A possible prebiotic synthesis of purine, adenine, cytosine, and 4(3H)-pyrimidinone from formamide: Implications for the origin of life

Article abstract of DOI:10.1016/S0968-0896(00)00340-0

The synthesis of prebiotic molecules is a major problem in chemical evolution as well as in any origin-of-life theory. We report here a plausible new prebiotic synthesis of naturally occurring purine and pyrimidine derivatives from formamide under catalytic conditions. In the presence of CaCO₃ and different inorganic oxides, namely silica, alumine, kaolin, and zeolite (Y type), neat formamide undergoes the formation of purine, adenine, cytosine, and 4(3H)-pyrimidinone, from acceptable to good yields. The role of catalysts showed to be not limited to the improvement of the yield but it is also relevant in providing a high selectivity in the products distribution.

Full text of DOI:10.1016/S0968-0896(00)00340-0

Bioorganic & Medicinal Chemistry 9 (2001) 1249±1253
A Possible Prebiotic Synthesis of Purine, Adenine, Cytosine,
and 4(3H)-Pyrimidinone from Formamide:
Implications for the Origin of Life
Ra€aele Saladino,a,* Claudia Crestini,b Giovanna Costanzo,c
Rodolfo Negrid and Ernesto Di Mauroc,d
a
Á
Dipartimento A.B.A.C., Universita della Tuscia, 01100 Viterbo, Italy
Á
b
Dipartimento di Scienze e Tecnologie Chimiche, Universita di Roma ``Tor Vergata'', 00133 Rome, Italy
cFondazione ``Istituto Pasteur, Fondazione Cenci Bolognetti'' c/o Dipartimento di Genetica e Biologia Molecolare,
Á
Universita di Roma ``La Sapienza'', 00185 Rome, Italy
dCentro di Studio per gli Acidi Nucleici, CNR, 00185 Rome, Italy
Received 3 November 2000; accepted 15 December 2000
AbstractÐThe synthesis of prebiotic molecules is a major problem in chemical evolution as well as in any origin-of-life theory. We
report here a plausible new prebiotic synthesis of naturally occurring purine and pyrimidine derivatives from formamide under
catalytic conditions. In the presence of CaCO3 and di€erent inorganic oxides, namely silica, alumine, kaolin, and zeolite (Y type),
neat formamide undergoes the formation of purine, adenine, cytosine, and 4(3H)-pyrimidinone, from acceptable to good yields.
The role of catalysts showed to be not limited to the improvement of the yield but it is also relevant in providing a high selectivity in
the products distribution. # 2001 Elsevier Science Ltd. All rights reserved.
Introduction
and in relatively low temperature. It was pointed out,6
that a major problem for the accumulation of the
nucleobases adenine (A), uracil (U), guanine (G), cyto-
sine (C), and thymine (T) on early Earth is in their rapid
rates of degradation at high temperatures, irrespective
of whether DNA or RNA were the original informa-
tional molecules. However, several considerations,
mainly based on the evidence indicating warm condi-
tions in the early evolution of the Earth, have led to
suggest that life originated under high-temperature
conditions (80±110 ꢀC or higher, up to 350 ꢀC).7À14
These considerations led to the conclusion that an ori-
gin of life based on a four-letter code and on the pre-
sent-day compounds would only be possible if it
occurred very rapidly (<100 yrs) at relatively low tem-
peratures, and in the presence of stabilizing micro-
environments. In addition to the condensation into
nucleobases, HCN hydrolyzes to formamide and then to
formic acid.15 Therefore, in a HCN-chemistry based
scenario, formamide could be a possible additional
down-the-line candidate for the synthesis of nucleo-
bases.5,16 Pioneering studies have shown that purine can
be synthesized from neat formamide when treated at
high temperature.17À19 Noteworthy, the introduction of
Any origin-of-life theory based on the evolution of a
self-replicating genetic system, that is accurate enough
to undergo Darwinian evolution, involves the use of
chemically stable compounds, and stabilizing micro-
environments. Stability is required to allow both low-
error template processes, and the accumulation of pre-
cursors in chemical conditions permitting polymeriza-
tion to oligomers. Several examples of synthetic
reactions producing prebiotic pregenetically-relevant
compounds have been described starting from CO, H2
and NH3 in the presence of Fe or Ni at 200±400 ꢀC,1
from ammonium cyanide,2 via cyanoacetylene by
3
sparkling mixtures of CH4+N2 and cyanoacetalde-
hyde.4 Hydrogen cyanate (HCN) chemistry provides a
preferential route for the prebiotic synthesis of purines
and pyrimidines.5 A set of conditions that would favour
the synthesis versus degradation has been described,6
essentially consisting in a high concentration of HCN
*Corresponding author. Tel.: +39-0761-357284; fax: +39-0761-
357242; e-mail: saladino@unitus.it
0968-0896/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00340-0

1250
R. Saladino et al. / Bioorg. Med. Chem. 9 (2001) 1249±1253
HCN in the reaction vessel gave adenine as major pro-
duct.19 13C NMR and 13C, 15N coupling NMR studies
of adenine prepared from doubly enriched potassium
cyanide, clearly indicated the presence of both mole-
solution are limited.24 The lowest temperature at which
formamide undergoes appreciable thermal decomposi-
tion is 180±190 ꢀC.24,25 Evaluation of the steady state
concentration of a formamide water solution at various
temperatures in prebiotic conditions is a procedure full
of uncertainty, because it would depend on which reac-
tion or on which combination of the many reactions
yielding formamide is considered, on the rate of hydro-
lysis (strictly dependent on pH), etc. In the ``drying-
lagoon'' model,4 a solution of formamide would easily
become highly concentrated upon heating, given its
solubility and non-volatility. In increasingly anhydrous
condition, hydrolysis of formamide becomes less rele-
vant. Thus, the most favorable set of conditions for the
synthesis would be high formamide concentration, the
presence of a catalytic system, a local stable micro-
environment, and a temperature higher than 100 ꢀC and
lower than 180 ꢀC. On the basis of these suggestions, all
experiments were performed with neat formamide 1
(5.7 g., 5 mL, 0.12 mol) at 160 ꢀC for 48 h in the presence
of selected inorganic catalysts (1.0% w/w). Besides low
amounts of unidenti®ed by-products (high temperature
reactions performed on a one carbon fragment yield
always very complex reaction mixtures) we focused our
attention on the identi®cation of the more abundant
purine and pyrimidine derivatives. The main reaction
products, purine 2, adenine 3, cytosine 4, and 4(3H)-
pyrimidinone 5, are shown in Figure 1. Their analysis
and identi®cation were performed by gas chromato-
graphy±mass spectrometry in comparison with samples
of authentic products (Table 1).26
cules of HCN and formamide in the adenine ring.20
A
new synthesis of purine and pyrimidine derivatives from
formamide under catalytic conditions would provide a
more ecient prebiotic route. Silica, alumina, and more
complicated metallic oxides such as perovskites, spinels,
clays and zeolites, were present in the early Earth. These
inorganic oxides have already shown to be of value in a
large range of abiotic catalytic transformations due to
acidic or basic properties, to cation-exchange capacity,
and to the possibility to accommodate copious quan-
tities of water or other polar molecules.21 In the latter
case, a network of cages and channels might provide a
local microenvironment to stabilize and template the
new formed prebiotic molecules into oligomers.22 As an
example, oligomers of glycine have been formed in the
presence of clays under experimental conditions simu-
lating natural prebiotic processes.23 Moreover, certain
purines and pyrimidines (such as thymine), which are
not normally intercalated into clays from aqueous solu-
tion, are taken up in the presence of adenine, with which
they may pair in the interlamellar region.21 In an e€ort
to investigate a possible role of metallic oxides on the
prebiotic synthesis of nucleic bases from formamide we
treated neat formamide in the presence of CaCO3, silica,
alumina, kaolin, zeolite (Y type) as representative
prebiotic models of heterogeneous catalysts.
Results
The temperature upper limit for the synthesis of
nucleobases from formamide is set by its boiling point
(211 ꢀC). Azeotropic e€ects for a 5±95% formamide
Table 2. Condensation of formamide
Entry Catalysta Temperature (ꢀC) Product (s)b Yield (mg/g)c
1
2
3
4
5
6
7
No catalyst
CaCO3
Kaolin
160
160
160
160
160
160
100
2
2
34.1
214.9
2, 3, 4
2, 3, 4
51.3, 0.6, 2.0
82.5, 0.6, 4.4
2, 3, 4, 5 39.7, 0.7, 1.4, 2.0
Zeolite
Alumina
Silica
Kaolin
2, 3, 4, 5
2
4.0, 0.9, 4.2, 1.5
3.0
aReactions were performed in the presence of 1.0% w/w of catalyst.
bProducts were identi®ed by comparison of their retention times and
mass spectra with those of authentic samples.
cquantitative evaluation was performed by capillary gas-chromato-
graphic analysis using a HP 5890III gas chromatograph with FID
detector equipped with a column SP-2380, using a temperature pro-
gramme of 100±250 ꢀC per min with He as carrier gas. 6-Metho-
xypurine was used as internal standard. Because of the uncertainty of
the number of formamide molecules involved in the synthesis of
recovered products the yield were calculated as mg of product formed
for gram of formamide.
Figure 1.
Table 1. Selected mass spectrometric dataa
Product
MS (m/z) data (%)
2
3
4
5
120 (M, 100), 93 (M-HCN, 37), 66 (M-2ÂHCN, 19)
135 (M, 100), 108 (M-HCN, 38), 81 (M-2ÂHCN, 26), 66 (M-69, 35), 54 (M-3ÂHCN)
111 (M, 100), 95 (M-NH2, 20), 83 (M-CO, 32), 69 (M-NCO, 45), 68 (M-HNCO, 35), 41 (M-HNCO-HCN, 58)
96 (M, 100), 69 (M-HCN, 25), 68 (M-CO, 35), 54 (M-NCO, 47), 53 (M-HNCO, 34)
aMass spectroscopy was performed with Hewlett-Packard 5971 mass-selective detector on a Hewlett-Packard 5890III gas chromatograph with FID
detector.

R. Saladino et al. / Bioorg. Med. Chem. 9 (2001) 1249±1253
1251
In the absence of any catalyst, purine 2 was obtained in
low yield as the only recovered product (Table 2,
entry 1). When the reaction was performed in the pre-
sence of CaCO3, we obtained a high increase in the yield
of 2 (Table 2, entry 2). In this case other purine and
pyrimidine derivatives were not obtained in appreciable
yield. A di€erent behavior was found in the presence of
kaolin, zeolite, alumina, and silica. In the presence of
kaolin and zeolite, an increase in the yield of 2 was
again observed, in addition to the presence of adenine 3
and cytosine 4 (Table 2, entries 3 and 4). In particular,
with zeolite, that is characterized by a micropourus
structure as a favourable microenvironment and dehy-
drating system, the yield of purine 2 was found second
only to CaCO3. The presence of adenine in the reaction
mixtures suggests that the dehydration of formamide to
HCN, an essential step in the synthesis of 3, is an
operative process in the latter two cases. Moreover, it is
reasonable that HCN might be involved also in the for-
mation of 4. In the presence of alumina there is not an
appreciable increase in the yield of 2. In this case,
besides the previously observed products 3 and 4,
4(3H)-pyrimidinone 5 was also obtained in the reaction
mixture (Table 2, entry 5). To the best of our knowledge
there are no reports in the literature dealing with the
direct synthesis of cytosine 4 and 4(3H)-pyrimidinone 5
from formamide. Formamide is widely used in the pri-
mary synthesis of pyrimidine derivatives involving for-
mation of three bonds with others reagents, for example
the consensation of two molecules of formamide with
ketone derivatives yielding 4-substituted pyrimidines via
N-formylformamidine as reactive intermediates.27,28
ably because of a less ecient generation of HCN.
Independently from the speci®c reaction conditions,
these results show that formamide is a plausible
prebiotic substrate for the synthesis of purine and
pyrmidine derivatives and that a major role could
have been played by inorganic catalysts in improving
the yield and changing the distribution of reaction
products.
Conclusions
The mixtures of the organic components of nucleic acids
that are formed from single molecules such as ammonia,
formaldehyde and HCN are usually very complex,5,32,33
so that the selection of particular derivatives is hard to
understand. Moreover, the rapid rates of degradation of
purine and pyrimidine derivatives at high temperature is
a relevant problem. Accordingly, it has been assumed
that the presence of inorganic components with layer or
porous cavity structure could provide the occurrence of
a favourable local microenvironment. This could intro-
duce a selectivity in the product distribution and possi-
bly increase the thermal stability of nucleobases through
molecular recognition processes or simply by their iso-
lation from the external environmental conditions. We
have hypothesized that formamide condensation per-
formed in the presence of inorganic oxides can provide a
framework for one plausible pathway of prebiotic
synthesis of purine and pyrimidine derivatives. This
hypothesis satis®es some of the accepted prerequisites
for prebiotic chemical evolution. Among the examples
studied, the reaction performed in the presence of
CaCO3 gave purine as the only recovered product in a
yield that is very high for a prebiotic synthesis. Alu-
mina, kaolin, and zeolite also catalyzed the formation of
purine, even if adenine, cytosine, and the previously not
reported 4(3H)-pyrimidinone were obtained in accep-
table yields. A di€erent reaction pattern was observed
with silica, in which case pyrimidine derivatives became
the main reaction products. Thus, a wide range of
selectivity and yield could be obtained depending on the
catalyst used. No matter how plausible and ecient the
described syntheses are (especially so for their qualita-
tive aspects), the problem remains of the measured low
stability of the nucleobases at temperatures of 100 ꢀC or
higher.6 Recently, we have reported that above 100 ꢀC
in a formamide solution, both purines and pyrimidines
can be degraded back to low molecular weight
fragments among which formamide. Adenine, for
example, undergoes nucleophilic opening of the imida-
zole ring producing pyrimidine derivatives such as, 6-
amino-5-formamido-4-[N-(20-deoxy-b-d-ribofuranosyl)]-
pyrimidine, 4,6-diamino-5-formamido pyrimidine, and
2,6-diamino-4-oxo-5-formamido pyrimidine.34,35 Con-
sequently, formamide, a product of degradation of
nucleobases, represents a substrate for their synthesis,
suggesting a prebiotic synthesis with possible elements
of cyclicity. The problem can also be solved invoking
the possibility that guanine might be produced under
di€erent reaction conditions, that the GC base-pair was
not used in the ®rst genetic material, or that life arose
quickly, or that a di€erent letter code was initially used.
Prebiotic syntheses of pyrimidines include ultraviolet
light induced dehydrogenation of dihydrouracil on
silica,29 the formation of orotic acid from cyanide poly-
merization,30 and the synthesis of cytosine and uracil
by condensation of cyanoacetaldehyde and urea.4,31
We have not studied in detail the mechanism of for-
mation of compounds 4 and 5 under our experimental
conditions.
It is interesting to note that 2-cyanoacetamide, a plau-
sible three carbon atom precursor of pyrimidine deriva-
tives, was identi®ed by mass-spectroscopy [m/z=84
(M+, 30%)] and by comparison with an authentic sam-
ple (data not shown). A di€erent reaction pathway was
observed in the presence of silica under analogous
experimental conditions. In this case, the yield of purine
2 was found to be lower with respect to the reaction
performed in the absence of catalyst, while cytosine 4
and 4(3H)-pyrimidinone 5 were recovered as the main
reaction products (Table 2, entry 6). Clearly, in the
presence of silica the formation of pyrimidine deriva-
tives is favoured. In part, a similar e€ect was obtained
also in the presence of alumina. In a selected repre-
sentative case (kaolin) the reaction was performed also
at 100 ꢀC to study the e€ect of the temperature on the
reaction pattern. Under these experimental conditions,
purine 2 was obtained in low yield as the only recovered
product (Table 2, entry 7). In this case the reaction
temperature appears to be a crucial factor for the
synthesis of adenine and pyrimidine derivatives, prob-

1252
R. Saladino et al. / Bioorg. Med. Chem. 9 (2001) 1249±1253
An additional solution could be provided by an as-yet-
undescribed, possibly catalyzed, competitively rapid
condensation of the nucleobases into more stable,
higher order molecular structures. In the absence of
experimentally veri®ed solutions for this bias, the for-
mamide-based catalyzed synthesis of nucleobases
described here should be considered only as a chemically
attractive possibility for prebiotic processes.
(CH), 145.10(CH), 147.02 (CH), 163.33 (C);%]; nmax
(Nujol) 3460, 3300, 2890, 1680, 1600, 1475 cmÀ1. Mass
spectrometry data are reported in Table 2.
Acknowledgements
This work was supported by a grant from the Italian
MURST.
Experimental
Mass Spectroscopy (MS) was performed with Hewlett-
Packard 5971 mass-selective detector on a Hewlett-
Packard 5890III gas chromatograph with FID detector.
NMR spectra were recorded on a Bruker (200 MHz)
spectrometer and are reported in d values. Infrared
spectra were recorded on a Perkin Elmer 298 spectro-
photometer using NaCl plates. All solvents were ACS
reagent grade and were redistilled and dried according
to standard procedures. Formamide (>99%, Fluka),
Kaolin, calcium carbonate, zeolite (Y type), silica, alu-
mina, and 6-methoxypurine (Aldrich) were used without
further puri®cation.
References
1. Hayatsu, R.; Anders, E. Top. Curr. Chem. 1981, 99, 1.
2. Oro, J.; Kimball, A. P. Arch. Biochem. Biophys. 1961, 94,
221.
3. Sanchez, R. A.; Ferris, J. P.; Orgel, L. E. J. Mol. Biol. 1967,
30, 223.
4. Robertson, M. P.; Miller, S. L. Nature 1995, 375, 772.
5. Miller, S. L. Cold Spring Harbor Symposia on Quantitative
Biology 1987, LII, 17.
6. Levy, M.; Miller, S. L. Proc. Natl. Acad. Sci. U.S.A. 1998,
95, 7933.
7. Pace, N. R. Cell 1991, 65, 531.
8. Schwartzman, D.; McMenamin, M.; Volk, T. Bioscience
1993, 43, 390.
9. Russel, M. J.; Daniel, R. M.; Hall, A. J.; Sherringham, J. A.
J. Molec. Evol. 1994, 39, 231.
Formamide condensation. General procedure
Formamide (5.7 g., 5 mL, 0.12 mol) in the presence of
the catalyst (1.0% w/w; calcium carbonate, kaolin, zeo-
lite Y type, and silica) was heated at 160 ꢀC for 48 h. The
reaction in the presence of kaolin was also performed at
100 ꢀC. The reaction mixture was ®ltered to remove the
catalyst and evaporated under high vacuum. Gas-chro-
matography and gas chromatography±mass spectro-
metry of the reaction products were performed using a
10. Kasting, J. F.; Ackerman, T. P. Science 1986, 234, 1383.
11. Kasting, J. F. Science 1993, 259, 920.
12. White, R. H. Nature 1984, 310, 430.
13. Miller, S. L.; Bada, J. L. Nature 1988, 334, 609.
14. Bada, J. L.; Miller, S. L.; Zhao, M. Origins Life Evol.
Biosphere 1995, 25, 111.
15. (a) Codrington, J. C. J. Chem. Soc. 1907, 91, 593. (b)
Sanchez, R.; Ferres, J.; Orgel, L. E. Science 1966, 153, 72. (c)
Miller, S. L. Cold Spring Harbor Symposia on Quantitative
Biology 1987, LII, 17.
16. Miller, S. L. J. Chem. Soc. 1955, 77, 2351.
17. Bredereck, H.; Gompper, R.; Schuh, H. G. V.; Theilig, G.
Angew. Chem. 1959, 71, 753.
SP-2380column (3m0
Â0.25 mm and 0.25 mm ®lm
thickness), and an isothermal temperature pro®le of
100 ꢀC for the ®rst 2 min, followed by a 10 ꢀC/min tem-
perature gradient to 250 ꢀC and ®nally an isothermal
period at 250 ꢀC for 30min. The injector temperature
was 280 ꢀC. Chromatography grade helium was used as
the carrier gas. The fragmentation patterns were com-
pared to those of authentic samples. Selected mass
spectrometry data are reported in Table 1. 6-Methoxy-
purine (from Aldrich) was used as internal standard.
Because of the uncertainty of the number of formamide
molecules involved in the synthesis of recovered pro-
ducts the yields were calculated as mg of product
formed for gram of formamide and are reported in
Table 2. In a selected case (kaolin) the reaction crude
was dissolved in chloroform:methanol (9.0:1.0) and
puri®ed by ¯ash-chromatography. The structures of
products were determined by usual spectroscopic tech-
niques (1H NMR, 13C NMR, FT-IR) and mass-spec-
troscopy (EI) and con®rmed by comparison with
authentic commercial samples. Data of 4(3H)-pyr-
imidinone 5 are reported as a selected case: mp 166±
18. Yamada, H.; Okamoto, T. Chem. Pharm. Bull. 1972, 20,
623.
19. Yamada, H.; Hirobe, M.; Higashiyama, K.; Takahashi,
H.; Suzuki, K. Tetrahedron Lett. 1978, 42, 4039.
20. Yamada, H.; Hirobe, M.; Higashiyama, K.; Takahashi,
H.; Suzuki, K.T. J. Am. Chem. Soc. 1978, 100.
21. (a) Thomas, J.M.; Thomas, W.J. In Principles and Prac-
tice of Heterogeneous Catalysis; VCH: Weinheim, 1997; pp
339±365, and 626±627. (b) Strasak, M.; Sitek, J.; Sersen, F.;
Janik, R.; Slosiarikova, H. Proc. Conf. Coord. Chem. 1991,
279. (c) Ristori, G. G.; Fusi, P. Environ. Impact Soil Compon.
Interact 1995, 1, 337.
22. Laszlo, P. Science 1987, 235, 1473.
23. Lahav, N.; White, D.; Chang, S. Science 1978, 201, 67.
24. Formic acid and derivatives (Formamide) In ECT Kirk-
Othmer Encyclopedia of Chemical Technology, Wiley Inter-
science, Ed., 1978, Vol. 11, pp 258±262.
25. Kirkpatrick, E.C. (to E.I. Dupont de Nemours & Co. Inc)
U.S. Patent, 2, 346,425, 1944.
26. (a) Goto, T.; Tatematsu, A.; Matsuura, S. J. Org. Chem.
1965, 30, 1844. (b) Tatematsu, A.; Goto, T.; Matsuura, S.
Nippon Kagaku Zasshi 1966, 87, 71. (c) Rice, J. M.; Dudek,
G. O.; Barber, M. J. Am. Chem. Soc. 1965, 87, 4569.
169 ꢀC; H NMR (DMSO-d6) d 6.35 (d, 1H, J=5.0Hz,
1
CH), 7.95 (d, 1H, J=5.0Hz, CH), 8.20 (s, 1H, CH),
12.8 (br. s, 1H, NH); 13C NMR (DMSO-d6) d 118.45

R. Saladino et al. / Bioorg. Med. Chem. 9 (2001) 1249±1253
1253
27. Davies, W.H.; Piggott, H.A. J. Chem. Soc. 1945, 347.
28. Bredereck, H.; Gompper, R.; Geiger, B. Chem. Ber. 1960,
1402.
29. Chittenden, G. J. F.; Schwartz, A. W. Nature 1976, 263,
350.
32. Orgel, L. E. Cold Spring Harbor Symposia on Quantitative
Biology 1987, LII, 9.
33. Ferris, J. P. Cold Spring Harbor Symposia on Quantitative
Biology 1987, LII, 29.
34. Saladino, R.; Mincione, E.; Crestini, C.; Negri, R.; Di
Mauro, E.; Costanzo, G. J. Am. Chem. Soc. 1996, 118,
5615.
30. Ferris, J. P.; Joshi, D. C.; Edelson, F. H.; Lawless, J. G. J.
Molec. Evol. 1978, 11, 293.
31. Ferris, J. P.; Zamek, O. S.; Altbuch, A. M.; Freiman, H. J.
J. Molec. Evol. 1974, 3, 301.
35. Saladino, R.; Crestini, C.; Mincione, E.; Costanzo, G.; Di
Mauro, E.; Negri, R. Bioorg. Med. Chem. 1997, 5, 2041.