Origins of Life and Evolution of Biospheres (2006) 36: 343–361
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION
OF RNA OLIGOMERS
, PRAKASH C. JOSHI
, MICHAEL J. GAFFEY
, CALLEN HYLAND
, TERESA ROSS
and JAMES P. FERRIS
Present address: RIBOMIC Inc., 3-16-3 Shirokanedai, Tokyo 108-0071, Japan;
Chemistry and New York Center for Studies on the Origins of Life, Rensselaer Polytechnic Institute,
Troy, NY 12180, USA;
Space Studies Department, University of North Dakota, Grand Forks,
author for correspondence, e-mail: firstname.lastname@example.org)
(Received 2 November 2005; accepted in revised form 4 January 2006)
Abstract. Activated mononucleotides oligomerize in the presence of montmorillonite clay to form
RNA oligomers. In the present study, effects of salts, temperature and pH on the clay-catalyzed syn-
thesis of RNA oligomers were investigated. This reaction is favored by relatively high concentration
of salts, such as 1 M NaCl. It was shown that the presence of divalent cations was not required for
this reaction. High concentrations of NH
and 0.01 M HPO
inhibit the reaction. The
yields of RNA oligomers decreased as the temperature was raised from 4
the major product at pH’s below 6. The catalytic activity of a variety of minerals and three meteorites
were investigated but none of them except galena catalyzed the oligomerization. ATP was generated
from ADP but it was due to the presence of HEPES buffer and not due to the minerals. Meteorites
catalyzed the hydrolysis of the pyrophosphate bonds of ATP. The results suggest that oligomers of
RNA could have formed in pH 7–9 solutions of alkali metal salts in the presence of montmorillonite
Keywords: RNA oligomers, catalysis, origins of life, montmorillonite, meteorite, mineral, ATP
Recent studies on the montmorillonite catalysis have emphasized the formation
of longer RNA oligomers. The reaction proceeds with imidazole (Weimann et al.,
1968) and 1- or 3-methyladenine activating groups (Figure 1) (Prabahar and Ferris,
1997; Huang and Ferris, 2003) in a reaction solution containing 0.1 M buffer, 0.075
and 0.2 M NaCl and at a temperature of 25
C. The reaction proceeds
equally as well with purine or pyrimidine nucleotides with a 3
bonds forming 67% of the time with purine and 20% of the time with pyrimidine
nucleotides (Ferris and Ertem, 1993; Ding et al., 1996; Prabahar and Ferris, 1997).
Not all montmorillonites catalyze the oligomerization reaction. The more effec-
tive catalysts include SPV-200 Volclay, a commercial product from the American
Colloid Company, montmorillonite 22-A from Wards Scientiﬁc, Belle Fourche,
North Dakota H-27 (Kerr et al., 1951) and SWA-1, a sample from the Clay
344 S. MIYAKAWA ET AL.
Figure 1. Activated RNA monomers. B= adenine, guanine, hypoxanthine, cytosine and uracil.
Minerals Society. A sample from Japan is somewhat less active than those men-
tioned above (Kawamura and Ferris, 1994) while montmorillonites with either
very low or very high iron content (nontronites) have little or no catalytic activ-
ity. Most montmorillonites have to be converted to a homoionic form in a proce-
dure that uses a cold acid wash and then adjusting the pH 6-7 before they exhibit
catalytic activity (Banin 1973; Banin et al., 1985). Catalytic activity is observed
when the exchangeable cations are alkali or alkaline earth cations with the one
exception being Mg
(Ferris and Ertem, 1992). The catalytic activity of mont-
morillonite is retained when ammonium ion is the exchangeable cation (Ferris and
Oligomers as long as 40–50 monomer units (Ferris et al., 1996; Ferris,
2002; Huang and Ferris, 2003), and an excess of homochiral products from
D, L-mixtures are formed by montmorillonite catalysis (Joshi et al., 2000).
The oligomers formed by the reaction of a mixture of ImpA with ImpC
exhibit sequence selectivity (Ertem and Ferris, 2000; Miyakawa and Ferris,
The scope of the montmorillonite catalyzed formation of RNA oligomers was
reported previously (Ferris and Ertem, 1992). These studies have been extended
in the present research where a variety of reaction conditions were examined to
determine the limits under which oligomers form. The effect of ionic strength, tem-
perature, and pH were investigated. Catalysis by other minerals and meteorites was
also investigated as well as nucleotides with activating groups other than imidazole
or adenine derivatives.
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 345
Adenosine, AMP, 2
-cyclic AMP, 3
-cyclic AMP, ADP, ATP, NH
pA and buffers were obtained from Sigma. Alkaline phosphatase
and phosphodiesterase I were obtained from Worthington Biochemical and ri-
was from Sigma. Montmorillonite [Volclay SPV-200, (Al, Fe
] was a gift from the American Colloid Company.
Murchison meteorite, which is a CM2 type carbonaceous chondrite, was a gift from
Dr. S. Pizzarello in Arizona State University. The Murchison meteorite collected
in Australia in 1969 is known to contain many types of bioorganic compounds
(Cronin and Chang, 1993; Gilmour, 2003). Yamato-791717 and Yamato-86751
were obtained from the Antarctic Meteorite Research Center, National Institute
of Polar Research in Japan. These Antarctic meteorites are CO3 and CV3 type
carbonaceous chondrites, respectively (Kaiden et al., 1997; Murakami and Ikeda
1994). Olivine ([Mg,Fe
), galena (PbS), calcite (CaCO
), magnetite (Fe
), sphalerite (ZnS), magnesite (MgCO
), siderite (FeCO
), brucite (Mg(OH)
, chalcosite (Cu
S), talc (Mg
and dolomite (CaMg(CO
), were obtained from Dr. M. J. Gaffey’s collection and
) and pyrite (FeS
) were purchased from Alfa Aesar.
An aqueous solution containing 0.015 M ImpA, salts and 0.1 M MOBS or HEPES
buffer was kept in the presence of Na
-montmorillonite clays, minerals or me-
teorites for several days at room temperature. ImpA was prepared following the
literature procedure (Joyce et al., 1984). The purity of ImpA was checked by anion-
exchange HPLC prior to starting reactions to make sure it was higher than 95%
and the A
ppA content was lower than 1%. pA and A
ppA may form during stor-
age of ImpA. Since A
ppA is a good initiator of oligomer formation (Ferris, and
Ertem 1993), care was taken to be sure it was not present as an impurity in the acti-
vated monomers. The Na
-montmorillonite was prepared from Volclay by titration
method (Banin et al., 1985). MOBS was favored as a buffer (Figure 2) because it
Figure 2. Structures MOBS and HEPES.
346 S. MIYAKAWA ET AL.
does not have a hydroxyl group that can react with ImpA. At the end of the reaction
time, the mixture was centrifuged and the supernatants were withdrawn. The clay
minerals and meteorites were extracted twice with a mixture of 0.1 M NaCl and
30% acetonitrile (Miyakawa and Ferris, 2003), 0.1 M, pH 9 pyrophosphate (Fer-
ris 2002), or ammonium acetate to elute the oligomers from the montmorillonite.
Ammonium acetate is less efﬁcient than the other eluants in removing the cyclic
dimers and long oligomers of AMP from montmorillonite. The supernatant and the
extracts were combined and incubated at pH 4 and 37
C for 4 h to hydrolyze the un-
reacted imidazole activating groups. Oligomers were analyzed by anion-exchange
HPLC (Biosphere GMB 1000Q column, Melcor Tech. Inc.) at 260 nm with a 2
mM Tris buffer (pH 8) (Buffer A) and 2 mM Tris (pH 8) and 400 mM NaClO
(Buffer B) as eluants, and by reversed-phase HPLC (Alltima C18 column, Alltech)
at 260 nm with a pH 2.7, 0.2% triﬂuoroacetic acid (Buffer A) and a pH 2.7, 0.2%
triﬂuoroacetic acid and 30% acetonitrile (Buffer B) as eluants (Kanavarioti, 1997).
Oligomers were identiﬁed by enzymatic hydrolysis with alkaline phosphatase and
phosphodiesterase I or ribonuclease T
and by coelution on the HPLC with authen-
tic samples. The hydrolysis details are described in Miyakawa and Ferris (2003).
The yields of oligomers were calculated from the uncorrected peak areas of HPLC
chromatograms following the procedure of Ferris and Ertem (1993). The length of
oligomers was obtained by counting the number of peaks on the anion-exchange
HPLC chromatograms (Ferris and Ertem, 1993).
The pH 8 solutions containing 0.015 M ImpA, 0.2 M NaCl, 0.075 M MgCl
0.1 M HEPES were added to the montmorillonite clays and were kept at 50
C for 2 days, at room temperature for 3 days, at 4
C for 1 week. The
oligomers formed were extracted with 0.1 M ammonium acetate and analyzed by
HPLC using a Biosphere anion exchange column.
The following buffers containing 0.35 M NaCl were used: hydrochloric acid for
pH 0–1, sodium fomate for pH 2–5, MES for pH 6, MOBS for pH 7–9, CHES
for pH 10, and sodium hydroxide for pH 13–14. The pH values were measured
with a pH meter both before and after the reactions to make sure it did not drift.
The buffers, with pHs that were lower than 2 and higher than 12, were carefully
prepared and were not measured with the pH meter. ImpA was added to the buffers
to give a ﬁnal concentration of 0.015 M. These solutions were then added to the
Since reaction products may hydrolyze if the incubation time is too long, differ-
ent incubation times were used at different pHs. The incubation times were 1 day
for pH 0–5, 7 days for pH 6–8 and 13–14, and 14 days for pH 9–10. The formation
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 347
ppA was monitored at a variety of times at pH 2.8 and 4.5. The yields were
maximum and nearly constant between 10 and 70 hours. The formation of pA
was monitored at a variety of times at pH 6 and 8. The yields were maximum and
nearly constant between 1 and 7 days. The yields of pA
pA and pA
pA at pH
10 were not much different from 14 days to 50 days. Unreacted ImpA (26%)was
detected after 50 days at pH 10.
ETEORITES AND MINERALS
Meteorites and minerals were powders that were washed with distilled water before
use. In different runs, meteorites were washed with ethanol, benzene or hexane, but
no difference in the reaction products was observed. In reactions, a pH 8 solution
containing 0.015 M activated AMP, 0.2 M NaCl, 0.075 M MgCl
, and 0.1 M HEPES
was kept in the presence of meteorites or minerals for 1 week. ImpA, ATP, ADP
and cyclic AMP were used as the activated AMP. The products were extracted with
0.1 M ammonium acetate or a mixture solution of 0.1 M NaCl and 30% acetonitrile,
and analyzed by HPLC.
HYDROLYSIS BY METEORITES
Reactions containing 1 mM ATP and 10 mM MOBS were adjusted to pH 7 and were
incubated at 37
C in the presence of meteorites. Small amount of samples were
withdrawn at various time intervals and the amounts of ATP left were determined by
anion-exchange HPLC chromatography. The decomposition rate of ATP followed
pseudo ﬁrst-order kinetics. The pseudo ﬁrst-order rate constants and the half-lives
were calculated from the slopes of graphs of Log ([ATP]
/[ATP]) as a function
INERAL CATALYSIS OF THE REACTION OF ATP, ADP, 3
AMP AND 2
The mineral catalysis of RNAoligomer formation was investigated using ATP, ADP,
-cyclic AMP and 2
-cyclic AMP as starting materials. These reactions were
performed under a variety of conditions: (1) at pH 8.0 using HEPES buffer and pH
5.8 using MES buffer (2) using the above buffers in the presence of 0.075 M MgCl
and 0.2 M NaCl. The reactions were performed in a 1 mL solution in the presence
of 50 mg of powdered mineral at 21
C and 37
C for 7 days. The reaction mixtures
were centrifuged and the supernatant was collected. The mineral was extracted with
pH 9, 0.1 M Na
to recover oligomers bound to the mineral and the combined
extracts were analyzed by anion-exchange HPLC. Some of the reaction products
were characterized further by hydrolysis using alkaline phosphatase followed by
HPLC analysis of the hydrolysis products on an Alltima reverse phase column.
348 S. MIYAKAWA ET AL.
Electrospray mass spectroscopy was used to conﬁrm the formation of AMP and
ATP from ADP. The product mixture obtained when talc was the mineral tested as a
catalyst, was separated on an ion-exchange HPLC column using a 0–45% gradient
of 2 mM ammonium acetate, pH 8 and a 2 M ammonium acetate, pH 8 in 0–40 min
at 1 mL/min with an Agilent 1100 series LC/MSD-SL ion trap mass spectrometer:
m/e 346 (M −1), and 348.1 (m +1) (calculated for C
P, 347.063), AMP;
426.0 (m −1), 428.0 (M +1) (calculated for C
, 427.029) ADP; 505.9
(m +1) and 507.9 (m −1) (calculated for C
, 506.996) ATP.
Experiments were performed in which no mineral was added and the reaction
was performed with ADP and varying amounts of 0.2M NaCl and 0.075 M MgCl
with 0.1M HEPES. The reactions were carried out at 25
C for 30, 38 and 63 days at
pH 8. Ion exchange HPLC indicated the presence AMP, ADP and ATP by retention
time and coinjection with authentic samples. The products formed from the reaction
of 0.015 M ADP, 0.1 M HEPES, 0.2 M NaCl and 0.075 M MgCl
by HPLC as described above and analyzed by electrospray mass spectrometry; m/e
345.9 (M −1) and 348 (M +1) (calculated 347.063) AMP; 425.9 (M −1) and 428
(M + 1) (calculated 427.029) ADP; 505.9 (M −1) and 507.9 (M +1) (calculated
3. Results and Discussion
FFECT OF ADDED SALTS
In the past the standard reaction mixture for oligomer formation was 0.2 NaCl,
0.075 M MgCl
. and 0.1 M buffer. Our previous results indicted that Mg
essential for reaction since the oligomers did not grow as long in its absence
(Kawamura and Ferris, 1994). However, investigation of ImpA oligomerization
in the presence of varying concentrations of NaCl revealed that oligomer formation
did occur in the presence of high concentrations of NaCl (Table I). A 1 M con-
centration of NaCl resulted in the formation of oligomers as long as 10 mers, an
amount comparable to that observed when our standard buffer electrolyte (0.2 M
NaCl and 0.075 M MgCl
) and 0.1 M HEPES was used. A 0.1 M NaCl solution
gave 6 mers. These studies establish that ionic strength and not coordination by
is the more important element in the reaction pathway.
3.1.2. Buffer Structure and Concentration
In previous studies PIPES and HEPES (Good and Izawa, 1972) were used
as buffers. HEPES contains a hydroxyl groups that may form phospho-
ester bonds by reaction with the activated monomers (Kanavarioti, 1997).
3-(N-Morpholino)butanesulfonic acid (MOBS) (Thiel et al., 1998) was evaluated
as an alternative buffer since it does not contain a hydroxyl group (Huang and
Ferris, 2003). The variation of oligomer chain length with the concentration of
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 349
Effects of cations and anions on the montmorillonite-catalyzed synthesis of RNA oligomers
Yields of RNA oligomers (%)
Strength 1 2 3 4 5 6 7 8 9 10 (mers)
1M NaCl 7.9 1.1 9.3 57 14 11 3.4 1.8 1.2 0.53 0.29 0.22
0.35M NaCl 7.9 0.42 11 66 13 7.4 1.6 0.61 0.39
0.1M NaCl 7.9 0.17 15 66 12 5.0 0.84 0.41
0.01M NaCl 7.9 0.077 24 64 8.3 2.2 0.25
7.9 0.49 11 ±050±014±010±1 5.2 ±0.1 3.2 ±0.4 2.0 ±0 1.0 ±0 0.61 ±0.08
8.0 0.50 28 ±245±114±1 6.9 ±0.2 2.7 ±0.1 1.5 ±0.1 0.43 ±0.07 0.16 ±0.06
0.35M KCl 7.9 0.42 30 46 16 5.7 1.6 0.49 0.26
Cl + 0.075M
+ 0.2M NaCl
8.0 0.50 13 49 14 9.9 5.0 3.2 1.9 1.3 0.71
Cl + 0.075M
+ 0.2M NaCl
8.0 0.51 14 48 14 10 4.8 3.0 1.9 1.0 0.71
7.8 0.41 59 31 6.2 2.0 0.33
0.35M HCOONa 7.8 0.41 16 ±160±016±1 6.6 ±0.2 1.6 ±0.1 0.36 ±0.1
COONa 7.9 0.42 20 62 12 5.1 0.58
7.9 0.42 12 68 12 6.1 1.2 0.30
8.0 0.50 14 49 14 9.5 5.1 2.8 2.0 0.98
8.0 0.51 21 47 14 8.1 3.8 1.9 1.0
(Continued on next page)
350 S. MIYAKAWA ET AL.
Yields of RNA oligomers (%)
Strength 1 2 345678910(mers)
8.1 0.43 100
8.0 0.60 14 67 12 5.5 0.85
+ 0.2M NaCl
8.0 0.50 49 36 9.4 3.1 1.7 0.72
+ 0.2M NaCl
8.0 0.53 85 14
8.0 0.60 100
Reaction: 0.015 M ImpA + Salts + 0.1 M MOBS + Volclay, 3 days.
Yields were obtained from the peak area of HPLC chromatograms.
About 20% of NH
pA was detectable both in the presence and absence of montmorillonite.
More than 60% of ImpA was detectable.
About 20% of ADP and 20% of ImpA were detectable.
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 351
Effects of MOBS on the clay-catalyzed synthesis of RNA
Yields of RNA oligomers (%)
Strength 1 2
3 4 (mers)
0 7.4 0 53 ±146±1 1.2 ±0.1 0.14 ±0.03
0.001 7.4 0.00038 53 ±445±4 1.3 ±0.1 0.14 ±0.03
0.01 7.7 0.0056 47 50 1.9 0.39
0.1 7.9 0.067 29 62 7.3 1.6
Reaction: 0.015M ImpA + MOBS + Volclay, 3 days.
The value of pH is an average number measured before and after the reaction. The pH
drifted in the reactions with low concentrations of MOBS.
Yields were obtained from the peak areas of HPLC chromatograms.
The most dominant dimer is cyclic A
MOBS had chain lengths comparable to those expected for a variation with ionic
strength (Table II). The chain lengths were comparable to those observed using
0.1 M PIPES or HEPES in the presence of 0.075 M MgCl
and 0.2 M NaCl indi-
cating that there is little reaction between the activated monomers and the buffers
that contain hydroxyl groups.
3.1.3. Some Soluble Salts That May Have Been Present on the Primitive Earth
The presence of other soluble compounds on the primitive Earth that could have
inhibited the oligomerization reactions by binding competitively to the montmo-
rillonite (Wang and Ferris, 2001) or by reacting with the activated monomers was
investigated. The effect of a series of simple inorganic compounds on the oligomer-
ization reaction was investigated (NaCl, MgCl
, KCl, NH
) (Table I). Substitution of
0.075 M CaCl
in reaction solutions containing 0.2 M NaCl gave slightly
lower yields of 4–8 mers than did MgCl
even though the ionic strengths of these
solutions were the same (0.5).
The chain length of oligomers resulting from reactant solutions with ionic
strengths of 0.4 of NaCl, KCl, HCOONa, and NaNO
were essentially the same
(Table I). Some inhibition was observed with NH
Cl and CH
COONa. Total inhi-
bition was observed with NaHCO
. The partial inhibition by NH
was due to the conversion of ImpA to a 20% yield of NH
pA by displacement of
the imidazole group with NH
. This displacement occurred in the presence and
absence of montmorillonite suggesting that the reaction took place in the aqueous
phase and not on the clay surface. It has been observed that NH
pA does not re-
act on montmorillonite to form oligomers (Prabahar et al., 1994; Miyakawa and
Ferris, 2003). The inhibitory effect of NaHCO
is apparently due to the binding of
to montmorillonite as suggested by the presence of more than 60% of
the starting ImpA remaining at the end of the reaction time. The inhibitory effect
352 S. MIYAKAWA ET AL.
was due in part to the reaction of ImpA with HPO
to give a 20%
yield of ADP. The recovery of 20% ImpA at the end of the reaction time suggests
may have also inactivated the clay catalyst.
The effect of salt concentration on the inhibitory effect was explored further
using inhibitor concentrations varying from 0.001 to 0.35 M (Table I). The ionic
strength was adjusted with MgCl
and NaCl. No inhibition was observed using
0.001 and 0.01 but was observed using 0.35 M NH
of 0.001 and 0.01 M resulted in the detection of 8 mer and 7 mers respectively
but total inhibition was observed with a 0.35 M solution. Use of 0.001 and 0.01 M
resulted in the formation of 6 and 2 mers respectively and no oligomers
were formed with a 0.175 M solution.
on the Primitive Earth
It is postulated that the highest concentration of NH
in the modern oceans is less
than 0.01 M (Bada and Miller, 1968). Ammonia could have been formed by the
hydrolysis of HCN and other nitrogen-containing organic compounds. It would
have been destroyed by solar UV radiation at wavelengths less than 200 nm (Ferris
and Nicodem, 1972) so that it is unlikely that there was sufﬁcient NH
primitive oceans to inhibit the reactions of activated RNA monomers.
There are large differences in the proposed concentrations of HCO
primitive oceans because there are large differences in the amounts of CO
posed to be in the atmosphere. It has been suggested that if the atmospheric pres-
sure of CO
was 0.3 atm. the concentration of HCO
was 0.07 M (Grotzinger
and Kasting 1993). Others have proposed a CO
partial pressure of ≤10 bars
(Walker, 1985). Since there are little data to constrain the partial pressure of CO
in the atmosphere of the primitive Earth, it is not possible to determine if enough
was present to inhibit the montmorillonite-catalyzed formation of RNA
Soluble phosphate may have had a central role in the prebiotic formation of
genetic material but at the same time it could have inhibited RNA oligomers forma-
tion. While the concentration of phosphate in the ocean today is 10
M because of
its biological conversion to the mineral apatite, this would not have occurred on the
primitive ocean since apatite only forms at a pH ≥ 8.5 in the absence of life. The
formation of the minerals whilockite and magnesium phosphate occurs in the pH
range 7–9 and brushite at pH 6–7 (Gedulin and Arrhenius, 1994). These minerals
are 100 times more soluble than apatite and that would have led to a phosphate
concentration of 10
M, a concentration too low to inhibit the montmorillonite-
catalyzed formation of RNA oligomers.
Reactions run at 4 and 25
C yielded 8 mers while 7 and 6 mers were formed at tem-
peratures of 37 and 50
C, respectively. The cause of the decrease in chain length
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 353
Effects of temperatures on the clay-catalyzed synthesis of RNA
Yields of RNA oligomers (%)
C) 1 2 3 4 5 6 7 8 (mers)
419±3.0 47 ±2.0 11 ±4.0 6.0 ±0.70 2.4 ±0.70 1.9 ±1 .0 1.0 ±0.51 0.48 ±0.33
25 26 ±4.0 45 ±10 9.0 ±0.40 5.7 ±1.1 2.2 ±0.30 0.85 ±0.55 0.31 ±0.31 0.15 ±0.15
37 28 ±4.0 43 ±3.0 9.0 ±1.6 4.3 ±1.5 2.9 ±0.5 1.6 ±0.50 1.0 ±0.50
50 55 ±10 25 ±5.0 4.9 ±2.9 1.3 ±0.90 0.78 ±0.81 0.18 ±0.18
Reaction: 0.015M ImpA + 0.2M NaCl + 0.075M MgCl
+ 0.1M HEPES + Volclay.
Yields were determined from the peak areas of the HPLC chromatograms.
may be due to the increased rate of hydrolysis of ImpA at higher temperatures (Ta-
ble III). The yield of 5
-AMP increases at higher temperatures, a ﬁnding consistent
with the decrease in the chain lengths of the oligomers.
FFECT OF pH
The effect of changes in pH on oligomer formation was investigated (Figure 3).
Dimeric products containing phosphodiester bonds were formed at pH 5–10 with
the optimal yield at pH 7–8. The longest oligomers were observed in this pH range
(Figure 4). The formation of A
ppA occurred in the pH 1-6 range with the optimal
yield at pH 2–3. The reactions were monitored at different times to determine if
Figure 3. pH proﬁle of the dimer yields. Reaction of 0.015M ImpA in 0.35 M NaCl, 0.1 M buffer,
-montmorillonite. The yields were determined from the uncorrected peak areas of the reverse
phase HPLC chromatograms.
354 S. MIYAKAWA ET AL.
Figure 4. Variation in the length of the longest oligomers formed with pH. See Figure 3 caption for
reaction conditions. The lengths of the oligomers were determined from the anion-exchange HPLC
hydrolysis resulted in a decrease in the products. The yields of products containing
phosphodiester bonds were constant over a 7-day reaction time at pH 7–8.
It was expected that RNA oligomers might have formed more efﬁciently under
acidic conditions, because the protonated ImpA would have bound directly to neg-
ative sites of interlayers of the clay (Dawson et al., 1969; Banin et al., 1985; Ertem
and Ferris, 1998). Instead large amounts of A
ppA formed around pH 2. This
is probably because protonated imidazole groups are displaced by the negatively
charged phosphate groups formed by the hydrolysis of the activated monomers
(Kanavarioti et al., 1989; Ruzicka and Frey, 1993).
It was also expected that RNA oligomers would form more efﬁciently around pH
10, because the hydrolytic rate of ImpG and ImpU are lowest at that pH (Kanavarioti
et al., 1989; Ruzicka and Frey, 1993). However, most of ImpA did not react at
ATALYSIS OF OLIGOMER FORMATION
3.4.1. Reaction of ImpA
Three meteorites and twelve minerals were investigated as potential catalysts for
the formation of RNA oligomers from ImpA. Meteorites were investigated because
they contain mineral assemblages with associated organic compounds (Cronin and
Chang, 1993). The meteorites were powdered samples of the Murchison meteorite
and two Antarctic meteorites. Minerals were selected on the basis of their potential
for binding to the phosphate group of the activated mononucleotides.
None of the meteorites and only one of the minerals catalyzed the formation
of RNA oligomers from ImpA. Galena (PbS) catalyzed the formation of pA
pA, a ﬁnding that was reported previously (Sleeper and Orgel, 1979). The
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 355
Reaction of ADP in the presence and absence of minerals for 4 days
Yield of reaction products (%)
Reaction mixture Temperature AMP ADP ATP Unknown
ADP + Brucite 37
C 15.4 79.4 3.6 1.0
ADP + Talc 37
C 14.5 80.0 3.8 1.0
ADP + Siderite 37
C 14.8 80.1 3.8 0.9
ADP + Magnetite 37
C 15.5 79.6 3.7 0.7
ADP + Galena 37
C 14.9 79.0 4.0 1.2
C 15.9 78.8 3.9 0.8
ADP + Olivine 37
C 15.1 79.6 3.9 0.9
ADP + Olivine 21
C 14.9 79.3 4.2 1.2
C 14.9 79.9 4.2 0.9
Reaction: 0.015M ADP + 0.2M NaCl + 0.075M MgCl
+ 0.1 M HEPES,
pH 8 + mineral (50 mg/mL)
Yields were determined from the peak areas of the HPLC chromatograms.
insoluble mineral and not the soluble Pb
was shown to be the catalyst because
a solution in contact with the powdered mineral, when ﬁltered to remove mineral
particles, did not catalyze oligomer formation. Galena was less effective than mont-
morillonite as a catalyst as shown by the observation of only dimers as products
from the reaction of ImpA.
3.4.2. Mineral Catalysis of the Reactions of ATP, ADP, 3
The minerals magnesite, brucite, talc, olivine, dolomite, calcite, hematite, goethite,
siderite, magnetite, galena, sphalerite and chalcosite were investigated as possible
catalysts for the oligomerization of the adenine nucleotides at pH 8.0 and 5.8 in
the presence of buffers alone and in the presence of buffers together with 0.075 M
and 0.2 M NaCl. The reactions were carried out for 4 days at 37
C and the
products were analyzed by anion exchange HPLC (Table IV).
Reactions performed only in the presence of buffer generally resulted in the
hydrolysis of ATP and ADP to AMP and adenosine (Figure 5a). While those where
the additional salts were added underwent little or no reaction. In the cases where the
HPLC retention times of some of the products were comparable to those expected
pA or pA
pA that have comparable retentions to ADP on a Biosphere anion
exchange column, the reaction mixtures were hydrolyzed with alkaline phosphatase
and analyzed for the presence of A
pA or A
pA by HPLC on a reverse phase
column. In every instance adenosine was the major product conﬁrming that these
reaction products were ADP and not the isomers of pApA.
-Cyclic AMP (Figure 5b) underwent hydrolysis to AMP in the presence
and absence of added minerals. 3
-Cyclic AMP (Figure 5b) was more resistant to
356 S. MIYAKAWA ET AL.
Figure 5. (a) Steps in the hydrolysis of ATP to adenosine. (b) Structures of cyclic phosphates.
hydrolysis and in most reactions yielded only about 5% AMP as a reaction product.
ATP did not oligomerize in the presence of meteorites Yamato-791717 (CO3) and
The reaction of ADP in the 0.1 M HEPES, 0.2 M NaCl and 0.075 M MgCl
the presence and absence of minerals yielded AMP and ATP (Table IV). The prod-
ucts were identiﬁed by coinjection with authentic samples using ion exchange and
reverse phase HPLC. The structural assignments were conﬁrmed by electrospray
mass spectrometry using an ion exchange column.
It was established that minerals did not have a role in the formation of ATP since
it was formed in their absence. It was established that the formation of ATP was
due to the presence of 0.1M HEPES in the reaction mixture. It does not appear to be
an ionic strength effect since ATP was not formed when MOBS replaced HEPES
(Figure 2) in the reaction (Table V). The functional groups of HEPES and MOBS
differ by the presence of a hydroxyl group in HEPES (Figure 2). This suggests that
the hydroxyl group in HEPES has a role in the reaction.
There are reports of the conversion of ADP to ATP in the presence of cyclodex-
trin, creatine and oxygen (Hattori et al. 1984) and with NADH and riboﬂavin
(Lozinova and Arutyunyan, 1990) but there are no reports of ATP formation in
the presence of a buffer-salt mixture. The yield of ATP decreased in the absence
of oxygen (Hattori et al. 1984). The formation of ATP from ADP by Lozinova
and Arutyunyan (Lozinova and Arutyunyan 1990) still occurred in the absence
of oxygen. Both reports propose that ATP formation proceeds via a free radical
3.4.3. Hydrolysis of ATP by Meteorites
The meteorites did not initiate the condensation of ATP to oligomers but rather
catalyzed the hydrolysis of ATP to ADP and AMP. The pseudo ﬁrst order rate
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 357
Formation of ATP from ADP
Yields of Products
Time (days) AMP ADP ATP Unknown
,HEPES 30 21.8 48.3 29.2 0.7
ADP, NaCl, MgCl
30 12.8 86.0 0.7 0.3
ADP, HEPES 30 30.7 65.3 3.3 0.3
30 13.5 85.7 0.4 0.2
ADP, NaCl 30 3.9 94.5 1.1 0.3
O 30 3.9 94.4 1.1 0.3
ADP, NaCl MgCl
, MOBS 38 6.4 90.6 1.7 0.8
ADP, NaCl MgCl
, MOBS 63 10.9 86.1 1.6 0.5
0.015 M ADP, 0.2 M NaCl, 0.075 M MgCl
, 0.1 M HEPES or 0.1 M MOBS,
pH 8, 25
C. Where no buffer was used the pH was adjusted daily using 0.1 M
Yields were determined from the peak areas of the HPLC traces.
Hydrolysis rates and half-life of ATP in the presence of meteorites
Murchison 3.2 5 h
Y7917127 (CO3) 0.17 4 day
Y86751 (CV3) 0.11 6 day
Blank 0.01 2 month
Hydrolytic conditions: 1 mM ATP, 10 mM MOBS, pH 7, 37
constants and half-lives were calculated from the slopes of the plots of the log of the
change in concentration of ATP with time (Table V). The meteorites were effective
catalysts as shown by 320-fold increase in the rate constant for the hydrolysis of
ATP by the Murchison meteorite over that observed in its absence.
Presence of salts in the reaction medium enhances the formation of RNA oligomers
in the montmorillonite-catalyzed reaction of nucleoside phosphorimidazolides. Pre-
viously it was believed that the reaction required the presence of Mg
with the activated phosphate group but the present studies indicate that the length
of the oligomers formed is mainly a function of the ionic strength of the reac-
tion solution. For example, it was demonstrated that the same length oligomers
were obtained using 1 M NaCl as was observed with a mixture of 0.075 MgCl
358 S. MIYAKAWA ET AL.
and 0.2 M NaCl. The chain lengths decreased as the concentration of NaCl was
decreased (Table I).
A variety of salts that may have been present in an evaporating body of wa-
ter, similar to the salt deposits discovered by the Opportunity Rover on Mars,
were investigated. Sodium sulfate (ionic strength 0.60) does not enhance the for-
mation of longer oligomers as much as does NaCl (ionic strength 0.42). Salts
that inhibit the reaction are those that displace the imidazole group from the ac-
tivated nucleotide or block binding of activated nucleotides to clays. These in-
(probably as NH
. The concentrations of NH
on the primitive Earth were probably lower than that was required
for inhibition. The concentration of HCO
would have been high enough to in-
hibit the reaction only if the early Earth had a CO
atmosphere of 10 bars or
higher. Higher salt concentrations would have been present in evaporating bodies of
A selection of minerals and meteorites were investigated as possible catalysts for
the formation of oligomers from ImpA. Only galena (PbS) catalyzed the formation
of the dimers pA
pA and pA
pA, a ﬁnding that was reported previously (Sleeper
and Orgel 1979). Our studies established that the reaction was catalyzed by the
mineral surface and not by soluble PbS. The dimer yields formed by PbS catalysis
were much lower than those formed by montmorillonite catalysis.
The pH of the reaction solution has a marked effect on the products formed
from the reaction of ImpA. Reactions carried out in the pH 6–10 range yielded
oligomers as reaction products with the optimal pH 7–8. Reactions performed at
pHs less than 6 yielded A
ppA, with an optimal pH of 2–3. The requirement for a
basic pH for oligomer formation differs from the highly acidic pH predicted for the
evaporated body of water found by the Opportunity Rover on Mars. It is proposed
that the Martian conditions were very acidic because crystals of the mineral Jarosite
] were detected (Madden, et al., 2004; Squires et al., 2004).
It is possible that an environment similar to that observed by the Opportunity
Rover was also present on the early Earth. This highly acidic environment would
have generated an acidic montmorillonite that was gradually converted to a neutral
catalytic montmorillonite as the acidity on the Earth decreased.
The lengths of the oligomers formed by the montmorillonite catalysis of the
reaction of ImpA decrease as the temperature is raised to 50
C. The yield of 5
formed by hydrolysis of ImpA increases at higher temperature. These observations
suggest that the increase in rate of hydrolysis results in a decrease in the relative
rate of oligomer formation and hence the lower yields of the longer oligomers.
The search for mineral or meteorite catalysis of the oligomerization of other
activated phosphate derivatives was not successful. Where a reaction was observed
the predominant product formed was due to the hydrolysis of the activating group
on the monomers. These studies did lead to the discovery that ADP was converted
to ATP and AMP at room temperature in the presence of HEPES buffer. It has been
proposed that similar transformation proceed by free radical processes but it seems
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 359
unlikely that free radicals would be generated under the reaction conditions used
in the present study.
We thank Dr. K. Kobayashi and Dr. H. Mita for helpful discussion about meteorites.
Montmorillonite (Volclay) was a gift from the American Colloid Company. The
Murchison meteorite was a gift from Dr. S. Pizzarello and Yamato-791717 and
Yamato-86751 were from the National Antarctic Laboratory in Japan. Dr. Dimitri
Zagorevski obtained mass spectra in the Department of Chemistry Mass Spectra
Facility. Research support was from NSF grant CHE-0413739 and NASA grant
NAG5-12750 to the NY Center for Studies on the Origins of Life.
Bada, J. L. and Miller, S. L.: 1968, Ammonium Ion Concentration in the Primitive Ocean, Science
Banin, A.: 1973, Quantitative Ion Exchange Process for Clays, U.S. Patent 3,725,528.
Banin, A., Lawless, J. G., Mazzurco, J., Church, F. M., Margulies, L. and Orenberg, J. B.:
1985, pH Proﬁle of the Adsorption of Nucleotides onto Montmorillonite, Origins Life 15, 89–
Cronin, J. R. and Chang, S.: 1993, Organic Matter in Meteorites: Molecular and Isotopic Analyses
of the Murchison Meteorite, in The Chemistry of Life’s Origins, Greenberg, J. M., et al. (eds.),
Kluwer Academic Publishers, Netherlands, pp. 209–258.
Dawson, R. M. C., Elliott, D. C., Elliott, W. H. and Jones, K. M. (eds): 1969, Data for Biochemical
Research, Oxford University Press, New York.
Ding, P. Z., Kawamura, K. and Ferris, J. P.: 1996, Oligomerization of Uridine Phosphorimidazolides
on Montmorillonite: A Model for the Prebiotic Synthesis of RNA on Minerals, Origins Life Evol.
Biosphere 26, 151–171.
Ertem, G. and Ferris, J. P.: 1998, Formation of RNA Oligomers on Montmorillonite: Site of Catalysis,
Origins Life Evol. Biosphere 28, 485–499.
Ertem, G. and Ferris, J. P.: 2000, Sequence- and Regio-Selectivity in the Montmorillonite-Catalyzed
Synthesis of RNA, Origins Life Evol. Biosphere 30, 411–422.
Ferris, J. P.: 2002, Montmorillonite Catalysis of 30-50 Mer Oligonucleotides: Laboratory Demon-
stration of Potential Steps in the Origin of the RNA World, Origins Life Evol. Biosphere 32,
Ferris, J. P. and Ertem, G.: 1992, Oligomerization Reactions of Ribonucleotides: The Reaction of
the 5’-Phosphorimidazolide of Nucleosides on Montmorillonite and Other Minerals, Origins Life
Evol. Biosphere 22, 369–381.
Ferris, J. P. and Ertem, G.: 1993, Montmorillonite Catalysis of RNA Oligomer Formation in Aqueous
Solution. A Model for the Prebiotic Formation of RNA, J. Am. Chem. Soc. 115, 12270–12275.
Ferris, J. P., Hill, A. R. Jr, Liu, R. and Orgel, L. E.: 1996, Synthesis of Long Prebiotic Oligomers on
Mineral Surfaces, Nature 381, 59–61.
Ferris, J. P. and Nicodem, D. E.: 1972, Ammonia Photolysis and the Role of Ammonia in Chemical
Evolution, Nature 238, 268–269.
360 S. MIYAKAWA ET AL.
Gedulin, B. and Arrhenius, G.: 1994, Sources and Geochemical Evolution of RNA Precursor
Molecules-The Role of Phosphate, in Nobel Symposium 84, Early Life on Earth, Columbia
University press, pp. 91–106.
Gilmour, I.: 2003. Structural and Isotopic Analysis of Organic Matter in Carbonaceous Chondrites.
In Structural and isotopic analysis of organic matter in carbonaceous chondrites, A. M. Davis,
(ed.) (Amsterdam, Elsevier), pp. 269–290.
Good, N. E. and Izawa, S.: 1972, Hydrogen Ion Buffers, Methods Enzymol. 24, 53–68.
Grotzinger, J. P. and Kasting, J. F.: 1993, New Constraints on Precambrian Ocean Composition, J.
Geol. 101, 235–243.
Hattori, K.,Takahashi, K. and Sasao, K.: 1984, ATP and AMP Formation From ADP In The Presence
of Cyclodextrin, J. Inclusion Phenomena 2, 683–688.
Huang, W. and Ferris, J. P.: 2003, Synthesis of 35–40 mers of RNA Oligomers from Unblocked
Monomers. A Simple Approach to the RNA World, Chem. Commun. 1458–1459.
Joshi, P. C., Pitsch, S. and Ferris, J. P.: 2000, Homochiral Selection in the Montmorillonite-Catalyzed
and Uncatalyzed Prebiotic Synthesis of RNA, Chem. Commun. 2497–2498.
Joyce, G. F., Inoue, T. and Orgel, L. E.: 1984, Non-Enzymatic Template-Directed Synthesis on RNA
Random Copolymers: Poly(C,U) Templates, J. Mol. Biol. 176, 279–306.
Kaiden, H., Mikouchi, T., Nonura, K. and Miyamoto, M.: 1997, Chemical Zoning of Olivines in the
Yamato-791717 CO3 Chondrite, Antarct. Meteorite Res. 10, 181–190.
Kanavarioti, A.: 1997, Dimerization in Highly Concentrated Solutions of Phosphoimidazolide Acti-
vated Mononucleotides, Origins Life Evol. Biosphere, 27, 357–376.
Kanavarioti, A., Bernasconi, C. F., Doodokyan, D. L. and Alberas, D. J.: 1989, Magnesium Ion
Catalyzed P-N Bond Hydrolysis in Imidazolide-Activated Nucleotides. Relevance to Template-
Directed Synthesis of Polynucleotides, J. Am. Chem. Soc. 111, 7247–7257.
Kawamura, K. and Ferris, J. P.: 1994, Kinetic and Mechanistic Analysis of Dinucleotide and Oligonu-
cleotide Formation from the 5
-Phosphorimidazolide of Adenosine on Na
Am. Chem. Soc. 116, 7564–7572.
Kerr, P. F., Hamilton,P. K. and Poll, R. J.: 1951,AmericanPetroleumInstitute;Clay Mineral Standards.
American Petroleum Institute, Project 49, Preliminary Report 7B, Chemical Analysis, Columbia
University, New York.
Lozinova, T. A. and Arutyunyan, A. E.: 1990, ATP Synthesis and Free Radicals in Aqueous Solutions
of ADP, Inorganic Phosphate, NADH and Riboﬂavin, Biophysical 35(6), 901–905.
Madden, M. E., Bodnar, R. J. and Rimstidt, J. D.: 2004, Jarosite as an Indicator of Water-Limited
Chemical Weathering on Mars, 2004, Nature 431, 821–823.
Miyakawa,S.andFerris,J.P.: 2003, Sequence- and RegioselectivityintheMontmorillonite-Catalyzed
Synthesis of RNA, J. Am. Chem. Soc. 125, 8202–8208.
Murakami, T. and Ikeda, Y.: 1994, Petrology and Mineralogy of the Yamato-86751 CV3 Chondrite,
Meteoritics 29, 379–409.
Prabahar, K. J. and Ferris, J. P.: 1997, Adenine Derivatives as Phosphate-Activating Groups for
the Regioselective Formation of 3’, 5
-Linked Oligoadenylates on Montmorillonite: Possible
Phosphate-Activating Groups for the Prebiotic Synthesis of RNA, J. Am. Chem. Soc. 119,
Prabahar, K. J., Cole, T. D. and Ferris, J. P.: 1994, Effect of Phosphate Activating Group on Oligonu-
cleotide Formation on Montmorillonite: The Regioselective Formation of 3’, 5
enylates, J. Am. Chem. Soc. 116, 10914–10920.
Ruzicka, F. J. and Frey, P. A.: 1993, The pH Dependence for the Hydrolysis of Uridine 5
Phosphoimidazolates, Bioorg. Chem. 21, 238–248.
Sleeper, H. L. and Orgel, L. E. 1979, Template-Directed Synthesis of Oligoadenylates Catalyzed by
ions, J. Mol. Evol. 13, 204–214.
Squires, S. W., Graettinger, J. P., Davidson, R. E., Bell, J. F. III, Calvin, W., Christensen, P. R., Clark,
B. C., Crisp, J. A., Farr, W. H., Herkenhoff, K. E., Johnson, J. R., Klingelhofer, G.’ Knoll, A. H.,
STUDIES IN THE MINERAL AND SALT-CATALYZED FORMATION OF RNA OLIGOMERS 361
McLennan, S. M., McSween, H.Y. Jr., Morris, R. V., Rice, J. W. Jr., Rieder, R. and Soderblom,
L. A. 2004, In situ Evidence for an Ancient Aqueous Environment at Meridiani Planum, Mars,
Science 306, 1709–1714.
Thiel, T., Liczkowski, L. and Bissen, S. T.: 1998, New Zwitterionic Butanesulfonic Acids That Extend
the Alkaline Range of Four Families of Good Buffers: Evaluation for Use in Biological Systems,
J. Biochem. Biophys. Meth. 37, 117–129.
Walker, J. C. G.: 1985, Carbon Dioxide on the Early Earth, Origins Life Evol. Biosphere 16, 117–127.
Wang, K.-J. and Ferris, J. P.: 2001, Effect of Inhibitors on the Montmorillonite Clay-Catalyzed
Formation of RNA: Studies on the Reaction Pathway, Origins Life Evol. Biosphere 31, 381–402.
Weimann, B. J., Lohrmann, R., Orgel, L. E.. Schneider-Bernloehr, H. and Sulston, J. E.: 1968,
Template-Directed Synthesis with Adenosine-5
-phosphorimidazolide, Science 161, 387–388.