Phosphate-mediated interconversion of Ribo- and Arabino-configured prebiotic nucleotide intermediates

Article abstract of DOI:10.1002/anie.201001662

(Represented Chemical Equation) Flipping stereochemistry: Inorganic phosphate catalyzes the interconversion of ribose and arabinose aminooxazolines, suggesting that the prebiotic synthesis of enantiopure pyrimidine ribonucleotides might have involved a single stereoinversion at Cl′ and a double stereoinversion at C2′.

Full text of DOI:10.1002/anie.201001662

Angewandte
Chemie
DOI: 10.1002/anie.201001662
Nucleotide Stereochemistry
Phosphate-Mediated Interconversion of Ribo- and Arabino-
Configured Prebiotic Nucleotide Intermediates**
Matthew W. Powner and John D. Sutherland*
We recently described a prebiotically plausible synthesis of
pyrimidine ribonucleoside-2’,3’-cyclic phosphates 1 that pro-
ceeds through the intermediacy of arabinose aminooxazoline
2 and involves inorganic phosphate in several of its steps
(Scheme 1).[1]
potentially giving rise to enormously complex diastereomeric
mixtures of polymers), but as it stands, the synthesis does not
yield enantiomerically pure ribonucleotides 1 unless the
starting glyceraldehyde 5 is enantiomerically pure. Whilst
enantioselectivity in the formation of glyceraldehyde 5 by
aldolization of glycolaldehyde 6 and formaldehyde is easy to
imagine if a suitable chiral catalyst could be found, formation
of enantiopure 5 seems unlikely. Fractional crystallization of
the ribonucleotides 1 or chiral intermediates in the synthesis
offers a potential means of further enantioenrichment. The
aminooxazolines 2 and 3 are less polar than 1 or the
anydronucleosides such as 9, and so offer the best hope of
such a crystallization. However, the arabino-configured 2
needed for ribonucleotide synthesis is more water-soluble
than the ribo-configured 3 along with which it is produced.
Indeed, upon cooling or concentration of a solution of all four
aminooxazolines, 3 selectively crystallizes.[2] Furthermore,
scalemic 3 is enantioenriched by crystallization such that
enantiopure 3 can be obtained from 5 with an enantiomeric
excess (ee) of 60%.[3] We have therefore sought a means
whereby 3, purified and enriched to enantiomeric purity in
such a way, might subsequently be partly[4] converted into 2.
Such a conversion would maintain the enantiopurity estab-
lished by crystallization but transfer it from the ribo series to
the arabino series and from there onwards to the ribo
nucleotides 1.
Scheme 1. Potentially prebiotic synthesis of activated pyrimidine nucle-
otides 1. Pi =inorganic phosphate.
Compound 2 is produced alongside ribose aminooxazo-
line 3 and its lyxo and xylo diastereoisomers when 2-amino-
oxazole 4 adds to glyceraldehyde 5. The synthesis of 4 from
glycolaldehyde 6 and cyanamide 7 only takes place efficiently
at neutral pH when phosphate was included as a general acid–
base catalyst. Furthermore, the subsequent reaction of 2 with
cyanoacetylene 8, giving the anhydronucleoside 9, is also
controlled by phosphate at pH 6.5, where it acts as a pH
buffer, thereby preventing hydrolysis of 9 to arabinocytidine.
Heating 9 with phosphate in formamide or in the dry state
then results in the formation of 1 (base = Cyt). In a final step,
1 (base = Cyt) is partly converted into 1 (base = Ura) by UV
irradiation in aqueous solution.
Phosphate has proved to be a key player in the systems
chemistry aspects of the synthesis, so we took the requirement
for phosphate to be present in the subsequent conversion of 2
into 9 as a clue that phosphate might also catalyze the
conversion of 3 into 2. Furthermore, we envisaged a potential
mechanism for the interconversion of 2 and 3 that appeared to
require general acid–base catalysis (Scheme 2), and we had
previously found phosphate to be an ideal mediator of such
catalysis.[1]
When exploring the effects of pH buffering by phosphate
on the reactions of aminooxazolines 2 and 3 with cyanoace-
tylene 8, we had not seen any interconversion of 2 and 3.
Enantiomerically pure monomers are considered essential
for the abiogenesis of RNA (racemic or scalemic monomers
[*] Dr. M. W. Powner, Prof. Dr. J. D. Sutherland
School of Chemistry, The University of Manchester
Oxford Road, Manchester M13 9PL (UK)
Fax: (+44)161-275-4939
E-mail: john.sutherland@manchester.ac.uk
[**] This work was supported by the Engineering and Physical Sciences
Research Council through provision of a studentship to M.W.P.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201001662.
Scheme 2. Proposed mechanism for the interconversion of 2 and 3.
Angew. Chem. Int. Ed. 2010, 49, 4641 –4643
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4641

Communications
These experiments involved limited exposure of the amino-
their fused tricyclic framework. As the conversion of 10 into 2
and 3 was believed to require general acid catalysis, we
thought that it should be possible to deprotect 17 by specific
acid catalysis without the 12 so-produced equilibrating with 2
and 3. Accordingly, we treated 17 with mineral acid at room
temperature and, gratifyingly, it was converted to 12 in
quantitative yield.
oxazolines to phosphate before addition of 8 and subsequent
rapid cyanovinylation chemistry. It therefore struck us that if
phosphate-catalyzed interconversion of 2 and 3 is possible, it
must be slow, and we planned our experimental approach
accordingly. In a preliminary experiment, 3 was dissolved in
deuterated 1m phosphate buffer at pD = 7, and this solution
was then incubated for three weeks at 608C with periodic
With an authentic sample of 12 in hand, we were able to
1
1
examination by H NMR spectroscopy. Over time, a singlet
compare H NMR spectral signals and confirm that 12 was
signal appeared at d = 6.54 ppm, the signal for H1’ of 3
collapsed from a doublet to a singlet, and three major new
singlet signals appeared in the chemical shift region for H1’ of
aminooxazolines and related species between d = 5.50 and
6.00 ppm. When this experiment was repeated in H2O with
samples being periodically removed and lyophilized prior to
indeed present in the mixture of compounds formed by
heating 3 in phosphate buffer. We then subjected a synthetic
sample of 12 to reaction in 1m phosphate buffer at 408C and
pH 7.0 to confirm the intermediacy of 12 in the equilibration
of 3 and 2. Samples were periodically withdrawn and
lyophilized, and the residues redissolved in D2O for
1H NMR analysis. This procedure revealed that 12 was
partially converted into 3 and 2, and, more slowly, the
hydrolysis products 10 and 11 (Figure 1).
1
dissolution in D2O for H NMR analysis,[5] the signal at d =
6.54 ppm was still a singlet, but the other four major
downfield signals were doublets. By sample spiking with
authentic standards, it was shown that the doublet signals
were due to 2 and 3,[6] and the oxazolidinones 10 and 11[7]
(Scheme 2). The singlet signal at d = 6.54 ppm was subse-
quently shown to be due to the pentose aminooxazole 12, a
presumed intermediate in the interconversion of 2 and 3.
The potential interconversion mechanism (Scheme 2)
involves furanose ring-opening of 3 to give the iminium
species 13 which can undergo phosphate-mediated deproto-
nation of C2’ to give 12. Phosphate-mediated reprotonation of
C2’ of 12 can then either regenerate 13, or give 14 from which
2 can be produced by ring-closure. It is thought that the
oxazolidinones 10 and 11 are either products of the direct
hydrolysis of 3 and 2, or the indirect hydrolysis via 13 and 14.
Supporting its validity, the interconversion mechanism
accounts for the behavior first observed in D2O solution, as
deuteronation of 12 would give 3 and 2 (and thence the
oxazolidinones 10 and 11) deuterated at C2’ and thus having
1
Figure 1. Time course of the reaction of 12 in phosphate buffer.
Amt=amount of each species as determined by 1H NMR integration.
singlet signals for H1’ in H NMR spectra.
To confirm the presence of 12 in reaction mixtures, and to
provide support for its presumed intermediacy in the inter-
conversion of 3 and 2, we synthesized 12 using conventional
chemistry, and then subjected it to the conditions of the
interconversion. The synthesis began with 4,6-O-benzylidene-
d-glucose 15, which was elaborated to the vinyl alcohol 16 by
diol cleavage with periodate[8] followed by addition of vinyl
magnesium bromide (Scheme 3).
Ozonolysis of 16 followed by reductive work-up and
condensation with cyanamide 7 then gave 17, the 3’,5’-O-
benzylidene derivative of 12—3’,5’-O-benzylidene amino-
oxazoline derivatives not being formed because of strain in
~
&
^
*
12, 3, 2, ꢀ 10, 11.
With the major products of this reaction now identified,[9]
and the intermediacy of 12 confirmed, we next sought to
assess the effect of temperature and pH on the product
distribution (Table 1).
The reaction of 3 is slow at room temperature, and needs
several days at higher temperatures for equilibration with 2
and 12 to occur. Hydrolysis to 10 and 11 always accompanies
this equilibration. At pH 6 and above, the apparent thermo-
Table 1: Products of incubation of 3 in 1m phosphate buffer.[a]
pH
T [8C]
Product distribution ratio[b]
12
2
3
11
10
5.0
7.0
8.0
6.0
7.0
7.0
60
60
60
40
40
20
0.1
0.5
0.4
0.3
0.3
0
0.5
0.4
0.3
0.3
0.3
0
1.0
1.0
1.0
1.0
1.0
1.0
1.4
3.0
1.0
0.2
0.2
0
4.1
9.1
2.9
0.5
0.5
0
=
Scheme 3. a) NaIO4, CH2Cl2/H2O, RT; b) CH2 CHMgBr, THF, 08C,
then NH4Cl, H2O, 59% from 15; c) O3, MeOH, ꢀ788C, then Me2S,
MeOH, ꢀ788C!RT; d) H2NCN, N,N-dimethylacetamide, 608C, 75%
from 16; e) HCl, dioxane/H2O, RT, quantatitive.
[a] After 6 days incubation. [b] Normalized relative to 3.
4642 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 4641 –4643

Angewandte
Chemie
dynamic stabilities decrease in the order 3 > 12 ꢁ 2, whereas
at pH 5, 12 is disfavored relative to both 3 and 2. Furthermore,
at this lower pH, the 2:3 ratio is at its highest.
Keywords: chirality · homogeneous catalysis · nucleotides ·
phosphate · prebiotic chemistry
.
As regards the overall synthesis of 9, the formation of
oxazolidinones 10 and 11 during the equilibration reaction is
not critical, as they do not react in the subsequent cyanovi-
nylation reaction. More important is the 2:3 ratio, as 3, like 2,
reacts with cyanoacetylene to give an anhydronucleoside
intermediate.[1] The indication is that the highest yield of 9
relative to its ribo-configured counterpart (but see [4]) will be
obtained if 3 is first allowed to equilibrate in phosphate buffer
at a pH between 5 and 6, and the resultant mixture then
allowed to react with cyanoacetylene 8 at pH 6.5. Whether or
not enantiomerically pure ribonucleotide monomers 1 can be
produced under prebiotically realistic conditions now
depends on a synthesis of enantioenriched glyceraldehyde 5
under similar conditions. If the synthesis of 5 with an ee of
more than 60% can be demonstrated, the prebiotic formation
[1] M. W. Powner, B. Gerland, J. D. Sutherland, Nature 2009, 459,
239.
[2] G. Springsteen, G. F. Joyce, J. Am. Chem. Soc. 2004, 126, 9578.
[3] C. Anastasi, M. A. Crowe, M. W. Powner, J. D. Sutherland,
Angew. Chem. 2006, 118, 6322; Angew. Chem. Int. Ed. 2006, 45,
6176.
[4] A partial conversion of 3 to 2 was sought because it is possible that
activated purine ribonucleotides 1 (base = Ade, Gua) might result
from nucleophilic displacement at C1’ of cyanovinylated and
phosphorylated derivatives of 3.
[5] Full experimental details are given in the Supporting Information.
[6] R. A. Sanchez, L. E. Orgel, J. Mol. Biol. 1970, 47, 531.
[7] J. Kovꢀcs, I. Pintꢁr, U. Lendering, P. Kꢂll, Carbohydr. Res. 1991,
210, 155; P. Kꢂll, J. Kovacs, D. Abeln, J. Kopf, Carbohydr. Res.
1993, 239, 245.
[8] L. A. Paquette, Q. Zeng, H-C. Tsui, J. N. Johnstone, J. Org. Chem.
1998, 63, 8491.
[9] We also identified low levels of diastereoisomeric hydrates of 13
and 14 (18 and 19, respectively; see Supporting Information)
formed by the reversible addition of water to C1’.
of enantiomerically pure pyrimidine ribonucleotides
1
(base = Pyr) must then be considered plausible.
Received: March 19, 2010
Published online: May 20, 2010
Angew. Chem. Int. Ed. 2010, 49, 4641 –4643
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 4643