, Ala with P3m in H218O: Amino acids (0.05 mmol)
and P3m (0.05 mmol) were dissolved in H218O ([18O]Ͼ96%,
0.5 mL) and heated to 40 °C. The pH of the solution was main-
tained at 11.0 by the frequent addition of Na18OH solution. The
reaction was quenched by Na18OH solution after 24 h, and the
quenched reaction mixture was mixed with D2O (0.05 mL, for lock-
ing and shimming in NMR) and tested by NMR spectroscopy.
Conclusions
The results described herein indicate that α-CAPA gener-
ated in situ from α-amino acids and P3m is a reactive inter-
mediate in alkaline media. The bielectrophilicity of α-
CAPAs (Gly-CAPA, Ala-CAPA) was identified by isotopic
analysis (18O, 15N) and further proved by trapping the α-
CAPAs with nucleophiles such as water, amino acids, phos-
phate and methanol in alkaline media, which yielded inter-
esting phosphorylated products. By comparison with the re-
Reaction of an Excess Amount of Gly, Ala with P3m in Water:
Amino acids (4 mmol) and P3m (0.4 mmol) were dissolved in water
(4 mL) at room temperature. The pH of the solution was main-
tained at 11.0 by adding NaOH solution. The reaction mixture was
then monitored by 31P NMR spectroscopy.
Reaction of Amino Acids with P3m in Na2HPO4 Buffer: Amino ac-
ids (2 mmol), P3m (2 mmol) and Na2HPO4·10H2O (20 mmol) were
activity of NCAs, the bielectrophilicity of CAPAs indicates dissolved in water (20 mL) at 40 °C. The pH of the solution was
3032 Eur. J. Org. Chem. 2009, 3026–3035
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

On the Electrophilicity of CAPAs Postulated as Intermediates
maintained at 11.0 by adding NaOH solution. The reaction mix-
ture was then test by 31P NMR spectroscopy after 12 h.
and sequentially washed with 10% HCl (3ϫ90 mL), 10% Na2CO3
(150 mL), distilled water (150 mL) and brine (2ϫ100 mL). After
drying the organic layer with MgSO4, the solvent was removed in
vacuo to give a yellow oil. The crude product was puried by
crystallization from ethyl acetate to give colourless tiny crystals
Isolation of N-Pyrophosphoglycine 9a: 31P NMR spectroscopy was
used to monitor the following isolation process. The reaction above
was quenched after 24 h by adding NaOH (20 mmol), and the solu-
tion was slowly cooled to 4 °C. The precipitate (inorganic phos-
phate) was filtered off. The filtrate was mixed with methanol
(20 mL) and then stored at –20 °C for 10 h to precipitate residual
inorganic phosphate, pyrophosphate and triphosphate, which were
filtered off. The filtrate was then condensed to a volume of 2 mL,
to which methanol (4 mL) was slowly added to precipitate the oily
crude product 9a. The oily product was separated and dissolved in
0.1  NaOH (2 mL) and again precipitated by adding methanol
(4 mL, repeat this step 10 times to remove the N-phosphoglycine
and unreacted glycine). Yield 32%, colourless oil. 1H NMR
(400 MHz, D2O): δ = 3.43 (d, J = 6.6 Hz, 2 H) ppm. 13C NMR
(100 MHz, D2O): δ = 45.7, 179.8 (d, J = 13.0 Hz) ppm. 31P NMR
(162 MHz, D2O): δ = –6.04 (d, J = 21.5 Hz, 1 P), –1.54 (d, J =
1
(28 g, 52%). M.p. 102.2–103.8 °C. H NMR (400 MHz, CDCl3): δ
= 4.11 (t, J = 6.5 Hz, 2 H), 4.21–4.31 (m, 4 H), 6.66 (d, J = 706 Hz,
1 H), 7.23–7.30 (m, 4 H), 7.32–7.41 (m, 4 H), 7.47–7.53 (m, 4 H),
7.69–7.73 (m, 4 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 47.9
(d, J = 6.7 Hz) 67.0 (d, J = 6.3 Hz), 120.0 (d, J = 4.5 Hz), 125.0,
127.1 (d, J = 2.2 Hz), 127.9 (d, J = 2.5 Hz), 141.3, 142.9 (d, J =
9.2 Hz) ppm. 31P NMR (162 MHz, CDCl3): δ = 8.43 ppm.
N-Bis(9-uorenylmethyl)phosphoryl Glycine Methyl Ester: To
a
stirred solution of glycine methyl ester (1.1 mmol) in acetonitrile
(5 mL) and N,N-diisopropylethylamine (392 µL, 2.2 mmol) was
added a solution of bis(9-uorenylmethyl) phosphite (1 mmol) in
acetonitrile (1 mL) and carbon tetrachloride (5 mL) by syringe un-
der an atmosphere of argon at 0 °C over 5°min. The reaction mix-
ture was stirred overnight at room temperature. The reaction mix-
ture was then diluted with ethyl acetate (50 mL) and sequentially
washed with 10% HCl (50 mL), 10% NaHCO3 (50 mL), distilled
water (50 mL) and brine (50 mL). After drying the organic layer
with MgSO4, the solvent was removed in vacuo to give a light yel-
low raw product, which was crystallized from dichloromethane
(10 mL) and hexane (35 mL) to give colourless tiny crystals (79%).
M.p. 130.6–130.9 °C. 1H NMR (400 MHz, CDCl3): δ = 3.14 (dt, J
= 10.8, 6.4 Hz, 1 H), 3.45 (dd, J = 10.4, 6.4 Hz, 2 H), 3.66 (s, 3
H), 4.13 (t, J = 6.6 Hz, 2 H), 4.20–4.34 (m, 4 H), 7.23–7.30 (m, 4
H), 7.33–7.41 (m, 4 H), 7.48–7.54 (m, 4 H), 7.70–7.74 (m, 4 H)
ppm. 13C NMR (100 MHz, CDCl3): δ = 42.4, 47.9 (d, J = 8.0 Hz),
52.3, 68.0 (d, J = 5.4 Hz), 119.9 (d, J = 5.7 Hz), 125.0 (d, J =
13.6 Hz), 127.0 (d, J = 3.1 Hz), 127.8 (d, J = 5.1 Hz), 141.3 (d, J
= 8.0 Hz), 143.3 (d, J = 12.8 Hz), 171.1 (d, J = 7.5 Hz) ppm. 31P
NMR (162 MHz, CDCl3): δ = 8.65 ppm. HRMS (ESI-Q-TOF+):
calcd. for C31H29NO5P+ 526.1778; found 526.1793.
–
21.5 Hz, 1 P) ppm. LC–HRMS (IT-TOF–): calcd. for C2H6NO8P2
233.9574; found 233.9559. LC–HRMS (IT-TOF–): calcd. for
–
C2H5NNaO8P2 255.9394; found 255.9390. LC–HRMS (IT-TOF–
–
): calcd. for C2H4NNa2O8P2 277.9213; found 277.9199. Note: no
attempt was made to maximize the yield. Attempts to evaporate
oily product 9a to dryness under reduced pressure at 20 °C resulted
in decomposition; a small portion of the oily product was weighed
and the solvents were evaporated to dryness under 120 °C and the
yield was estimated on the basis of the weight of dried material.
Reaction of Gly with P3m in Methanol/Water: Glycine (0.4 mmol)
and P3m (0.4 mmol) were dissolved in methanol/water (1:4, 4 mL)
at 40 °C. The pH of the solution was maintained at 11.0 by adding
NaOH solution. The reaction mixture was then monitored by 31P
NMR spectroscopy.
General Procedures for the Synthesis of the Sodium Salt of N-Phos-
phoamino acids: Sodium salts of N-phosphoamino acids were syn-
thesized according to a literature method[37] with minor modifica-
tions (Scheme 4). The sodium salt of the N-phosphoamino acids
was obtained through deprotection of the corresponding N-bis(9-
uorenylmethyl)phosphoryl amino acid methyl esters in dioxane/
H2O (1:1) containing NaOH (4 equiv.).
N-Bis(9-uorenylmethyl)phosphoryl Alanine Methyl Ester: Title com-
pound was synthesized from ()-alanine by using the same syn-
thetic procedure of the glycine derivative. Colourless tiny crystals.
Yield: 84%. [α]2D0 = –5.8 (c = 0.01, CHCl3). M.p. 141.8–142.2 °C.
1H NMR (400 MHz, CDCl3): δ = 1.20 (d, J = 7.1 Hz, 3 H), 3.25
(t, J = 10.2 Hz, 1 H), 3.60 (s, 3 H), 3.68–3.78 (m, 1 H), 4.10–4.35
(m, 6 H), 7.23–7.31 (m, 4 H), 7.33–7.41 (m, 4 H), 7.47–7.56 (m, 4
Bis(9-uorenylmethyl) Phosphite: To a stirred solution of diisopropyl
phosphoramidous dichloride (24.74 g, 122.5 mmol) in dichloro-
methane (200 mL) was added a solution of 9-uorenylmethanol
(30.6 g, 196 mmol) in dichloromethane (300 mL) under an atmo-
sphere of argon at 0 °C. The reaction mixture was then stirred for
3 h at room temperature. The solvent was evaporated under re-
duced pressure to give a viscous residue, which was subsequently
dissolved in acetonitrile (250 mL) and chilled to 0 °C. A solution of
acetonitrile (50 mL), 1-H-tetrazole (5.35 g, 76.4 mmol) and distilled
water (20 mL) was then added, and the reaction mixture was stirred
overnight. The solution was then diluted in ethyl acetate (400 mL)
13
H), 7.70–7.74 (m, 4 H) ppm. C NMR (100 MHz, CDCl3): δ =
20.9 (d, J = 4.9 Hz), 47.9 (dd, J = 4.1 Hz), 49.8, 52.3, 68.0 (d, J =
5.6 Hz), 119.9 (m), 125.0 (m), 127.0, 127.8 (m), 141.3 (m), 143.3
31
(m), 174.1 (d, J = 7.1 Hz) ppm. P NMR (162 MHz, CDCl3): δ
= 7.97 ppm. HRMS (ESI-Q-TOF+): calcd. for C32H31NO5P+
540.1934; found 540.1939.
Sodium Salt of N-Phosphoglycine (5a): To a stirred solution of
bis(9-uorenylmethyl)phosphoryl glycine (249 mg, 0.5 mmol) in di-
Scheme 4. Synthetic path of N-phosphoamino acids.
Eur. J. Org. Chem. 2009, 3026–3035
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3033

F. Ni, X. Gao, Z. X. Zhao, C. Huang, Y. F. Zhao
FULL PAPER
oxane/H2O (1:1, 10 mL), NaOH (2 mmol) was added. The reaction
mixture was stirred overnight at room temperature. The reaction
solution was evaporated to a volume of about 1 mL under reduced
pressure and mixed with water (4 mL), the solids were ltered out
and the ltrate was evaporated to dryness under reduced pressure to
give the sodium salt product. White solid, m.p. Ͼ300 °C.1 H NMR
(400 MHz, D2O): δ = 3.23 (d, J = 6.1 Hz, 2 H) ppm. 13C NMR
(100 MHz, D2O): δ = 46.28, 180.1 (d, J = 12.5 Hz) ppm. 31P NMR
(162 MHz, D2O): δ = 8.38 ppm. MS (ESI): m/z = 221.8 [M + H]+.
5 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 42.9, 52.3, 53.7 (d,
J = 5.6 Hz), 120.1 (d, J = 5.9 Hz), 124.9, 129.7, 150.7 (d, J =
6.6 Hz), 170.9 (d, J = 8.5 Hz) ppm. 31P NMR (162 MHz, CDCl3):
δ = 5.38 ppm. HRMS (ESI-Q-TOF+): calcd. for C10H15NO5P+
260.0682; found 260.0696. HRMS (ESI-Q-TOF+): calcd. for
C10H14NaNO5P+ 282.0502; found 282.0500.
Sodium Salt of N-Monomethoxyphosphoryl Glycine (11a): Title
compound was synthesized according to a literature method.[39]
A
solution (2 mL) of 0.8  NaOH (MeOH/H2O, 1:1) was added to
N-(methoxyphenoxyphosphanyl)glycine methyl ester (0.2 mmol)
and a deprotection reaction was carried out at 60 °C while stirring
under an atmosphere of argon for 12 h. The course of the reaction
was monitored by 31P NMR spectroscopy until the disappearance
of the starting material. Once N-(methoxyphenoxyphosphinyl)gly-
cine methyl ester disappeared, the solvent was removed under re-
duced pressure to give an oily crude product, which was recrys-
tallized from 95% ethanol (1 mL). The precipitate was evaporated
to dryness to give the product as a white solid (88%). 1H NMR
(400 MHz, D2O): δ = 3.31 (d, J = 8.2 Hz, 2 H), 3.43 (d, J =
10.9 Hz, 3 H) ppm. 13C NMR (100 MHz, D2O): δ = 45.2, 52.0 (d,
J = 5.3 Hz), 179.2 (d, J = 9.3 Hz) ppm. 31P NMR (162 MHz, D2O):
δ = 11.88 ppm. LC–HRMS (IT-TOF–): calcd. for C3H7NO5P–
168.0067; found 168.0081. LC–HRMS (IT-TOF–): calcd. for
C3H6NNaO5P– 189.9887; found 189.9889.
Sodium Salt of N-Phosphoalanine (5b): Title compound was synthe-
sized by using the same synthetic procedure used for 5a. White
solid, m.p. Ͼ300 °C. [α]2D0 = –2.7 (c = 0.01, H2O). 1H NMR
(400 MHz, D2O): δ = 1.18 (d, J = 7.0 Hz, 3 H), 3.42–3.50 (m, 1 H)
ppm. 13C NMR (100 MHz, D2O): δ = 21.6 (d, J = 3.5 Hz), 52.8,
184. 1 (d, J = 8.6 Hz) ppm. 31P NMR (162 MHz, D2O): δ =
8.07 ppm. MS (ESI): m/z = 235.8 [M + H]+.
General Procedures for the Synthesis of the Lithium Salt of N-Phos-
phoglycylglycine (6a): Title compound was synthesized according
to a literature method[37] with minor modification, as shown in
Scheme 4. Because the amide bond is labile in the presence of
strong bases such as NaOH, 6a was obtained through deprotection
of the corresponding N-bis(9-uorenylmethyl)phosphoryl glycylgly-
cine methyl ester in dioxane/H2O (1:1) containing LiOH (4 equiv.).
Detailed synthetic procedures were similar to that of N-phos-
phoamino acids except NaOH was replaced by LiOH in the depro-
tection step.
Methyl Phosphate Disodium Salt (MeP): Methyl phosphate was
synthesized according to the literature.[40] White solid. 1H NMR
(400 MHz, D2O): δ = 3.36 (d, J = 10.0 Hz, 3 H) ppm. 31P NMR
(162 MHz, D2O): δ = 8.59 ppm.
N-Bis(9-uorenylmethyl)phosphoryl Glycylglycine Methyl Ester:
Yield: 87%, white solid, m.p. 139.6–139.9 °C. 1H NMR (400 MHz,
CDCl3): δ = 3.28 (dd, J = 11.9, 7.0 Hz, 2 H), 3.39 (dt, J = 10.8,
7.4 Hz, 1 H), 3.66 (s, 3 H), 3.86 (d, J = 5.4 Hz, 2 H), 4.10 (t, J =
6.1 Hz, 2 H), 4.23–4.34 (m, 4 H), 6.73 (t, J = 5.3 Hz, 1 H), 7.21–
7.29 (m, 4 H), 7.32–7.40 (m, 4 H), 7.43–7.52 (m, 4 H), 7.68–7.73
(m, 4 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 40.9, 44.2, 48.0
(d, J = 7.8 Hz), 52.3, 68.1 (d, J = 5.6 Hz), 120.0 (d, J = 4.5 Hz),
124.9 (d, J = 8.7 Hz), 127.1, 127.8 (d, J = 5.3 Hz), 141.4 (d, J =
4.8 Hz), 143.2 (d, J = 6.0 Hz), 168.4, 169.9 (d, J = 6.1 Hz) ppm.
31P NMR (162 MHz, CDCl3): δ = 7.62 ppm. HRMS (ESI-Q-
TOF+): calcd. for C33H3N2NaO6P+ 605.1812; found 605.1810.
N-Phosphoglycylglycine Lithium Salt (6a): White solid. 1H NMR
(400 MHz, D2O): δ = 3.40 (d, J = 9.5 Hz, 2 H), 3.71 (s, 2 H) ppm.
13C NMR (100 MHz, D2O): δ = 43.2, 45.6, 176.4 (d, J = 8.0 Hz),
176.9 ppm. 31P NMR (162 MHz, D2O): δ = 10.48 ppm. HRMS
(ESI-Q-TOF–): calcd. for C4H8N2O6P– 211.0125; found 211.0120.
HRMS (ESI-Q-TOF–): calcd. for C4H7LiN2O6P– 217.0207; found
217.0191.
Supporting Information (see also the footnote on the first page of
this article): Comparison of the 1H NMR spectra of 5a and 5b; 31
P
NMR spiking spectra of 6a, 9a, 11a, MeP; 31P NMR spectra of 7b
and N-pyrophosphorylated derivatives of Ala, Ser, Phe; HRMS–
1
MS spectra of 9a; 31P, H and 13C NMR spectra of 6a and 9a.
Acknowledgments
We acknowledge financial support from the Ministry of Science
and Technology (2006DFA43030) and the Chinese National Natu-
ral Science Foundation (20732004). We appreciate Dr. G. M. Black-
burn for useful discussions and suggestions.
[1] a) L. J. Leman, L. E. Orgel, M. R. Ghadiri, J. Am. Chem. Soc.
2006, 128, 20–21; b) L. J. Leman, L. E. Orgel, M. R. Ghadiri,
Science 2004, 306, 283–286.
[2] G. Danger, L. Boiteau, H. Cottet, R. Pascal, J. Am. Chem. Soc.
2006, 128, 7412–7413.
[3] a) H. R. Kricheldorf, α-Amino Acid–N-Carboxyanhydrides and
Related Heterocycles: Syntheses, Properties, Peptide Synthesis,
Polymerization, Springer, Berlin, 1987; b) H. R. Kricheldorf,
Angew. Chem. Int. Ed. 2006, 45, 5752–5784.
[4] a) J. Rabinowitz, J. Flores, R. Krebsbach, G. Rogers, Nature
1969, 224, 795–796; b) J. Rabinowitz, Helv. Chim. Acta 1970,
53, 1350–1355; c) N. M. Chung, R. Lohrmann, L. E. Orgel, J.
Rabinowitz, Tetrahedron 1971, 27, 1205–1210.
[5] a) L. Jaenicke, J. Koch, Justus Liebigs Ann. Chem. 1963, 663,
50–58; b) G. W. Smithers, H. Jahansouz, J. L. Kofron, R. H.
Himes, G. H. Reed, Biochemistry 1987, 26, 3943–3948; c) H.
Jahansouz, K. B. Mertes, M. P. Mertes, R. H. Himes, Bioorg.
Chem. 1989, 17, 207–216.
N-(Methoxyphenoxyphosphanyl)glycine Methyl Ester: Title com-
pound was synthesized according to a literature method.[38,39] Di-
phenyl phosphite (2 mmol) and a catalytic amounts of Et3N (about
5 drops) were dissolved in dry THF (5 mL) and cooled to –5 °C.
Under an argon atmosphere, CH3OH (1 mmol) in dry THF (5 mL)
was added dropwise to the solution. The solution was stirred for
about 30 min and warmed from –5 °C to room temperature during
that time. The progress of the reaction was monitored by 31P NMR
spectroscopy. At 0 °C, glycine methyl ester hydrochloride (2 mmol),
Et3N (TEA; 0.31 g, 3 mmol) and CCl4 (3.6 mmol) in CH2Cl2
(5 mL) were sequentially added to the resulting solution, and the
solution was stirred for about 2 h. The solvent was removed under
reduced pressure, and the remaining residue was purified by silica
gel column chromatography (ethyl acetate/petroleum ether, 3:1) to
afford the target product as a colourless oil (56%). 1H NMR
(400 MHz, CDCl3): δ = 3.14 (dt, J = 11.2, 5.9 Hz, 1 H), 3.74 (s, 3
H), 3.77–3.81 (m, 2 H), 3.83 (d, J = 11.4 Hz, 3 H), 7.14–7.35 (m,
[6] a) F. Lipmann, L. C. Tuttle, J. Biol. Chem. 1944, 153, 571–582;
b) R. Bentley, J. Am. Chem. Soc. 1949, 71, 2765–2767; c) D. E.
Koshland, J. Am. Chem. Soc. 1952, 74, 2286–2296; d) G. Di Sa-
bato, W. P. Jencks, J. Am. Chem. Soc. 1961, 83, 4393–4400; e)
3034
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3026–3035

On the Electrophilicity of CAPAs Postulated as Intermediates
G. Di Sabato, W. P. Jencks, J. Am. Chem. Soc. 1961, 83, 4400–
4405; f) J. P. Klinman, D. Samuel, Biochemistry 1971, 10, 2126–
2131; g) D. Herschlag, W. P. Jencks, J. Am. Chem. Soc. 1986,
108, 7938–7946.
a) A. Katchalsky, M. Paecht, J. Am. Chem. Soc. 1954, 76,
6042–6044; b) M. Paecht, A. Katchalsky, J. Am. Chem. Soc.
1954, 76, 6197–6198.
a) K. Moldave, P. Castelfranco, A. Meister, J. Biol. Chem. 1959,
234, 841–848; b) N. S. M. D. Wickramasinghe, M. P. Staves,
J. C. Lacey, Biochemistry 1991, 30, 2768–2772; c) D. W. Mul-
lins, J. C. Lacey, Biochem. Biophys. Res. Commun. 1980, 96,
491–497; d) D. W. Mullins, J. C. Lacey, J. Mol. Evol. 1980, 15,
339–345; e) A. L. Weber, J. C. Lacey, J. Mol. Evol. 1975, 6, 309–
320; f) S. Tyagi, C. Ponnamperuma, J. Mol. Evol. 1990, 30,
391–399; g) J. C. Lacey, N. Senaratne, D. W. Mullins, Origins
of Life 1984, 15, 45–54.
P. V. Attwood, M. J. Piggott, X. L. Zu, P. G. Besant, Amino
Acids 2007, 32, 145–156.
a) Y. F. Wei, H. R. Matthews, Methods Enzymol. 1991, 200,
388–414; b) A. H. West, A. M. Stock, Trends Biochem. Sci.
2001, 26, 369–376; c) P. G. Besant, E. Tan, P. V. Attwood, Int.
J. Biochem. Cell Biol. 2003, 35, 297–309.
M. E. Jones, F. Lipmann, Proc. Natl. Acad. Sci. USA 1960, 46,
1194–1205.
a) I. S. Kulaev, V. M. Vagabov T. V. Kulakovskaya, The Bio-
chemistry of Inorganic Polyphosphates, 2nd ed., John Wiley &
Sons, Chichester, 2004; b) A. Kornberg, N. N. Rao, D. Ault-
Riche, Annu. Rev. Biochem. 1999, 68, 89–125; c) M. R. W.
Brown, A. Kornberg, Trends Biochem. Sci. 2008, 33, 284–290.
[21] F. C. Lightstone, T. C. Bruice, J. Am. Chem. Soc. 1996, 118,
2595–2605.
[22] J. P. Biron, A. L. Parkes, R. Pascal, J. D. Sutherland, Angew.
Chem. Int. Ed. 2005, 44, 6731–6734.
[23] With high concentration of reactants, our preliminary results
showed the formation of the corresponding N-mono-(5Ј-urid-
ine) phosphorylglycine (data not shown here).
[24] A. C. Farthing, R. J. W. Reynolds, Nature 1950, 165, 647–647.
[25] a) P. D. Bartlett, R. H. Jones, J. Am. Chem. Soc. 1957, 79,
2153–2159; b) P. D. Bartlett, D. C. Dittmer, J. Am. Chem. Soc.
1957, 79, 2159–2160.
[26] J. L. Bailey, J. Chem. Soc. 1950, 3461–3466.
[27] J. P. Biron, R. Pascal, J. Am. Chem. Soc. 2004, 126, 9198–9199.
[28] a) M. Bergmann, L. Zervas, W. F. Ross, J. Biol. Chem. 1935,
111, 245–260; b) B. F. Erlanger, E. Brand, J. Am. Chem. Soc.
1951, 73, 3508–3510; c) M. Frankel, E. Katchalski, J. Am.
Chem. Soc. 1943, 65, 1670–1674.
[29] H. R. Kricheldorf, α-Amino Acid–N-Carboxyanhydrides and
Related Heterocycles: Syntheses, Properties, Peptide Synthesis,
Polymerization, Springer, Berlin, 1987, pp. 63–64.
[30] K. D. Kopple, J. Am. Chem. Soc. 1957, 79, 6442–6446.
[31] H. Fu, Z. L. Li, Y. F. Zhao, G. Z. Tu, J. Am. Chem. Soc. 1999,
121, 291–295.
[32] a) Y. M. Li, Y. W. Yin, Y. F. Zhao, Int. J. Pept. Protein Res.
1992, 39, 375–381; b) Y. F. Zhao, P. S. Cao, J. Biol. Phys. 1994,
20, 283–287; c) W. H. Zhou, Y. Ju, Y. F. Zhao, Orig. Life Evol.
Biosph. 1996, 26, 547–560; d) C. M. Cheng, X. H. Liu, Y. M.
Li, Y. Ma, B. Tan, R. Wan, Y. F. Zhao, Orig. Life Evol. Biosph.
2004, 34, 455–464.
[7]
[8]
[9]
[10]
[11]
[12]
[33] a) F. Mathey, Angew. Chem. Int. Ed. 2003, 42, 1578–1604; b)
K. B. Dillon, F. Mathey, J. F. Nixon in Phosphorus, The Carbon
Copy, Wiley, Chichester, 1998.
[13]
A. W. Schwartz, Phil. Trans. R. Soc. B 2006, 361, 1743–1749.
[14] a) Y. Yamagata, H. Watanabe, M. Saitoh, T. Namba, Nature
1991, 352, 516–519; b) M. A. Pasek, T. P. Kee, D. E. Bryant,
A. A. Pavlov, J. I. Lunine, Angew. Chem. Int. Ed. 2008, 47,
7918–7920.
[34] D. P. Drontle, C. R. Wagner, Mini-Rev. Med. Chem. 2004, 4,
409–419.
[35] a) Y. F. Zhao, P. S. Cao, Pure Appl. Chem. 1999, 71, 1163–1166;
b) R. Pascal, L. Boiteau, A. Commeyras, Top. Curr. Chem.
2005, 259, 69–122.
[15] X. Gao, Y. Liu, P. X. Xu, Y. M. Cai, Y. F. Zhao, Amino Acids
2008, 34, 47–53.
[16] H. Inoue, Y. Baba, T. Furukawa, Y. Maeda, M. Tsuhako,
Chem. Pharm. Bull. 1993, 41, 1895–1899.
[36] I. Devedjiev, K. Petrova, I. Glavchev, Synth. Commun. 2000,
30, 4411–4415.
[17] F. Ni, S. T. Sun, C. Huang, Y. F. Zhao, Green Chem. 2009, 11,
[37] L. Y. Wu, C. E. Berkman, Tetrahedron Lett. 2005, 46, 5301–
5303.
569–573.
[18] M. Cohn, A. Hu, Proc. Natl. Acad. Sci. USA 1978, 75, 200–
203.
[19] R. Krishnamurthy, S. Guntha, A. Eschenmoser, Angew. Chem.
Int. Ed. 2000, 39, 2281–2285.
[38] A. Kers, I. Kers, J. Stawinski, M. Sobkowski, A. Kraszewski,
Synthesis 1995, 427–430.
[39] X. S. Li, H. Fu, Y. Y. Jiang, Y. F. Zhao, J. Y. Liu, Synlett 2004,
14, 2600–2602.
[20] a) D. Saboulard, L. Naesens, D. Cahard, A. Salgado, R. Pathir-
ana, S. Velazquez, C. Mcguigan, E. De Clercq, J. Balzarini,
Mol. Pharmacol. 1999, 56, 693–704; b) S. J. Hecker, M. D.
Erion, J. Med. Chem. 2008, 51, 2328–2345.
[40] P. K. Grzyska, P. G. Czyryca, J. Golightly, K. Small, P. Larsen,
R. H. Hoff, A. C. Hengge, J. Org. Chem. 2002, 67, 1214–1220.
Received: March 3, 2009
Published Online: May 11, 2009
Eur. J. Org. Chem. 2009, 3026–3035
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3035