Cyanamide as a prebiotic phosphate activating agent-catalysis by simple 2-oxoacid salts

Article abstract of DOI:10.1039/c7cc07517k

Cyanamide is a prebiotically plausible compound that has previously been invoked as a phosphate activating agent. However, its reactions with phosphate monoesters are very slow and tend to be low yielding. We now report a fast and efficient phosphate activation reaction using cyanamide in the presence of glyoxylate or pyruvate. These simple 2-oxoacid salts are shown to function as catalysts and in an optimised system, adenosine-3′-phosphate was converted to adenosine-2′,3′-cyclic phosphate in 95% yield.

Full text of DOI:10.1039/c7cc07517k

ChemComm
View Article Online
View Journal
COMMUNICATION
Cyanamide as a prebiotic phosphate activating
agent – catalysis by simple 2-oxoacid salts†
Cite this:DOI: 10.1039/c7cc07517k
Maria Tsanakopoulou
and John D. Sutherland
*
Received 26th September 2017,
Accepted 12th October 2017
DOI: 10.1039/c7cc07517k
rsc.li/chemcomm
Cyanamide is a prebiotically plausible compound that has previously and inherently favoured reactions and pathways to such molecules
been invoked as a phosphate activating agent. However, its reactions that either incorporate 1, or are catalysed by derivatives thereof.
with phosphate monoesters are very slow and tend to be low Thus, cyanamide 1 features large in the prebiotic reaction network
yielding. We now report a fast and efficient phosphate activation that leads to the formation of nucleotides, amino acids and lipid
reaction using cyanamide in the presence of glyoxylate or pyruvate. precursors. More specifically, it reacts with glycolaldehyde to form
These simple 2-oxoacid salts are shown to function as catalysts and 2-aminooxazole, which is a precursor for the synthesis of nucleo-
in an optimised system, adenosine-30-phosphate was converted to tides. It also undergoes phosphate-catalysed hydration to urea,
adenosine-20,30-cyclic phosphate in 95% yield.
which in turn is a catalyst and medium for phosphorylation
reactions. Further, cyanamide 1 reacts with b-aminopropionitrile
Before the advent of biology on the early Earth, the linking of and b-aminopropionaldehyde en route to arginine.7 Accordingly,
nucleotides to give oligonucleotides must have occurred in a cyanamide 1 has been thoroughly investigated in the context of
non-enzymatic way. Phosphate activation of some sort was phosphate activation, but despite its likely abundance and the
presumably crucial to drive phosphodiester bond formation, favourable thermodynamics of its hydration, it has proven to be too
a process that is otherwise unfavourable in water. Although we unreactive kinetically. Thus, Orgel found that uridine-30-phosphate
have previously described two prebiotically plausible syntheses could only be transformed to the 20,30-cyclic phosphate derivative in
of the canonical pyrimidine nucleotides,1,2 and others have good yields if high concentrations of cyanamide 1 were employed
described syntheses of purine nucleotides,3,4 we still need to and the reactions were heated for several days (e.g. 73% with
8
´
uncover efficient prebiotic activation chemistry and energy 800 mM 1, 1 M acetate buffer, pH 5, 65 1C, 6 days). Oro showed
sources. In this regard, our attention, like that of others over the that treatment with cyanamide 1 led to the prebiotic condensation
years, has been drawn to cyanamide 1, whose fleeting tautomer of mononucleotides, but in very low yields.9 Recently, Richert
(HNQCQNH) is reminiscent of the N,N0-dialkylcarbodiimide reported that even low level conversion of a nucleoside-50-
condensing agents of conventional synthetic organic chemistry. phosphate and glycine to the corresponding phosphoramidate
The feedstocks for the cyanosulfidic chemistry that underpins our required very high concentrations of 1 and extremely long
nucleotide syntheses have been suggested to arise through thermal reaction times (75 days).10 Given that cyclic nucleotides are
metamorphosis of ferrocyanide salts. The products of such meta- labile, we sought conditions under which they could be synthe-
morphosis are contingent upon the nature of the cation(s) of sised far faster than they degrade. This led us to search for
these salts and heating calcium and magnesium ferrocyanide catalysts of the activation chemistry. We have now discovered
leads to calcium cyanamide and magnesium cyanamide or nitride that the activation of monophosphates by cyanamide 1 – as
(depending on the temperature).5–7 Hydrolysis of these cyanamide assessed by the conversion of adenosine-30-phosphate 2 to
salts produces cyanamide 1 and the metal hydroxide. Other adenosine-20,30-cyclic phosphate 3 – is dramatically accelerated
evidence that cyanamide 1 is a plausible prebiotic compound is by the prebiotically plausible 2-oxoacid salts, glyoxylate 4 and
more circumstantial and is based on the constitutional presence of pyruvate 5,11–15 especially so in the presence of magnesium or
cyanamide-derivable groupings of atoms in several key biomolecules, calcium ions.‡
At first, cyanamide 1 (to 1 M) was added as a solid to a
solution of adenosine-30-phosphate 2 (50 mM) and glyoxylic
acid 4 (10 mM) at room temperature and at different pH values in
MRC Laboratory of Molecular Biology, Francis Crick Avenue, Cambridge Biomedical
Campus, CB2 0QH, UK. E-mail: johns@mrc-lmb.cam.ac.uk
the range pH 4.5–6.9 (pKa of glyoxylic acid 4 B3.3), whereupon it
was found that pH 5 gave the best yield of adenosine-20,30-cyclic
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
c7cc07517k
This journal is ©The Royal Society of Chemistry 2017
Chem. Commun.

View Article Online
Communication
ChemComm
Scheme 1 Conversion of 2 to 3 with cyanamide 1 and 2-oxoacid salts 4
or 5.
phosphate 3 – 11%, according to 31P NMR integration, after
only 24 h (Scheme 1 and Table 1). This encouraged us because
activation by 1 has typically been associated with more pro-
longed reaction times.
We then proceeded to optimise the reaction and found that
the yield was significantly improved by increasing the amount
of glyoxylate 4, by slowly adding the cyanamide 1 portionwise
and by increasing the temperature to 40 1C (Scheme 1 and
Table 1). Pyruvate 5 showed similar reactivity to 4 although
yields were slightly lower. Notably, in the absence of the 2-oxoacid
salts, 4 or 5, adenosine-30-phosphate 2 underwent cyclisation to
adenosine-20,30-cyclic phosphate 3 in only 1% yield suggesting a
significant role for the 2-oxoacid salts in promotion or catalysis.
Analysis of the reaction mixture by NMR revealed the presence
of glyoxylurea 6 and urea 7 along with the cyclic nucleotide 3
(Scheme 2). Glyoxylurea 6, which has been studied in depth
because of its biological relevance,16 exists in equilibrium with
urea 7 and glyoxylate 4 and the rate of equilibration is accelerated
both by heating and by catalysis by divalent metal cations such as
Mg2+ and Ca2+. Reasoning that availability of glyoxylate 4 in our
reaction might be limited by a sluggish breakdown of glyoxylurea 6,
we added divalent metal cations. Yields increased significantly
(maximum yield of 95%) bolstering this supposition (Scheme 1
and Table 1).
Scheme 2 Proposed mechanistic pathway for the reaction in presence of
glyoxylate 4.
intramolecularity, or because the carbodiimide tautomer of 8 is
additionally somehow favoured. As shown in the supplementary
information (Fig. S6, ESI†), the reaction displays an initial rapid
burst phase and the rate then levels off as an intermediate
reaches steady-state. O-Acylisourea 9 probably functions as the
active intermediate for the phosphate activation, and in so doing
is converted to glyoxylurea 6, which we identified as the steady-
state intermediate. As already mentioned, the interconversion of
this latter adduct with urea 7 and glyoxylate 4, is catalysed by
divalent metal cations. As regards the actual phosphate activation
step, we suspect that the nucleotide monoanion 2 and neutral
O-acylisourea 9 are in acid–base equilibrium with a more reactive
pairing of nucleotide dianion and O-acyluronium cation that
combine to form the glyoxylurea nucleotide mixed anhydride
derivative 10. Ordinarily, related but simpler mixed anhydrides
transfer the acyl group to the 20-hydroxyl group in preference to
cyclisation,17 but in this case, the latter reaction appears to
predominate forming the cyclised product 3 (Scheme 2).
Regarding the mechanism of the reaction (Scheme 2), we
propose that cyanamide 1 and glyoxylate 4 reversibly associate to
give the carbonyl adduct 8 which then cyclises to the O-acylisourea
9. We do not know whether the addition of the glyoxylate carboxyl
group to the cyanamide carbon is fast simply due to induced
Table 1 Yield in different conditions for the conversion of 2 to 3 at pH 5
Divalent
cations
Yield
2-Oxoacid salt
Cyanamide 1
Temperature
(50 mM)
of 3 (%)
4 (10 mM)
4 (50 mM)
4 (50 mM)
4 (100 mM)
4 (100 mM)
5 (100 mM)
5 (100 mM)
4 (100 mM)
4 (100 mM)
5 (100 mM)
5 (100 mM)
4 (100 mM)
4 (100 mM)
5 (100 mM)
5 (100 mM)
1 Ma
r.t.
r.t.
r.t.
r.t.
40 1C
r.t.
40 1C
r.t.
40 1C
r.t.
40 1C
r.t.
—
—
—
—
—
—
—
11
23
35
57
68
36
43
84
95
69
80
85
92
64
77
1 Ma
1 Mb
According to the 13C NMR data, cyanamide 1 is consumed as
the cyclic nucleotide product 3 is produced. Glyoxylurea 6 is
formed in the mixture, but is interconverted in the presence of
magnesium cations with glyoxylate 4 and urea 7. In Fig. 1, the
clear conversion of 2 to 3 along with the formation of glyoxylurea 6
and urea 7 can be observed (Fig. 1a–d). It is notable that the
intensity of the carbonyl peak of urea 7 at 162.8 d increases as the
reaction progresses, thus it can be concluded that the consumption
of cyanamide 1 results in the production of urea 7. In a control
experiment, where 1 eq. of cyanamide 1 was added slowly to
glyoxylate 4 in the absence of any nucleotide, extensive conversion
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
4 ꢀ 100 mMb
Mg2+
Mg2+
Mg2+
Mg2+
Ca2+
Ca2+
Ca2+
Ca2+
40 1C
r.t.
40 1C
a
b
Fast addition. Slow addition.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2017

View Article Online
ChemComm
Communication
Fig. 1 13C NMR spectra showing the progress of the reaction in comparison with standards and control experiments; (a) standard sample of adenosine-30-
phosphate 2, (b) reaction mixture when the yield of cyclisation of 2 was 50%, (c) reaction mixture when the yield of cyclisation of 2 was 90%, (d) standard
sample of adenosine-20,30-cyclic phosphate 3, (e) control experiment of glyoxylate 4 + cyanamide 1, (f) control experiment of glyoxylate 4 + urea 7.
to glyoxylurea 6 was observed (Fig. 1e). Another control reaction
Because of favourable Franck–Condon factors and the strength of
between glyoxylate 4 and urea 7 confirms the equilibration of its triple bond, the cyanide radical is a likely initial product of high
these two entities with glyoxylurea 6 (Fig. 1f) that has already energy chemistry in C, H, O, and N containing atmospheres subject
been reported.18 When 2 eq. of cyanamide 1 were quickly added to electrical discharge or impact plasma.19 According to a possible
to a solution of glyoxylate 4, another product in addition to multistage scenario, some of this energy could have ended up stored
glyoxylurea 6 was formed. This additional product is tentatively in the form of the triple bond of cyanamide 1. However, absent
assigned as the 5-ureidoacylisourea 11, which could have catalysis, this triple bond energy is not kinetically accessible to drive
resulted from an attack of a second equivalent of cyanamide the phosphate activation chemistry necessary to produce RNA from
1. This observation perhaps explains why a slow addition of nucleotides. Complex enzymatic machinery of the sort biology now
cyanamide 1 is required to produce the 20,30-cyclic nucleotide 3 employs to produce and utilise activate phosphates would not have
in the highest yields (Scheme 3).
been available on the early Earth, but simple 2-oxoacid salts such as
The phosphate activation reaction described herein could glyoxylate 4 and pyruvate 5 could easily have been. Indeed, other
have taken place on early Earth if a stream containing cyanamide authors have described so much other potential chemistry of 4 that
1 were to have flowed slowly into a body of water containing the phrase ‘glyoxylate scenario’ has been coined.20–22 Further given
glyoxylate 4 and nucleotides such as 2. The enhanced catalysis in the simplicity and small size of glyoxylate 4, it is especially note-
the presence of magnesium or calcium ions is intriguing given worthy that it can function as such an effective catalyst for phosphate
that our proposed geochemical source of cyanamide 1 involves activation using cyanamide 1. Over time, the urea 7 that is produced,
heating of magnesium and/or calcium ferrocyanide to give the as cyanamide 1 is consumed by such activation chemistry,
corresponding cyanamide salt followed by carbonation to give a would hydrolyse to ammonia and carbon dioxide, gaseous products
solution of cyanamide 1 and magnesium and/or calcium cations. that would not therefore accumulate in the system. Clean,
To further explore this possible geochemical scenario, we repeated catalytic nucleotide activation chemistry to drive RNA synthesis via
the activation reactions of adenosine-30-phosphate 2 in presence of nucleoside-20,30-cyclic phosphates using energy persisting from violent
the 2-oxoacid salts 4 or 5 under optimal conditions (MgCl2 and earlier events might thus have been possible on the early Earth.
40 1C), but in this case cyanamide 1 (to 400 mM) was added
This work was supported by the Medical Research Council
continuously over 5 h using a syringe pump. The yields for these (No. MC_UP_A024_1009), and a grant from the Simons Foun-
last reactions are remarkable, since both glyoxylate 4 and pyruvate 5 dation (No. 290362 to J. D. S.).
catalysed the conversion of 2 very efficiently, affording the
product adenosine-20,30-cyclic phosphate 3 in 92% and 83%
yield respectively.
Conflicts of interest
There are no conflicts to declare.
Notes and references
‡ We use numbers to denote compounds regardless of their ionisation
state.
1 M. W. Powner, B. Gerland and J. D. Sutherland, Nature, 2009, 459,
239–242.
ˇ
2 J. Xu, M. Tsanakopoulou, C. J. Magnani, R. Szabla, J. E. Sponer,
ˇ
Scheme 3 Formation of product 11 when the reaction was performed
with fast addition of excess cyanamide 1 to glyoxylate 4.
´
J. Sponer, R. W. Gora and J. D. Sutherland, Nat. Chem., 2017, 9,
303–309.
This journal is ©The Royal Society of Chemistry 2017
Chem. Commun.

View Article Online
Communication
ChemComm
ˇ
3 S. Stairs, A. Nikmal, D.-K. Bucar, S.-L. Zheng, J. W. Szostak and M. W. 13 G. D. Cody, N. Z. Boctor, T. R. Filley, R. M. Hazen, J. H. Scott,
Powner, Nat. Commun., 2017, 8, 15270, DOI: 10.1038/ncomms
15270.
A. Sharma and H. S. Yoder Jr, Science, 2000, 289, 1337–1340.
14 M. I. Guzman and S. T. Martin, Astrobiology, 2009, 9, 833–842.
4 S. Becker, I. Thoma, A. Deutsch, T. Gehrke, P. Mayer, H. Zipse and 15 A. J. Coggins and M. W. Powner, Nat. Chem., 2017, 9, 310–317.
T. Carell, Science, 2016, 352, 833–836.
16 E. J. Gravenmade, G. D. Vogels and C. Van der Drift, Biochim.
Biophys. Acta, 1970, 198, 569–582.
17 F. R. Bowler, C. K. W. Chan, C. D. Duffy, B. Gerland, S. Islam, M. W.
Powner, J. D. Sutherland and J. Xu, Nat. Chem., 2013, 5, 383–389.
18 D. Ben-Ishai, J. Altman and Z. Bernstein, Tetrahedron, 1977, 33(10),
1191–1196.
5 G. B. Seifer, Russ. J. Inorg. Chem., 1962, 7, 640–643.
6 G. B. Seifer, Russ. J. Inorg. Chem., 1962, 7, 1187–1189.
7 B. H. Patel, C. Percivalle, D. J. Ritson, C. D. Duffy and J. D. Sutherland,
Nat. Chem., 2015, 7, 301–307.
8 R. Lohrmann and L. E. Orgel, Science, 1968, 161, 64–66.
´
´ ´
19 M. Ferus, P. Kubelık, A. Knıˇzek, A. Pastorek, J. Sutherland and
9 J. D. Ibanez, A. P. Kimball and J. Oro, Science, 1971, 173, 444–446.
ˇ
10 H. Griesser, P. Tremmel, E. Kervio, C. Pfeffer, U. E. Steiner and
S. Civis, Sci. Rep., 2017, 7, 6275, DOI: 10.1038/s41598-017-06489-1.
C. Richert, Angew. Chem., Int. Ed., 2017, 56, 1219–1223.
20 A. Eschenmoser, Tetrahedron, 2007, 63, 12821–12844.
11 F. S. Mohammed, K. Chen, M. Mojica, M. Conley, J. W. Napoline, 21 V. N. Sagi, V. Punna, F. Hu, G. Meher and R. Krishnamurthy, J. Am.
C. Butch, P. Pollet, R. Krishnamurthy and C. L. Liotta, Synlett, 2017,
93–97.
12 D. J. Ritson and J. D. Sutherland, Unpublished results.
Chem. Soc., 2012, 134, 3577–3589.
22 J. L. Andersen, C. Flamm, D. Merkle and P. F. Stadler, Isr. J. Chem.,
2015, 55, 919–933.
Chem. Commun.
This journal is ©The Royal Society of Chemistry 2017