Silica Metal Oxide Vesicles Catalyze Comprehensive Prebiotic Chemistry

Article abstract of DOI:10.1002/chem.201706162

It has recently been demonstrated that mineral self-assembled structures catalyzing prebiotic chemical reactions may form in natural waters derived from serpentinization, a geological process widespread in the early stages of Earth-like planets. We have s

Full text of DOI:10.1002/chem.201706162

A Journal of
Accepted Article
Title: Silica metal-oxide vesicles catalyze comprehensive prebiotic
chemistry
Authors: Bruno Mattia Bizzarri, Lorenzo Botta, Maritza Iveth Pérez-
Valverde, Raffaele Saladino, Ernesto Di Mauro, and Juan
Manuel Garcia Ruiz
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201706162
Link to VoR: http://dx.doi.org/10.1002/chem.201706162
Supported by

10.1002/chem.201706162
Chemistry - A European Journal
FULL PAPER
Silica metal-oxide vesicles catalyze comprehensive prebiotic
chemistry
Bruno Mattia Bizzarri,1 Lorenzo Botta,1 Maritza Iveth Pérez-Valverde,2 Raffaele Saladino,1,*, Ernesto
Di Mauro,1 Juan Manuel García-Ruiz2,*
1Ecological and Biological Sciences Department (DEB), University of Tuscia, Via S. Camillo de Lellis
snc, 01100, Viterbo, Italy.2Laboratorio de Estudios Cristalográficos, Instituto Andaluz de Ciencias de la
Tierra, Consejo Superior de Investigaciones Científicas–Universidad de Granada, Avenida de las
Palmeras 4, Armilla, Granada 18100, Spain.
Abstract: It has recently been demonstrated that mineral self-
assembled structures catalyzing prebiotic chemical reactions may
form in natural waters derived from serpentinization, a geological
process widespread in the early stages of Earth-like planets. We have
synthesized self-assembled membranes by mixing microdrops of
metal solutions with alkaline silicate solutions in the presence of
formamide (NH2CHO), a single carbon molecule, at 80ºC. We found
that these bilayer membranes, made of amorphous silica and metal
oxide-hydroxide nanocrystals, catalyze the condensation of
formamide, yielding the four nucleobases of RNA, three aminoacids
and several carboxylic acids in a single pot experiment. Besides
manganese, iron and magnesium, two abundant elements in the
earliest Earth crust that are key in serpentinization reactions, are
enough to produce all these biochemical compounds. These results
suggest that the transition from inorganic geochemistry to prebiotic
organic chemistry is common on a universal scale and, most probably,
earlier than ever thought for our planet.
efficient reaction providing these molecular compounds and the
proper physico-chemical environment is the interaction of water
with olivine minerals, the so-called serpentinization reaction.[15, 16]
Serpentinization, along with the Sabatier reaction and Fischer-
Tropsch-type reactions, is the most productive chemical scenario
for the synthesis of organic molecules. It is a scenario in which
mineral/organic chemistry works today in few specific geological
sites but it is thought to have been working on a global scale
during the early Earth at Hadean times. It is also a scenario where
that singular chemistry plays and has been playing beyond the
Earth, at least in Earth-like planets, moons, and interstellar dust,
where formamide is fully available. Interestingly, serpentinization
environments may form alkaline waters rich in silica. It has been
experimentally demonstrated that these alkaline waters induce
the growth of self-assembled mineral structures,[17] including
silica/carbonate biomorphs mimicking the morphology of primitive
organisms, calcium carbonate mesocrystals similar to those
formed in biominerals, and silica/oxy-hydroxide metal bilayer
membranes, also known as silica gardens, with interesting
catalytic properties.[18-20] We have synthesized these mineral
membranes in alkaline silica-rich solutions similar to those
deriving from serpentinization reactions in the presence of
formamide at 80ºC. Rather than pellets of salts used in classical
silica garden experiments, we used microdrops of solution of
different soluble metal salts, thus mimicking a more realistic
geochemical environment. We show that the four nucleobases of
RNA, three aminoacids and several carboxylic acids relevant to
key metabolic cycles, can be obtained in a robust and simple one-
pot reaction even in the absence of radiation.
Introduction
It has long been thought that minerals have a significant catalytic
role in the synthesis of life’s building blocks.[1-3] Pioneering work
successfully demonstrated the formation of aminoacids,
nucleobases and other molecular bricks of life from simple organic
molecules, with hydrogen cyanide (HCN) being the precursor
compound most used in prebiotic chemistry experiments so far.[4-
6] Among the alternatives to HCN that meet prebiotic requirements
of stability and reactivity, formamide (NH2CHO), a single-carbon
organic molecule, has been proven to be very successful both in
terms of the variety of relevant molecules synthesized and in yield.
For instance, formamide has been demonstrated to condense into
Results and Discussion
carboxylic acids, several aminoacids, as well as into purines and
8]
pyrimidines.[7,
The condensation reaction is catalyzed by
The experiments to explore the formation of membranes using
microdrops were performed with solutions of FeCl2, FeCl3 and
MnCl2. The formation of membranes was observed at pH as low
as 10.4 and silica concentration of 0.75 mol/L (Figure 1a). Upon
mixing the microdrops of the salt solution with the silicate solution,
a metal-silica membrane is formed surrounding the drop. The
initial microdrops expand due to osmotic forces until reaching a
critical size beyond which they break, releasing the inner metal-
rich acidic content into the alkaline silicate solution (Figure 1b, and
video S1). The jet of solution arising from the breaking point
creates a new tubular structure that grows into the silicate solution
temperature, by several minerals, and by high-energy proton
irradiation.[9] Formamide is present in a variety of star-forming
environments,[10, 11] and it has also been found on several comets
of the solar system.[12, 13] In fact, formamide appears to be a critical
intermediary in Miller-type reactions.[14]
No matter the route, whether HCN or NH2CHO, in addition to
minerals, the geological niche(s) for prebiotic chemistry needs to
be fed with molecular hydrogen, carbon and nitrogen, within a
reduced or anoxic environment where the formation of stable
organic molecules can thrive towards complexity. The most
This article is protected by copyright. All rights reserved.

10.1002/chem.201706162
Chemistry - A European Journal
FULL PAPER
eventually breaking and forming new tubular membranes. The
complexity of the tubular structure and the degree of breaking
depends on the concentration and viscosity of salt and silicate
solutions. The lower the pH and the silica concentration the larger
the degree of disruption of the membrane. This behavior, which
was observed for all the tested salts, ensures that formamide,
initially located in the alkaline silica solution, enters also in contact
with the metal-rich solution during the formation of the silica/metal
membrane. This allows the catalysis of the condensation of
formamide by the inner side of the membrane, made of metal-
oxide-hydroxide nanoparticles, and by the outer part of the
membrane made of porous silica.[21,22] The growth of the
structures is visible for several minutes but the chemical reaction
is known to last for several hours.[18] Therefore, unlike previous
experiments in which passive membranes were used, the
condensation of formamide occurs while the reaction leading to
the formation of the membranes is active.
trimethylsilyl ethers (TMS). GC-MS chromatograms are reported
in SI#2. The analysis was limited to products ≥1 ng/mL, and the
yield was calculated as milligrams of product per milliliter of
starting NH2CHO. The results of the experiments are shown in
Figure 2 and Table 1. The reaction of formamide (10% v/v) with
sodium silicate solution (pH 12) without inorganic membranes
afforded only pyruvic acid 13, lactic acid 14, guanidine 22, and
urea 23, in traces or in very low amount (micrograms).[20]
As a general trend, we observed the formation of a large panel of
compounds of biological relevance, including nucleobases and
their analogues, carboxylic acids, amino acids and low molecular
weight intermediates (Scheme 1). The values and the peak
abundances of products and the original m/z fragmentation
spectra are in SI #3 and SI # 4, respectively.
Figure 1: A) Screening of experimental conditions yielding metal-silica
membranes. The red dot indicates the selected conditions for formamide
condensation experiments; B: Frames of a time sequence of Fe+2 and Fe+3
membranes from video S1 and video S2 (SI#1) recorded in the absence of
formamide. Video S4 and video S5 describing the Fe+2 and Fe+3 membranes
growth in the presence of formamide are in SI#1. Scale bar 250 microns; C) X-
ray diffraction patterns of Fe2+, Fe3+, and Mn2+ membranes made with pellets
and drops of the corresponding metals chlorides in the absence of formamide.
Figure 2. Schematic representation of products obtained during the thermal
condensation of formamide and water in the presence of mineral vesicles. The
products are classified on the basis of their biological role: nucleobases and
heterocycles, acids, amino acids and condensing agents.
Accordingly, to optimize the catalysis in the formamide
experiments we selected low silica concentration (1.65 mol/L) and
pH (11.5). The condensation experiments were performed by
pouring drops of the saturated solutions of the appropriate metal
salt [ZnCl2, FeCl2·4H2O, CuCl2·2H2O, MnCl2 Fe2(SO4)3·9H2O,
Cu(NO3)2 and MgSO4] to sodium silicate containing 10% (v/v) of
NH2CHO at room temperature. At the end of the reaction the gel
and the membrane were mechanically ground with a spatula in
the presence of formamide (500 l) to extract the products. Then
the residue was filtered washing again with formamide (500 l),
and the solution distilled. The experiments were reproduced three
times. The products were analyzed by gas chromatography–
mass spectrometry (GC-MS) after formation of the corresponding
Nucleobases. The complete set of nucleobases of RNA [adenine
(2), uracil (3), cytosine (4), and guanine (6)] and some nucleobase
analogues isocytosine (5), 4(3H)-pyrimidinone (7) and
hypoxanthine (12), which are of relevance in the context of the
origin of life and of chemiomimesis,[23, 24, 25] were obtained (Table
1). Purine (11), 6-hydroxy-2,4-diamino pyrimidine (8), 5-carboxy-
2,4-diaminopyrimidine (9) and 3,5-diamino-1,2,4-triazole (10)
were also detected in appreciable amount. The mechanism of
formation of nucleobases from NH2CHO is reviewed,[26] guanidine
(22), urea (23) and diamino malonitrile (24) being well-known key
This article is protected by copyright. All rights reserved.

10.1002/chem.201706162
Chemistry - A European Journal
FULL PAPER
intermediates in the synthesis of both purine and pyrimidine
derivatives.
under thermal conditions alone. According to data previously
described[7], adenine 2 and guanine 6 are produced from
NH2CHO by the same reaction pathway. This includes the initial
formation of the HCN tetramer diaminomaleonitrile (DAMN), the
cyclization to aminoimidazole carbonitrile (AICN) and the closure
of the pyrimidine ring (by further addition of NH2CHO and
ammonia). The fact that different catalysts yield the same
products hints at the robustness of formamide chemistry.
Carboxylic acids are key intermediates of numerous processes
and metabolic cycles required in the cell for the production of
energy and for the biosynthesis of primary and secondary
metabolites. Mineral vesicles afforded six carboxylic acids
including pyruvic acid (13), lactic acid (14), oxalic acid (15),
succinic acid (16), oxaloacetic acid (17) and parabanic acid (18)
(Scheme 1, Table 1). In terms of biological relevance, oxalic acid
and lactic acid are involved in the glyoxylate cycle,[29] in the Cori
cycle and in gluconeogenesis,[30] whereas pyruvic acid, succinic
acid and oxaloacetic acid are intermediates of the citric acid cycle
[tricarboxylic acid cycle or Krebs cycle (KC)], one of the most
ancient processes in cell metabolism. The KC is, in its reversed
Table 1. Products obtained by thermal condensation of NH2CHO in the
presence of active silica/metal-oxide-hydroxide membranes a
Productb,c ZnCl2
FeCl2
CuCl2 MnCl2 Fe2(SO4)3 MgSO4
Cu(NO3)2
Adenine
2
Uracil
3
1.20
(0.01)
0.27
(0.22)
1.69
(0.18)
8.63
(0.11)
0.62
(0.01)
4.35
(0.23)
4.9
(0.15)
0.55
(-)
0.43
(trace)
0.09
(-)
0.5
(-)
-
-
-
-
-
-
3.85
(3.8)
-
Cytosine
Traces
(-)
Traces
(-)
Traces
(-)
-
-
4
Isocytosine
0.06
(-)
0.05
(-)
-
-
5
Guanine
0.02
-
-
-
-
-
1.12
0.02
0.03
0.04
6
4(3H)-Pyr
7
6(OH)-2,4-
DAP 8
traces
(-)
1.33
(0.3)
0.83
3.5
(traces)
7.45
(trace)
0.27
(0.05)
-
0.12
-
-
-
(0.06)
3.4
2,4-DAP-
5COOH 9
6.7x10-3
(-)
3.65
(-)
0.45
(traces)
3.08
(0.14)
-
-
-
(traces)
1.13
Triazole
10
0.23
(-)
0.05
(-)
-
(-)
0.39
(-)
-
Purine
11
Hypoxantine
12
4.65
(-)
0.01
(-)
0.02
(-)
0.38
(-)
0.05
(-)
0.75
(-)
Traces
(-)
-
-
-
-
-
1.68
-
version, a prebiotic device for the production of energy and
Pyruvic ac.
13
Lactic ac.
14
0.01
(0.01)
0.81
(0.63)
0.21
2.60
(-)
0.08
(0.07)
1.65
(0.21)
0.85
(0.38)
0.23
(0.21)
0.91
(0.71)
0.37
(0.28)
0.01
(-)
2.75
(-)
1.65
(0.18)
0.25
(-)
3.5x10-4
traces
[31]
organics in the carbon-oxide-rich primitive atmosphere.
The
0.14
(0.15)
2.85
(-)
formation of carboxylic acids 13–17 has been previously
explained by the oligomerization of DAMN (actually detected in
the reaction mixture), followed by hydrolysis and successive
redox processes.[32] Parabanic acid 18 is a product of the
degradation of both purine and pyrimidine nucleobases. [33, 34] All
salts were effective catalysts for the production of carboxylic acids.
In addition to CuCl2, Fe2(SO4)3·9H2O and Cu(NO3)2 were the best
catalysts.
Amino acids. The formation of three amino acids: glycine (20),
alanine (21) and N-formyl glycine (19) was observed. Amino acids
are probably synthesized through a Strecker-like mechanism
(Strecker-cyanohydrin), in accordance with the recent theoretical
observations on the key role of NH2CHO in the Miller-Urey
synthesis of amino acids (14). N-formylglycine 19 is produced
from 20 by a NH2CHO based formylation process involving “in situ”
generated carbodiimide (not isolated in this case), guanidine 22
and urea 23.[35] Aminoacids were synthesized (with the sole
exception of alanine) only in the presence of Fe2(SO4)3·9H2O,
Cu(NO3)2 and MgSO4, the iron salt being the most reactive
catalyst.
These results show the higher reactivity of the silica/metal
oxyhydroxide membranes of the mineral vesicles when compared
with the data previously obtained using preformed metal silicate
membranes prepared with classical silica garden experiments. [20]
For an equal amount of salt, both the yield and the variety of
synthesized products are enhanced: guanine, hypoxanthine,
oxaloacetic acid and alanine have been obtained using only active
mineral vesicles (Table 1). In classical chemical garden
experiments, previously grown membranes were added to the
-
-
Oxalic ac.
15
0.05
0.32
(0.10)
1.15
(0.16)
3.05
(-)
0.5
(-)
0.31
(-)
1.19
(-)
0.50
(0.12)
0.95
(2.8 x10-3
)
(0.18)
Succinic ac. 8.6 x10-3
-
16
(-)
(0.24)
Oxaloacetic
2.3
(-)
-
0.16
0.22
17
Parabanic
0.03
(-)
1.35
(-)
0.17
(-)
0.05
(-)
0.95
(0.9)
-
-
18
N-Formylgly
19
Glycine
20
Alanine
21
Traces
(-)
1.67
(9.0x10-3
0.62
(0.57)
0.03
0.12
(-)
0.04
(-)
Traces
(-)
-
-
-
-
-
1.21(-)
0.68 (0.53)
0.02 (-)
)
Traces
(-)
-
-
0.07
(- )
traces
(-)
-
(-)
0.03
Guanidine
0.04
(-)
0.03
(-)
0.8
(0.67)
(5.2 x10-
-
-
-
22
3
d
)
0.21
(5.0
0.58
Urea
23
7.4 x10-3
(2.0 x10-3
2.5
(2.6)
0.18
(-)
-
-
-
-
)
x10-3
)
(0.02)
traces
(-)
DAMN
24
Glycol
aldehyde
Dimer 25
1.125
(-)
Traces
(-)
0.04
(-)
-
-
2.6 x10-3
(-)
0.02
(-)
traces
(-)
Traces
(-)
Traces
(-)
-
a
The yield is reported as amount (mg) of isolated products. The data are the
mean values of three experiments with standard deviations of less than 0.1%. b
Values in parentheses refer to results obtained in the presence of preformed
metal silicate membranes under similar experimental conditions. c The reaction
of formamide (10% v/v) with sodium silicate solution (pH 12) without inorganic
membranes afforded only pyruvic acid 13, lactic acid 14, guanidine 22, and urea
23, in traces or in very low amount (micrograms).[20]
ZnCl2 and MnCl2 provided only guanine 6, and CuCl2 only
uracil 3. However, Fe2(SO4)3·9H2O, MgSO4 and CuN2O6 afforded
the complete panel of natural nucleobases 2-4 and 6. As for the
order of reactivity, MgSO4 afforded the highest total amount of
nucleobases (9.9 mg), followed by Fe2(SO4)3·9H2O (3.18 mg) and
Cu(NO3)2 (1.06 mg). Remarkably, the four nucleobases are here
obtained in a one-pot reaction and in a synthetic set-up fed only
by thermal energy. This observation is unprecedented.
acidic and alkaline solutions mimicking the inner and outer
[20]
solutions.
Different panels of products were observed for the
experiments, with carboxylic acids prevailing in the experiment
with tubular structures within the outside solutions, while
nucleobases prevailed for the inside solutions. During the
formation of the mineral vesicles, NH2CHO experiences
Previous studies have reported the prebiotic synthesis of guanine
[9, 27, 28] using UV or proton irradiated solution, but this is the first
time for obtaining the complete set of four RNA nucleobases
This article is protected by copyright. All rights reserved.

10.1002/chem.201706162
Chemistry - A European Journal
FULL PAPER
successively the two different reaction environments, i.e., the
internal acidic metal-rich and the external alkaline silicate-rich
solution, the reaction products of the inner part of the membrane
being released in the bulk of the solution upon the breaking of the
membrane. Moreover, NH2CHO condensation occurs during the
self-assembles, rather than being local and bizarre, appear to be
universal and geologically conventional.
Experimental Section
Preparation of silica-metal oxide mineral vesicles with
microdrops: The synthesis of mineral vesicles using drops
instead of solid pellets of the acidic salt has been investigated by
a) pouring, and b) injecting drops of saturated solutions of FeCl3
(>97% Sigma-Aldrich) pH 0.68(3), FeCl2 (≥ 99% Sigma-Aldrich)
pH 1.5 and MnCl2.4H2O (ACS reagent >98% Sigma-Aldrich) pH
2.06(5) on and into sodium silicate solutions. Screening on three
main variables, namely silica concentration, pH of the silicate
solution, and the volume of the iron solution drop (Figure 1), was
performed. The pH values tested range from 10 to 14. The SiO2
concentration ranges from 0.75 to 7.75 mol/L. The formation of
membranes separating an inner metal-rich solution from the outer
silica-rich solution depends on the concentration of silica and on
pH (Figure 1). At high silica concentration and low pH, the gelling
of the silica solution precludes the formation of the mineral
membranes. As shown in Figure 1a, in all the other conditions
tested (black dots) we observed the formation of membranes. The
formation of silica/metal oxide hydroxide membranes was video-
recorded using a Nikon AZ10 microscope and a Nikon DSFi1
camera (Figure 1b). Videos S1-S6 describing the growth of
membranes in the presence or in the absence of formamide, in
the case of different salts, are in SI#1. The pH was measured with
a pH electrode (Mettler Toledo InLab Expert Pro_15M, tip
diameter: 3 mm), and a laboratory pH meter (Eutech Intruments
pH 510). The precipitating membranes were analyzed by X-ray
powder diffraction in the absence of formamide. They were first
harvested and rinsed thoroughly with water and ethanol. They
were dried and ground to fine powder. Powder diffraction
very formation of the membrane, thus benefiting of ionic exchange
[18]
and electric field processes.
However, the most important
factor explaining the better yield of the active microdrops
experiments can be the lower crystallinity of the metal-oxide-
hydroxide phases found in microdrop-driven experiments with
respect to silica gardens made with pellets (Figure 1C). While the
yield obtained with microdrops is notably high it must be
considered that they were obtained in the absence of irradiation.
The irradiation of these experiments with protons or ultraviolet
light is expected to enhance the yield and increase the number of
biochemically relevant molecules. The study of the effect of UV
radiation will be particularly interesting because iron/silica
precipitates are supposed to screen it. [36]
As shown in Table 1, the best catalysts for the NH2CHO
condensation among the different salts tested are
Fe2(SO4)3·9H2O, and MgSO4 (Table 1). Iron and magnesium are
the two cations in the mineral composition of olivine, and they are
also found in pyroxenes. Olivine and pyroxene are the main rock-
forming minerals of the ultramafic and komaititic crust of the
earliest Earth. The serpentinization reaction taking place when
these minerals interact with water [15] is responsible for most of the
compounds produced abiotically on the planet. Consequently,
those geological sites where serpentinization is occurring are
considered among the most likely niches for the transition from
inorganic to organic geochemistry, and perhaps for the
emergence of life. Along with Mn and Cu, the oxy-hydroxides of
these four metals account for the four nucleobases of the RNA
and three aminoacids obtained by condensation of NH2CHO.
measurements were performed on
a PANalytical X'Pert
diffractometer operating at a wavelength of 1.54 Å (Cu Kα
radiation). The measured 2Theta range was chosen from 10-90°
in steps ranging from 0.02°. Integration time per step was set to
22 s. Assignment and identification of detected reflexes to
crystalline matter was accomplished using the reference library of
the X’Pert HighScore Plus-PDF2. Occurring phases of FeCl3
include lepidocrocite γ-FeO(OH) and goethite α-FeO(OH). In the
case of FeCl2, the phases identified were akagaenite β-FeO(OH)
and maghemite γ-Fe2O3. For MnCl2 membranes the identified
phases were hausmannite Mn3O4, kempite Mn2Cl(OH)3. The
crystallinity of the mineral vesicles formed by microdrops is lower
compared to the ones made with solid pellets, as seen from X-ray
diffraction study (Figure 1c).
Conclusions
We conclude that the four nucleobases required for RNA
synthesis, three amino acids (glycine, alanine, and N-formyl
glycine), and six carboxylic acids can be synthesized from
formamide in a single geochemical scenario at 80ºC, without
irradiation. The condensation of formamide is catalyzed during
the formation of silica-oxyhydroxides membranes of iron,
magnesium, manganese, and copper, common metals in the
ultramafic and komaititic rocks of the earliest crust of the planet,
as clearly highlighted by the comparison with the reaction
performed in the absence of mineral vesicles. These membranes
have been demonstrated to form within alkaline waters derived
from serpentinization reaction,[20] a common phenomenon on
primitive Earth, and in Earth-like planets and moons. It is
reasonably that the enhanced catalytic properties of the mineral
vesicles obtained with microdrops with respect to silica gardens
might be due their smaller size (and consequent higher surface
area), and to the fact that they are actively formed during
formamide condensation. Our results suggest that the conditions
required for the synthesis of the molecular bricks from which life
Thermal condensation of formamide: Formamide (Fluka,
>99%) was used without further purification. Fresh commercial
sodium silicate solution (Sigma-Aldrich, reagent grade, containing
about 13.8 wt % Na and 12.5 wt % Si) was used as the silica
source after 1:4 (v/v) dilution with Millipore water. 1.8 mL of this
sodium silicate solution was mixed with 200ul of formamide. The
presence of formamide did not change the pH of the alkaline
silicate solution but it accelerated the gelling process. Before
gelling, 10 microdrops (5ul each) of saturated solution of the metal
salts were added by pouring. We tested ZnCl2, FeCl2·4H2O,
CuCl2·2H2O, MnCl2 Fe2(SO4)3·9H2O,CuN2O6 and MgSO4.
This article is protected by copyright. All rights reserved.

10.1002/chem.201706162
Chemistry - A European Journal
FULL PAPER
Microdrops were observed to expand immediately and the
reaction mixture was heated at 80°C for 24h. After some hours (4-
8 hrs, depending on the nature of the salt), a low-density gel
appears and the reaction products are recovered by the following
procedure: a) the gel and the membrane are broken off with a
spatula adding formamide (500 l); b) filtration of the inorganic
material and the gel by Büchner funnel (washing with (500 l) of
formamide and recovering of the liquid phase); c) High vacuum
distillation of the liquid phase to obtain a crude (SiO2 + reaction
products); d) Direct analysis of the reaction products by silylation
of the crude. The crude was analyzed by gas chromatography
associated with mass spectrometry (GC-MS) after treatment with
N,N-bis-trimethylsilyl trifluoroacetamide in pyridine (620 μL) at
60 °C for 4 h in the presence of oleic acid as the internal standard
(0.2 mg). Mass spectrometry was performed through the following
program: injection temperature 280 °C, detector temperature
280 °C, gradient 100 °C for 2 min, and 10 °C/min for 60 min To
identify the structure of the products, two strategies were followed.
First, the spectra were compared with commercially available
electron mass spectrum libraries such as NIST (Fison,
Manchester, U.K.). Second, GCMS analysis was repeated with
standard compounds. All products have been recognized with a
similarity index (SI) greater than 98% compared to that of the
reference standards. The analysis was limited to products of ≥1
ng/mL, and the yield was calculated as milligrams of isolated
products. Mass-to-charge ratio (m/z) value and the abundance of
mass spectra peaks of compounds 2-25 are reported in Table 2.
Products[a]
m/z (%)
Adenine[c] (2)
Guanine 3
Uracil[c] (4)
279 (27) [M], 264 (100) [M-CH3], 249 (1) [M-(CH3)2], 192 (17)
367 (100) [M], 352 [M-CH3]294 [M-HSi(CH3)3]
256 (35) [M], 241 (100) [M-CH3], 225 (15) [M-CH3-CH4], 182 (7) [M-Si(CH3)3-H2], 142 (70), 113 (55)
Cytosine[c] (5)
255 (49) [M], 254 (100) [M-H], 240 (72) [M-CH3], 182 (5) [M-HSi(CH3)3]
Isocytosine[d] (6)
327 (18) [M], 312 (100) [M-CH3], 282 (9) [M-(CH3)3], 255 (6) [M-Si(CH3)3], 240 (7) [M-HSi(CH3)3-CH3], 183 (2) [M-2xSi(CH3)3]
280 (30) [M-Si(CH3)3] 265 (100) ) [M-Si(CH3)3-CH3]
168 (25) [M], 153 (100) [M-CH3], 123 (5) [M-(CH3)3], 99 (100)
192(100) [M], 177 (100) [M-CH3]
Hypoxantine (7)
4(3H)pyrimidinone[b] (8)
Purine(9) [b]
3,5(NH2)1,2,4 Triazole[c] (10)
2,4-DAP-5COOH [d] (11)
6(OH)-2,4-DAP [c] (12)
Piruvic acid[c] (13)
243 (65) [M], 230 (100) [M-CH3]
370 (11) [M], 355 (100) [M-CH3]
270 (35) [M], 255 (100) [M-CH3]
160 (10) [M], 145 (7) [M-CH3], 88 (14) [M-Si(CH3)3], 71 (12) [M-Si(CH3)3-OH], 43 (100) [M-HSi(CH3)3-CO2]
Lactic acid[c] (14)
Oxalic acid[c] (15)
219 (6) [M-CH3], 190 (14) [M-CO2] , 147 (71) [M-Si(CH3)3-CH3], 133 (7), 117 (76) [M-Si(CH3)3-(CH3)3]
219 (3) [M-CH3], 189 (5) [M-(CH3)3], 147 (78) [M-Si(CH3)3-CH3], 117 (1) [M-Si(CH3)3-3xCH3], 73 (100)
Succinic acid[c] (16)
Oxaloacetic ac. 17
247 (16) [M-CH3], 173 (5) [M-HOSi(CH3)3], 147 (100), 73 (80)
346 (10) [M], 333 (100) [M-CH3]
Parabanic acid[c] (18)
Glycine[b] (19)
258 (15) [M], 243 (35) [M-CH3], 215 [M-2xCH3], 100 (100), 73 (23)
147 (11) [M], 132 (28) [M-CH3], 88 (9),73 (100)
Alanine[c] (20)
N-Formylglycine[b] (21)
218 (4) [M- CH3], 190 (6), 147 (13), 116 (100) [M-HOSi(CH3)3-CO], 73 (60)
160 (38) [M-CH3], 147 (5) [M-CO], 131 (22) [M-CONH2], 102 (11) [M-Si(CH3)3], 73 (100)
Guanidine[c] (22)
188 (11) [M-CH3], 173 (10) [M-2xCH3], 171 (100), 73 (33)
204 (7) [M], 189 (73) [M-CH3], 147 (100), 73 (35)
Urea[c] (23)
DAMN[c] (24)
252 (5) [M], 153 (18) [M-Si(CH3)3-HCN], 138 (3) [M-NHSi(CH3)3-HCN], 73 (100) [Si(CH3)3]
264 (4) [M], 191 (95) [M-Si(CH3)3]
Glycol aldehyde Dimer[c] 25
Table 2. Mass-to-charge ratio (m/z) value and the abundance of mass spectra peaks of compounds (2-25).
a]Mass spectroscopy was performed by using a GC-MS. Samples were analyzed after treatment with N,N-bis-trimethylsilyltrifluoroacetamide and pyridine. The peak
abundance is reported in parentheses.[b] Product analyzed as the monosilyl derivative;[c] Product analyzed as the bis-silyl derivative;[d] Product analyzed as the tris-
silyl derivative.
We acknowledge funding from the European Research Council
under the European Union’s Seventh Framework Programme
(FP7/2007-2013)/European Research Council grant agreement
AEI/FEDER. MIUR Ministero dell'Istruzione, dell'Università della
Ricerca and Scuola Normale Superiore (Pisa, Italy), project PRIN
2015 STARS in the CAOS - Simulation Tools for Astrochemical
Reactivity and Spectroscopy in the Cyberinfrastructure for
Astrochemical Organic Species, cod. 2015F59J3R, is
no. 340863 (Prometheus). Ministerio de Economia
y
Competitividad is acknowledge for Project CGL2016-78971-P,
This article is protected by copyright. All rights reserved.

10.1002/chem.201706162
Chemistry - A European Journal
FULL PAPER
[10] A. López-Sepulcre et al. Mon. Not. R. Astron.Soc.2015, 449, 2438-2458.
[11] G. R. Adande, N. J. Woolf, L. M. Ziurys. Astrobiology 2013,13, 439-453.
[12] N. Biver et al. Astron. Astrophys. 2014, 566, L5.
[13] D. Bockelée-Morvan et al. Astron. Astrophys. 2000, 353, 1101-1114.
[14] A. M. Saitta, F. Saija. PNAS 2014, 111, 13768-13773.
[15] N. H. Sleep, A. Meibom, T. Fridriksson, R. G. Coleman, D. K. Bird. PNAS
2004, 101, 12818-12823.
acknowledged. This work was supported by COST Action TD
1308. M. I. Pérez Valverde acknowledges financial support from
CONACyT (CVU 557501/300081) and VIEP-BUAP. Author
contributions. E.D.M., R.S., B.M.B., designed the prebiotic
chemistry experiments. B.M.B and L.B. performed the prebiotic
chemistry experiments. M.I. P.V. designed and performed the
experiments of mineral self-assembly and analyzed them.
J.M.G.-R. conceived the work and designed experiments of
mineral self-assembly. J.M.G.-R., E. D.M. and R.S. analyzed and
discussed the results, and wrote the paper. Competing interest.
The authors declare that they have no competing interests. Data
and materials availability. All data needed to evaluate the
conclusions in the paper are present in the paper and/or the
Supporting information.
[16] A. Neubeck, N. T. Duc, D. Bastviken, P. Crill, N. G. Holm. Geochem.
Trans. 2011, 12, 6 .
[17] J. M. García-Ruiz, E. Nakouzi, E. Kotopoulou, L. Tamborrino, O.
Steinbock. Science Advances 2017, 3, e1602285.
[18] F. Glaab, M. Kellermeier, W. Kunz, E. Morallon, J. M. García‐Ruiz,
Angew. Chem. 2012,124, 4393-4397.
[19] L. M. Barge et al. Angew. Chem. Int. Edit. 2015, 54, 8184-8187.
[20] R. Saladino, G. Botta, B. M. Bizzarri, E. Di Mauro, J. M. Garcia Ruiz.
Biochemistry 2016, 55, 2806-2811.
[21] F. Glaab et al. ChemPhysChem 2017, 18, 338-345.
[22] M. Kellermeier, F. Glaab, E. Melero-García, J. M. García-Ruiz. Method
Enzymol. 2013, 532, 225-256.
Keywords: prebiotic chemistry, • formamide • silica micro-drops
• catalysis• biomolecules and condensing agents
[23] R. Saladino, G. Botta, S. Pino, G. Costanzo, E. Di Mauro,. Front. Biosci.
2013, 18, 1275-1289 .
[24] H. D. Bean et al. J Am. Chem. Soc. 2007,129, 9556-9557.
[25] I. Hirao, M. Kimoto, R. Yamashige. Accounts Chem. Res. 2012, 45,
2055-2065.
References
[26] M. Ferus et al. J Am. Chem. Soc. 2012, 134, 20788-20796.
[27] S. Senanayake, H. Idriss. PNAS 2006, 103, 1194-1198.
[28] H. L. Barks et al. ChemBioChem 2010, 11, 1240-1243.
[29] K. Storrey. Wiley, Hoboken,NJ, 2004.
[1]
A. G. Cairns-Smith, Genetic takeover: and the mineral origins of life.
1982, pp 278.
[2]
[3]
[4]
[5]
V. Goldschmidt. New Biol 1952,12, 97-105.
J. Bernal, The Origin of Life 1967.
[30] W. D. McArdle, F. I. Katch, V. L. Katch. Exercise physiology: nutrition,
energy, and human performance. Lippincott Williams & Wilkins, 2010.
[31] E. Smith, H. J. Morowitz. PNAS 2004, 101, 13168-13173.
[32] A. Eschenmoser. Chem. Biodiv. 2007, 4, 554-573.
[33] N. Jena, P. Mishra. Free Radical Bio. Med. 2012, 53, 81-94 .
[34] C. Menor‐Salván, M. R. Marín‐Yaseli. Chem. Eur. J. 2013, 19, 6488-
6497.
O. Leslie E, Crit. Rev. Biochem. Mol. 2004, 39, 99-123.
M. Ferus, F. Pietrucci, A. M. Saitta, A. Knížeka, P. Kubelíka, O. Ivaneka,
V. Shestivska, Svatopluk Civiš. PNAS 2017, 114, 4306-4311.
a)J. Oró, S. Kamat. Nature 1961, 190, 442-443. b) Common origins of
RNA, protein and lipid precursors in a cyanosulfidic protometabolism", B.
H. Patel, C. Percivalle, D. J. Ritson, C. D. Duffy, and J. D. Sutherland,
Nat Chem. 2015; 7(4): 301-307.
[6]
[35] R. Saladino, G. Botta, S. Pino, G. Costanzo, E. Di Mauro. Chem. Soc.
Rev. 2012, 41, 5526-5565.
[7]
[8]
M. Ferus. PNAS 2015, 112, 657-662.
E. Carota, G. Botta, L. Rotelli, E. Di Mauro, R. Saladino. Curr. Org. Chem.
2015, 19, 1963-1979.
[36] V. Phoenix, K. Konhauser, D. Adams, S. Bottrell. Geology 2001, 29, 823-
826.
[9]
R. Saladino et al. PNAS 2015, 112, E2746-E2755.
This article is protected by copyright. All rights reserved.

10.1002/chem.201706162
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Silica metal-oxide vesicles
catalyze the thermal
condensation of formamide to
molecules of biological
relevance.
Bruno Mattia Bizzarri, Lorenzo
Botta, Maritza Iveth Pérez-
Valverde, Raffaele Saladino,*
Ernesto Di Mauro, Juan Manuel
García-Ruiz,*
Page No. – Page No.
Silica metal-oxide vesicles
catalyze comprehensive
prebiotic chemistry
This article is protected by copyright. All rights reserved.